Starinar LXIV

ARHEOLO[KI INSTITUT BEOGRAD INSTITUT ARCHÉOLOGIQUE BELGRADE UDK 902/904 (050) ISSN 0350-0241 ([tampano izd.) ISSN 2406-0739 (Online) STARINAR LXIV/2014, 1–270, BEOGRAD 2014. INSTITUT ARCHÉOLOGIQUE BELGRADE STARINAR Nouvelle série volume LXIV/2014 BELGRADE 2014 ARHEOLO[KI INSTITUT BEOGRAD STARINAR Nova serija kwiga LXIV/2014 BEOGRAD 2014. STARINAR Nova serija kwiga LXIV/2014 IZDAVA^ Arheolo{ki institut Kneza Mihaila 35/IV 11000 Beograd, Srbija e-mail: [email protected] Tel. 381 11 2637191 UREDNIK Slavi{a Peri}, direktor Arheolo{kog instituta REDAKCIONI ODBOR Miloje Vasi}, Arheolo{ki institut, Beograd Rastko Vasi}, Arheolo{ki institut, Beograd Noel Dival, Univerzitet Sorbona, Pariz Bojan \uri}, Univerzitet u Qubqani, Filozofski fakultet, Qubqana Mirjana @ivojinovi}, Srpska akademija nauka i umetnosti, Beograd Vasil Nikolov, Nacionalni arheolo{ki institut i muzej, Bugarska akademija nauka, Sofija @ivko Miki}, Univerzitet u Beogradu, Filozofski fakultet, Beograd Ivana Popovi}, Arheolo{ki institut, Beograd Marko Popovi}, Arheolo{ki institut, Beograd Nikola Tasi}, Srpska akademija nauka i umetnosti, Beograd SEKRETAR REDAKCIJE Jelena An|elkovi} Gra{ar, Arheolo{ki institut, Beograd LEKTOR ZA SRPSKI JEZIK Mirjana Radovanovi} PREVODILAC I LEKTOR ZA ENGLESKI JEZIK Dejv Kalkat GRAFI^KA OBRADA Danijela Paracki D_SIGN, Beograd [TAMPA ALTA NOVA, Beograd TIRA@ 400 primeraka SEKUNDARNA PUBLIKACIJA COBISS SCIndeks STARINAR Nouvelle série volume LXIV/2014 EDITEUR Institut archéologique Kneza Mihaila 35/IV 11000 Belgrade, Serbie e-mail: [email protected] Tél. 381 11 2637191 RÉDACTEUR Slavi{a Peri}, directeur de l’Institut archéologique COMITÉ DE RÉDACTION Miloje Vasi}, Institut archéologique, Belgrade Rastko Vasi}, Institut archéologique, Belgrade Noël Duval, Université Paris Sorbonne, Paris IV Bojan \uri}, Université de Ljubljana, Faculté des Arts, Ljubljana Mirjana @ivojinovi}, Académie serbe des sciences et des arts, Belgrade Vasil Nikolov, Institut archéologique national et Musée, Académie bulgare des sciences, Sofia @ivko Miki}, Université de Belgrade, Faculté des Arts, Belgrade Ivana Popovi}, Institut archéologique, Belgrade Marko Popovi}, Institut archéologique, Belgrade Nikola Tasi}, Académie serbe des sciences et des arts, Belgrade SECRÉTAIRE DE RÉDACTION Jelena An|elkovi} Gra{ar, Institut archéologique, Belgrade LE LECTEUR POUR LA LANGUE SERBE Mirjana Radovanovi} TRADUCTEUR ET LECTEUR POUR LA LANGUE ANGLAIS Dave Calcutt RÉALISATION GRAPHIQUE Danijela Paracki D_SIGN, Belgrade IMPRIMEUR ALTA NOVA, Belgrade TIRAGE 400 exemplaires PUBLICATION SECONDAIRE COBISS SCIndex ^asopis je objavqen uz finansijsku pomo} Ministarstva prosvete, nauke i tehnolo{kog razvoja Republike Srbije Ce périodique est publié avec le soutien du Ministère de l’éducation, de la science et du développement technologique de la République Serbie SADR@AJ – SOMMAIRE RASPRAVE – ETUDES Miljana Radivojevi}, Julka Kuzmanovi}-Cvetkovi} Copper minerals and archaeometallurgical materials from the Vin~a culture sites of Belovode and Plo~nik: overview of the evidence and new data . . . . . . . . . . . . . . . . . . . . . . . . . 7 Miqana Radivojevi}, Julka Kuzmanovi}-Cvetkovi} Minerali bakra i arheometalur{ki materijali sa vin~anskih lokaliteta Belovode i Plo~nik: pregled materijala i novi podaci . . . . 29 Mario Gavranovi}, Aleksandar Kapuran Über einige Tüllenbeilvarianten im Zentralbalkan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Mario Gavranovi}, Aleksandar Kapuran O nekim varijantama bronzanih sekira na centralnom Balkanu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Aleksandar P. Bulatovi} New finds as a contribution to the study of the Early Bronze Age in the southern part of the Central Balkans . . . . . . . . . . . . . . . . . . . . . . 57 Aleksandar P. Bulatovi} Novi nalazi kao prilog prou~avawu ranog bronzanog doba u ju`nom delu centralnog Balkana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Ivana Popovi} New discoveries of marble sculptures in the Sirmium imperial palace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Ivana Popovi} Novi nalazi mermernih skulptura u carskoj palati u Sirmijumu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Sofija Petkovi}, Nata{a Miladinovi}-Radmilovi} Military graves from the Late Roman necropolis at Slog in Ravna (Timacum Minus) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Sofija Petkovi}, Nata{a Miladinovi}-Radmilovi} Vojni~ki grobovi sa kasnoanti~ke nekropole Slog u Ravni (Timacum Minus) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Vujadin Ivani{evi}, Michel Kazanski Illyricum du Nord at les Barbares à l’époque des Grandes Migrations (Ve–VIe siècle) . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Vujadin Ivani{evi}, Mi{el Kazanski Severni Ilirik i Varvari u vreme seobe naroda (4–6. vek) . . . . . . . . . . . 160 Miroslav Vujovi} Ranovizantijski kantar iz Beograda . . . . . . . . . . . . . . . . . . . . . . . . . 161 Miroslav Vujovi} Early Byzantine steelyard from Belgrade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 PRILOZI – APERCUS Milo{ P. Spasi}, Adam N. Crnobrwa Vin~anske zdele sa protomama . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Milo{ P. Spasi}, Adam N. Crnobrnja Vin~a bowls with protoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Rastko Vasi} Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen . . . . 205 Rastko Vasi} Prilog PBF sveskama koje se odnose na centralni Balkan . . . . . . . . . . . . . 216 Vujadin Ivani{evi}, Sonja Stamenkovi} Late Roman fortifications in the Leskovac basin in relation to urban centres. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Vujadin Ivani{evi}, Sowa Stamenkovi} Kasnoanti~ka utvr|ewa u Leskova~koj kotlini i wihov odnos prema urbanim centrima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230 PREGLEDNI RADOVI – ARTICLES SOMMAIRES Miloje Vasi}, Gordana Milo{evi}, Nade`da Gavrilovi} Iskopavawa Medijane u 2010. i 2011. godini . . . . . . . . . . . . . . . . . . . 231 Miloje Vasi}, Gordana Milo{evi}, Nade`da Gavrilovi} Excavations of Mediana in 2010 and 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 KRITIKE I PRIKAZI – COMPTES RENDUS Ivan Bugarski Rachel S. Opitz, David C. Cowley (eds.), INTERPRETING ARCHAEOLOGICAL TOPOGRAPHY, 3D DATA, VISUELISATION AND OBSERVATION, Occasional Publication of the Aerial Archaeology Research Group No. 5, Oxbow Books, Oxford – Oakville 2013. . . . . . . . . . . . . . . 265 @arko Petkovi} Ifigenija Radulovi}, Gordan Mari~i}, ANDOKID BESEDNIK KOJI JE JEDNOM POGRE[IO, Beograd – Novi Sad 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 MILJANA RADIVOJEVI], UCL Institute of Archaeology, London and Institute for Balkan Studies, Serbian Academy of Arts and Sciences, Belgrade JULKA KUZMANOVI]-CVETKOVI], Homeland Museum in Toplica, Prokuplje COPPER MINERALS AND ARCHAEOMETALLURGICAL MATERIALS FROM THE VIN^A CULTURE SITES OF BELOVODE AND PLO^NIK: OVERVIEW OF THE EVIDENCE AND NEW DATA e-mail: [email protected] UDK: 903:669.3"634"(497.11) ; 903.4"634"(497.11) DOI: 10.2298/STA1464007R Received: February 18, 2014 Accepted: July 17, 2014 Original research article Abstract. – The Vin~a culture sites of Belovode and Plo~nik have been attracting scholarly attention for decades now, due to numerous discoveries indicative of copper mineral and metal use in these settlements, which are confirmed as, currently, the earliest worldwide and very likely developed independently in Eurasia.1 The authors attempt to give an overview of already published data along with new results stemming from the recently completed doctoral research of the primary author.2 All materials related to copper mineral use and pyrometallurgical activities are presented through the concept of metallurgical chaîne opératoire, following the established sequence of operations,3 which is adjusted for this specific case study and divided into three categories: copper mineral processing, (s)melting debris, and the making and working of finished metal objects. The qualitative overview of available data is therefore focused mainly around the material side of the studied samples and provides an insight into the technological choices for making copper mineral ornaments and copper metal artefacts in the sites of Belovode and Plo~nik. Accordingly, it provides a model for the understanding of similar material assemblages that occur in other Vin~a culture sites, or beyond. Key words. – Belovode, Plo~nik, Vin~a culture, malachite, copper metal, tin bronze, malachite beads, chaîne opératoire, slag, Serbia. V in~a culture metallurgy has gained renewed interest through international scholarship in the past few years, owing to the productive publication activity of Serbian researchers.4 This activity recently led to one of the largest international collaborative archaeology research projects ever conducted in Serbia. Funded by the UK government, the partners come from three countries, the United Kingdom, Germany and Serbia.5 Its main goal is to investigate the rise of the earliest known metallurgy in Eurasia with specific focus on the earliest occurrence of this technology at three Vin~a culture sites of Belovode, Plo~nik and Jarmovac. All three sites have thus far yielded evidence that covers the entire chaîne opératoire of metallurgical activities: mining, production and consumption. The archaeometallurgical assemblage originating from Belovode and Plo~nik is reviewed here in light of the 1 Radivojevi} et al. 2010. Radivojevi} 2012. 3 cf. Ottaway 2001. 4 e.g. Antonovi} 2002, 2006; [ljivar, Kuzmanovi}-Cvetkovi} 2009; [ljivar et al. 2012; Radivojevi} et al. 2010, 2013. 5 http://www.ucl.ac.uk/rise-metallurgy-eurasia 2 * Author is participating in the project Society, spiritual and material culture and communications in the prehistory and early history of the Balkans (no. 177012), funded by The Ministry of Education, Science and Technological Development of the Republic of Serbia. 7 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) recently completed analytical work.6 The Jarmovac materials will be studied in detail in the future. The focus of this paper is on a qualitative overview of the available evidence for activities related to copper mineral use and archaeometallurgical activities at Belovode and Plo~nik. A detailed quantitative study is beyond the scope of this paper, and will be attempted separately elsewhere. All materials considered here were discovered before 2009 at the sites of Belovode and Plo~nik. The majority of them originate from detailed investigations carried out by the National Museum in Belgrade (D. [ljivar) and the Museum in Toplica (J. Kuzmanovi}Cvetkovi}) from 1993 and 1996 respectively. The careful consideration of archaeometallurgical evidence from these two sites has already been communicated through national and international scholarship.7 To this, we now add a set of 48 newly analysed materials, which provides a more nuanced picture of the chaîne opératoire of early copper making at the sites of Belovode and Plo~nik, and the Vin~a culture in general. The majority of the studied collection consists of copper minerals and malachite beads, while the rest of the assemblage includes individual slag samples, slagged ceramic sherds and copper metal artefacts. In comparison with the amount of technological debris (and slags in particular) in later prehistoric periods, the sample size in this study appears small by any standard. However, it targets the crucial period in the evolution of metallurgy in Europe and, as a coherent assemblage, is unprecedented in size, quality and resolution. In terms of chronology, all metallurgical activities at the sites of Belovode and Plo~nik are placed within the first half of the 5th millennium BC. The suggested start of the occupation of Belovode is c. 150 years earlier than in Plo~nik, which is set at c. 5350 BC and c. 5200 BC respectively. Both sites were abandoned at about c. 4650 BC.8 The chaîne opératoire of metallurgical activities consists of complex operations that altogether provide information on solutions shaped by three different spheres: the physical, social and environmental worlds.9 In order to assess how and why some particular technological choices in the metal production process were made over others, this process needs to be deconstructed into distinctive components.10 Furthermore, these components require additional division due to differences in the temperature treatment they underwent, which helps us distinguish “cold” from “hot” (extractive) metallurgy, and accordingly differentiate between copper mineral use and pyrometallurgy.11 All components of the metal STARINAR LXIV/2014 production process represent variables prone to mutation, replication and innovation within the transmission process, and significantly influence the final outcome of the metal making activity. In order to address metal production in the Vin~a culture, activities related to copper mineral use and pyrometallurgical activities are described here in three distinctive stages of the process: copper mineral processing, (s)melting debris, and the making and working of finished metal objects. All technological analyses were carried out at the Wolfson Archaeological Science Laboratories at the UCL Institute of Archaeology, by the primary author of this paper, and under the supervision of Professor Thilo Rehren. THE PROCESSING OF (ARCHAEOLOGICAL) COPPER MINERALS All copper minerals studied here are recognised as archaeological since they are originating from archaeological sites (in contrast to geological minerals that come from the mines). Although bead minerals and ores in this study are typically malachite, the rationale for a distinction between these has been developed in the previous study of material from Belovode,12 and refers to their differentiation in the technological treatment that was applied in their processing. Thus, minerals most likely used for bead making at the sites of Belovode and Plo~nik (i.e. “cold” processing) (Fig. 1a) will be referred to as copper minerals only, while those most likely used for copper smelting (or “hot” processed) will be termed copper ores. In this study, copper ores are assumed to contain significant manganese content, as first indicated by the analyses of copper production evidence from Belovode.13 Macroscopically, these ores appear green and black, where green comes from the colour of a secon- 6 Radivojevi} et al. 2010; Radivojevi} 2012; Radivojevi} et al. 2013. 7 [ljivar 1996; [ljivar, Jacanovi} 1996b; [ljivar, Kuzmanovi}Cvetkovi} 1998a; [ljivar 2006; [ljivar et al. 2006 (to cite just a few). 8 Bori} 2009. 9 Roux 2003. 10 Ottaway 2001; Ottaway, Roberts 2008. 11 Radivojevi} et al. 2010, 2779–2780. 12 Radivojevi} et al. 2010. 13 Radivojevi} et al. 2010. 8 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 their wider cultural, environmental and physical surroundings (Fig. 3). dary copper mineral (malachite) and black from the manganese content (Fig. 1b). ARCHAEOLOGICAL SITES AND RELATED MATERIALS (S)MELTING DEBRIS The smelting of copper ores refers to the primary extraction process, where the produced metal was usually further purified by refining, or melting. The common smelting debris are installations, slagged ceramics and slags. Technical ceramics (crucibles, furnace remains, or tuyères) are particularly interesting for studying past metallurgical processes as they reflect technological choices in this craft. The understanding of their shape, fabric and size is crucial for distinguishing whether the typically fragmented ceramic pieces belong to a crucible container or to the lining of the smelting installation.14 However, the most informative part of a crucible or an installation fragment is the slag attached to its walls, as is the case with slagged sherds from the site of Belovode in this research (Fig. 2a). One of the most valuable materials for studying metal production at the site of Belovode are small slag pieces (8 in total), discovered together with slagged sherds in a single trench (No. 3) in this site (for typical example see Fig. 2b). Belovode The site of Belovode (Fig. 4) lies on a windy plateau with the eponymous spring running through the settlement, located near the village of Veliko Laole, c. 140 km southeast of Belgrade. The location of Belovode is typical of Vin~a culture settlements: it rests on a large plateau of ellipsoidal shape at an altitude of c. 200 m, in surroundings suitable for agricultural activities, cattle breeding and pastures.17 The nearby Mlava River runs deep into the volcanic mountain range called Homolje, which belongs to the zone of primary copper mining and metallurgy.18 It has been excavated since 1993 by the National Museum of Belgrade and the Museum in Po`arevac.19 Within an estimated 100 hectares covered by up to 3 m of cultural layers, four building horizons have been documented (Belovode A–D), correlating with the entire Vin~a culture sequence.20 The Vin~a culture sequence of Belovode is established on the basis of its distinctive ceramic typology,21 as well as locally recognised pottery variations.22 In its earliest stages, Belovode was most likely inhabited by Star~evo groups, as indicated by the, so-far occasional, finds of late Star~evo pottery. This potential brief occupation by the Star~evo groups was followed by the Vin~a culture occupation, which ended around the mid 5th millennium BC. By the end of the 4th millennium BC a section of this site was briefly re-occupied by the Late Chalcolithic Kostolac culture but this is limited to a single intrusion in Trench No. 6;23 we have only MAKING AND WORKING: COPPER MINERALS AND METAL ARTEFACTS This group of artefacts consists of malachite beads and copper metal artefacts, both of which include subgroups of semi-finished and finished artefacts. Malachite beads, as one of the earliest artefacts made of copper minerals, were used for ornamental purposes in both the Balkans and the Near East, due to their distinctive green colour.15 These objects were commonly processed in a series of “cold” shaping techniques, applied in different technological steps. The ornament (bead) making process starts with a raw nodule of mineral, which, once roughly shaped (but not perforated), is recognised as a roughout; following sessions of fine chipping it reaches the stage of a preform or blank, commonly followed by drilling before it reaches the final, bead shape.16 Notably, the order of sequences varies culturally but also in relation to materials used for ornament making. The metal artefacts group is particularly informative for the metallurgical chaîne opératoire in the Vin~a culture, as the variety allows the investigation of different sequences of production and interpretation within 14 Rehren 2003. Bar-Yosef Mayer, Porat 2008. 16 Lankton et al. 2003, 16. 17 [ljivar et al. 2006, 251–252. 18 Krajnovi}, Jankovi} 1995. 19 [ljivar, Jacanovi} 1996a; [ljivar, Jacanovi} 1996b; [ljivar, Jacanovi} 1997; Jacanovi}, [ljivar 2003; [ljivar et al. 2006. 20 [ljivar et al. 2006, 251. 21 Gara{anin 1951; Gara{anin 1973. 22 [ljivar, Jacanovi} 1996b; Arsenijevi}, @ivkovi} 1998. 23 Bori} 2009, 208. 15 9 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 b a Fig. 1. a) Malachite bead from Belovode (23). Note the pure green colour of this sample in comparison with the Mn-rich one on the right; b) Typically green-and-black Mn-rich copper mineral from Belovode Sl. 1. a) Malahitna perla sa Belovoda (23). Primetiti ~isto zelenu boju ovog uzorka u pore|ewu sa mineralom (desno) koji je oboga}en manganom; b) Tipi~an zeleno-crni bakarni mineral sa Belovoda oboga}en manganom b a Fig. 2. a) A slagged ceramic sherd from Belovode (31b); b) An individual slag sample from Belovode (134) Sl. 2. a) Kerami~ki fragment sa zgurom sa Belovoda (31b); b) Individualni uzorak zgure sa Belovoda (134) b a Fig. 3. a) Folded copper metal sheet (Plo~nik 75); b) massive copper implement (Plo~nik 143) Sl. 3. a) Presavijeni bakarni lim (Plo~nik 75); b) Masivna bakarna alatka (Plo~nik 143) 10 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 Fig. 4. Plan of the site of Belovode (by J. Pendi}) Sl. 4. Plan lokaliteta Belovode (pripremio J. Pendi}) included here sample M12 (Table 1) after careful examination of its firm contextualisation within the Vin~a culture occupation. By 2009, c. 400 m² had been excavated through 14 trenches, usually 25 m² in size, all of which are concentrated in the southern part of this settlement. The site of Belovode has recently received nine accelerator mass spectrometry (AMS) radiocarbon dates, obtained from animal bones from Belovode, which confirmed the expected Vin~a culture chronological span.24 The probability distribution for the beginning of the Vin~a occupation in Belovode indicates a date of c. 5350 BC, while the boundary for the end of the Vin~a culture use of the site is set at c. 4650 BC. Of particular significance here is the dating of the earliest stratigraphic evidence for extractive metallurgy in Belovode, which starts at around 5000 BC; this is currently the earliest secure date for copper metal production anywhere.25 Importantly, it coincides with the intensive mining activities in Rudna Glava, also from around c. 5000 BC.26 However, Rudna Glava does not appear to have been exploited by the inhabitants of Belovode as another copper source, discovered in @drelo (Fig. 5), c. 10 km from Belovode, makes the most likely candidate according to lead isotope analysis.27 Nine Vin~a culture settlements in total have been found in this area, prompting scholars to propose their association with the mining and metallurgical activities in the area,28 although this remains to be explored in future research. 24 25 26 27 28 11 Gläser 1996; Bori} 2009. Radivojevi} et al. 2010. Bori} 2009, 206. Radivojevi} et al. 2010, 2781, Fig. 10. [ljivar, Jacanovi} 1996b. RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 b a Fig. 5. a) Open-cast mine (?) at @drelo; b) Entrance to the shaft in @drelo (photos by M. Radivojevi}) Sl. 5. a) Povr{inski rudnik (?) u @drelu; b) Ulaz u okno na @drelu (fotografije: M. Radivojevi}) COPPER MINERAL USE Installations/slagged sherds Charred surfaces with malachite, copper mineral powder adhering to fragmented ceramic sherds and grooved stone mallets are common situations in household contexts in Belovode. There are also a few smallsized pottery vessels of conical shape and coarse fabric from Trenches Nos. 7 and 8, Vin~a B1 horizon, which had green minerals attached to the outer surface, however, analyses have shown that these were not crucibles (samples Belovode M26, M29 and M30).34 In addition, a fragment of ceramic mould discovered on the site surface is thought to originate from the latest layer of occupation.35 Two shallow pits rimmed with ceramic sherds and a burnt layer of clay from Trenches 10 and 13, a Vin~a Malachite beads, pendants and “green” copper minerals appeared from the earliest occupation in Belovode, and continued throughout all building horizons. These most numerous finds in Belovode are usually uncovered mixed with ash and pieces of charcoal. Other contexts include house floors, storage jars, ceramic sherds (with malachite adhered to them), or workshops within a household. Two such areas in Trenches 12 and 13 yielded together c. 2.5 kg of copper minerals throughout all building horizons, which is almost one third of the total weight of malachite finds discovered at this site.29 Ten samples belonging to the copper minerals group have been selected for this study, three of which come from the so-called metallurgical Trench No. 3,30 while the rest originates from various household contexts (see Table 1). All samples, barring M3, M13, M17, 33b and 3) are confirmed to be malachite with significant levels of manganese in their composition. The remaining set of copper minerals was sourced from a vein containing paragenesis of cuprite with copper sulphides.31 The small size of these minerals (c. 1–3 cm) may imply that they were beneficiated, or in other words crushed to facilitate smelting. Two stone mallets with a groove in the middle, discovered in the context of workshops in Trenches 1 and 7,32 could offer clues about tools used for this process. Such tools may have been used for mining as well, according to the analogy with similar finds from Rudna Glava.33 29 The primary author weighed all malachite from Belovode for the purpose of her PhD research. 30 Radivojevi} et al. 2010. 31 Radivojevi} 2012. 32 [ljivar et al. 2006, 259, Pl. I/4 33 cf. Jovanovi} 1982. 34 Radivojevi} 2007. 35 [ljivar et al. 2006, 259, Pl. I/3. 36 [ljivar et al. 2006, 253, 260, Pl. II/4. 37 [ljivar 2006; [ljivar et al. 2006, 253. 38 The investigation with portable XRF carried out by Prof. Thilo Rehren (UCL Qatar) and the primary author of this paper did not show remains of potentially adhering slag, copper contamination or excessive vitrification of these cylindrical objects and is, thus, considered as not related to the archaeometallurgical process. 12 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 with traces of convincing remains are yet to be discovered at the site of Belovode.38 Pyrometallurgical activities at the site of Belovode are only represented by eight individual copper slag samples in total and four slagged ceramic sherds from Trench No. 3, all of which demonstrate sustained smelting activities on the outskirts of this site. Another B1 building horizon, are identified as early furnaces by the excavator.36 Elongated cylindrical ceramic forms (so-called chimneys) with a diameter of about 20 cm and a reconstructed height of up to 80 cm and open at both ends (Figure 6), have been tentatively linked to the rimmed pits in the ground and, thus, the smelting operation.37 Nevertheless, “true” smelting installations Table 1: The newly studied collection from Belovode, arranged by analytical numbers. Samples No. 23–29 have already been studied and presented in Radivojevi} (2007) Tabla 1: Nova analizirana kolekcija materijala sa Belovoda, ure|ena prema analiti~kim brojevima. Uzorci 23–29 su prethodno analizirani i prezentovani u Radivojevi} (2007) No. Analytical No. 1 2 3 4 5 6 7 8 9 Belovode 3 Belovode 9 Belovode 10 Belovode 12 Belovode 13 Belovode 18 Belovode 23 Belovode 30a, 30c Belovode 31a, 31b 10 Belovode 33b Excavation year 2007 1995 2001 2007 2003 2007 2001 1995 1995 2008 11 Belovode 34a (1,2,3) 2008 12 13 14 15 16 17 18 19 20 Belovode 40 Belovode 131 Belovode 134 Belovode 136 Belovode 154 Belovode M3 Belovode M6 Belovode M10 Belovode M14 2002 1995 1995 1995 1999 1995 1995 1995 2002 21 Belovode M32 1995 22 23 24 25 26 27 28 29 1995 1998 2000 2004 1995 1995 1995 1995 Belovode M35 Belovode M12 Belovode M13 Belovode M17 Belovode M20 Belovode M21 Belovode M22 (a,b) Belovode M23 Field label Field context Results: type of material Trench 13, spit 14 Trench 3, spit 12 Trench 8, spit 22 Trench 13, spit 10 Trench 10, spit 27 Trench 13, spit 10 Trench 8, spit 23 Trench 3, spit 5 Trench 3, spit 6 Trench 14, spit 15, surface 4 Trench 14, spit 3 Household Household Household Household- workshop? Household Household Household (Building ) waste pit (Building ) waste pit Copper mineral Malachite bead Malachite bead Malachite bead Malachite bead Copper mineral Malachite bead Slagged ceramic sherd Slagged ceramic sherd Household Trench 9, spit 18 Trench 3, spit 6 Trench 3, spit 7 Trench 3, spit 5 Trench 7, spit 6 Trench 3, spit 8 Trench 3, spit 10 Trench 3, spit 19 Trench 9, spit 11 Trench 3, spit not known Trench 3, spit 17 Trench 6, spit 10 Trench 7, spit 18 Trench 10, spit 8 Trench 3, spit 2 Trench 3, spit 4 Trench 3, spit 5 Trench 3, spit 7 Household-pits (Building ) waste pit (Building ) waste pit (Building ) waste pit Household-workshop? (Building ) waste pit (Building ) waste pit Household Household Copper mineral Copper minerals from an amphora Lead-based slag cake Copper slag Copper slag Copper slag Malachite bead Copper mineral Copper metal droplet Copper mineral Copper metal droplet Household Malachite bead Household Household Household Household (Building ) waste pit (Building ) waste pit (Building ) waste pit (Building ) waste pit Copper mineral Copper mineral Copper mineral Copper mineral Copper slag Copper slag Copper slag Copper slag 13 Household RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 b a Fig. 6. a) Reconstructed ceramic “chimneys” from Belovode (after [ljivar et al. 2006, Plate II/3); b) Tentative reconstruction of the smelting installation at Belovode (after [ljivar et al., 2006, Plate II/5) Sl. 6. a) Rekonstruisani kerami~ki „dimwak“ sa Belovoda (prema: [ljivar et al. 2006, Tabla II/3); b) Pretpostavqena rekonstrukcija instalacije za topqewe sa Belovoda (prema: [ljivar et al. 2006, Tabla II/5) pyrometallurgical situation has been recovered in Trench No. 9; both will be presented in more detail below. These materials were usually discovered in areas filled with ash, charcoal, charred wood or stone constructions and mainly represent an outdoor activity. rior, or tri- and four-partite vessels with a cone-shaped neck and protruding shoulder, usually accompanied by typically incised ribbon decorative ornaments.43 The slag pieces collected from this trench are vitrified, strongly magnetic and green-stained droplets, not exceeding 1 cm in length (example in Fig. 2b). They appear to have been highly viscous and very rich in copper metal, however with no signs of crushing in pursuit of copper metal prills. This may well have been due to their small size and weight, since all samples in total weigh less than 4 g. Given that, in appearance, these slag pieces resemble (green) malachite, as a result of the corrosion of the copper metal prills entrapped in them, it is possible that the green colour facilitated their recognition in the field excavations, leading to a Trench No. 3 Trench No. 3 (dimensions 8 x 2 m) yielded evidence for copper smelting activities throughout the final, D horizon of occupation of the site of Belovode, which coincides with the entire Gradac Phase sequence starting in c. 5000 BC.39 This phase of the Vin~a culture is known to have lasted longer in the Morava Valley settlements than in those situated nearer to the Danube,40 and, at this site, most likely covers the late Vin~a culture sequence, dated to c. 5000–4650 BC.41 The Belovode D horizon, represented in this Trench by materials from a waste pit,42 includes all finds coming from spits 1–11, and amongst others various archaeometallurgical debris. Thousands of ceramic finds were unearthed in this horizon alone, many of which are diagnostic of the Gradac Phase in general, like the conical bowls with a thickened rim channelled on the inte- 39 40 41 42 43 2000. 14 Bori} 2009, 206; Radivojevi} et al. 2010. Jovanovi} 1994. Jacanovi}, [ljivar 1995. [ljivar, Jacanovi} 1996a. Arsenijevi}, @ivkovi} 1998; see also Schier 1996; Schier RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 Sl. 7. Veliki komad zgure sa Belovoda, datiran u posledwe vekove 6. milenijuma pre n. e. (foto: M. Radivojevi}) Fig. 7. Large slag cake from Belovode, dated to the last centuries of the 6th millennium BC (photo by M. Radivojevi}) The use of copper minerals occurred regularly throughout building horizons Belovode A–C, excluding Belovode D and the successive, not yet well-defined, horizon. Of particular interest here is spit 11, which yielded a copper metal droplet. The copper metal droplet (M14) was discovered in the context of the regular appearance of ceramic pedestal bowls, typical of the Vin~a A to Gradac Phase. Chronologically, and in relative stratigraphic terms, it can be seen as correlated with the early Gradac Phase and, effectively, with the start of metallurgical activities in Trench No. 3, dated to c. 5000 BC.47 Interestingly, early Belovode horizons in this trench yielded Vin~a culture figurines with modelled appliques: necklaces with perforated disc-pendants,48 resembling gold applications from the late 5th millennium BC burials and settlements in Bulgaria and Ukraine. Similar examples have been discovered at the site of Vin~a–Belo Brdo as well.49 An unusually large round slag cake was unearthed in spit 18 (Belovode 40, Fig. 7), and argued as being biased recovery in favour of more copper-rich pieces and overlooking those without green staining.44 The green staining on fragmented slagged sherds comes from the contact of these samples with the metallurgical process. The slagged mass is followed by heavily vitrified areas, which appear along the edges of the studied samples (Fig. 2a), but also extend across their cross-sections. The latter implies that these sherds were most likely used fragmented during the metallurgical process. All metallurgy-related materials were discovered sealed with building waste, such as the remains of house daub, domestic pottery and animal bones. Notably, in spit 10, which belongs to this building horizon, two shallow rock-lined constructions, indicated as fireplaces, were identified as potentially linked with metallurgical debris in excavation records. The stratigraphic evidence related to the earliest slag piece is dated to c. 5000 BC; the smelting evidence, according to the excavation reports, continued until the abandonment of the site, in c. 4650 BC. Trench No. 9 Pyrometallurgical activities are recorded in Trench No. 9 as well (dimensions 5x5 m).45 In this trench, spits 21–7 yielded typical Vin~a culture material that corresponds with all four building horizons in Belovode (A–D). Archaeological materials found in spits 6–1 belonged, most probably, to the latest manifestation of the Vin~a D phase.46 44 45 46 47 48 49 15 Radivojevi} et al. 2010, 2779. Jacanovi}, [ljivar 2003. cf. Jovanovi} 1994. cf. Radivojevi} et al. 2010. [ljivar et al. 2012, 31, Pl. III/1–4. Tasi} 2008, 151, Fig. 58. RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) firmly contextualised within the Vin~a B1 phase.50 Preliminary analyses conducted by the primary author of this paper revealed that this artefact has significant concentrations of lead, which may suggest the production of this metal. However, future analysis is expected to shed more light on this unusual archaeometallurgical evidence. In terms of absolute dating, a metal droplet (M14) could be dated to c. 5000 BC, while the slag cake (Belovode 40) may be ascribed the date of c. 5200 BC, since it is argued to come from an undisturbed context of spit 18.51 STARINAR LXIV/2014 sites with archaeometallurgical installations have been identified around toponymic places like Suvo Rudi{te (in Serbian: Dry Mine) or Bakarnja~a (in Serbian: Copper-rich), indicating intensive metallurgical activities in the past. The archaeological settlement of Plo~nik was first identified in 1926, when the first group of metal artefacts was discovered during the building of the Yugoslav railway. Excavation campaigns commenced in 1927, and then continued 1960–1978, under the jurisdiction of M. Grbi} and B. Stalio respectively, both of the National Museum Belgrade.58 Most recently, field excavation resumed in 1996 under the joint supervision of D. [ljivar (National Museum Belgrade) and J. Kuzmanovi}-Cvetkovi} (Homeland Museum in Toplica, Prokuplje),59 and are still ongoing. Grbi}’s and Stalio’s campaigns uncovered an area of c. 1800 m² which, when added to another c. 550 m² from an ongoing campaign, adds up to a total of c. 0.2 ha explored thus far. The estimated size of the Vin~a village in Plo~nik is c. 100 ha, which refers to the size of the top cultural horizon.60 The unique and abundant ceramic finds in the site inspired Gara{anin to name the late Vin~a culture after Plo~nik (I, IIa and IIb),61 which corresponds to Vin~a C, D1 and D262. Twenty massive copper implements, discovered by chance in two groups during the 1927 campaign, prompted Grbi}63 to title the first site publication: Plo~nik, eine Prähistorische Ansiedlung aus der Kupferzeit (Plo~nik, COPPER MINERAL AND COPPER METAL ARTEFACTS Malachite beads occur throughout all horizons in Belovode and vary in size, from 4 mm to 1.5 cm in diameter, with the exception in size and form (deltoid pendant with perforation) being the one found in Trench No.1, Vin~a A1 horizon.52 Beads selected for this study were recovered as related to various copper minerals, slag pieces, workshop activities or dwelling structures (here Belovode 9, 10, 12, 13, 23, 154, M32). All beads are present as finished artefacts only, either as fragmented or whole pieces (see example in Fig. 1a). Some Belovode beads have undergone mineralogical study, which confirmed the presence of malachite, with traces of tenorite and kolwezite.53 The only copper metal artefacts in Belovode were found in the vicinity of the site: a copper metal chisel and a bun-shaped metal ingot.54 Their cultural provenance is only assumed to be of the Vin~a culture due to the vicinity of the site. However, one needs to bear in mind the Late Chalcolithic occupation by the bearers of the Kostolac culture, also detected in Belovode.55 50 [ljivar et al. 2012, 33–34, Pl. VIII/1. Bori} 2009, 207–209; [ljivar et al. 2012, 33–34. 52 [ljivar et al. 2006, Pl. II/1. 53 Jovi} 1996. 54 [ljivar et al. 2006, 252, 269, Pl. I/1–2. 55 Bori} 2009, 208. 56 Stalio 1960, 34; Stalio 1962, 21. 57 Bogosavljevi} et al. 1988; Bogosavljevi} 1989; Bogosavljevi}-Petrovi}, Tomovi} 1993; Bogosavljevi}-Petrovi} 1995; Mrkobrad 1989; Mrkobrad et al. 1989. 58 Grbi} 1929; Stalio 1960; Stalio 1962; Stalio 1964; Stalio 1973. 59 [ljivar 1996; [ljivar 1999; [ljivar 2006; [ljivar, Kuzmanovi}Cvetkovi} 1997; [ljivar, Kuzmanovi}-Cvetkovi} 1998a; [ljivar, Kuzmanovi}-Cvetkovi} 1998b; [ljivar et al. 2006; Kuzmanovi}Cvetkovi} 1998. 60 [ljivar, Kuzmanovi}-Cvetkovi} 1998b, 80. 61 Gara{anin 1951, 12. 62 cf. Miloj~i} 1949. 63 Grbi} 1929. 51 Plo~nik The site of Plo~nik (Fig. 8) is situated underneath the eponymous modern village, 19 km west of the town of Prokuplje in south Serbia and 300 km south of the capital, at c. 300 m asl. It is set on the left bank of the Toplica river, whose shifted bed is presently eroding the c. 3.60 m thick cultural layer of this site.56 Plo~nik is surrounded by good quality agricultural land and thermal springs, but also with good communication routes along the Toplica, the major river stream in this part of Serbia. It springs from Kopaonik, a mountain c. 50 km from the site, whose rich iron veins were exploited in Roman and medieval times.57 More than 50 16 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 Fig. 8. Plan of the site of Plo~nik (by J. Pendi}) Sl. 8. Plan lokaliteta Plo~nik (pripremio J. Pendi}) a prehistoric settlement of the Copper Age, in German). He assumed that Plo~nik was an important metallurgical centre which existed at the dawn of the Copper Age and maintained rich exchange networks with other contemporary settlements in the region, such as Vin~a–Belo Brdo, Gradac or Butmir.64 Nevertheless, these copper implements, having been unique and isolated finds from a location remote from the Near East, did not appeal to Gara{anin as being of genuinely Vin~a culture origin.65 He referred to them as intrusive hoards belonging to the Middle Eneolithic/Chalcolithic Bubanj–Hum culture which, in his opinion, succeeded the Vin~a culture sequence at the site of Plo~nik. Stalio, coming across two more groups of massive metal artefacts in Plo~nik, also agreed that they belonged to the Bubanj–Hum culture.66 In 1978, a single well-contextualised copper implement was discovered associated with the Gradac Phase feature (house?), but was published almost two decades later.67 The resumed excavation campaign (from 1996) brought more evidence for dating metallurgical activities at Plo~nik firmly within the Vin~a culture sequence, which was uninterrupted at this site. A total of three building horizons were identified, all of which belonged solely to the Vin~a culture.68 Horizon III (the Gradac Phase) yielded a well-contextualised copper chisel typologically resembling some of the previous chance metal finds from Plo~nik.69 This, and the chisel from year 197870 demonstrated the Vin~a culture origins for metallurgical activities in Plo~nik. The site of Plo~nik has recently received its first AMS dates.71 The probability distribution for the start of the Vin~a culture occupation in Plo~nik was 5290–5140 cal BC, with the highest probability around 5200 BC. As for the boundary end, the highest probability is at c. 4650 BC, suggesting the use of the settlement for c. 600 64 65 66 67 68 69 70 71 17 Grbi} 1929, 7, 18. Gara{anin 1951; Gara{anin 1973. Stalio 1960, 35; Stalio 1964, 39–40. [ljivar 1996; [ljivar, Kuzmanovi}-Cvetkovi} 1997. [ljivar 1996, 94. [ljivar et al. 2006, 255, Pl. VIII/3. [ljivar 1996. Bori} 2009, 211–215. STARINAR LXIV/2014 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) years.72 Significantly, one of the AMS dates is closely related to the copper chisel (here Plo~nik 216);73 the preceding context is dated between 5040–4860 BC (95% probability) thus marking the terminus ante quem for this and other metal artefacts in Plo~nik, along with the start of the Gradac Phase at this site. This is consistent with the beginning of the Gradac Phase at other sites as well (Belovode, Rudna Glava).74 Three building horizons in Plo~nik correspond with Vin~a A, B1 and B2 respectively.75 Horizons I and II (c. 2.5 m thick altogether) bore massive remains of dwelling structures (up to 6.5 m in length) and related postholes, including wide pits filled with ash, charcoal and bone fragments in the top layers, and ceramic fragments characteristic for Vin~a A and B1 phases.76 The Gradac Phase is represented with a c. 1 m thick cultural layer at the site of Plo~nik which, at c. 0.4 m.77 Here, dwelling structures are intersected with wide pits filled with charcoal and ash in several successive layers, some with bone and ceramic fragments in the upper layers (Figs. 9, 10).78 The materials selected for this study originate from campaigns in 1998–2009, and come from 7 trenches in total (see Table 2). The most extensively sampled is Trench No. 20, which in the Gradac Phase occupation unveiled exceptional evidence for the Vin~a culture metallurgy.79 COPPER MINERAL USE The use of copper minerals is evident from the early formation of this site: green lumps are found scattered across the settlement in the same manner as in Belovode, usually outside potential dwelling features in so-called workshop areas. These workshops usually consisted of stone structures and amorphous remains of floors for which insufficient evidence is present to ascribe them to either economic or dwelling structures. [ljivar and collaborators reported finds of lumps of copper minerals of varying sizes, with an altered structure which was porous and mixed with Table 2: Newly studied collection of materials from the site of Plo~nik Tabla 2: Nova analizirana kolekcija materijala sa lokaliteta Plo~nik Field label Field context Result: type of Material 1 Plo~nik 43 2 Plo~nik 51 3 Plo~nik 52 Excavation year 2007 2006 2000 Trench 20, spit 10 Trench 19, spit 23 Trench 14, spit 10 Workshop Household Household 4 Plo~nik 54 (b, m) 2002 Trench 16, spit 19 Household 2006 2008 2008 2004 2007 2007 2007 Trench 19, spit 13 Trench 21, spit 5 Trench 21, spit 6 Trench 18, spit 7 Trench 20, spit 7 Trench 20, spit 4 Trench 20, spit 7 Household Dwelling structure Dwelling structure Stone structure Workshop Workshop Workshop 12 Plo~nik 72 (b, m) 2007 Trench 20, spit 3 Workshop 13 14 15 16 17 18 19 2007 2007 2004 2007 2009 2009 2000 Trench 20, spit 7 Trench 20, spit 7 Trench 18, spit 7 Trench 20, spit 7 Trench 22, spit 17 Trench 22, spit 15 Trench 14, spit 7 Workshop Workshop Stone structure Workshop Household Household Stone structure Malachite bead Copper mineral Copper metal droplet Copper mineral and a malachite bead (blank) Copper mineral Tin bronze foil Malachite bead Malachite bead Copper metal artefact Copper metal droplet Copper mineral Copper mineral and malachite bead Copper metal artefact Copper metal artefact Copper metal artefact Copper metal artefact Malachite bead Copper mineral Copper metal chisel No. Analytical No. 5 6 7 8 9 10 11 Plo~nik 57 Plo~nik 63 Plo~nik 65 Plo~nik 66 Plo~nik 67 Plo~nik 69 Plo~nik 71 Plo~nik 73 Plo~nik 75 Plo~nik 143 Plo~nik 145 Plo~nik 207 Plo~nik 209 Plo~nik 216 18 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 metal objects in the same structure could indirectly imply its use for casting or melting, for example. charcoal, hence assuming its association with metallurgical activities.80 Six copper minerals in total have been selected for this study (Plo~nik 51, 54 m, 57, 71, 72, 209, see Table 2), two of which come from Trench No. 20 (71 and 72), which covers the entire Gradac Phase section. Other copper mineral finds were sampled in order to provide information on their use in earlier phases of Plo~nik occupation. All minerals (barring 72) are black and green Mn- rich copper minerals. COPPER MINERAL AND METAL ARTEFACTS The most significant discoveries in Plo~nik were four groups of exceptionally massive copper metal implements, 34 in total.85 To this, four more implements are added, all well-contextualised within the Gradac Phase occupation of this settlement.86 The term hoard was introduced by Grbi}87; however, to avoid confusion they will be referred to here as groups.88 Group I was a donation of the Directorate of Yugoslav Railways in 1926, discovered during the building works for the railway station in Plo~nik, at a depth of c. 0.8–1 m.89 He also noticed that the find originated from the close proximity of an oven; however, he did not elaborate further on this. The find consisted of nine objects, seven of which were made of copper metal: two hammer axes (Plo~nik type), two chisels, two complete and one fragmented bracelet and two stone axes made of magnesite (Fig. 14). Group II consisted of 18 objects, 13 of which were copper metal tools: one hammer-axe (Plo~nik type), 12 chisels and five stone axes made of magnesite (Fig. 14). Grbi}90 reported that they were found in the vicinity of a destroyed furnace, scattered over an area 5 m wide, at a depth of c. 1 m.91 METALLURGICAL ACTIVITIES Stalio81 discusses the appearance of “bronze” and various metallurgical debris from the latest building horizon as being related to the production of Plo~nik metal. Regrettably, no such archaeometallurgical debris was found in the archives of the National Museum in Belgrade. Nevertheless, the most recently uncovered situation in Trench 20 yielded evidence of a potential metallurgical workshop in Plo~nik, which is described in more detail below. Trench No. 20 A structure whose contours were followed over a 6 x 6 m area appeared at a relative depth of 0.8 m, with a surface filled with rubble, stones, numerous pottery fragments and several metal artefacts and metal casting debris82 (Fig. 11). A rectangular fireplace (possibly a furnace), measuring 1.4 x 1.4 m dominated this structure, with massive walls preserved up to 0.5 m in height, visibly repaired several times (several clay layers), and traces of intense firing.83 This, along with the discovery of a massive copper chisel (Plo~nik 145), a fragment of a tool or ornament (Plo~nik 67), a fragmented bracelet (?) (Plo~nik 73), a folded metal sheet (Plo~nik 75) and copper minerals (Plo~nik 71 and 72) (Table 2), led excavators to assume that the structure represented a metallurgical workshop (see Fig. 12).84 Other finds in this structure included stone tools and large ceramic vessels, such as amphorae, jugs and similar liquid containers. Particularly interesting were ceramic “tubes”, which did not contain indications of use in metallurgical processes, but resembled the shape of the “chimneys” from Belovode (Figs. 6, 13). The potential furnace remains, although impressive, did not offer sufficient data to assume its metallurgical function; however, the conjunction with fragmented 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 19 Bori} 2009, 212. [ljivar et al. 2006, 255, Pl. VIII/3. Bori} 2009, 214. [ljivar 1996. [ljivar, Kuzmanovi}-Cvetkovi} 1997, 106. [ljivar, Kuzmanovi}-Cvetkovi} 1997, 104. [ljivar. Kuzmanovi}-Cvetkovi} 1998a, 5–6. [ljivar, Kuzmanovi}-Cvetkovi} 2009. [ljivar et al. 2006, 256 ff. Stalio 1960, 34. [ljivar, Kuzmanovi}-Cvetkovi} 2009, 59. See figures in: [ljivar, Kuzmanovi}-Cvetkovi} 2009, 58. [ljivar, Kuzmanovi}-Cvetkovi} 2009, 61. Grbi} 1929; Stalio 1964; [ljivar 1999; [ljivar et al. 2006. [ljivar 1996. Grbi} 1929. Radivojevi} 2006. Grbi} 1929. Grbi} 1929, 8–9. [ljivar et al. 2006, 255. RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 Fig. 9. The profile of the site of Plo~nik facing the river (November 2012). Note the massive cultural layer bearing remains of pits (photo by J. Pendi}) Sl. 9. Profil lokaliteta Plo~nik prema reci (novembar 2012). Primetiti masivni kulturni sloj sa ostacima jama (foto: J. Pendi}) Fig. 10. Southeast profile of Plo~nik: a) dark brown soil; b) light brown soil; c) sterile soil; d) daub; e) ash (after [ljivar, Kuzmanovi}-Cvetkovi} 1998a, 4) Sl. 10. Jugoisto~ni profil Plo~nika: a) tamnobraon zemqa; b) svetlobraon zemqa; c) zdravica; d) lep; e) pepeo (prema: [ljivar, Kuzmanovi}-Cvetkovi} 1998a, 4) 20 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 metal: one hammer-axe (Plo~nik type), five chisels, a bracelet, a pin with a forked end, a copper ingot bar and two stone axes made of magnesite (Fig. 14). The resumed excavation in 1996 took place in the area of the Similar conditions were recorded for Group III, where copper and stone tools were uncovered, scattered in an area 5 x 0.5 m, at a depth of 0.7 m.92 The find consisted of 11 objects, 9 of which were made of copper Fig. 11. Dwelling structure in Trench No. 20, Plo~nik (excavations in 2007) Sl. 11. Naseobinska struktura u sondi 20, Plo~nik (istra`ivawa iz 2007. godine) Fig. 12. A copper chisel (Plo~nik 145) in Trench No. 20 in situ (photo by J. Kuzmanovi}-Cvetkovi}) Fig. 13. A ceramic “tube” (far left) from Trench 20 (photo by J. Kuzmanovi}-Cvetkovi}) Sl. 12. Bakarno dleto (Plo~nik 145) iz sonde 20 in situ (foto: J. Kuzmanovi}-Cvetkovi}) Sl. 13. Kerami~ka „tuba“ (levo) iz sonde 20 (foto: J. Kuzmanovi}-Cvetkovi}) 21 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 Fig. 14. Four Plo~nik groups (clockwise): Group I (complete), Group II (only 13 metal implements), Group IV (complete) and Group III (complete) (after [ljivar et al. 2006, 261–265) Sl. 14. ^etiri grupe nalaza sa Plo~nika (u smeru kazaqke na satu): Grupa I (cela), Grupa II (samo 13 metalnih alatki), Grupa IV (cela) i Grupa III (cela) (prema: [ljivar et al. 2006, 261–265) 22 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 tools. All these were discovered scattered across the same surface, in spit 7 and in the vicinity of a potential rectangular furnace (see Figs. 9, 10).100 Plo~nik 216 is a small fragmented chisel discovered within a stone structure outside the potential dwelling feature, and chronologically falls close to the date obtained from the preceding context of a collapsed burnt structure in spits 10–11 at c. 5040–4840 BC.101 These metal artefacts were made of pure copper, with only traces of other impurities.102 Malachite beads of different sizes and shapes were recovered throughout all horizons of occupation in Plo~nik, six of which have been selected for this study (Table 2, Fig. 16). A bead roughout (72) and a whole bead (43) were discovered in Trench 20, and together with two other whole beads (65, 66) can be dated within the Gradac Phase occupation in Plo~nik. Bead blank (54) and two whole beads (43, 207) can also be dated to the Gradac Phase occupation at this site. A tin bronze foil (Plo~nik 63) was excavated from an undisturbed context, on the floor of a dwelling structure next to the copper metal workshop at the site, about 1 m from a fireplace and was enclosed in several late Vin~a culture pottery vessels.103 Compositional analysis demonstrated that this metal foil was made of a complex alloy of copper and tin, along with significant concentrations of elements like As, Sb, Co, Ni, Pb and Fe. This securely contextualised find comes from a single and undisturbed occupation horizon at the site of Plo~nik, dated to c. 4650 BC. This date is, according to the field evidence, terminus ante quem for the Plo~nik foil. The tin bronze foil from the site of Plo~nik is, therefore, the earliest known tin bronze artefact anywhere in the world.104 Group III discovery and unearthed dozens of ceramic materials belonging to the early Gradac Phase, a small rectangular stone structure and one stone axe made of magnesite, identical to the one accompanying metal objects in Group III.93 The owner of the excavated land on this occasion donated a copper chisel to the National Museum in Belgrade which he found, during the building of his family house, in the vicinity of the Group III location. Group IV was found in 1968 during the building of the Nis–Pri{tina railway, at a depth of c. 0.3 m. It includes 14 objects, 5 of which are copper chisels, eight stone axes made of magnesite and one miniature pottery vessel (a casting pot?) (Fig. 14).94 The casting (?) pot, however, did not show assumed traces of metalworking upon inspection by the primary author at the National Museum in Belgrade. Four Plo~nik groups of metal finds in total amount to: 4 Plo~nik type hammer-axes, 25 chisels, four bracelets, a copper ingot bar and a pin, all together weighing c. 16 kg.95 As such, they are a unique and exceptional find of copper artefacts and, according to the most recent AMS dating, one of the earliest in this part of Eurasia.96 Seventeen copper metal tools from Plo~nik were studied earlier for their chemical composition and lead isotopes, which revealed an unexpected complexity of ore/metal exchange networks: at least three different copper deposits from east Serbia, Macedonia and across Bulgaria provided metal for their production.97 Significantly, all finds originated from a single area in Plo~nik, which excavators also call the “industrial” zone due to the thick ash-and-charcoal deposit frequently encountered in Horizon III. An excellently preserved figurine head from the “industry zone” was discovered with half of the face painted lengthwise with a light grey slip (Fig. 15). The excavators believe that the colour effect on its face symbolised the light (surface) and dark (underground) environments of early miners’ lives in Plo~nik.98 For this study two out of the four most recently discovered and well-contextualised copper artefacts were sampled, labelled as Plo~nik 143 and 145 (see Table 2). Plo~nik 143 was discovered within a stone structure, together with a stone axe made of magnesite and a small pottery vessel, all of which again correspond with the beginning of the Gradac Phase in Plo~nik.99 A massive copper chisel (Plo~nik 145) was unearthed in a similar workshop setting in Trench 20 (Fig. 10), along with two fragmented pieces of a tool and a bracelet respectively (Plo~nik 67, 73), a folded metal sheet (Plo~nik 75), a copper mineral (Plo~nik 71) and several stone 92 Stalio 1964, 35. [ljivar, Kuzmanovi}-Cvetkovi} 1998b, 82. 94 Stalio 1964. 95 [ljivar et al. 2006, 255. 96 Bori} 2009, 214. 97 Pernicka et al. 1997, 93–94, T. 3. 98 [ljivar, Kuzmanovi}-Cvetkovi} 1998a; [ljivar, Kuzmanovi}Cvetkovi} 1998b. 99 [ljivar et al. 2006, 255–256. 100 [ljivar, Kuzmanovi}-Cvetkovi} 2009, 58–59. 101 Bori} 2009, 213–214, Fig. 24. 102 Radivojevi} 2012. 103 Radivojevi} et al. 2013, 1033, Fig. 2. 104 Radivojevi} et al. 2013, 1032. 93 23 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 CONCLUSIONS The overview of activities related to copper mineral use and extractive metallurgy at the Vin~a culture sites of Belovode and Plo~nik suggests that metal smiths at these settlements were covering different stages of metallurgical chaîne opératoire. Both sites, however, exhibit overlapping activities, such as the collection and potential beneficiation of distinctively coloured copper (black and green Mn-rich) minerals as well as pure green ones. This differentiation shows that Vin~a culture metal smiths understood the material properties of these two types of copper minerals, and very likely used them for different purposes. The pure green mineral (malachite) seems to have been used for malachite bead making, as seen in Fig. 1a, while Mn-rich copper ores at these sites were used for copper smelting, as already suggested by Radivojevi} and collaborators.105 Although these distinctively coloured minerals occur at both settlements, evidence for copper smelting has only been recovered from Belovode. Besides the five copper slag samples published in Radivojevi} et al.,106 more slag pieces and slagged sherds from the same trench (No. 3) have advanced our understanding of the early smelting process. The slagged sherds were probably used fragmented, given that the slagged mass spilled over the cross sections (Fig. 2b). These were very likely used to line a hole in the ground, where the metal could have been smelted. It is not yet clear if production evidence from Belovode represents an in situ workshop, or a pit where metal production waste was discarded; future publications on the settlement formation will reveal more information on the character of the context of these finds. However, it is important to emphasise that this context has not been disturbed and, as such, is firmly placed at the beginning of the 5th millennium BC.107 The presence of the black and green (Mn-rich) minerals at the site of Plo~nik may lead one to assume that these were also prepared for copper smelting at this site. The smelting evidence, however, has not yet been discovered, but the presence of macroscopically and compositionally analogous minerals to those smelted in Belovode indirectly suggests that evidence for copper production in Plo~nik remains to be recovered in future investigations. The presence of an in situ metallurgical workshop has only been archaeologically confirmed at the site of Plo~nik (Trench 20).108 Remains of a fragmented tool/ ornament (67), a fragmented bracelet (73), and a folded Fig. 15. A figurine head from Plo~nik, divided lengthwise with one half painted dark and the other whitish, dimensions (in cm): height 20, face 10.7, face width 16.3 and neck width 11.6 (photo by J. Kuzmanovi}-Cvetkovi}) Sl. 15. Glava figurine oslikana po du`ini, dimenzija (u cm): visina 20, lice 17, {irina lica 16,3 i {irina vrata 11,6 (foto: J. Kuzmanovi}-Cvetkovi}) metal sheet (75), together with a well preserved massive copper implement (145) in a single dwelling structure indeed implies that this object could have been occupied by a metal smith. This workshop was, nevertheless, not used for primary metal production but, according to the present evidence, only for the casting and/or repair of metal tools. This copper workshop from Trench 20 was also used for malachite bead making, judging by the presence of a bead roughout (Plo~nik 72, Fig. 16a). The bead blank 105 106 107 108 24 Radivojevi} et al. 2010, 2784. Radivojevi} et al. 2010. Radivojevi} et al. 2010, 2779. [ljivar, Kuzmanovi}-Cvetkovi} 2009. RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 b a c Fig. 16. All stages of malachite bead processing at the site of Plo~nik: a) Bead roughout (72); b) bead blank (54); c) whole bead (66) Sl. 16. Sve faze procesuirawa malahitnih perlica na lokalitetu Plo~nik: a) roughout – jezgro (72); b) blank – nedovr{eno (54); c) cela perlica (66) (Plo~nik 54, Fig. 16b) from another trench (No. 16) a few meters from the metal workshop in Trench 20, may suggest the presence of yet another bead making workshop. Significantly, the assumption on the presence of several workshop areas at the site of Plo~nik does not exclude the possibility of their part-time use by Plo~nik craftsmen. The concept of a specialised metallurgical workshop appears later in prehistory (e.g. Bronze Age), and earlier suggestions of specialised workshops need to be treated with caution until they are supported by a more detailed publication on site formation, which is currently beyond the scope of this paper. The chaîne opératoire of copper mineral use and metallurgical activities at the sites of Belovode and Plo~nik, despite the overlap in copper mineral selection, differs when it comes to production (which is identified in Belovode only) and consumption activities (detected in Plo~nik only). Nonetheless, the suggestion of the potential existence of smelting activities in the site of Plo~nik based on the presence of black and green minerals, as well as a few (insufficiently contextualised) metal artefacts from the catchment area of Belovode may prove this division on the production and consumption site as unsubstantiated in future investigations. Thus far, it appears that present materials from these two sites complement each other: the first step of the process, primary metal production, was conducted in Belovode, while the second, casting and finishing of metal artefacts, took place at the site of Plo~nik. We still lack evidence of the (re)melting of primary produced copper, and a crucible that was most likely used during this particular process. Evidence from the two different sites, however, can still help to reconstruct the Vin~a culture metallurgy chaîne opératoire. Although more nuanced discussions on this process remain to be published, this qualitative and macroscopic overview of the present evidence provides a good foundation for understanding the metal production process in the 5th millennium BC Vin~a culture. Despite the fact that both Belovode and Plo~nik were carefully excavated, information on their formation processes is not always clear, as already pointed out above. Since the majority of field documentation remains unpublished, relevant contexts for this study are limited to relative spits within individual trenches. This has provided only restricted information on the spatial distribution of metallurgical debris and its relation to the spatially closest settlement features. Importantly, most of these features originated from arbitrary units, where the relationships are obscured and not always straightforward. Thus, this paper focused on the material properties of a variety of excavated artefacts, attempting to point to important connections among them from this particular perspective, rather than to rely on a plan of distribution of small-sized trenches scattered across both Belovode and Plo~nik. Another problem in relation to the sites of Belovode and Plo~nik arises with the lack of AMS data on specific contexts related to diverse metallurgical activities. Bori}109 made a detailed study on the relation of metallurgical activities and the newly available AMS data. Nevertheless, these dates are not sufficient for the fine separation of copper mineral use and pyrometallurgical processes in the Vin~a culture settlements, nor 109 25 Bori} 2009. RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 products, cover nearly a complete chaîne opératoire of metallurgy. This production chain comes across as sufficiently elaborate in the earliest known stage of pyrometallurgical activity in this part of the world, implying that it had probably developed a few generations earlier from the moment we spotted it in archaeological records. Also, the cultural transmission of this once precious skill seems to take place across two relatively distant settlements, implying close economic and potentially social connections among their inhabitants. The craftsmen from Belovode and Plo~nik were, based on current evidence, covering different ends of the same metal making process, which strengthens the likelihood of their potential collaboration. The lead isotope match of Belovode slag samples with the Plo~nik copper chisel (216) strengthens this assumption;113 this will be explored in detail in future publications. Finally, the quantity and quality of collected and sampled materials is currently unprecedented in academic work, and provides an excellent case for studying the emergence, evolution and transmission of metallurgical skills within the Vin~a culture, and across the Balkans. do they provide direct dating evidence for the majority of metallurgy-related finds. Therefore, the temporal analysis of metallurgical activities here has been mainly dependant on the relative chronology, based on specific pottery forms and the conventional periodisation of the Vin~a culture across the entire region. Such an approach, although not without errors, currently provides the only feasible resolution for the context of the majority of metallurgy-related samples presented here. The relation of the Gradac Phase to the rise of metallurgy is undoubtedly crucial for our understanding of this process in the Vin~a culture: the earliest stratigraphic evidence for copper smelting is discovered at the beginning of this phase. It is also contemporary with the intensified activities in Rudna Glava, as well as the earliest dated copper implement from Plo~nik. The changes in material culture that follow this phase are a phenomenon common for the Balkans in general,110 and require particular attention in interpreting the origins of metallurgy in this part of Eurasia in future discussions. The introduction of metallurgy evidently influenced other aspects of material culture at the time. The most important association is, beyond doubt, pottery production, such as the appearance of black burnished ware, whose conjunction with the emergence of metallurgy has been discussed at length;111 nevertheless, not yet sufficiently studied. A valuable observation comes from Belovode, where modelled applications on figurines resemble contemporaneous metal jewellery from sites situated along the lower Danube, and further towards the northern Black Sea coast.112 This analogy highlights the importance of the Danubian communication route for the spread of metallurgy across the Balkans. However, the main point of the contextual and qualitative survey of available artefacts in this overview is the diversity of metallurgical debris identified across the sites of Belovode and Plo~nik. Various technological sequences, indicated by crushed minerals, fragments of installations, smelting slags, casting debris or final Acknowledgements This paper represents a part of M. Radivojevi}’s PhD thesis, conducted at the UCL Institute of Archaeology and funded generously under the UK’s EPSRC Dorothy Hodgkin Postgraduate Award Scheme, Freeport McMoRan Gold and Copper Foundation (via the Institute for Archaeo-Metallurgical Studies in London), Serbian Ministry of Culture and Fund for Young Talents, British Federation of Women Graduates, Soros Foundation, Prokuplje City Council and “Planinka” Kur{umlija. M. Radivojevi} is indebted to her supervisors Professors Th. Rehren, S. Shennan and E. Pernicka for their immense help during her research, as well as to B. Jovanovi} for encouragement. Both authors remain grateful to D. [ljivar for his kind help with the field documentation. 110 111 112 113 26 cf. Gara{anin 1995. e.g. Renfrew 1969; Gimbutas 1976. See examples in: Dumitrescu 1961; Sergeev 1963. Radivojevi} 2012. RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 BIBLIOGRAPHY: Gimbutas 1976 – M. Gimbutas, Neolithic Macedonia as Reflected by Excavations at Anza, Los Angeles 1976. Gläser 1996 – R. Gläser, Zur absoluten Datierung der Vin~a-Kultur anhand von C-Daten, in: Draºovean, F. (ed.), The Vin~a culture, Its Role and Cultural Connections, Timisoara 1966, 141–162. Grbi} 1929 – M. Grbi}, Plo~nik, eine Prähistorische Ansiedlung aus der Kupferzeit, Belgrade 1929. Jacanovi}, [ljivar 1995 – D. Jacanovi}, D. [ljivar, Rezultati arheolo{kih istra`ivanje neolitskog naselja Konju{ica u Vite`evu, Glasnik Srpskog Arheolo{kog Dru{tva 10, 1995, 181–185. Jacanovi}, [ljivar 2003 – D. Jacanovi}, D. [ljivar, Preliminarni izve{taj sa arheolo{kog istra`ivanja od 1998 do 2002, Veliko Laole, Belovode, Viminacium 13–14, 2003, 297–302. Jovanovi} 1982 – B. Jovanovi}, Rudna Glava, najstarije rudarstvo bakra na centralnom Balkanu, Beograd, Bor 1982. Jovanovi} 1994 – B. Jovanovi}, Gradac Phase in the relative chronology of the late Vin~a culture, Starinar XLIII–XLIV, 1994, 1–12. Jovi} 1996 – V. Jovi}, Mineralogical study of malachite and azurite from the Belovode locality (Veliko Laole), Préhistorie Européene 8, 1996, 191–196. Krajnovi}, Jankovi} 1995 – D. Krajnovi}, S. Jankovi}, Copper mineralisation as potential raw material source of ancient copper metallurgy in Serbia, in: Jovanovi}, B. (ed.) Ancient Mining and Metallurgy in Southeast Europe, Donji Milanovac, Belgrade, Bor 1995, 21–28. Kuzmanovi}-Cvetkovi} 1998 – Prokuplje, grad sv. Prokopija, Prokuplje 1998. Kuzmanovi}-Cvetkovi}, [ljivar 1998 – J. Kuzmanovi}-Cvetkovi}, D. [ljivar, Die Göttin aus Plo~nik, Starinar XLIX, 1998, 173–178. Lankton et al. 2003 – J. W. Lankton, J. Diamanti, J. M. Kenoyer, A Bead Timeline: A Resource for Identification, Classification and Dating, Washington D.C. 2003. Miloj~i} 1949 – V. Miloj~i}, Chronologie der Jüngeren Steinzeit Mittel- und Südosteuropas, Berlin 1949. Mrkobrad 1989 – D. Mrkobrad, Istra`ivanje rudarske topionice u Donjoj Zaplanini, Glasnik Dru{tva Konzervatora Srbije 13, 1989, 91–92. Mrkobrad et al. 1989 – D. Mrkobrad, V. Bogosavljevi}, D. Bogosavljevi}, Prilog istra`ivanju kopaoni~kog rudarstva, Glasnik Dru{tva Konzervatora Srbije 13, 1989, 93–95. Antonovi} 2002 – D. Antonovi}, Copper processing in Vin~a: new contributions to the thesis about metallurgical character of the Vin~a culture, Starinar LII, 2002, 27–45. Antonovi} 2006 – D. Antonovi}, Malachite finds in the Vin~a culture: evidence for early copper metallurgy in Serbia, Metalurgija – Journal of Metallurgy 12, 2006, 85–92. Arsenijevi}, @ivkovi} 1998 – M. Arsenijevi}, J. @ivkovi}, Methodology and the results of statistical-typological analysis of pottery from the Vin~a culture settlement of Belovode, near Petrovac-upon-Mlava, Glasnik Srpskog Arheolo{kog Dru{tva 14, 1998, 281–291. Bar-Yosef Mayer, Porat 2008 – D. E. Bar-Yosef Mayer, N. Porat, Green stone beads at the dawn of agriculture, Proceedings of the National Academy of Sciences 105, 2008, 8548–8551. Bogosavljevi}-Petrovi} 1995 – V. Bogosavljevi}Petrovi}, Arheometalur{ki kompleksi na Kopaoniku, Glasnik Srpskog Arheolo{kog Dru{tva 10, 1995, 58–71. Bogosavljevi}-Petrovi}, Tomovi} 1994 – V. Bogosavljevi}-Petrovi}, M. Tomovi}, Kasnoanti~ki rudarsko-metalur{ki kompleks na zapadnom Kopaoniku, Glasnik Srpskog Arheolo{kog Dru{tva 9, 1994, 236–240. Bogosavljevi} 1989 – V. Bogosavljevi}, Uvod u rudarsku arheologiju na Kopaoniku (oblici starog rudarstva i metalurgije), Glasnik Srpskog Arheolo{kog Dru{tva 9, 1989, 86–93. Bogosavljevi} et al. 1988 – V. Bogosavljevi}, D. Mrkobrad, D. Bogosavljevi}, Istra`ivanje srednjovekovnog rudarstva na zapadnom Kopaoniku (okolina Ki`evaka), Na{a Pro{lost 3, 1988, 9–48. Bori} 2009 – D. Bori}, Absolute dating of metallurgical innovations in the Vin~a culture of the Balkans, in: T. L. Kienlin, B. W. Roberts (eds.), Metals and Societies. Studies in honour of Barbara S. Ottaway, Verlag Dr. Rudolf Habelt GmbH, Bonn 2009, 191–245. Dumitrescu 1961 – H. Dumitrescu, Connections between the Cucuteni–Tripolie cultural complex and the neighbouring Eneolithic cultures in the light of the utilization of golden pendants, Dacia 5, 1961, 69–93. Gara{anin 1951 – M. Gara{anin, Hronologija vin~anske grupe, Ljubljana 1951. Gara{anin 1973 – M. Gara{anin, Praistorija na tlu Srbije, Beograd 1973. Gara{anin 1995 – M. Gara{anin, Die Gradac-Stufe der Vin~a-Gruppe und der Beginn des Aeneolithikums, Dacia 38–39, 1995, 9–17. 27 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 Schier 2000 – W. Schier, Measuring change: the Neolithic pottery sequence of Vin~a–Belo Brdo, Documenta Praehistorica 27, 2000, 187–197. Sergeev 1963 – G. P. Sergeev, Rannetripolskii klad u s. Karbuna. Sovetskaja arheologija 1, 1963, 135–151. [ljivar 1996. – D. [ljivar, The Eastern settlement of the Vin~a culture at Plo~nik: a relationship of its stratigraphy to the hoards of copper objects, Starinar XLVII, 1996, 85–98. [ljivar 1999 – D. [ljivar, Plo~nik, Naselje vin~anske kulture i problem najranije metalurgije bakra, in: Marinkovi}, D. (ed.) Prokuplje u Praistoriji, Antici i Srednjem veku, Prokuplje 1999, 31–44. [ljivar 2006 – D. [ljivar, The earliest copper metallurgy in the Balkans, Metalurgija-Journal of Metallurgy 12, 2006, 93–104. [ljivar, Jacanovi} 1996a – D. [ljivar, D. Jacanovi}, Veliko Laole–Belovode, the settlement of the culture of Vin~a, Journal of Serbian Archaeological Society 12, 1996, 55–60. [ljivar, Jacanovi}, 1996b – D. [ljivar, D. Jacanovi}, Veliko Laole, Belovode – Vin~a culture settlement in northeastern Serbia, Préhistorie Européene 8, 1996, 175–188. [ljivar, Jacanovi} 1997 – D. [ljivar, D. Jacanovi}, Veliko Laole – Belovode, the settlement of the Vin~a culture, Journal of Serbian Archaeological Society 13, 1997, 115–125. [ljivar et al. 2006 – D. [ljivar, D. Jacanovi}, J. Kuzmanovi}-Cvetkovi}, New contributions regarding the copper metallurgy in the Vin~a culture, in: Tasi}, N. and Grozdanov, C. (eds.) Homage to Milutin Gara{anin, Belgrade 2006, 251–266. [ljivar, Kuzmanovi}-Cvetkovi}, 1997 – D. [ljivar, J. Kuzmanovi}-Cvetkovi}, Plo~nik near Prokuplje, the Vin~a culture settlement, Journal of the Serbian Archaeological Society 13, 1997, Beograd, 103–113. [ljivar, Kuzmanovi}-Cvetkovi} 1998a – D. [ljivar, J. Kuzmanovi}-Cvetkovi}, Najstarija metalurgija bakra na Plo~niku kod Prokuplja, naselju vin~anske kulture, Arheometalurgija 6, Beograd 1998, 1–18. [ljivar, Kuzmanovi}-Cvetkovi}, 1998b – D. [ljivar, J. Kuzmanovi}-Cvetkovi}, Plo~nik near Prokuplje, excavation in 1997, Journal of the Serbian Archaeological Society 14, 1998, Beograd, 79–85. [ljivar, Kuzmanovi}-Cvetkovi} 2009 – D. [ljivar, J. Kuzmanovi}-Cvetkovi}, Plo~nik, archaeology and conservation, Diana 13, 2009, 56–61. [ljivar et al. 2012 – D. [ljivar, J. Kuzmanovi}Cvetkovi}, J. @ivkovi}, Belovode, Plo~nik: on copper Ottaway 2001 – B. S. Ottaway, Innovation, production and specialisation in early prehistoric copper metallurgy, European Journal of Archaeology 4, 2001, 87–112. Ottaway, Roberts 2008 – B. S. Ottaway, B. Roberts, The emergence of metalworking, in: Jones, A. (ed.) Prehistoric Europe: Theory and Practice, Chichester 2008, 193–225. Pernicka et al. 1997 – E. Pernicka, F. Begemann, S. Schmitt-Strecker, H. Todorova, I. Kuleff, Prehistoric copper in Bulgaria. Its composition and provenance, Eurasia Antiqua 3, 1997, 41–180. Radivojevi} 2006 – M. Radivojevi}, A contribution to the typology and distribution of hammer axes of Plo~nik type in southeast Europe, Journal of Serbian Archaeological Society 22, 2006, 211–224. Radivojevi} 2007 – M. Radivojevi}, Evidence for early copper smelting in Belovode, a Vin~a culture settlement in Eastern Serbia, Unpublished MSc Thesis, UCL Institute of Archaeology, London 2007. Radivojevi} 2012 – M. Radivojevi}, On the Origins of Metallurgy in Europe: Metal Production in the Vin~a Culture. Unpublished PhD Thesis, UCL Institute of Archaeology, London 2012. Radivojevi} et al. 2013 – M. Radivojevi}, Th. Rehren, J. Kuzmanovi}-Cvetkovi}, M. Jovanovi}, J. P. Northover, Tainted ores and the rise of tin bronze metallurgy, c. 6500 years ago, Antiquity 87, 2013, 1030–1045. Radivojevi} et al. 2010 – M. Radivojevi}, Th. Rehren, E. Pernicka, D. [ljivar, M. Brauns, D. Bori}, Origins of extractive metallurgy: new evidence from Europe, Journal of Archaeological Science 37, 2010, 2775–2787. Rehren 2003 – Th. Rehren, Crucibles as reaction vessels in ancient metallurgy, in P. T. Craddock & J. Lang (eds.) Mining and Metal Production Through the Ages, London 2003, 207–215. Renfrew 1969 – C. Renfrew, The autonomy of the south-east European Copper Age, Proceedings of the Prehistoric Society 35, 1969, 12–47. Roux 2003 – V. Roux, A dynamic systems framework for studying technological change: application to the emergence of the potter’s wheel in the southern Levant, Journal of Archaeological Method and Theory 10, 1969, 1–30. Schier 1996 – W. Schier, The relative and absolute chronology of Vin~a: new evidence from the type site, in: Draºovean, F. (ed.) The Vin~a Culture, its Role and Cultural Connections, Timiºoara 1996, 141–162. 28 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) Stalio 1973 – B. Stalio, ^etvrti nalaz bakarnog i kamenog oru|a sa Plo~nika kod Prokuplja, Zbornik Narodnog Muzeja 7, 1973, 157–161. Tasi} 2008 – N. Tasi}, Nemi svedoci jednog vremena: figuralna umetnost Vin~e, in: Nikoli}, D. (ed.) Sto godina istra`ivanja arheolo{kog lokaliteta Vin~a–Belo Brdo, Belgrade 2008, 139–163. metallurgy in the Vin~a culture, Zbornik Narodnog Muzeja 20, 2012, 27–46. Stalio 1960 – B. Stalio, Plo~nik–Prokuplje – naselje, Arheolo{ki pregled 2, 1960, 33–36. Stalio 1962 – B. Stalio, Plo~nik, Prokuplje – naselje vin~anske grupe, Arheolo{ki pregled 4, 1962, 19–25. Stalio 1964 – B. Stalio, Novi metalni nalaz iz Plo~nika kod Prokuplja, Zbornik Narodnog Muzeja 4, 1962, 35–41. Rezime: STARINAR LXIV/2014 MIQANA RADIVOJEVI], UKL Arheolo{ki institut, London i Balkanolo{ki institut Srpske akademije nauka i umetnosti, Beograd JULKA KUZMANOVI]-CVETKOVI], Narodni muzej Toplice, Prokupqe MINERALI BAKRA I ARHEOMETALUR[KI MATERIJALI SA VIN^ANSKIH LOKALITETA BELOVODE I PLO^NIK: PREGLED MATERIJALA I NOVI PODACI Kqu~ne re~i. – metalurgija, vin~anska kultura, Belovode, Plo~nik, bakar, zgura. li, koji su se koristili za pravqewe malahitnih perlica, i druge, kojoj pripadaju crno-zeleni, tj. bogati manganom, i za koji je prethodna studija ve} pokazala da su topqeni kako bi se dobio metal. Ova druga vrsta minerala je nazvana rudama, za razliku od prve, koju i daqe nazivamo mineralima, a razlikujemo ih po temperaturnom tretmanu. Obe vrste minerala na|ene su na oba lokaliteta. Ostaci proizvodwe metala su na|eni samo na Belovodama i to najvi{e u jednoj sondi, br. 3 koja je iznedrila ukupno 8 nalaza bakarnih zgura i 4 nalaza keramike sa zalepqenom zgurom (tako|e od topqewa bakra). Za ove analize je ve} potvr|eno da pripadaju primarnoj produkciji metala. Isti zakqu~ak va`i i za veliki komad zgure iz sonde br. 9 na Belovodama, za koji su analize pokazale da ima zna~ajnu koncentraciju olova. Iako na lokalitetu Plo~nik nisu prona|eni ostaci proizvodwe metala, prisustvo zeleno-crnih ruda bakra bogatih manganom, koje su se na Belovodama topile, ukazuju na mogu}nost da }e mo`da sli~an dokaz biti prona|en i na Plo~niku u budu}im istra`ivawima. Ono ~ega na Belovodama nema jesu gotovi proizvodi od bakra, koji su prona|eni u zna~ajnom broju na lokalitetu Plo~nik. Radionica iz sonde 20 sadr`ala je uzorke savijenog lima, fragmentovanih alatki ili nakita, kao i gotove metalne alatke, {to ukazuje da se u woj najverovatnije obra|ivao metal. Ovaj metal je primarno proizveden negde drugde, po{to su izotopske analize pokazale da se zgure sa Belovoda i bakarno dleto sa Plo~nika sla`u po mestu porekla rude bakra od koje su napravqeni. Radionica u sondi 20 na Plo~niku tako|e je iznedrila nezavr{ene Belovode i Plo~nik, lokaliteti vin~anske kulture koji se nalaze u isto~noj i ju`noj Srbiji, ve} decenijama privla~e pa`wu {iroke akademske javnosti zbog otkri}a jedinstvenih nalaza koji potvr|uju postojawe metalur{ke aktivnosti u ovoj kulturi. Ovi nalazi datiraju sa po~etka 5. milenijuma pre n.e., {to ih ~ini trenutno najstarijim poznatim dokazima prerade metala u svetu, koji ujedno i o`ivqavaju teoriju o nezavisnom nastanku metalurgije na Balkanu. Sa novim podacima se nametnula potreba da se sa~ini pregled dostupnih informacija o na~inu proizvodwe metala i o tehnolo{kim izborima vin~anskih kova~a tokom tog procesa. Autori su sastavili kvalitativan pregled iskopanog materijala koji je u vezi sa upotrebom bakarnih minerala i sa arheometalur{kim aktivnostima na ova dva lokaliteta. Za ovu priliku je napravqen pregled materijala koji je ve} publikovan, ali i onog ~ije su analize skoro zavr{ene u sklopu doktorske teze primarnog autora (ukupno 48 analiziranih uzoraka). Sav materijal je podeqen u tri glavne kategorije: arheolo{ki minerali, ostaci (primarne) proizvodwe metala i izrada i obrada gotovih metalnih artefakata, koji zajedno ~ine lanac metalur{ke proizvodwe. Rezultati studije ukazuju na prisustvo raznovrsnih nalaza koji se zajednouklapaju u jedan skoro kompletan metalur{ki lanac operacija. Ovaj lanac po~iwe odabirom bakarnih minerala i wihovom pripremom (usitwavawe) za proces topqewa. Primetili smo da su vin~anski metalurzi znali da razlikuju materijalne kvalitete dveju vrsta bakarnih minerala: jedne kojoj pripadaju ~isto zeleni minera- 29 RADIVOJEVI], KUZMANOVI]-CVETKOVI], Copper minerals and archaeometallurgical… (7–30) STARINAR LXIV/2014 sti obrade metala bakra. Ova vrsta tehnolo{kog pristupa u izu~avawu procesa proizvodwe metala pokazala se kao jedeina koja u datom trenutku mo`e da izvede zakqu~ke o ovoj aktivnosti na lokalitetima Belovode i Plo~nik. Terenska dokumentacija, koja je samo parcijalno publikovana i koja se oslawa na specifi~ne kontekste u okviru malih sondi, jo{ ne pru`a zadovoqavaju}u sliku o prostornom odnosu analiziranog materijala, pa je stoga analiti~ki pristup jedini koji trenutno mo`e me|usobno da pove`e bakarne minerale, rude, zgure i metalne predmete u generalnu sliku razvoja metalurgije na ovim vin~anskim naseqima. Kolekcija nalaza, iako mala po uobi~ajenim standardima predstavqa najvredniju kolekciju te vrste u svetu, jer je po kvalitetu unikatna u hronolo{kim okvirima u kojima se pojavila. komade bakarnih perli, {to je defini{e kao radionicu za „hladnu“ obradu bakarnih minerala, naspram visokotemperaturnog tretmana koji je primewivan za izradu metalnih alatki. Na lokalitetu Plo~nik je prona|en i lim na~iwen od prirodne bronze, tj.topqewem kompleksne rude bakra i kalaja, a ne legurom ta dva metala. Ovaj izvanredan nalaz je trenutno najstariji nalaz ovakve vrste na svetu. Lanac metalur{kih operacija sa lokaliteta Belovode i Plo~nik se stoga me|usobno dopuwavao – dok se na Belovodama ekstrahovao metal, na Plo~niku se on kovao i obra|ivao. Iako ovaj lanac operacija nije kompletan, ova studija je demonstrirala da on postoji u razvijenoj formi, sa tehnolo{kim izborima koji odra`avaju odli~no poznavawe hemijskih i fizi~kih karakteristika razli~itih minerala, kontrole procesa topqewa metala i tehni~kih mogu}no- 30 MARIO GAVRANOVI], Berlin ALEKSANDAR KAPURAN, Archaeologisches Institut, Belgrad ÜBER EINIGE TÜLLENBEILVARIANTEN IM ZENTRALBALKAN e-mail: [email protected] UDK: 903.21"637"(497) DOI: 10.2298/STA1464031G Abgegeben: Februar 21, 2014 Akzeptiert: Juli 19, 2014 Originelle wissenschaftliche Arbeit Abstract. – Im Rahmen der Vorbereitung für die Gesamtdarstellung der bronzenen Beile aus Serbien konnten einige lokale Formen mit einem begrenzten Verbreitungsradius verzeichnet werden. Es handelt sich in erster Linie um mehrere Varianten der Tüllenbeile mit einer arkadenförmigen Facettierung auf dem Beilkörper, einer seitlichen Öse und einem verstärkten Tüllenwulst mit horizontalen Rippen darunter. Im Bezug auf die Typologie, Verbreitung und den kultur-historischen Kontext sind die meisten der hier präsentierten Funde bislang unzureichend beschrieben. Die charakteristischen Tüllenbeile lassen sich vermutlich als Produkte der Werkstätte im Zentralbalkan interpretieren. Wichtig in diesem Zusammenhang sind noch prähistorische Kupferbergwerke, die im Verbreitungsraum der umschriebenen Beile liegen. Key words. – Tüllenbeile aus Bronze, Spätbronzezeit, Zentralbalkan, Hortfunde, Belegi{ II – Gava, Metallurgie. I m Laufe der Fundaufnahme für das Sammelband über die bronzezeitlichen Beile Serbiens lie¼en sich mehrere Typen und Varianten der spätbronzezeitlichen Tüllenbeile lokalen Charakters erkennen, deren Verbreitungsraum hauptsächlich Serbien umfasst (Karten 1–3). In der vorliegenden Arbeit wird der Schwerpunkt auf die Tüllenbeile mit einem verdickten Mündungswulst mit Horizontalrippen und einem facettierten arkadenförmigen oder dreieckförmigen Feld auf dem Körper. Auch wenn einige der hier präsentierten Funde bereits publiziert sind, steht eine typologischchronologische Gliederung dieser Beilgruppe für Serbien noch immer aus. Dabei lassen sich durchaus Ähnlichkeiten mit den Varianten C6 bis C10 der siebenbürgischen Tüllenbeile nach der Typologie M. Rusu1 feststellen, obwohl die Exemplare aus Serbien im Bezug auf die horizontale Rippen und Wulstprofilierung sowie durch zusätzliche dekorative Elemente als eigenständige lokale Varianten aufzufassen sind. Tüllenbeile mit arkadenförmigen Feld kommen zahlreich in den Regionen an der unteren Donau vor, mit einer besonders starken Konzentration im bulgarischen Raum zwischen den Flüssen Iskar und Jantra.2 Doch, trotz der gro¼en Anzahl an Typen und Varianten sind die ostbalkanischen Tüllenbeile in der Regel nicht mit einer seitlichen Öse ausgestattet. Die für Serbien charakteristische Kombination aus einem durch horizontale 1 2 Rusu 1966, 27. ^ernùh 1978, 187; Derga~ev 2011, Sl. 5. * Die Arbeit resultiert aus dem Projekt: Archäologie Serbiens: Kulturidentität, Integrationsfaktoren, technologische Prozesse und die Rolle des Mittelbalkans in der Entwicklung europäischer Vorgeschichte (Nr. OI 177020), des Ministeriums für Bildung, Wissenschaft und Technologische Entwicklung der Republik Serbien. 31 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 oberen Ende regelhaft an den etwas verdickten Mündungswulst ansetzt. Gelegentlich kommen auch unregelmä¼ige Rippen vor, die sich vorzugsweise zwischen dem Tüllenmund und dem facettieren Feld befinden. In Serbien ist dieser Typ durch mehrere Exemplare aus dem Depot Urovica12 sowie in den Horten Mesi}–[upaja13, Leskovo14 in Se~anj III15 vertreten. Hinzu kommen noch zwei Einzelfunde aus der Umgebung der Stadt Jagodina.16 Folgt man die Datierung der genannten Horte, so kann eine lange Nutzungszeit angenommen werden. Die Depots aus Mesi} und Urovica stammen aus dem zweiten Horizont der serbischen Hortfunde nach Vasi} (Ha A1), während Leskovo und Se~anj deutlich jünger sind und in die Zeit der Stufen Ha A2 bis Ha B1 datieren.17 • Tüllenbeile mit kräftigem Mündungswulst und einer mandelförmigen Zier innerhalb des arkadenförmigen Feldes. Der einzige Fund aus Serbien liegt im Depot Mali Izvor bei Zaje~ar vor. Dieses Beil wurde bei Derga~ev dem Typ Verbica A, Variante B zugeschrieben,18 obwohl die meisten Vergleichsfunde aus Bulgarien keine seitliche Öse aufweisen.19 Die Tüllenbeile vom Typ Verbica A, Variante B kommen in den Hortfunden der Zeit Ha A1 und Ha A2 vor.20 Der Fund aus Zaje~ar stellt derzeit den westlichsten Verbreitungspunkt dieser ostbalkanischen Form dar, die vor allem für dem Raum Nordbulgariens charakteristisch ist. Die zweite Gruppe der bislang ausführlich umschriebenen Beile setzt sich aus den Funden zusammen, die den Varianten C1 bis C10 der siebenbürgischen Tüllenbeile nach der Klassifikation Rippen profilierten Tüllenrand, einem arkadenförmigen Feld und einer Öse ist unter der Masse der bulgarischen Funde so gut wie nicht vorhanden. Erschwerend bei der chronologischen Einordnung der hier definierten Varianten ist die Tatsache, dass die meisten Beile ohne entsprechenden Kontext bzw. als Einzelfunde überliefert sind. Deshalb wird man auf die Vergleiche mit den geschlossenen Funden, in erster Linie mit den Depots aus Serbien und benachbarten Ländern Südosteuropas, nicht verzichten können. TYPOLOGISCHE GLIEDERUNG Betrachtet man alle Tüllenbeile mit einem arkadenförmigen Feld aus Serbien, so lassen sich zunächst einige Grundtypen aussondern, die den Funden aus Rumänien oder Bulgarien mehr oder weniger entsprechen und die in der bisherigen Literatur ausführlich behandelt worden sind. Sie werden in dieser Arbeit nicht detailliert dargestellt, dennoch sollen sie zwecks besserer Übersicht kurz erwähnt werden. Die erste grö¼ere Gruppe umfasst die Tüllenbeile mit einem leicht verdickten Mündungswulst ohne jegliche Horizontalrippen. Unter den Funden aus Serbien sind drei unterschiedliche Typen zu erkennen: • Tüllenbeile ohne Öse mit einem dreieckigen bzw. arkadenförmigen Feld auf dem Beilkörper. Typische Vertreter sind die Funde aus Futog3, [etonje4 (mit dreieckigem Feld) oder aus Privina Glava5, Dobrinci6, Dubravica7 oder Mali Izvor8 (mit arkadenförmigen Feld). Die Funde aus Serbien wurden in der umfangreichen Monographie V. Derga~evs über die Tüllenbeile der unteren Donau mitberücksichtigt und dem Typ Okniþa zugeschrieben.9 Dieser Typ zeichnet sich durch eine geringe Länge, die selten 10 cm übertrifft. Der Typ Okniþa tritt in Serbien, Ungarn, Rumänien und Bulgarien hauptsächlich in den Hortfunden des Zeithorizontes Ha A1 auf, obwohl Exemplare aus deutlich jüngeren Zusammenhängen (Ha B2/3) ebenfalls bekannt sind.10 Der universelle Charakter dieses relativ einfachen Typs spiegelt sich auch in seiner weiten Verbreitung wider. Eine stärkere Konzentration lässt sich im Norden Bulgariens und auf der gegenüberliegenden Donauseite bzw. in der Kleinen Walachei (Oltenien) im Süden Rumäniens beobachten.11 • Tüllenbeile mit einem dreieckigen bzw. arkadenförmigen Feld auf dem Beilkörper und einer seitlichen Öse. Der einzige Unterschied zu dem vorherigen Typ ist die seitlich angebrachte vertikale Öse, die mit dem 3 Bori} 1997, T. 4/35. Jacanovi}, Radoj~i} 2003, br. 2. 5 Gara{anin 1975, T. 63/4. 6 Popovi} 1994, T. 8/3. 7 Gara{anin 1954, T. 36/4. 8 Vasi} 1982, Abb. 3/2. 9 Derga~ev 2011, 94, Sl. 51–52 und Karte 3 im Katalog. 10 Ibid., 100–103. 11 Ibid., Sl. 55. 12 Srejovi} 1975, T. 81/1–8, 11–14; 62/1.5. 13 Ra{ajski 1975, T. 61/5. 14 Todorovi} 1975, T. 71/8. 15 Marinkovi} 1991, T. 1/4. 16 Stoji} 1986, T. 21/4.6. 17 Vasi} 1982, 268, 282. 18 Derga~ev 2011, 80, Sl. 41 B. 19 Ibid., Sl. 41 B/18.19. 20 Ibid., Sl. 46. 4 32 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) von M. Rusu angehören.21 Das wesentliche Merkmal aller C-Varianten ist eben die arkadenförmige, facettierte Fläche auf dem Beilkörper. Die Varianten C1 bis C4 umfassen Exemplare ohne besondere Verdickung am Tüllenwulst und ohne ausgeprägte Profilierung darunter, während die Varianten C5 bis C10 Beile mit horizontalen Rippen unter dem mehr oder weniger verstärkten Wulst einschlie¼en. Unter Berücksichtigung der in dieser Arbeit vorgestellten Varianten, sollten in erster Linie die Funde der Varianten C5 bis C10 (falls vorhanden) aus Serbien kurz angeführt werden: • Die Variante C6 umschreibt Beile mit einem verstärkten Tüllenmund, der an den beiden Enden bogenförmige bzw. „hornartige“ Zipfeln aufweist. Der Gesamtumriss ist meistens dreieckig mit einem sanften Übergang zwischen der Schmal- und Breitseite. Die Schneide ist in der Regel breit fächerförmig, während die Öse unter den Verstärkungen sitzt. Typische Vertreter der Variante C6 in Serbien sind die Funde aus Vin~a22, zwei Beile aus dem Hort Boljetin bei Donji Milanovac23 so wie die Beile aus den ostserbischen Horten Alun bei Brza Palanka24 und Mali Izvor bei Zaje~ar.25 Alle Exemplare aus Serbien weisen recht einheitliche Ma¼en mit einer Länge zwischen 9,5 und 10,5 cm auf. Auffällig bei der Verbreitung ist, dass alle bislang bekannten Tüllenbeile der Variante C6 östlich der Velika Morava bzw. in Ostserbien verzeichnet wurden. Die serbischen Funde widerlegen somit die Annahme Rusus über die ausschlie¼liche Verbreitung der Variante C6 in den Regionen nördlich der Donau (Muntenien, Oltenien, Banat). Ausgehend von den Hortfunden Boljetin, Alun und Mali Izvor, lässt sich die Variante C6 in Serbien in eine lange Zeitspanne zwischen Ha A1 und Ha B2 datieren.26 • Der Variante C7 nach Rusu können aus Serbien ein Beil aus dem Hort Mali Izvor27, ein Einzelfund aus Dobrnje bei Petrovac na Mlavi28 sowie ein Exemplar aus dem Hort Se~anj III im Banat29 zugeordnet werden. Es handelt sich um Tüllenbeile mit einem kräftigen, gerundeten Mündungswulst und einer horizontalen Rippe direkt darunter. Die vertikale Öse setzt mit dem oberen Ende an den verdickten Wulst an, während das untere Ende unterhalb der horizontalen Rippe liegt. Im Bezug auf die Grundform mit den leicht ausbreitenden Seiten und einer halbkreisförmigen Schneide, sind sich alle serbischen Funde sehr ähnlich. Auch bei der Länge (zwischen 10 und 10,5 cm) gibt es wenig Abweichungen. Die Tüllenbeile der Variante C7 STARINAR LXIV/2014 kommen schon in den Depots der zweiten Stufe nach Vasi} (Ha A1 – Mali Izvor) vor, erscheinen jedoch, wie z. B. in Se~anj III, offenbar auch in den jüngeren Stufen (Ha A2 – Ha B1).30 • Zu erwähnen ist noch, dass die hier definierte Variante F letztlich identische Verzierung wie die siebenbürgische Variante C5 besitzt.31 Der einzige Unterschied ist die stärkere Rippenprofilierung unter dem Tüllenwulst bei den serbischen Funden, weswegen sie als eine eigenständige Variante aufzufassen sind. Ebenfalls, die Tüllenbeile der Variante E entsprechen mehr oder weniger den siebenbürgischen Varianten C9 und C10, auch wenn es auch hier gewisse Abweichungen hinsichtlich der Tüllenprofilierung gibt, die eine Umschreibung der eigenen Varianten gestatten. Neben den aufgezählten Tüllenbeilen mit einem arkadenförmigen Feld, bietet der Fundstoff aus Serbien die Möglichkeit auch weitere Varianten zu unterscheiden. Da die meisten von ihnen in der Regel einen kleinen Verbreitungsradius aufweisen, ist ihre Entstehung wahrscheinlich auf die Entwicklung einer lokal orientierten Metallurgie mit sehr begrenztem Distributionsraum zurückzuführen. Insgesamt werden sieben Varianten definiert, die mit gro¼en lateinischen Buchstaben (A bis G) gekennzeichnet sind. Die Voraussetzung für eine eigenständige Variante ist, dass es mindestens zwei Funde gibt, die sich durch die Grundform oder die spezifische Verzierung sehr ähneln. Bei einigen Varianten gibt es noch einzelne Beile, die nicht jedes Kriterium erfüllen, dennoch aber der Variante offenbar sehr nahe stehen. Bei jeder Variante erfolgt zunächst die Beschreibung mit den wichtigsten typologischen und morphologischen Merkmalen. Danach werden die bekannten Exemplare aufgelistet. Abschlie¼end folgen die Anmerkungen über die chronologische Einordnung, die Verbreitungstendenzen sowie mögliche Vergleichsfunde. 21 22 23 24 25 26 27 28 29 30 31 33 Rusu 1966, 10–11. Vukmanovi}, Radoj~i} 1995, 87, Sl. 153. Ibid., 168, Sl. 333, 335. Srejovi} 1975a, T. 30, 7. Vasi} 1982, Abb. 3/7. Ibid., 268. Ibid., Abb. 3,6. Kondi} (ur.) 1992, br. 24. Marinkovi} 1991, T. I/5. Vasi} 1982, 268. Rusu 1966, fig. 3. GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) Variante A Erkennungsmerkmale sind eine ringförmige, gerundete Verdickung des Mündungswulstes und drei bis fünf vertikale Rippen innerhalb der arkadenförmigen Fläche. Die Öse setzt mit ihrem oberen Ende am Wulst an. Die Breitseiten setzten sich fächerförmig ab und die Schneide ist leicht gerundet. Die Variante A und die nahe stehenden Funde umfassen Tüllenbeile unterschiedlicher Länge zwischen 6 und 10,6 cm. Die beiden kleinsten Exemplare kommen aus SüdSüdostserbien. 1. Atenica, Gde. ^a~ak, Westserbien. Einzelfund entdeckt 1974 auf dem Gutshof des Vlastimir Mili}, nahe der Brücke über Ateni~ka reka (T. I/1). Länge: 10,6 cm. Narodni Muzej ^a~ak (inv. A61). 2. Mali Izvor, Gde. Zaje~ar, Ostserbien. Gefunden zusammen mit drei weiteren Tüllenbeilen (Varianten C6 und C7 nach Rusu und Typ Verbica, Variante B nach ^ernjih) (T. I/2). Länge: 9 cm. Narodni Muzej Zaje~ar (inv. 158).32 3. Unbekannter Fundort, Serbien. Tüllenbeil aus dem Nationalmuseum in Belgrad ohne Angaben über die Zeit oder den Ort der Auffindung (T. I/3). Am Wulst verbleibende Reste des Materials. Wahrscheinlich ungefertigt oder missglückt. Länge: 8,4 cm. Narodni Muzej Beograd (inv.7075).33 4. Svinji{te, Gde. Pre{evo, Südserbien. Tüllenbeil aus der Siedlung, gefunden am Boden eines Hausobjekts (T. I/4). Länge: 7,7 cm. Narodni Muzej Vranje.34 STARINAR LXIV/2014 Variante A nahe stehend Das Tüllenbeil aus dem Hort Se~anj III besitzt ebenfalls einen verdickten Wulst und vertikale Rippen in der arkadenförmigen Fläche, weist jedoch im Gegensatz zu den übrigen Funden der Variante A noch zwei zusätzliche Horizontalrippen auf. Sie befinden sich dicht beieinander zwischen dem Wulst und des unteren Ösenansatzes. Der Grundform nach entspricht das Tüllenbeil aus Se~anj den rumänischen Funden der Variante C9 nach Rusu, nur dass diese wiederum keine Vertikalrippen in der arkadenförmigen Facettierung haben.38 6. Se~anj, Gde. Se~anj, Banat. Gefunden in einem Hort zusammen mit fünf weiteren Tüllenbeilen und Arm- und Beinschmuck (T. I/5). Länge: 9 cm. Narodni Muzej Zrenjanin (inv. 2102).39 Variante A und die verwandten Exemplare umfassen eine im Hinblick auf die Grö¼e und Datierung sehr heterogene Gruppe. Das Verzierung der Beile mit Vertikalrippen innerhalb des Arkadenfeldes scheint weder räumlich (Karte 1) noch chronologisch besonders charakteristisch zu sein. Der älteste Fund ist das Beil aus Atenica, das aufgrund seiner Grö¼e und zweier durch typische Gusskerne entstanden Seitenlöcher, zweifelsohne der Zeit der Stufe Ha A1 bzw. dem zweiten Depothorizont der serbischen Horte nach Vasi} zuzuschreiben ist. 40 Es ist auch das einzige Beil aller hier präsentierten Varianten mit Seitenlöchern. Der Hort Mali Izvor mit einem kleineren Beil datiert ebenfalls in den zweiten Depothorizont.41 Die Datierung von Mali Izvor stützt sich auf den Vergleich mit bulgarischen Funden ähnlicher Form und Verzierung,42 die jedoch regelhaft keine Öse besitzen. Gut vergleichbar mit den bulgarischen Funden, die vor allem in den Depotfunden des dortigen zweiten Horizonts (Lesura–Verbica nach Hänsel43) erscheinen, ist das kleine Beil aus Pirot. Der Ohne Öse Tüllenbeile ohne Öse mit einem vergleichbaren Wulst und Vertikalrippen in der arkadenförmigen Facettierung kommen vor allem in Bulgarien vor, wo sie als Typ K–1035 bzw. Typ Verbica, Variante E36 umschrieben sind. Im Unterschied zu den meisten Tüllenbeilen mit einem arkadenförmigen Feld, die sich im Raum Nordbulgariens konzentrieren, ist dieser Typ auch im Süden bzw. Sudwesten des Landes verbreitet. Das einzige Beil aus Serbien stammt aus der Gemeinde Pirot an der bulgarischen Grenze. Vergleichbar mit dem Stück aus Pirot, zeigen die südbulgarischen Beile ebenfalls eine geringe Länge zwischen 6 und 8 cm. Die geographische Nähe der südbulgarischen Funde und die vergleichbaren typologischen Merkmale weisen darauf hin, dass es sich um eine spezifische Lokalform handelt. 5. Pirot, Gde. Pirot, Ostserbien. Einzelfund ohne weitere Angaben (T. I/6). Der verdickte Tüllenwulst ist beschädigt. Länge: 6 cm. Muzej Poni{avlja Pirot.37 32 Lalovi} 1975, 147, T. 8/4; Vasi} 1982, Abb. 3/8. Gara{anin 1954, 55 (ohne Abb.); Vukmanovi}-Radoj~i} 1995, 85, Sl. 150. 34 Bulatovi}, Kapuran 2007, T. 8/14. 35 ^ernùh 1978, 187, T. 31/4–7, Sl. 97. 36 Derga~ev 2011, 82, Sl. 42. 37 Bulatovi}, Kapuran 2007, T. 8/14. 38 Rusu 1966, 27. 39 Marinkovi} 1991, T. 1/3. 40 Mayer 1977, 207; König 2004, 99. 41 Vasi} 1982, 286. 42 ^ernùh 1978, 188, T. 31/4–6. 43 Hänsel 1976, 39. 33 34 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Karte 1. 1) Atenica; 2) Mali izvor; 3) Unberkannt; 4) Svinji{te; 5) Pirot; 6) Se~anj III; 7) Gornji Ra~nik; 8) Jagodina; 9) Juhor Karta 1. 1) Atenica; 2) Mali izvor; 3) Unberkant; 4) Sviwi{te; 5) Pirot; 6) Se~aw III; 7) Gorwi Ra~nik; 8) Jagodina; 9) Juhor Variante B Variante B zeichnet sich durch zwei typologische Merkmale aus: zwei Rippen am Tüllenrand mit einer zwischen diesen Verstärkungen platzierten Öse und Vertikalrippen innerhalb der arkadenförmigen Facettierung. Die obere Horizontalrippe stellt gleichzeitig die Verstärkung des Tüllenwulstes dar. Auch wenn die vertikalen Rippen am Beilkörper eine Ähnlichkeit mit der Variante A suggerieren, haben die Beile der Variante B nicht nur einen andersartigen Tüllenbereich, sondern auch einen unterschiedlichen Umriss mit nahezu gerade Fund aus der Siedlung Svinji{te wurde von den Autoren in die Zeitspanne Ha A2 – Ha B1 datiert.44 Das gleiche gilt auch für das sehr ähnliche Beil aus dem Nationalmuseum in Belgrad. In die Zeit Ha A2 – Ha B1 bzw. in den dritten Horizont der Depotfunde gehört auch Se~anj III.45 Das der Variante A nahe stehende Beil aus diesem Hort entspricht im Gro¼em und Ganzen (bis auf vertikale Rippen) der Variante C9 nach Russu, die in Rumänien hauptsächlich in den Horten der Stufe Miograd (Ha B1) vorkommt.46 Es kann demnach konstatiert werden, dass die Beile der Variante A in eine lange Zeitspanne zwischen Ha A1 und Ha B1 datieren. Die kleineren, gedrungenen Exemplare der Übergangszeit Ha A2 – Ha B1 scheinen sich im Raum Ost- und Südserbiens zu konzentrieren, obwohl die Zahl der bekannten Stücke eigentlich zu gering ist um die Rückschlüsse über die spezifische Verbreitung zu ziehen. 44 Bulatovi}, Kapuran 2007, 8. Vasi} 1982, 268. 46 Rusu 1966, 27, vergleiche Petrescu–Dîmboviþa 1978,Taf. 225 B/4; 248 B/1; 253/15. 45 35 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Sîmbãta Nouã I in Dobrudscha, der schon in die Depotstufe Miograd (Ha B1) datiert.60 Ausgehend von den Analogien lässt sich annehmen, dass die Beile der Variante B spätestens seit der Zeit des zweiten Hortfundhorizonts (Ha A1) erscheinen. Die Vergleichsfunde aus Rumänien weisen allerdings darauf hin, dass die Hauptzeit dieser Beile in die Zeitspanne Ha A2 – Ha B1 bzw. ins 12. und 11. Jh. v. Chr. fällt.61 Im Hinblick auf die Verbreitung ist die relative Nähe der drei bislang gefundenen Exemplare nicht zu übersehen. Alle drei Beiel kommen aus einem 30 km gro¼en Radius zwischen den Flüssen Velika und Zapadna Morava (Karte 1). Es handelt sich wahrscheinlich um eine lokale Form, die zwar auf Anregung der bulgarischen/ siebenbürgischen Produktionsstätten entstanden ist, aber dem örtlichen Trend mit einer Seitenöse angepasst wurde. abfallenden Seiten und einer nur mä¼ig ausbreitenden Schneide. Hinzu kommt noch, dass die Beile der Variante A nicht über 10 cm Länge hinausgehen, während die der Variante B in der Regel 10 cm oder länger sind. 1. Gornji Ra~nik, Gde. Jagodina, Zentralserbien. Einzelfund aus dem Jahr 1966 (Finder Mom~ilo Kosti}) (T. I/7). Länge: 10 cm. Zavi~ajni Muzej Jagodina (inv. 412).47 2. Jagodina, Gde. Jagodina, Zentralserbien. Einzelfund ohne nähere Angaben (T. I/8). Die Schneide weist einen Bruch auf. Zwischen den zwei Rippenverstärkungen am Tüllenrand befindet sich eine eingeritzte Zickzack Linie. Über der Öse sind noch Materialreste, entstanden wahrscheinlich beim Gie¼en. Länge: 12,2 cm. Zavi~ajni Muzej Jagodina (ohne inv.) 3. Juhor, Gde. Para}in, Zentralserbien. Einzelfund, entdeckt auf dem Berg Juhor ohne weitere Beschreibung (T. I/9). Länge: 11,5 cm. Narodni Muzej Beograd.48 (ohne inv.) Bei alle drei Beilen der Variante B handelt es sich um Einzelfunde ohne überlieferten Fundkontext. Die Grundkonturen der Variante B mit geraden Seiten und einer leicht ausbreitenden Schneide entsprechen gut den Tüllenbeilen aus den nordbulgarischen Horten Verbica I/II und Stra`ica, die in der Derga~evs Studie als Typ Verbica zusammengefasst worden sind.49 Einige der bulgarischen Funde zeigen auch eine ähnliche Verstärkung mit einer zweiten Rippe unter dem Tüllenwulst, wobei der Bereich dazwischen mit kurzen schrägen oder vertikalen Ornamenten in Manier der „falschen Tordierung“ verziert ist. Die Verwandtschaft der Variante B und der bulgarischen Beile unterstreicht auch ein Exemplar aus dem Hort Verbica I mit einer identischen Zickzackzier50 wie beim Fund aus Jagodina. Die Horte Verbica I/II und Stra`ica datieren in den zweiten Hortfundhorizont nach Hänsel51 bzw. in die Zeit der Horizonte Kisapáti–Lengyeltóti im Karpatenbecken52, Suseni im Siebenbürgen53, der Phasen II in Kroatien54 und Bosnien55 und schlie¼lich des zweiten Horizontes der serbischen Horte nach der Chronologie Vasi}s.56 In Bulgarien dauern die Beile des Typs Verbica offenbar bis in die Zeit des dritten Horizonts bzw. bis in die Zeit Ha A2 – Ha B1 an.57 Im Unterschied zur Variante B sind die bulgarischen Funde jedoch nicht mit einer seitlichen Öse ausgestattet. Vergleichbare, Exemplare mit Öse sind noch in den siebenbürgischen Depots Guºteriþa II (Horizont Susueni)58 und Predeal II aus dem darauf folgenden Horizont Jupalnic (Ha A2)59 bekannt. Ein ähnliches Beil findet sich auch im Hort Variante C Einen identischen Tüllenbereich mit zwei Rippen und einer dazwischen verbindenden Öse wie bei der Variante B, besitzen auch alle Vertreter der Tüllenbeilvariante C. Der einzige wesentliche Unterscheid ist das Fehlen der Vertikalrippen und jeglicher zusätzlichen Verzierungen. Die Verstärkung am Tüllenmund ist dabei etwas kräftiger als die zweite, darunter liegende Horizontalrippe. Die Seiten sind gerade und breiten sich bei der Schneide etwas aus. Die Schneide selbst ist leicht bogenförmig. Alle bisher bekannten Exemplare der Variante C haben eine vergleichbare Länge zwischen 9,5 und 10,5 cm. 1. Dvori{te, Gde. ]uprija, Zentralserbien. Tüllenbeil aus einem zerstörten Depotfund, gefunden zusammen mit einer Sichel (T. II/1). Weitere Angaben fehlen. Länge: 9,6 cm. Narodni Muzej Beograd (inv. 22821).62 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 36 Stoji} 1986, 55, T. 21/1. Idem 2003, 26/3. Derga~ev 2011, 83–84, Variante E und besonders F. Ibid., Sl. 43/4. Hänsel 1976, 38–39. v. Brunn 1968, 38. Petrescu–Dîmboviþa 1978, 88. Vinski-Gasparini 1973, 125. König 2004, Tabelle 2. Vasi} 1982, 268. Derga~ev 2011, Sl. 56. Petrescu–Dîmboviþa 1978, Taf. 104/34–35. Ibid., Taf. 221A/2. Ibid., Taf. 252/13. König 2004, 23. GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Variante D Mit einem verstärkten Tüllenwulst, einer zusätzlichen horizontalen Rippe sowie einer seitlichen Öse, die beide Verstärkungen miteinander verbindet, steht die Variante D den Varianten B und C nahe. Ein Charakteristikum der Variante D ist die Verzierung aus Vertikalrippen zwischen der horizontalen Rippe an der Tülle und dem arkadenförmigen Feld. Trotz der Tatsache, dass aus Serbien bislang nur zwei Exemplare bekannt sind, berechtigt die signifikante Kombination aus Verzierung und Tüllenmundverstärkung mit einer Öse die Umschreibung einer eigenständigen Variante. Eine ähnliche Zier ist auch bei einigen Tüllenbeilen aus Bulgarien festzustellen, die ^ernjih als Typ K–48 bezeichnete.71 Im Unterschied zur Variante D besitzen die bulgarischen Beile jedoch keine Horizontalrippe unter dem verstärkten Tüllenmund. 1. Bunar, Belica, Gde. Jagodina, Zentralserbien. Einzelfund aus dem Jahr 1977 (T. II/5). Länge: 7,4 cm. Zavi~ajni Muzej Jagodina (inv. 1095).72 2. Trnjane, Gde Po`arevac, Nordostserbien. Einzelfund ohne weitere Angaben (T. II/6). Länge: 8,8 cm. Narodni Muzej Po`arevac (inv. 407). Das Tüllenbeil aus Bunar ist aufgrund der spezifischen Tüllenprofilierung und der Lage des Fundortes wahrscheinlich in die gleiche Gruppe wie die Varianten B und C einzuordnen und demnach in die Zeitspanne Ha A2 bis Ha B1 zu datieren. Ähnlichkeiten mit den bulgarischen Funden sind zwar offensichtlich, helfen jedoch bei der chronologischen Einordnung nicht sonderlich viel, da alle vier Exemplare des dortigen Typs K–48 ohne weiteren Kontext und in zwei Fällen ohne Fundortangabe überliefert worden sind.73 Interessant ist in diesem Zusammenhang der Verweis von ^ernjih auf die Gussformhälfte aus Schicht VII B 2. Donje [tiplje, Gde. Jagodina, Zentralserbien. Einzelfund, erworben im Jahr 1972 von Mido Vasi} (T. II/2). Länge: 9,5 cm. Zavi~ajni Muzej Jagodina (inv. 829).63 3. Kopaonik, Zentralserbien. Einzelfund aus dem Jahr 1912, ohne weitere Angaben (T. II/3). Am Tüllenmund zusätzlich zu horizontalen Verstärkung befinden sich noch zwei vertikale Zipfel, einmal über Öse und einmal auf der gegenüberliegenden Mundseite. Länge: 10,5 cm. Narodni Muzej Beograd (inv. 1685).64 4. Sekuri}, Gde. Rekovac, Zentralserbien. Einzelfund von der Fundstelle „Ko{nica–Bakovo“ (T. II/4). Länge: 10 cm. Narodni Muzej Kragujevac (inv. 3175).65 Bei der Veröffentlichung des Fundes aus Donje [tiplje wies Stoji} die vergleichbaren Beile der Zeit zwischen Ha A und Ha B zu, führte jedoch nur Vergleichsfunde für die Exemplare ohne zweite horizontale Rippenverstärkung unter dem Mündungswulst auf.66. In dem Metallfundkatalog des Belgrader Museums wurde der Fund aus Dvori{te mit einem Beil aus dem Depot Mesi}–[upaja in Banat verglichen, das jedoch ebenfalls keine zweite Rippe unter der Mündungswulst aufweist. In der gleichen Publikation wurde das Beil aus Dvori{te in die Zeit der Stufe Ha A2 datiert.67 Selbst im umfangreichen Fundkorpus der Tüllenbeile aus Siebenbürgen lassen sich kaum brauchbare Analogien für die hier definierte Variante C mit spezifischer Kombination der arkadenförmigen Facettierung und einer zwischen zwei Verstärkungen an der Tüllen angebrachten Öse ausmachen. Der beste Vergleich liegt in dem slowakischen Depot Levice vor, der in den Horizont Kisapáti– Lengyeltóti (Ha A1) datiert.68 Zu berücksichtigen ist auch die gewisse Ähnlichkeit der serbischen Variante C mit den siebenbürgischen Varianten C7 und C8 nach Rusu, die vor allem in den Depotfunden der Zeitstufen Ha A2 und Ha B1 belegt worden sind.69 Vergleichbar mit der Variante B, die einen identischen Tüllenrandaufbau besitzt, ist die Variante C ebenfalls ausschlie¼lich in dem Raum westlich von Velika Morava verbreitet (Karte 2). Eine Konzentration ist vor allem in der Region zwischen Zapadna und Velika Morava festzustellen. Offensichtlich ist die spezifische Tüllenrandverstärkung mit zwei Rippen und einer verbindenden Öse für diese Region Zentralserbiens charakteristisch. Darauf weisen auch Funde mit einer andersartigen Verzierung hin, wie z. B. ein Tüllenbeil mit trapezförmigen Facettierung aus Medojevac bei Jagodina70 sowie ein Exemplar der hier umschriebenen Variante D aus der gleichen Gemeinde. 62 Trbuhovi} 1960, 183, Sl. 5; Vukmanovi}, Radoj~i} 1995, 88, br. 155. 63 Stoji} 1986, 55, T. 21/2. 64 D. Gara{anin 1954, 54; T. 36/2. 65 Kondi} (ur.) 1992, 40, 51/25. 66 Stoji} 1986, 56. 67 Vukmanovi}, Radoj~i} 1995, 88. 68 Novotná 1978, Taf. 42/784. 69 Rusu 1966, 11. 70 Stoji} 1986, T. 21/3. 71 ^ernùh 1978, 200, T. 39/4–7. 72 Stoji} 1986, 55; T. 21/5. 73 ^ernùh 1978, 200. 37 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Karte 2. 1) ]uprija; 2) Donje [tiplje; 3) Kopaonik; 4) Sekuri}; 5) Belica; 6) Trnjane; 7) Hisar; 8) Stari Kostolac; 9) Vojilovo; 10) Se~anj Karta 2. 1) ]uprija; 2) Dowe [tipqe; 3) Kopaonik; 4) Sekuri}; 5) Belica; 6) Trwane; 7) Hisar; 8) Stari Kostolac; 9) Vojilovo; 10) Se~aw 2 in Troja.74 Es handelt sich um eine Form für ein Tüllenbeil mit Öse, einem verstärkten Mündungswulst und einem arkadenförmigen Feld. Zwischen dem Rand und der Facettierung befinden sich unregelmä¼ige Zickzacklinien. Aus der gleichen Region stammt wohl noch eine Gussform für ein sehr ähnliches Beil.75 Falls die Funde aus Kleinasien tatsächlich in einem Zusammenhang mit den balkanischen Tüllenbeilen stehen (K–48 in Bulgarien und Variante D in Serbien), so würde keine chronologische Diskrepanz dazwischen liegen, da die erwähnte Schicht in Troja ins 11. Jh. v. Chr. datiert. Hinzuweisen ist noch auf die vergleichbar verzierte Tüllenbeile aus den siebenbürgischen Depots ªpãlnaca II76 i Aiud77 (Horizont Suseni) sowie auf ein Beil unbekannten Fundortes aus dem Museum in Bratislava.78 Die genannten Beispiele haben jedoch keine zusätzliche Verstärkung des Tüllenmundes, wie dies bei der Variante D der Fall ist. Die geringe Zahl der bekannten Exemplare ist nicht ausreichend um generelle Verbreitungstendenzen der Variante D erkennen zu können (Karte 2). Auch hier handelt es sich vermutlich um eine lokale Interpretation und Anpassung einer offenbar nicht häufigen, aber geographisch weit verbreiteten Zierweise. Variante E Umfasst mit der Variante E sind die Tüllenbeile mit einem verstärkten Tüllenmund und zwei horizontalen Rippen gleich darunter. Der einzige Unterschied zur 74 75 76 77 78 38 ^ernùh 1978, 200; Wanzek 1989, Taf. 3/3. Wanzek 1989, Taf. 3/7. Petrescu–Dîmboviþa 1978, Taf. 144/136. Petrescu–Dîmboviþa 1977, Pl. 101/4. Novotná 1978, Taf. 43/788. GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Vergleichsfunde aus Rumänien stammen vor allem aus den Depotfunden der Stufe Miograd (Ha B1).86 Vereinzelt lassen sie sich noch in dem darauf folgenden Horizont Fizeºu Gherlii nachweisen.87 Eine ähnliche chronologische Stellung nehmen die vergleichbaren Beile auch in der von Metzner-Nebelsick dargelegten Entwicklungsschema ein.88 Die Tüllenbeile der Variante E können demnach in Ha A2 – Ha B1 datiert werden, mit der Tendenz zum jüngeren Abschnitt. Die Hauptunterschiede zu den siebenbürgischen Varianten gleicher Zeitstellung (C9 und C10) sind eine grö¼ere Rippenzahl und andersartige Profilierung des Tüllenwulstes. Im Hinblick auf die Gesamtverbreitung der Tüllenbeile mit mehreren Rippen unter dem Rand und arkadenförmigen Feld stellen die Funde aus Serbien den südwestlichen Randbereich dar. Die bisherigen Funde aus Serbien konzentrieren sich im östlichen Teil des Landes bzw. östlich der Linie Thei¼ – Donau – Velika Morava (Karte 2). Tüllenprofilierung der Varianten B, C und D ist also eine zusätzliche Horizontalrippe. Der obere Ansatz der Öse schlie¼t sich an den verstärkten Wulst an, während das untere Ende bei der letzten Rippe angebracht ist. Gemeinsame Merkmale der Variante E sind relativ geringe Ma¼e zwischen 8 und 9,5 cm sowie leicht bogenförmig abgesetzte Breitseiten. 1. Hisar, Gde. Leskovac, Südserbien. Tüllenbeil aus der mehrschichtigen Siedlung Hisar, gefunden im Sektor III (T. II/7). Länge: 9,4 cm. Narodni Muzej Leskovac.79 2. Stari Kostolac, Gde. Kostolac, Nordostserbien – Brani~evo. Einzelfund, aufgelesen bei der Fundstelle ^air (T. II/8). Länge: 9,5 cm. Narodni Muzej Po`arevac (inv. 860). 3. Vojilovo, Gde. Golubac, Nordostserbien – Brani~evo. Tüllenbeil aus dem Depot, entdeckt im Jahr 1952 (T. II/9). Unter den sonstigen Funden befanden sich u. a. noch ein Tüllenbeil der siebenbürgischen Variante C6 nach Rusu sowie ein trapezförmiges, längliches Tüllenbeil. Das Tüllenbeil der Variante E war nahezu unversehrt. Länge: 8,5 cm. Narodni Muzej Po`arevac (inv. 59).80 4. Unbekannte Fundstelle, Serbien. Einzelfund ohne weitere Angaben (T. II/10). Ohne merkliche Beschädigungen. Länge: 8 cm. Muzej Grada Beograda (inv. 4164). 5. Unbekannte Fundstelle, Serbien. Sammlung „Dunji}“. Ohne Angaben über den genauen Fundort (T. II/11). Narodni Muzej Kraljevo (A1268). Variante F Bezeichnend für die Variante F sind zwei bis drei horizontale Rippen unter dem verstärkten Tüllenmund und eine Vertikalrippe, die vom Scheitelpunkt der arkadenförmigen Fläche nach unten fällt. Zwei Exemplare weisen noch eine zusätzliche Seitenverzierung mit lappenförmig angeordneten Rippen auf. Die Öse befindet sich zwischen dem verdickten Tüllenmund und der untersten Horizontalrippe. Trotz der kleineren Abweichungen in der Länge (zwischen 8 und 10,8 cm), zeigen alle vier Beile der Variante F eine ähnliche Form mit gerade oder leicht bogenförmig abfallenden Seiten und einer fächerförmig verbreiteten Schneide. 1. Banatski Karlovci, Gde. Vr{ac, Banat. Tüllenbeil aus dem im Jahr 1896 entdeckten Depotfund (T. III/1). Im Depot wurden noch 16 andere Tüllenbeile, Variante E nahe stehend 6. Se~anj, Gde. Zrenjanin, Banat. Tüllenbeil aus dem im Jahr 1957 entdeckten Depot in der Ziegelfabrik „Radovanov“. Teile des Depots sind zerstört bzw. verschollen. Unter den anderen Funden konnte noch ein Tüllenbeil mit Y – förmigen Rippen identifiziert werden. Das Tüllenbeil der Variante E ist stark abgenutzt mit kleineren Bruchstellen an der Schneide (T. II/12). Länge: 10 cm. Narodni Muzej Zrenjanin (inv. 4559).81 Nach der chronologischen Einteilung Vasi}s datieren die Depots Vojilovo und Se~anj in den dritten Horizont der serbischen Hortfunde (Ha A2 – Ha B1).82 In Se~anj fand sich u. a. auch ein Tüllenbeil mit Yförmigen Rippen und abgesetztem Schneidenteil.83 Diese zwei Merkmale charakterisieren die Tüllenbeile der Zeit Ha A2 – Ha B1 in allen angrenzenden Gebieten.84 In die gleiche Zeit datiert auch die Siedlungsetappe in Hisar bei Leskovac mit einem weiteren Exemplare der Variante E.85 Zahlreiche 79 Stoji} 2011. Todorovi} 1956/1957, Sl. 1A; Idem 1975a, 75, T. 70/1. 81 Radi{i} 1958, T. 1/4. 82 Vasi} 1982, 286. 83 Radi{i} 1958, T. 1/5. 84 v. Brunn 1968, 58; Vinski-Gasparini 1973, 137; König 2004, 99; Petrescu–Dîmboviþa 1978, 92. 85 Stoji} 2011, 9. 86 Petrescu–Dîmboviþa 1978, Taf. 225 C/3; 236 B/8; 248 B/1; 253/15. 87 Ibid., Taf. 259 C/8; 260 B/2–3. 88 Metzner-Nebelsick 2002, 62–63, Abb. 14. 80 39 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) Fu¼- und Armringe, Lanzenspitzen und Helmteile gefunden. Ein Teil des Hortfundes, einschlie¼lich das Tüllenbeil der Variante F, befindet sich im Musuem in Timiºoara, während der Rest im Musuem in Vr{ac aufbewahrt wird. Länge: 10,8 cm. Museul regional al Banatului – Timiºoara (1488).89 2. Futog, Gde. Novi Sad, Ba~ka. Tüllenbeil aus einem Depot, entdeckt im Jahr 1913. Im Hort wurden insgesamt 193 Bronzegegenstände verzeichnet, darunter auch ein Tüllenbeil vom Typ Okniþa sowie mehrere Exemplare der siebenbürgischen Varianten A und B nach Rusu. Das Beil der hier umschriebenen Variante F besitzt zwei bogenförmige Seitenrippen, die zusammen einen Lappendekor ergeben (T. III/2). Länge: 10,9 cm. Muzej Vojvodine Novi Sad (inv. 3303).90 3. Klenje, Gde. Golubac, Nordostserbien – Brani~evo. Das Tüllenbeil stammt aus einem 1982 entdeckten Depot mit 129 Objekten, darunter auch ein weiteres Beil der gleichen Variante sowie ähnliche Exemplare mit einer trapezförmigen Facettierung auf dem Beilkörper. Das Beil aus Klenje ist, wie das Stück aus Futog, ebenfalls mit seitlichen Bogenrippen in Form eines Lappens versehen (T. III/3). Länge: 9,8 cm. Narodni Muzej Po`arevac (inv. 1008).91 4. Klenje, Gde. Golubac, Nordostserbien – Brani~evo. Aus dem Hort mit 129 Bronzeobjekten (s. o). Das Tüllenbeil ist etwas kleiner als der andere Vertreter der Variante F und besitzt keine bogenförmige Seitenrippen (T. III/4). Länge: 8 cm. Narodni Muzej Po`arevac (inv. 1007).92 Beile der Variante F stammen aus den Depotfunden, die bei ihrer Veröffentlichung in die Zeit Bz D – Ha A1 datiert wurden.93 Abweichend davon haben R. Ra{ajski und M. Gara{nin die Funde aus Banatski Karlovci und Futog etwas jünger eingestuft und in die Zeit der Stufe Ha A2 datiert.94 Bei der Auflistung der Vergleichfunde für Tüllenbeile aus Klenj und Futog wurden einige Funde aus Nordostungarn und Siebenbürgen angeführt, die entweder gar kein arkadenförmiges Feld oder einen nicht profilierten Tüllenrandbereich aufweisen.95 Viel näher der Variante F stehen die Beile aus siebenbürgischen Depots Taut96, „Criºana I“97 i Bãdeni98, die eine Zeit zwischen Ha A1 (Taut ) und Ha A2 – Ha B1 (Criºana I“ i Bãdeni) umspannen. Folglich ist anzunehmen, dass die Tüllenbeile der Variante F spätestens in der Stufe Ha A1 auftauchen (Funde aus Futog und Banatski Karlovci) und offensichtlich mindestens bis Ha A2 fortdauern. In diesem Zusammenhang ist noch ein Beil mit sehr ähnlicher STARINAR LXIV/2014 Rippenzier im arkadenförmigen Feld aus dem ostserbischen Depot Leskovo zu erwähnen. Dieser Fund datiert in den dritten Hortfundhorizont Serbiens (Ha A2 – Ha B1).99 Angesichts der verhältnismä¼ig geringen Zahl der Vergleichsfunde aus Siebenbürgen stellt sich die Frage, ob diese Form tatsächlich aus diesem Raum stammt, wie von den meisten Autoren angenommen. Die gleiche Anzahl der rumänischen und serbischen Funde deutet eher darauf hin, dass es sich um eine regionale Variante handelt, die zwischen Donau im Westen, Siebenbürgen im Osten und Brani~evo im Süden verbreitet ist (Karte 3). Variante G Das wesentliche Merkmal der Variante G sind mehrfache arkadenförmige Rippen um den Schneidenteil. Der Tüllenrand mit der vertikalen Öse ist durch Wulstverstärkungen und Rippen profiliert. Über dem oberen Ösenende und auf der gegenüberliegenden Tüllenmundseite lassen sich kleine, zipfelförmige Verstärkungen erkennen. Das Beil aus @agubica ist zusätzlich mit einem tropfenförmigen Ornament innerhalb der des arkadenförmigen Feldes verziert. Beide bislang bekannte Exemplare der Variante G haben nahezu identische Konturen mit einem profilierten Tüllenrand, gerade abfallenden Seiten im Mittelteil und einer fächerförmigen Schneide. Die Tatsache, dass die beiden Beile nicht nur eine sehr ähnliche Kontur, sondern auch fast identische Länge aufweisen, spricht für die Umschreibung einer eigenständigen Variante. 1. Banatski Karlovci, Gde. Vr{ac, Banat. Tüllenbeil aus dem 1896 entdeckten Hort (siehe oben) (T. III/5). Länge 10,8 cm. Narodni Muzej Vr{ac (inv. 926).100 2. @agubica, Gde. @agubica, Nordostserbien – Brani~evo. Angeblich stammt das Tüllenbeil aus einem 89 90 91 92 93 94 Ra{ajski 1975a, T. 72/5 Bori} 1997, T. 4/40. Jacanovi} 1986, T. I/7. Ibid. 1986, T. I/6. Vasi} 1982, 268; Jacanovi} 1986, 171; Bori} 1997, 71. M. Gara{anin 1973, 431; Ibid. 1983, 685; Ra{ajski 1975a, 90. 95 Jacanovi} 1986, 168; Bori} 1997, 64. Vergleiche Petrescu– Dîmboviþa 1977, Pl. 101/1–15; 148/9; 193/10. 96 Petrescu–Dîmboviþa 1978, Taf. 159A/1. 97 Petrescu–Dîmboviþa 1977, Pl. 288/6. 98 Ibid., Pl. 299/12. 99 Todorovi} 1975, T. 71/3; Vasi} 1982, 268. 100 Holste 1951, T. 17/15; Ra{ajski 1975a, T. 76/4. 40 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Karte 3. 1) Banatski Karlovci; 2) Futog; 3–4) Klenje; 5) Banatski Karlovci; 6) @agubica Karta 3. 1) Banatski Karlovci; 2) Futog; 3–4) Klewe; 5) Banatski Karlovci; 6) @agubica Hortfund, der in Suvi Do-Ponikve zu Tage kam. Die übrigen Objekte aus dem Depot sind nicht auffindbar. Die Öse des Beiles ist abgebrochen und die Schneide ist ebenfalls beschädigt (T. III/6). Länge: 11,1 cm. Narodni Muzej Po`arevac (inv. 856). Bei der Vorlage des Depots aus Banatski Karlovci datierte Ra{ajski diesen Fund in die Stufe Ha A2. Dabei stützte er sich auf die Datierung der Tüllenbeile der siebenbürgischen Varianten B und C nach Rusu. Das Beil der hier umschriebenen Variante G (T. III/5) wurde bei Ra{ajski als eine Sonderform der siebenbürgischen Variante C bezeichnet. R. Vasic hingegen datiert den Fund aus Banatski Karlovci in den ersten Horizont der serbischen Hortfunde (Bz D).101 Ausschlaggebend war dabei ein gro¼er, für die entwickelte Mittelbronzezeit typischer, Armring mit Spirallenden. Gute Parallelen im Hinblick auf die Ornamentierung der Beile mit mehrfachen arkadenförmigen Rippen lassen sich bei dem nordbulgarischen Typ Pelovo (mit und ohne Öse) finden. Typ Pelovo erscheint hauptsächlich in den Horten des zweiten Horizonts (Ha A1) mit vereinzelten jüngeren Funden (Ha A2 – Ha B1).102 Es ist allerdings zu beachten, dass sich die Beile der Variante F hinsichtlich der Grundform und Länge deutlich unterscheiden. Typ Pelovo umfasst nämlich fast nur kleinere Beile mit einer Länge zwischen 7 und 9 cm. Eine ähnliche Zier wie der Variante F ist auch auf einem nur zur Hälfte erhaltenen Tüllenbeil aus dem Depotfund Guºteriþa II (Horizont Susueni – Ha A1) in Siebenbürgen zu verzeichnen.103 101 102 103 41 Vasi} 1982, 274. Derga~ev 2011, 86, Sl 45. Petrescu–Dîmboviþa 1978, Taf. 105/58. GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Ausbreitung der Gruppe Belegi{ II – Gava im Raum Zentralbalkans Im südlichen und östlichen Banat wurden bislang 23 Fundstellen (Nekropolen und Siedlungsplätze) der Belegi{ II – Gava Gruppe sowie 22 Hortfunde mit über 1000 Bronzeobjekten aus der gleichen Periode (Ha A1) registriert.113 In seiner Studie über den Gava-Komplex hat Lj. Bukvic für den gesamten Banat 67 Fundstellen und 34 Depotfunde aufgelistet.114 Im benachbarten syrmischen Raum ist die Zahl nicht wesentlich geringer. Insgesamt wurden dort bisher 51 Fundplätze der Belegi{ II – Gava Gruppe festgestellt. Das typische Fundmaterial (kannelierte Keramik) fand sich noch in 13 weiteren Siedlungen der sog. Bosut-Gruppe in Syrmien.115 Die Fundstellen konzentrieren sich meistens entlang der grö¼eren Flussläufe und im Gebiet von Fru{ka Gora. Im Norden Vojvodinas (Ba~ka) scheint vor allem der Einzugsgebiet von Tami{ zum bevorzugten Siedlungsraum zu gehören.116 Während man bei den Nekropolen im Hinblick auf die Topographie bislang keinen Unterschied feststellen kann, scheinen sich die Siedlungen der in der Forschung zusammengelegten Gruppen Gava und Belegi{ doch voneinander abzusetzen.117 Eine Ausnahme bilden lediglich die gro¼en Tellsiedlungen wie Gomolava und Feudvar. Insgesamt kann man für den Raum Vojvodinas während der Belegi{ II – Gava Phase an der Donau eine Grenze erkennen mit stärker ausgeprägten mittelbronzezeitlichen Traditionen in den Siedlungen südlich des Flusses, während im Norden die älteren Elemente offenbar nicht überliefert wurden.118 Unter den sonstigen Tüllenbeilen aus Banatski Karlovci lassen sich tatsächlich einige Formen erkennen, die vor allem für den ersten Hortfundhorizont (Bz D) in Siebenbürgen charakteristisch sind,104 aber auch solche Exemplare, die den zweiten Horizont bzw. die Zeit der Stufe Ha A1 repräsentieren.105 Folgt man der Regel, dass geschlossene Funde immer nach dem jüngsten Objekt zu datieren sind, ist die Variante F chronologisch in die Zeit der Stufe Ha A1 einzuordnen. Beide Funde aus Serbien stammen aus dem östlichen Teil des Landes, einmal nördlich und einmal südlich der Donau (Karte 3). Sie lassen sich demnach vermutlich mit dem ostbalkanischen oder siebenbürgischen Tüllenbeilkreis in Verbindung bringen, worauf auch die meisten Vergleichsfunde hinweisen. Übergangsperiode zwischen der Bronzeund Eisenzeit im serbischen Donauraum und im Moravatal (Pomoravlje) Die Übergangsprozesse zwischen der Bronze- und Eisenzeit im Zentralbalkan lassen sich nur schwer durch das Prisma der mitteleuropäischen Entwicklung erklären. Entscheidend bei dieser Transformation sind vielfältige lokale, in der bronzezeitlichen Tradition verwurzelte, Gruppen106 sowie das im Verhältnis zu Mitteleuropa früher aufkommendes Eisen.107 Die Stufen der älteren Urnenfelderzeit mit vielen Depotfunden wurden jedoch auch in Serbien, vor allem in den Arbeiten R. Vasi}s, als Spätbronzezeit aufgefasst.108 Die ersten urnenfelderzeitlichen Funde im serbischen Donauraum sind die Gräberfelder und Siedlungen der Gruppe Belegi{ in der Vojvodina und das Auftreten des Gava-Komplexes in der Zone um das Eiserne Tor (Karte 4).109 Nach N. Tasi} kann der Verbreitungsweg des Gava-Komplexes bis in den serbischen Donauraum nicht genau skizziert werden, da ähnliche Funde sowohl im Norden als auch im Süden bzw. im Raum der Flüsse Morava i Ni{ava festzustellen sind.110 In seiner Studie über die Schalen mit einziehendem und tordiertem Rand zeigte jedoch A. Bulatovi}, dass diese typische Gefä¼form sich von der Slowakei über den Karpatenbecken bis ins Moravatal und weiter südlich bis Makedonien in unterschiedlichen Etappen verbreitet hat.111 M. Gara{anin bezeichnete wiederum die gesamte urnenfelderzeitliche Periode provisorisch als „Vojvodinische Gruppe der Übergangsperiode“, da man bis dahin weder verlässliche stratigraphische Kontexte noch ausgedehnte und aufgearbeitete Gräberfelder kannte.112 104 Ra{ajski 1975a, T. 77/4.6; vergleiche Rusu 1966, Varijante A5 i B3. 105 Ra{ajski 1975a, T. 77/7–8; vergleiche Rusu 1966, Varijante B4 i B5. 106 107 108 109 110 111 112 113 114 115 116 117 118 42 Gara{anin 1973, 403. Vasi} 1982a, 1. Vasi} 1982a, 1. Tasi} 1983, 106. Ibid., 104. Bulatovi} 2010. Gara{anin 1973, 409. Uzelac 1996, 35–36, Map 7. Bukvi} 2000, 102–103, Karta 3/4. Popovi} 1994a, Karta 1/2. Popovi} 1994a, Karta 1/2. Bukvi} 2000, 96. Medovi} 1994, 46. GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Karte 4. Spätbronzezeitliche Gruppen in Serbien und Zentren der Erzverarbeitung Karta 4. Grupe kasnog bronzanog doba u Srbiji i metalur{ki centri u praistoriji Im Raum um das Eiserne Tor sind auf den beiden Seiten der Donau mehrere Siedlungen und Nekropolen mit charakteristischer Keramik der Belegi{ II – Gava Gruppe registriert. Aufgrund einiger spezifischen Merkmale der Keramik und Metallbeigaben aus den Brandbestattungen, haben M. Jevti} und M. Vukmanovi} diese Fundstellen als Gruppe Mala Vrbica–Hinova–Balta Verde zusammengefasst und ins 11. und frühe 10 Jh. v. Chr. datiert.119 Die wichtigsten Fundplätze auf der serbischen Donauseite sind Mala Vrbica–Livade120, Vajuga –Korobovo121 und Vajuga–Pesak122. Im serbischen Hinterland des Eisernen Tores können diesem Zeitraum noch einige Bronzefunde (Nadeln, Messer) aus Zlotska Pe}ina zugeschrieben werden. Die noch ausstehende Aufarbeitung und Präsentation des umfangreichen Keramikmaterials aus diesem Höhlenfundplatz wird mit Sicherheit ein besseres und vollständigeres Bild der Zeit zwischen Ha A und Ha B für die Region Ostserbiens darbieten. In diesem Zusammenhang sind noch vereinzelte Fragmente der schwarzen kannelierten Keramik aus Ku~ajno bei Bor zu erwähnen.123 Weiter nach Süden lässt sich die kannelierte Keramik in erster Linie entlang der Flusses Velika Morava verfolgen, der in allen Perioden die Haupt- 119 120 121 122 123 43 Jevti}, Vukmanovi} 1996, Map 3. Vukmanovi}, Popovi} 1986. Krsti} 1986. Popovi}, Vukmanovi}, Radoj~i} 1986; Radoj~i} 1986. Jevti} 2004, 141. GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Höhensiedlung Hisar in Leskovac mit Siedlungsphasen der späten Bronze- und frühen Eisenzeit. Bezeichnend für die jüngere, früheisenzeitliche Phase sind die Funde der P{eni~evo-Gruppe. Aus Hisar stammen auch einige Bronzefunde, einschlie¼lich ein Tüllenbeil der Variante E (T. II/7) sowie eine Gussform für ein weiteres Tüllenbeil.135 In der Region um die Stadt Vranje, im äu¼ersten Süden Serbiens, hat A. Bulatovi} deutliche Unterschiede in der Topographie der Plätze mit kannelierter Keramik und der alten, bronzezeitlichen Siedlungen feststellen können.136 Wichtig für diese Mikroregion ist die befestigte Höhensiedlung Svinji{te in der u. a. zusammen mit der kannelierter Ware ein Tüllenbeil der Variante A gefunden wurde.137 Mit Hisar und Svinji{te liegen demnach für Raum Südserbiens zwei Tüllenbeile aus einem Siedlungskontext vor. Über die Übergangsperiode in Kosovo ist wenig bekannt. Repräsentativ sind nach wie vor die prominenten Nekropolen Donja Brnjica138, Gra{tica und Ulpijani139 sowie die Siedlungen Lju{ta i Gladnice. kommunikation durch den Zentralbalkan darstellte (Karte 4). Am Mittellauf der Velika Morava hat M. Stoji} 99 Siedlungsplätze, 8 Nekropolen sowie 14 Einzelfunde (Tüllenbeile und Mei¼el, Nadeln) und 1 Sammelfund aus der Zeit der Stufen Ha A und Ha B aufgelistet.124 Die Gro¼zahl der hier präsentierten Tüllenbeile aus Zentralserbien stammt ebenfalls aus dieser Region.125 Die Fundkontexte sind jedoch meistens nicht bekannt, weswegen die Beile vor allem als Zufallsfunde in die Museen kamen. Bei den Siedlungen lässt sich nach Stoji} vor der Phase mit der kannelierten Keramik ein deutlicher Hiatus feststellen, der auf eine schwache Besiedlung hinweist. Die ersten Siedlungsplätze mit kannelierter Keramik befinden sich vorwiegend auf strategisch gut positionierten Stellen in den Mündungsgebieten der kleineren Flüsse in die Velika Morava, während etwas später offensichtlich auch die Täler der Nebenflüsse besiedelt wurden. Die wichtigsten Siedlungen im Zusammenflussgebiet der Zapadna und Ju`na Morava (vgl. Karten 2–3) sind Gologlava, Kru{evac i Konopljara.126 Die Einflüsse der Gruppen mit der kannelierten Keramik und Urnenbestattungen in Zentralserbien lassen sich auch in der Brandgräbernekropole ^itluk deutlich erkennen.127 Im Raum Westserbiens sind die Funde, die im weiteren Sinne mit der Urnenfelderkultur zu verbinden sind, nur sporadisch vertreten. Dennoch gibt es einige Fundkomplexe, die auf einen Kontakt oder Austausch hindeuten. Es handelt sich um einen Sammelfund aus Konju{a (wahrscheinlich Grab)128, einen Tumuls in Aluge bei U`ice und die befestigte Siedlung Likodra bei Krupanj.129 Gut erkennbar ist vor allem die schwarze, polierte Keramikware aus Likodra, die besten Vergleiche in den Urnengräber der Gruppe Belegi{ II – Gava besitzt130. Trotz der Tatsache, dass dies bislang der einzige Siedlungsplatz in Westserbien mit solcher Keramik ist, sind die deutlichen Beziehungen Westserbiens mit dem Becken der Velika Morava nicht zu bestreiten.131 In Südserbien wurden die meisten für Belegi{ II – Gava typischen Funde in dem Siedlungsraum um die Mündung des Flusses Ni{ava in die Ju`na Morava mit bislang 36 bekannten Fundstellen festgestellt.132 An besten erforscht ist Medijana bei Ni{ka Banja. Die Funde der Gava Gruppe (kannelierte Keramik und urnenförmige Gefä¼e) sind nach M. Gara{anin vor allem für den Horizont Medijana II charakteristisch.133 In dem benachbarten Becken um die Stadt Leskovac konnten noch 24 weitere Fundstellen mit vergleichbarer Keramik identifiziert werden.134 Hervorzuheben ist die befestigte BRONZEZEITLICHE METALLURGIE Die Funde aus Kastanas und Troja140 weisen darauf hin, dass die kannelierte Keramik bis in den Ägäischen Raum entweder direkt durch die Anwesenheit der Bevölkerungselemente oder indirekt durch den Kontakt mit den nördlichen Regionen vermittelt wurde. Vergleichbar gut erforschte Siedlungen aus dem Raum Zentralbalkans sind bislang nicht bekannt, so dass man über die ökonomischen und sozialen Verhältnisse der 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 44 Stoji} 1986, 27. Ibid., 56. Stoji}, ^a|enovi} 2006, 34. Ibid. Gara{anin 1973, 435–438; Idem, T. CVIII. Gara{anin 1983a, 779. Dmitrovi}, Lju{tina 2013, 159. Ibid. Stoji}, Joci} 2006, 44–45. Gara{anin 1996, 211, 213. Bulatovi}, Jovi} 2010, 45. Stoji} 2011, 17, Sl. 25/25. Bulatovi} 2007, 40/41. Bulatovi}, Kapuran 2007; Stoji} 2011. Gara{anin 1983b, 775. Quci 1998, 134, 136, 138. Stefanovich 1973; Hänsel 1982, 34. GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 vermutlich mehr Informationen über die Art und Weise der Metallverarbeitung bringen und womöglich den Zusammenhang mit der nahe liegenden Nekropolen in Trnjane und bei Borsko Jezero bestätigen. Die genannten Beispiele zeigen deutlich, dass die Tradition der Kupfererzverarbeitung im nordöstlichen und südwestlichen Serbien sowie in [umadija (Zentralserbien) noch vor dem Auftauchen der Gruppen mit der kannelierten Keramik zurückreicht. Die Tüllenbeilgussformen aus Boljetin151, Planinica bei Zaje~ar, Medijana bei Ni{152, Hisar 153, Klinovac bei Vranje154 und aus Ra{ka bei Novi Pazar weisen darauf hin, dass die lokale Metallverarbeitung auch in der Zeit der Stufen Ha A und Ha B mit Sicherheit weiter betrieben wurde. Gruppen mit charakteristischer kannelierter Ware und dem Brandbestattungsritus derzeit nur spekulieren kann. Im Zusammenhang mit der gro¼en Zahl der Depotfunde und einer offensichtlich intensiven Bronzeproduktion, stellt sich jedoch die Frage, inwiefern man die spätbronzezeitlichen Gruppen mit den zahlreichen Kupferlagerstätten Serbiens bzw. deren Abbau in Verbindung bringen kann. Viele von diesen Lagerstätten waren bereits Jahrtausende zuvor bekannt und abgebaut. Das bekannteste Bespiel ist „Dnevni Kop“ in Rudna Glava bei Majdenpek mit 40 vorgeschichtlichen Stollen und Funden der Vin~a Kultur.141 Die naturwissenschaftlichen Untersuchungen haben darauf hingewiesen, dass bei der Herstellung kupferzeitlicher Objekte aus der Region sehr wahrscheinlich das Erz aus den Lagerstätten im Hinterland des Eisernen Tores (Bor – Majdanpek) verwendet wurde.142 Durch die Erforschung der letzten Jahrzehnte in den erzreichen Regionen im südwestlichen, nordöstlichen und zentralen Teil Serbiens, konnten auch einige neue Erkenntnisse zur prähistorischen Metallurgie gewonnen werden. Erzverarbeitende Aktivitäten während der Zeitspanne zwischen der Kupferzeit und der Spätbronzezeit sind in den Fundstellen um die Stadt Bor wie z. B. Praurije (Majdanpek)143, in Zentralserbien in Prlju{a– Mali [turac bei Rudnik144 sowie in Jarmovac bei Priboj145 im Südwesten des Landes zu verzeichnen (Karte 4). Zu nennen ist noch die Fundstelle Ravni Kopaonik – Suvo Rudi{te in Zentralserbien, wo man die Metallproduktion aufgrund der dort gefundenen Gerätschaften annehmen kann.146 Einen guten Einblick in die Struktur einer primär auf der Erz- und Metallverarbeitung basierenden Gemeinschaft bieten die Untersuchungen um die Ortschaft Trnjane bei Brestova~ka Banja (Bor) im Hinterland des Eisernen Tores.147 Bei der Fundstelle Trnjane handelt es sich um einen Siedlungsplatz und die dazugehörige Brandnekropole, die man der sog. Para}in -Gruppe des späten 2. Jahrtausends v. Chr. zuschreiben kann. Einen hohen Stellenwert der Erzverarbeitung unterstreicht die Schlacke aus den Urnen in der Nekropole.148 Vergleichbare Grabbeigaben (Schlacke) wurden auch im benachbarten Brandgräberfeld bei Borsko Jezero festgestellt.149 Eine weitere Produktionsstätte in dieser Region wurde vor kurzem in Ru`ane beim heutigen Dorf Banjsko Polje entdeckt.150 Die ältesten Funde aus Ru`ane stammen aus der Spätbronzezeit, die Metallverarbeitung setzte sich jedoch offenbar auch in der Eisenzeit und in der Antike fort. Die zukünftigen Analysen der Schlacke werden SCHLUSSWORT Bezeichnend für die Übergangsperiode in Serbien ist die gro¼e Zahl der Depotfunde und die Ausbreitung der verschiedenen Gruppen mit kannelierter Keramik, die zusammenfassend als Belegi{ II – Gava Komplex bezeichnet werden. Es ist zu vermuten, dass diese zwei Phänomene mit verstärkten metallurgischen Aktivitäten in den erzreichen Regionen in einen Zusammenhang zu bringen sind. Die Existenz einer lokalen Metallurgie wird u. a. durch die Gussformen und insbesondere durch viele Formen mit einem kleinen Verbreitungsradius (z. B. in dieser Arbeit umschriebene Tüllenbeilvarianten) bekräftigt. Es bleibt zu hoffen, dass die 141 Jovanovi} 1971; Idem 1984; Bori} 2009. Krajnovi} at all 1995; Begemann et al. 1995. 143 Jovanovi} 1995; 144 Jovanovi}, Mini}, Mrkobrad 1987; Jovanovi} 1988; Bogosavqevi} 1989; Idem 2005; Antonovi}, Vukadinovi} 2012. 145 Daevies 1937; Bogosavqevi} 1989; Idem 2005; Derikowi} 2005. 146 Bogosavqevi} 1989. 147 Jovanovi}, Jankovi} 1987–1990; Idem 1996; Srejovi}, Lazi} 1997. 148 Krajnovi} at all 1995; Kapuran, Miladinovi} 2011. 149 Jankovi} et al. 1980, Fig. 11. 150 Kapuran, Jovanovi} 2013. 151 Jevti} 1982. 152 Gara{anin 1983, T. CVII/2 153 Stoji} 2011, Sl. 25. 154 Bulatovi} 2003. 142 45 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 mit den Lagerstätten können einen wichtigen Beitrag zur Argumentation über die Verwendung der einheimischen Erzquellen leisten. Bis dahin bleibt es die bekannten Bronzeformen lediglich stilistisch-typologisch zu vergleichen und in einen chronologisch – räumlichen Zusammenhang zu bringen. rezenten Entdeckung der Kupferlagerstätten und Produktionsstätten auch eine entsprechende Intensivierung der archäometallurgischen Analysen initiieren. Die Erschaffung einer Datenbank mit der chemischen Signatur der lokalen Formen, insbesondere im Hinblick auf Kupfer und Zinn, sowie anschlie¼ende Vergleiche 46 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 LITERATURVERZEICHNIS: Antonovi}, Vukadinovi} 2012 – D. Antonovi}, M. Vukadinovi}, Geofizi~ka i arheolo{ka istra`ivanja na Malom [turcu u 2011. godini, u V. Biki} at all (ur.) Arheologija u Srbiji. Projekti Arheolo{kog instituta u 2011. godini, Arheolo{ki institut, Beograd 2012, 26–29. Begemann et al. 1995 – F. Begemann, E. Pernicka, S. Schmitt-Strecker, Searching for the ore sources of Eneolithic and EBA Copper artefacts from Serbia, in B. Jovanovi} (ed.) Ancient minning and metalurgy in southeast Europe, International symposium in Bor, Bor–Belgrade 1995, 143–150. Bogosavqevi} 1989 – V. Bogosavqevi}, Uvod u rudarsku arheologiju na Kopaoniku (oblici starog rudarstva i metalurgije), GSAD 5, 1989, 86–93. Bogosavljevi} 1995 – V. Bogosavljevi}, Minning hammerstones of Prlju{a Mali [turac, in B. Jovanovi} (ed.) Ancient minning and metalurgy in southeast Europe, International symposium in Bor, Bor–Belgrade 1995, 37–44. Bogosavqevi} 2005 – V. Bogosavqevi}, Praistorijski rudnici na centralnom Balkanu, Zbornik Narodnog muzeja XVIII–1, 2005, 79–113. Bori} 1997 – D. Bori}, Ostava kasnog bronzanog doba iz Futoga, PMB 39, Novi Sad 1997, 41–92. Bori} 2009 – D. Bori}, Absolute Dating of Metallurgical Innovations in the Vin~a Culture of the Balkans. In T. Keinlin and B. Roberts (eds.) Metals and Societies, Studies in honour of Barbara S. Ottaway, Verlag Dr. Rudolf Habelt GMBH, Bonn 2009, 191–245. v. Brunn 1968 – W. A. von Brunn, Mitteldeutsche Hortfunde der jüngeren Bronzezeit, Römisch-Germasniche Forschungen 29, Berlin 1968. Bukvi} 2000 – Lj. Bukvi}, Kanelovana keramika Gava kompleksa u Banatu, SANU ogranak u Novom Sadu, Novi Sad 2000. Bulatovi} 2003 – A. Bulatovi}, Kalup iz Klinovca, Starinar LII, 2003, 139–144. Bulatovi} 2007 – A. Bulatovi}, Vrawe, Kulturna stratigrafija praistorijskih lokaliteta u Vrawskoj regiji, Arheolo{ki institut i Narodni muzej u Vrawu, Beograd–Vrawe 2007. Bulatovi} 2010 – A. Bulatovi}, poreklo i distribucija blagobikoni~nih zdela fasetiranog ili kanelovanog oboda sa kraja bronzanog i po~etka gvozdenog doba na Balkanskom poluostrvu, Starinar LIX, 2010, 89–108. Bulatovi}, Jovi} 2010 – A. Bulatovi}, S. Jovi}, Leskovac, Kulturna stratigrafija praistorijskih lokaliteta u Leskova~koj regiji, Arheolo{ki institut i Narodni muzej u Leskovcu, Beograd–Leskovac 2010. Bulatovi}, Kapuran 2007 – A. Bulatovi}, A. Kapuran, The Early Iron Age Hillforth at Gradina site near Pre{evo in South Serbia, Archaeologia Bulgarica IX 3, 2007, 1–24. ^ernùh 1978 – E. H. ^ernùh, Gornoe delo i metalurgiý v drevneç{eç Bolgarii, Sofija 1978. Davies 1937 – O. Davies, Praehistoric copper-mine at Jarmovac near Priboj na Limu, Glasnik Zemaljskog muzeja u BiH XLIX 1, 1937, 1–3. Dmitrovi}, Lju{tina 2013 – K. Dmitrovi}, M. Lju{tina, Some observations on the borderes between the Bronze Age cultural groups in the region of the West Morava Velley, Central Serbia, in V. Sîbu and R. Stefanescu (eds.) The Thracians and their Neighbors in the Bronze and iron Age, Necropolises, Cult places, Religion, Mythology, Museul Brailei and Editura istros, Brasov 2013, 153–171. Derga~ev 2011 – B. A. Derga~ev, Kelúðù i Seríù ni`neáo Podunavøý (Socket Axes and Sickels from Lower Danube Area), Ki{inûu 2011. Derikowi} 2005 – S. Derikowi}, Arheometalur{ki i rudarski centar Jarmovac kod Priboja na Limu, preliminarni izve{taj o arheolo{kim iskopavawima u 2003. i 2004. Godini, Glasnik dru{tva konzervatora Srbije 29, 2005, 33–36. Gara{anin 1954 – D. Gara{anin, Katalog metala I, Narodni muzej, Beograd 1954. Gara{anin 1973 – M. Gara{anin, Praistorija na tlu SR Srbije, Srpska kwi`evna zadruga, Beograd 1973. Gara{anin 1975 – D. Gara{anin, Ostava iz Privine Glave, Praistorijske ostave u Srbiji i Vojvodini I, Srpska akademija nauka i umetnosti, Beograd 1975, 63–68. Gara{anin 1983 – M. Gara{anin, Medijana grupa, u A. Benac (ur.) PJZ IV, Bronzano doba, ANUBiH i Svjetlost, Sarajevo 1983, 761–772. Gara{anin 1983a – M. Gara{anin, Period polja sa urnama u zapadnoj Srbiji, u A. Benac (ur.) PJZ IV, Bronzano doba, ANUBiH i Svjetlost, Sarajevo 1983, 779–785. Gara{anin 1983b – M. Gara{anin, Bronzano doba na Kosovu, u A. Benac (ur.) PJZ IV, Bronzano doba, ANUBiH i Svjetlost, Sarajevo 1983, 754–759. Gara{anin 1994 – M. Gara{anin, Predgovor, Praistorijske ostave u Srbiji i Vojvodini II, Srpska akademija nauka i umetnosti, Beograd 1994. 47 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Jovanovi} 1982 – B. Jovanovi}, Rudna Glava, najstarije rudarstvo na podru~ju centralnog Balkana, Muzej rudarstva i metalurgije u Boru i Arheolo{ki institut u Beogradu, Beograd 1982. Jovanovi} 1988 – B. Jovanovi}, Prqu{a–Mali [turac, praistorijski rudnik bakra i gorskog kristala na Rudniku, Zbornik radova Narodnog muzeja XVIII, 1988, 5–12. Jovanovi} 1995 – B. Jovanovi}, Continuity of the Praehistoric Minning in the Central Balkans, in B. Jovanovi} (ed.) Ancient minning and metalurgy in southeast Europe, International symposium in Bor, Bor–Belgrade 1995, 29–38. Jovanovi}, Jankovi} 1987–1990 – B. Jovanovi}, I. Jankovi}, Nekropola para}inske grupe u Trnjanima kod Brestova~ke banje, Zbornik radova muzeja rudarstva i metalurgije u Boru 5/6, 1987–1990, 1–20. Jovanovi}, Jankovi} 1996 – B. Jovanovi}, I. Jankovi}, Die Keramik der Nekropole der Para}in Kultur – Trnjane bei Bor, in N. Tasi} (ed)., The Yugoslav Danube basin and the neighbouring regions in the 2nd Millenium B. C., International symposium in Vr{ac 1995, SANU Institute for Blakan studes, Beograd 1996, 185–200. Jovanovi}, Mini}, Mrkobrad 1987 – B. Jovanovi}, D. Mini}, D. Mrkobrad, Spomenici starog rudarstva i metalurgije na Rudniku, GSAD 4, 1987, 47–55. Kapuran, Jovanovi} 2013 – A. Kapuran, I. Jovanovi}, Ru`ana – new Bronze Age Metallurgical Center in North Eastern Serbia, in N. [trbac, D. @ivkovi} and S. Nestorovi} (eds.), Poroceedings of 45th International October Conference on Mining and Metallurgy, 16–19 October on Bor Lake, University of Belgrade, Technical Faculty in Bor and Mining and Metallurgical Institute in Bor, Bor 2013, 831–834. Kapuran, Miladinovi}-Radmilovi} 2011 – A. Kapuran, N. Miladinovi}-Radmilovi}, Nekropola na Borskom jezeru, Novi prilozi o sahrawivawu u bronzanom dobu, Starinar LXI, 2011, 114–153. Kondi} (ur.) 1992 – V. Kondi} (ur.), Praistorijski metal Pomoravqa, Arheolo{ki institut, Beograd 1992. Krajnovi} at all 1995 – D. Krajnovi}, B. Jankovi}, B. Jovanovi}, I. B. lorenz, M. K. Pavi}evi} and G. A. Wagner, in B. Jovanovi} (ed.) Ancient minning and metalurgy in southeast Europe, International symposium in Bor, Bor–Belgrade 1995, 59–67. Krsti} 1986 – D. Krsti}, Vajuga–Korbovo, Compterendu des fouilles exécutées en 1981, \erdapske sveske III, Arheolo{ki institut, Beograd 1986, 148–167. Gara{anin 1996 – M. Gara{anin, Die kulturelle und chronologishe stellung der Mediana-Gruppe, in N. Tasi} (ed.) The Yugoslav Danube Basin and the Neighbouring regions in the 2nd Millennium, SANU and Institute for Balkan Studies, Belgrade–Vr{ac 1996, 201–218. Hänsel 1976 – B. Hänsel, Beiträge zur regionalen und chronologischen Gliederung der älteren Hallstattzeit an der unteren Donau, Beiträge zur ur- und frühgeschichtlichen Archäologie des Mittelmeer-Kulturraumes, Bonn 1976, 16–17. Hänsel 1982 – B. Hänsel, Sudosteuropa zwischen 1600 und 1000 v. Chr., in B. Hänsel (ed.) Südosteuropa zwishen 1600 und 1000 v. Chr., Moreland editions-Band Bramstedt, Berlin 1982, 1–38. Holste 1951 – F. Holste, Hortfunde Südosteuropas, Marburg/Lahn 1951. Jankovi} et al. 1980 – S. Jankovi}, M. Terzi}, S. Karameta, T. Spaspv, M. Jovanovi}, M. Mili~i}, V. Mi{kovi}, A. Grubi}, I. Antonijevi}, Metallogenic features of copper deposits in the Volcano-Intrisive complexes of the Bor District, Yugoslavia, in S. Jankovi} and R. H. Sillitoe (eds.) European Copper Deposits, Society for Geology Applied to Mineral Deposits, UNESCO–IGCP Projects Nos 169 and 60, Copper Minning, Smelting and Refining Corporation Bor, Department of Economic Geology, Faculty of Minning and Geology Belgrade University, Belgrade 1980, 42–48. Jacanovi} 1986 – D. Jacanovi}, Praistorijska ostava iz Klewa, Starinar XXXVII, 1986, 153–173. Jacanovi}, Radoj~i} 2003 – D. Jacanovi}, H. Radoj~i}, Praistorijska ostava metalnih predmeta iz sela [etowe kod Petrovca na Mlavi, Viminacivm 13–14, 2003, 7–45. Jevti} 1982 – M. Jevti}, praistorijsko naseqe kod Boqetina, Starinar XXXII, 1982, 19–30. Jevti} 2004 – M. Jevti}, Gvozdeno doba u okolini Bora, u M. Lazi} (ur.) Bor i okolina u praistoriji, antici i sredwem veku, Muzej rudarstva i metalurgije u Boru i Centar za arheolo{ka istra`ivawa Filozofskog fakulteta, Bor–Beograd 2004, 127–164. Jevti}, Vukmanovi} 1996 – M. Jevti}, M. Vukmanovi}, Late Bronze and Early Iron Age in the Danube Valley, in N. Tasi} (ed.) The Yugoslav Danube Basin and the Neighbouring regions in the 2nd Millennium, SANU and Institute for Balkan Studies, Belgrade–Vr{ac 1996, 283–291. Jovanovi} 1971 – B. Jovanovi}, Metalurgija eneolitskog perioda Jugoslavije, Arheolo{ki institut, Beograd 1971. 48 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Popovi}, Vukmanovi}, Radoj~i} 1986 – P. Popovi}, M. Vukmanovi}, N. Radoj~i}, Fouilles de sondage sur la localite Vajuga–Pesak, \erdapske sveske III, Arheolo{ki institut, Beograd 1986, 168–183. Radi{i} 1958 – R. Radi{i}, Bronzani nalaz iz Se~nja, PMB 12–13, Novi Sad 1958, 115–122. Radoj~i}, N. 1986 – N. Radoj~i}, Les Fouilles du site “Pesak” a Korbovo en 1981, \erdapske sveske III, Beograd 1986, 133–142. Ra{ajski 1975 – R. Ra{ajski, Ostava Mesi}– [upaja kod Vr{ca, Praistorijske ostave u Srbiji i Vojvodini I, Srpska akademija nauka i umetnosti, Beograd 1975, 63–68. Ra{ajski 1975a – R. Ra{ajski, Ostava iz Banatskih Karlovaca, Praistorijske ostave u Srbiji i Vojvodini I, Srpska akademija nauka i umetnosti, Beograd 1975, 86–90. Rusu 1966 – M. Rusu, Depozitul de bronzuri de la Balþa, Sargetia 4, 1966, 17–40. Srejovi} 1975 – D. Srejovi}, Ostava iz Urovice, Praistorijske ostave u Srbiji i Vojvodini I, Srpska akademija nauka i umetnosti, Beograd 1975, 96–100. Srejovi} 1975a – D. Srejovi}, Ostava iz Aluna, Praistorijske ostave u Srbiji i Vojvodini I, Srpska akademija nauka i umetnosti, Beograd 1975, 93–96. Srejovi}, Lazi} 1997 – D. Srejovi}, M. Lazi}, Naseqa i nekropole bronzanog doba u Timo~koj krajini, u M. Lazi} (ur.), Arheologija Isto~ne Srbije, Centar za arheolo{ka istra`ivawa Filozofskog fakulteta, Beograd 1997, 225–244. Stefanovich 1973 – R. Stefanovich, Some Balkan Elements in the Aegeian Migration, in M. Gara{anin, A. Benac and N. Tasi} (eds.) Actes du VIIIe congres International des Sciences Prehistoriques et protohistoriques, Beograd 1973, 148–162. Stoji} 1986 – M. Stoji}, Gvozdeno doba u basenu Velike Morave, Centra za arheolo{ka istra`ivanja Filozofskog fakulteta u Beogradu, Beograd–Svetozarevo 1986. Stoji} 2003 – M. Stoji}, Veliki Vetren, Arheolo{ki institut, Beograd 2003. Stoji} 2011 – M. Stoji}, Odnos sredweg Podunavqa i basena ju`ne Morave u gvozdeno doba I (pribli`no 1350–1100. Godine pre n. e.) na osnovu metalnih nalaza sa lokaliteta Hisar u leskovcu, Leskova~ki zbornik LI, 2011, 9–30. Stoji}, Joci} 2006 – M. Stoji}, M. Joci}, Ni{, kulturna startigrafija praistorijskih lokaliteta u Ni{koj regiji, Arheolo{ki institut i Narodni muzej u Ni{u, Beograd–Ni{ 2006. König 2004 – P. König, Spätbronzezeitliche Hortfunde aus Bosnien und der Herzegowina, Prähistorische Bronzefunde (PBF) XX, Band 11, 2004. Lalovi} 1975 – A. Lalovi}, Praistorijske ostave iz Narodnog Muzeja u Zaje~aru Starinar XXVI, 1975, 143–149. Quci 1998 – K. Quci, Bronzano doba, u N. Tasi} (ur.) Arheolo{ko blago Kosova i Metohije, od neolita do ranog sredweg veka, SANU i Muzej u Pri{tini 1998, 116–147. Marinkovi} 1991 – S. Marinkovi}, Bronzana ostava iz Se~wa III, RVM 33, Novi Sad 1991, 17–22. Mayer 1977 – E- F. Mayer, Die Äxte und Beile in Österreich, Prähistorische Bronzefunde (PBF) IX, Band 9, 1977. Medovi} 1994 – P. Medovi}, Geneza kultura starijeg gvozdenog doba u Jugoslovenskom podunavqu, u N. Tasi} (ur.) Kulture gvozdenog doba Jugoslovenskog podunavqa, Balkanolo{ki institut SANU i Gradski muzej u Somboru, Beograd 1994, 45–50. Mani} 2010 – A. Mani}, Praistorijska nalazi{ta u gornjem Poni{avlju, Habilitacioni rad za muzejsko zvanje, nepublikovan, 2010. Medovi} 1994 – P. Medovi}, Geneza kultura starijeg gvozdenog doba u Jugoslovenskom podunavqu, u N. Tasi} (ur.) Kulture gvozdenog doba Jugoslovenskog podunavqa, Balkanolo{ki institut SANU i Gradski muzej u Somboru, Beograd 1994, 45–50. Metzner-Nebelsick 2002 – C. Metzner-Nebelsick, Der „Thrako-Kimmersiche“ Formenkreis aus der Sicht der Urnenfelder- und Hallstattzeit im südöstlichen Pannonien. Vorgeschichtliche Forschungen 23, 2002. Novotná 1978 – M. Novotná, Die Äxte und Beile in der Slowakei, Prähistorische Bronzefunde (PBF) IX, Band 3, 1978. Petrescu-Dîmboviþa 1977 – M. Petrescu-Dîmboviþa, Depozitele de bronzuri din Romania, 1977. Petrescu-Dîmboviþa 1978 – M. Petrescu-Dîmboviþa, Die Sicheln in Rumänien mit Corpus der jung – und spätbronzezeitlicher Horte Rumäniens, Prähistorische Bronzefunde (PBF), XVIII, Band 1, 1978. Popovi} 1994 – D. Popovi}, Ostava iz Dobrinaca, Praistorijske ostave u Srbiji i Vojvodini II, Srpska akademija nauka i umetnosti, Beograd 1994, 8–25. Popovi} 1994a – D. Popovi}, Novi nalazi starijeg gvozdenog doba u Sremu, u N. Tasi} (ur.) Kulture gvozdenog doba Jugoslovenskog podunavqa, Balkanolo{ki institut SANU i Gradski muzej u Somboru, Beograd 1994, 63–72. 49 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Danube Basin and the Neighbouring regions in the 2nd Millennium, SANU and Institute for Balkan Studies, Belgrade–Vr{ac 1996, 23–42. Vasi} 1982 – R. Vasi}, Spätbronzezeitliche und Älterhallstattzeitliche Hortfunde im östlichen Jugoslawien. Südosteuropa zwischen 1600 und 1000 v. Chr, Prähistorische Archäologie in Südosteuropa, Band 1, Berlin 1982, 267–285. Vasi} 1982a – R. Vasi}, O po~etku gvozdenog doba u Srbiji, Starinar XXXII, 1982, 1–7. Vinski-Gasparini 1973 – K. Vinski-Gasparini, Kultura polja sa `arama u sjevernoj Hrvatskoj, Zadar 1973. Vukmanovi}, Popovi} 1986 – M. Vukmanovi}, P. Popovi}, Recherches archéologiques sur la localité “Livade” près de Mala Vrbica. \erdapske sveske III, Beograd 1986, 7–13. Vukmanovi}, Radoj~i} 1995 – M. Vukmanovi} i N. Radoj~i}, Katalog metala II, Narodni muzej, Beograd 1995. Wanzek 1989 – B. Wanzek, Die Gu{model für Tüllenbeile im südöstlichen Europa, Universitätsforschungen zur Prähistorischen Archäologie (UPA) 2, 1989. Stoji}, ^a|enovi} 2006 – M. Stoji}, G. ^a|enovi}, Kru{evac, kulturna stratigrafija praistorijskih lokaliteta u zoni sastava Zapadne i Ju`ne Morave, Arheolo{ki institut i Narodni muzej u Kru{evcu, Beograd–Kru{evac 2006. Tasi} 1983 – N. Tasi}, Jugoslovensko Podunavqe od Indoevropske seobe do prodora Skita, Matica Srpska i Balkanolo{ki institut, Beograd 1983. Todorovi} 1956/1957 – J. Todorovi}, Bronzana ostava iz Vojilova, Strinar VII–VIII, 1956/1957, 275–278. Todorovi} 1975 – J. Todorovi}, Bronzana ostava iz Leskova, Praistorijske ostave u Srbiji i Vojvodini I, Srpska akademija nauka i umetnosti, Beograd 1975, 78–79. Todorovi} 1975a – J. Todorovi}, Bronzana ostava iz Vojilova, Praistorijske ostave u Srbiji i Vojvodini I, Srpska akademija nauka i umetnosti, Beograd 1975, 75–78. Trbuhovi} 1960 – V. Trbuhovi}, Iz praistorije doline Resave, Starinar HI, 1960, 181–184. Uzelac 1996 – J. Uzelac, Bronze Age in the South of the Yugoslavian Banat, in N. Tasi} (ed.) The Yugoslav 50 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) Rezime: STARINAR LXIV/2014 MARIO GAVRANOVI], Berlin ALEKSANDAR KAPURAN, Arheolo{ki institut, Beograd O NEKIM VARIJANTAMA BRONZANIH SEKIRA NA CENTRALNOM BALKANU Kqu~ne re~i. – Bronzana sekira – kelt, bronzano doba, centralni Balkan, ostave bronzanih predmeta, Belegi{ II–Gava, metalurgija. 1. Atenica (T. I/1). Du`ina: 10,6 cm. Muzej ^a~ak (A 61). 2. Mali Izvor (T. I/2). Du`ina: 9 cm. Muzej Zaje~ar (ib. 158). 3. Nepoznato nalazi{te, Srbija (T. I/3). Du`ina: 8,4 cm. Narodni Muzej Beograd (ib. 7075). 4. Sviwi{te (T. I/4). Du`ina: 7,7 cm. Muzej Vrawe. Tokom rada na sakupqawu gra|e za opse`nu publikaciju koja se bavi bronzanim sekirama na tlu Srbije, uo~eno je nekoliko karakteristi~nih tipova i varijanti bronzanih sekira – keltova sa specifi~nom, lokalno ograni~enom, prostornom redistribucijom (Karte 1–3). Ovom prilikom }e biti detaqnije predstavqena grupa keltova koje odlikuje poja~ani ili, horizontalnim rebrom, profilisani obod, i ~ije je se~ivo fasetirano u vidu arkade ili blago zaobqenog trougla. Mada je jedan deo nalaza prikazanih u ovom radu objavqen, jasna tipolo{ko-hronolo{ka kvalifikacija ove vrste bronzanih sekira jo{ uvek nedostaje. Bez u{ice Keltovi bez u{ice sa sli~nim zadebqanim zavr{etkom oboda i vertikalnim rebrima unutar arkadnog se~iva poznati su iz Bugarske, gde su opisani kao tip K–10, odnosno kao tip Vrbica A, Varijanta E. 5. Pirot (T. I/6). Du`ina: 6 cm. Muzej Pirot (ib. 3). Tipolo{ka podela Ako se posmatraju svi keltovi sa arkadno fasetiranim se~ivom sa teritorije Srbije, mogu}e je definisati nekoliko osnovnih tipova, koji se, u ve}oj ili mawoj meri, podudaraju sa spomenutim nalazima, pre svega iz Rumunije i Bugarske. Prvu ve}u grupu ~ine keltovi sa mawe ili vi{e izra`enim zadebqawem oboda i bez dodatnih rebara ili ukrasa. Unutar ove grupe mogu se izdvojiti tri osnovna tipa, koja su u dosada{wim radovima uglavnom dobro definisana i iscrpno obra|ena, te ih ovde ne}emo detaqnije analizirati. Radi se o slede}im tipovima: • Keltovi bez u{ice sa trouglastim odnosno arkadnim poqem na se~ivu. Tipi~ni predstavnici ovog tipa su keltovi iz ostava Futog, [etowe (oba sa trouglastim poqem) ili Privina Glava, Dobrinci, Dubravica i Mali Izvor (sa arkadnim poqem). • Keltovi sa u{icom i sa trouglastim odnosno arkadnim poqem na se~ivu. Jedina su{tinska razlika u odnosu na prethodne keltove je postojawe u{ice sa strane, ~iji se gorwi kraj gotovo uvek nastavqa na zadebqawe na obodu, te sporadi~na pojava nepravilnih i neujedna~enih ukrasa na delu izme|u oboda i poqa se~iva. • Keltovi sa zadebqanim obodom i bademastim ukrasom na arkadnom poqu se~iva. • C6 po Rusu. • C7 po Rusu. Blisko varijanti A Istaknuto zadebqawe na obodu i vertikalna rebra poseduje i kelt iz ostave Se~aw III. Razliku od ostalih primeraka varijante A predstavqaju dva, me|usobno blisko postavqena, horizontalna tanka rebra. 6. Se~aw III (T. I/5). Du`ina: 9 cm. Muzej Zrewanin (ib. 2102). Varijanta B Varijantu B defini{u dva tipolo{ka obele`ja: obod sa dva horizontalna rebra i u{icom izlivenom upravo izme|u ta dva poja~awa, i vertikalna rebra unutar arkadnog poqa se~iva (Karta 1). Zadebqano gorwe rebro ujedno je i poja~awe ivice otvora. 1. Gorwi Ra~nik (T. I/7). Du`ina: 10 cm. Muzej Jagodina (ib. 412). 2. Jagodina (T. I/8). Du`ina 12,2 cm. Muzej Jagodina. 3. Juhor (T. I/9). Du`ina 11,5 cm. Narodni Muzej Beograd. Varijanta C Identi~an oblik oboda kao kod varijante B, sa dva horizontalna rebra i u{icom koja ih zatvara, poseduju i keltovi varijante C (Karta 2). Jedina zna~ajna razlika je nedostatak vertikalnih rebara odnosno bilo kakvih dodatnih ukrasa. Rebro na zavr{etku oboda je ne{to ja~e izra`eno nego dowe rebro. 1. Dvori{te, op{tina ]uprija (T. II/1). Du`ina: 9,6 cm. Narodni Muzej Beograd (ib. 22821). 2. Dowe [tipqe (T. II/2). Du`ina: 9,3 cm. Muzej Jagodina (ib. 829). 3. Kopaonik (T. II/3). Du`ina: 10,5 cm. Narodni Muzej Beograd (ib. 1685). Varijanta A Prepoznatqive karakteristike ove varijante su prstenasto zadebqawe na rubu i vertikalna rebra unutar arkadnog se~iva (Karta 1). Po~etak u{ice odnosno wen gorwi kraj izveden je iz zadebqawa na rubu. Dowi deo se~iva se kod svih primeraka pro{iruje i prelazi u lepezastu formu. Sam rub se~ice je vi{e ili mawe zaobqen. 51 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) 4. Sekuri} (T. II/4). Du`ina: 10 cm. Narodni Muzej Kragujevac (ib. 3175). STARINAR LXIV/2014 u ju`noj Panoniji, s jedne strane, i Gava kompleksa u zoni \erdapa, s druge (Karta 4). Tasi} smatra da za Gava kompleks i daqe nije jasno da li je u Podunavqe do{ao sa severozapada ili sa juga, odnosno Pomoravqa i Poni{avqa, mada je Bulatovi}, analizom pojave zdela sa uvu~enim i tordiranim obodom od mati~nih oblasti u Slova~koj, kroz Panoniju i Pomoravqe do Makedonije, odnosno, od wihove pojave u Br D/Ha A1 pa do Ha B–C, ukazao na mogu}nost obrnutog pravca kretawa. Zbog nedostatka sinteza kao i zbog odre|enih nejasno}a u stratigrafiji vezanih za distribuciju naseqa i nekropola tranzicionog perioda, M. Gara{anin je ceo ovaj period nazvao „Vojvo|anska grupa prelaznog perioda“. Varijanta D Kao i dve prethodne varijante, keltovi varijante D tako|e poseduju obod sa dva horizontalna rebra od kojih je gorwe rebro, sme{teno na ivici otvora, nepravilno zadebqano (Karta 2). Specifi~nost ove varijante su vertikalna rebra izme|u arkadnog poqa se~ice i horizontalnih rebara na poja~anom obodu. Iako su dosada iz Srbije poznata samo dva primerka ovog tipa, karakteristi~an raspored vertikalnih rebara u kombinaciji sa obodom i u{icom opravdava definisawe posebne varijante. 1. Bunar, Belica (T. II/5). Du`ina: 7,4 cm. Muzej Jagodina (ib. 1095). 2. Trwane (T. II/6). Du`ina: 8,8 cm. Narodni Muzej Po`arevac (ib. 407). Prostorna distribucija i pravci kretawa nosilaca kulture Belegi{ II – Gava kroz centralni Balkan Na prostoru ju`nog i isto~nog Banata otkriveno je vi{e naseqa i nekoliko nekropola Belegi{ II – Gava kulturnog kompleksa (ukupno 23), kao i 25 ostava, sa preko 1000 bronzanih predmeta, koje su u najve}em broju publikovane i, sem tri, sve pripadaju periodu Na A1. Me|utim, nakon studije Q. Bukvi}a o Gava kulturnom kompleksu, ovaj broj se znatno uve}ao, tako da obuhvata 67 lokaliteta i 34 ostave. Tako|e se i u Sremu nalazi ve}i broj naseqa i nekropola (kulturno homogenih 51 i 13 onih koji se poklapaju sa Bosutskim naseqima), a koncentrisani su uz obale velikih reka i na Fru{koj Gori, dok se u Ba~koj najvi{e javqaju oko Tami{a. Zanimqiva je pojava da se pozicije naseqa Belegi{ I i Gava kulture ne poklapaju, iako je to slu~aj sa nekropolama. Ovo jedino ne va`i za velika telska naseqa poput Gomolave i Feudvara. Na podru~ju \erdapske klisure konstatovano je vi{e naseqa i nekoliko nekropola Gava kulture koje su pozicionirane sa obe strane Dunava. Analiziraju}i tipove kanelovanih posuda i metalnih nalaza iz grobova, M. Jevti} i M. Vukmanovi} su ovu grupu nazvali Mala Vrbica – Hinova–Balta Verde, i datovali je u 11. vek i po~etak 10. veka pre n. e. Najzna~ajnije lokalitete na srpskoj strani \erdapa predstavqaju Mala Vrbica – Livade, Vajuga–Korbovo i Vajuga–Pesak. Prodor kulture kanelovane keramike na jug najverovatnije je i{ao glavnom transferzalom kroz centralni Balkan, dolinom Velike Morave (Karta 4). M. Stoji} je u regionu centralne Morave konstatovao 99 naseqa i 8 nekropola, 1 grupni nalaz i 14 pojedina~nih nalaza (bronzanih sekira i dleta, igala itd.). Bronzane sekire su prete`no dospevale u Muzeje bez jasne lokacije nalaza, a datuju se u Na A–V. Stoji} tako|e zakqu~uje da je nastanak ovih naseqa izuzetno va`an zbog ~iwenice da je ova teritorija dugo bila nenastawena. Dok su u po~etku naseqavane iskqu~ivo strate{ki lako odbrawive lokacije na u{}ima mawih reka u Veliku Moravu, u kasnijim fazama naseqa se {ire i na wene pritoke. Na podru~ju sastava Zapadne i Ju`ne Morave tokom prelaznog perioda iz bronzanog u starije gvozdeno doba konstatovan je mawi broj naseqa, od kojih su najva`nija Gologlava, Kru{evac i Konopqara. Uticaji nosilaca kulture kanelovane keramike naro~ito se vide na nekropoli spaqenih pokojnika u ^itluku. Uvre`eno je mi{qewe da je na teritoriji zapadne Srbije uticaj poqa sa urnama sporadi~an, mada jedan skupni nalaz, jedna sahrana pod tumulom i jedan gradinski lokalitet Varijanta E Varijanta E obuhvata neukra{ene keltove sa arkadnim ili trouglasto zaobqenim poqem, koji ispod zadebqane ivice otvora poseduju jo{ dva horizontalna rebra (Karta 2). 1. Hisar, op{tina Leskovac (T. II/7). Du`ina: 9,4 cm. Narodni Muzej Leskovac. 2. Stari Kostolac (T. II/8). Du`ina: 9,5 cm. Narodni Muzej Po`arevac (ib. 860). 3. Vojilovo (T. II/9). Du`ina 8,5 cm. Narodni Muzej Po`arevac (ib. 59). 4. Nepoznato nalazi{te, Srbija (T. II/10). Du`ina: 8 cm. Muzej Grada Beograda (ib. 4164). 5. Nepoznato nalazi{te, Srbija (T. II/11). Muzej Kraqevo (A 1268). Blisko varijanti E 6. Se~aw (T. II/12). Du`ina: 10 cm. Narodni Muzej Zrewanin (ib. 4559). Varijanta F Keltovi sa vi{e horizontalnih rebara na obodu i jednim vertikalnim rebrom na arkadnom se~ivu (Karta 3). 1. Banatski Karlovci (T. III/1). Du`ina: 10,8 cm. Museul regional al Banatului – Timiºoara (1488). 2. Futog (T. III/2). Du`ina: 10,9 cmm. Muzej Vojvodine, Novi Sad (ib. 3303). 3. Klewe (T. III/3). Du`ina: 9,8 cm. Narodni Muzej Po`arevac (ib. 1008). 4. Klewe (T. III/4). Du`ina: 8 cm. Narodni Muzej Po`arevac (ib. 1007). Varijanta G Keltovi sa horizontalnim rebrima na obodu i vi{estrukim arkadnim rebrima na se~ivu (Karta 3). 1. Banatski Karlovci (T. III/5). Du`ina: 10,4 cm. Narodni Muzej Vr{ac (ib. 926). 2. @agubica (T. III/6). Du`ina: 11,1 cm. Narodni Muzej Po`arevac (ib. 856). Prelazni period iz bronzanog u gvozdeno doba u srpskom Podunavqu i Pomoravqu Kao po~etak dominacije urnenfelder kulture u srpskom Podunavqu smatra se formirawe Belegi{ II a–b faze 52 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 do bronzanog doba, identifikovani su lokaliteti u okviru Borskog revira i Praurije (Majdanpek), lokaliteti Prqu{a – Mali {turac (Rudnik) i Jarmovac (Priboj na Limu). Postoje odre|ene indikacije koje ukazuju da se i Ravni Kopaonik – Suvo rudi{te tako|e mo`e ukqu~iti u listu praistorijskih metalur{kih centara, s obzirom na izvesna otkri}a praistorijskih alatki koje poti~u sa ovog lokaliteta. Istra`ivawa praistorijske nekropole i naseqa Trwane kod Brestova~ke bawe, dovelo je do novih saznawa o `ivotu metalur{kih zajednica Para}inske kulture poznog bronzanog i ranog gvozdenog doba u zale|u \erdapa. Ostaci metali~nih {qaka koji su se mogli na}i u urnama na Trwanima i na susednoj nekropoli sa spaqenim pokojnicima na Boskom jezeru, dovoqno govore o primarnim privrednim aktivnostima koje su se odvijale u toj izrazito rudonosnoj oblasti krajem II milenijuma pre n. e. U prilog ovome govori najnovije otkri}e metalur{kog centra Ru`ana, koji se nalazi u neposrednoj blizini Trwana, na prostoru dana{weg naseqa Bawsko Poqe kod Brestova~ke Bawe. Ovaj metalur{ki centar je nastao krajem bronzanog doba i `iveo je tokom celog gvozdenog doba, i u kasnoj antici. Analize metali~nih {qaka tek treba da poka`u kakve su se sve metalur{ke aktivnosti odvijale na obalama potoka Ru`ana. U svakom slu~aju, na prostorima severoisto~ne i jugozapadne Srbije, kao i [umadije pre dolaska kultura kanelovane keramike ve} je postojala hiqadugodi{wa rudarskometalur{ka tradicija eksploatacije i prerade bakra. O ovome svedo~e i nalazi liva~kih kalupa koji pripadaju raznim tipovima {upqih bronzanih sekira otkrivenih na Boqetinu, Planinici kod Zaje~ara, Medijani kod Ni{a, Hisaru u Leskovcu, Klinovcu kod Vrawa i u okolini Ra{ke kod Novog Pazara. jasno pokazuju da je do wega ipak do{lo. Radi se o, verovatno, grobnom prilogu iz Kowu{e, tumulu u Alugama i gradinskom utvr|ewu Likodra. [to se ti~e prodora kulture Gava daqe na jug, izgleda da se oko u{}a Ni{ave u Ju`nu Moravu formirao kompleks naseqa, kojih ima oko 36. Od najva`nijih naseqa u ovoj regiji svakako se izdvaja Medijana kod Ni{ke Bawe. Prema M. Gara{aninu, nalazi Gava kulture na ovom lokalitetu vezuju se za horizont Medijana II, u kome se nalazi odre|en broj urni i drugih posuda sa kanelovanim ornamentom. U susednoj Leskova~koj kotlini identifikovana su 24 lokaliteta na kojima je otkrivena kanelovana keramika sa odlikama Belegi{ II – Gava kulturnog kompleksa. Me|u wima najzna~ajnije mesto zauzima gradinsko utvr|ewe Hisar u Leskovcu, na kojem se izdvajaju dve faze starijeg gvozdenog doba, od kojih mla|a faza najverovatnije pripada P{eni~evo kulturnoj grupi. Nalazi bronzanih predmeta otkriveni na ovom lokalitetu, kao i jedan kalup za sekire – keltove, datovani su u period Na A. Daqe ka jugu, u Vrawskoj regiji Ju`ne Morave, A. Bulatovi} je, prema stilsko-tipolo{kim karakteristikama kanelovane keramike koja je otkrivena tokom istra`ivawa, konstatovao promene u topografiji naseqa domoroda~kih i novoprido{lih zajednica. Prilikom istra`ivawa gradinskog naseqa u Sviwi{tu, kod Pre{eva, na podnici ku}e je in situ otkrivena jedna bronzana sekira koja, zajedno sa onom otkrivenom na Hisaru, predstavqa retke nalaze koji poti~u sa arheolo{kih istra`ivawa i iz zatvorenih celina. O tranzicionom periodu iz bronzanog u gvozdeno doba na Kosovu nema mnogo podataka, i on je predstavqen prvenstveno nekropolama u Dowoj Brwici, Gra{tici i Ulpijani, kao i naseqima Qu{ta i Gladnice. Zakqu~ak Pojava velikog broja ostava bronzanih predmeta koje se vezuju za tranzicioni period iz bronzanog u starije gvozdeno doba i za dominaciju kultura Belegi{ II – Gava, morala je imati vezu sa tradicionalnom eksploatacijom i metalurgijom bakra na teritoriji centralnog Balkana. O postojawu metalurgije za sada svedo~e jedino nalazi kalupa za livewe, kao i posebne varijante bronzane sekire karakteristi~ne za ovo podru~je. Nakon du`eg perioda, ponovo se javio interes za prou~avawe arheometalurgije u praistoriji na tlu Srbije. Zbog toga verujemo da }e otkri}a novih rudnih okana i metalur{kih centara inicirati projekte koji }e se baviti strukturalnim, fizi~kim i hemijskim analizama predmeta od bronze, naro~ito izotopa bakra i kalaja, a koje bi mogle da doka`u eksploataciju mineralnih sirovina na teritoriji Srbije. Do tada se mo`emo jedino osloniti na stilsko-tipolo{ke analize postoje}ih proizvoda od bronze, na koje nailazimo re|e tokom arheolo{kih istra`ivawa a ~e{}e van arheolo{kog konteksta, u zbirkama muzeja i privatnih kolekcija. Metalurgija bakra i bronze Uticaji nosilaca kulture kanelovane keramike {irili su se na prostoru izme|u centralne Evrope sve do Egeje, gde su identifikovani na nalazima sa Kastanasa i Troje. Mali stepen istra`enosti naseqa u kojima su oni `iveli nije dovoqan da rekonstrui{emo kulturno-ekonomske prioritete na kojima se zasnivao `ivot tih zajednica. Za wih se vezuje pojava velike koncentracije ostava sa bronzanim predmetima na na{em prostoru, i zbog toga smatramo da se mogu dovesti u vezu sa metalurgijom bakra koja je u Srbiji ve} postojala nekoliko milenijuma pre wihovog dolaska. Predistorija rudarstva i metalurgije bakra na teritoriji centralnog Balkana vezuje se za kraj VI i po~etak V milenijuma pre n. e. Tokom posledwih nekoliko decenija do{lo se do novih saznawa o praistorijskoj metalurgiji, prvenstveno zahvaquju}i istra`ivawima u okvirima metalogenetskih zona na jugozapadu, severoistoku i u centralnoj Srbiji. Kao primeri rudarskih i metalur{kih aktivnosti od ranog eneolita 53 STARINAR LXIV/2014 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) Variante A 3 1 2 3 4 5 6 Variante B 7 8 Tafel I – Tüllenbeile der Varianten A und B Tabla I – Bronzani keltovi varijanti A i B 54 9 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Variante C 1 2 3 4 5 6 7 8 9 10 11 12 Variante D Variante E Tafel II – Tüllenbeile der Varianten C, D und E Tabla II – Bronzani keltovi varijanti C, D i E 55 GAVRANOVI], KAPURAN, Über einige Tüllenbeilvarianten im Zentralbalkan (31–56) STARINAR LXIV/2014 Variante F 1 2 3 5 6 Variante G 4 Tafel III – Tüllenbeile der Varianten F und G Tabla III – Bronzani keltovi varijanti F i G 56 ALEKSANDAR P. BULATOVI] Institute of Archaeology, Belgrade NEW FINDS AS A CONTRIBUTION TO THE STUDY OF THE EARLY BRONZE AGE IN THE SOUTHERN PART OF THE CENTRAL BALKANS UDK: 903"637"(497) DOI: 10.2298/STA1464057B e-mail: [email protected] Received: February 18, 2014 Accepted: July 28, 2014 Original research article Abstract. – In this paper, new sites are presented which were explored throughout the protective excavations conducted along the E 75 motorway, as well as new indicative finds from the sites in north-eastern Macedonia. These results will complete the picture of life in the Early Bronze Age communities of the territory of Southern Pomoravlje and Gornja P~inja, as well as that of their relationship with the neighbouring populations. Of particular importance are the results of the investigation of this territory’s first explored necropolis from this period, which enriched our knowledge of the spiritual life of these communities. Key words. – Early Bronze Age, Southern Pomoravlje, Gornja P~inja, Armenochori culture, Bubanj–Hum III culture. T into one monograph, as well as in a number of articles.2 Shortly after that, however, during the protective explorations of the sites endangered by the construction of the E 75 motorway, a few very significant sites were discovered on the Leskovac–Bujanovac section of the road,3 which offer extremely important information about the material and spiritual culture of the communities who, in the Early Bronze Age, populated the South Morava basin and the P~inja valley, as well as their he south-eastern part of Serbia, comprising the basin of the upper course and middle course of the South Morava was, for a long time, considered to be a mainly unexplored area, at least in terms of sites from the Early Bronze Age. A small number of explored sites were known, such as Bubanj in Novo Selo, Velika Humska ^uka in Hum and ^esma in Vrti{te, whilst the majority of the finds from this period consisted of accidental or individual finds registered at unexplored sites, or at unmarked or unreliable locations.1 Only in recent years, by investigating the wider area of the E 75 and E 80 motorways and by resuming the explorations at Bubanj and in Hum, along with other archaeological research, has the number of sites from this period increased and, thereby, the knowledge of this period has been enhanced. All these results, including the results of older research, were combined 1 Gara{anin, Ajdi}, Dejanovi} 1971, kat. br. 226; Gara{anin 1973, 164–206 and quoted literature; Stoji}, Joci} 2006, 67–77. 2 Bulatovi}, Stankovski 2012; Bulatovi} 2011, 1–14; Bulatovi} 2013, 1–14. 3 Bulatovi}, Kapuran 2013, 70–79. * The article results from the project: Archaeology of Serbia: cultural identity, integration factors, technological processes and the role of the Central Balkans in the development of European prehistory (no OI177020) funded by the Ministry of Education, Science and Technological development of the Republic of Serbia. 57 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 List of sites: 1. Meani{te site, Ranutovac; 2. Golo Rebro site, Donji Neradovac; 3. Kova~ke Njive site, Pavlovac; 4. Gradi{te site, Davidovac; 5. Taticev kamen site, Kokino Selo, Republic of Macedonia; 6. Tri Kru{ke site, Klinovac; 7. Dve Mogili site, Pelince, Republic of Macedonia; 8. Banjka–Kaldrma site, Miratovac; 9. Pribovce site, Lopate, Republic of Macedonia. Spisak lokaliteta: 1. Meani{te, Ranutovac 2. Golo rebro, Dowi Neradovac 3. Kova~ke wive, Pavlovac 4. Gradi{te, Davidovac 5. Tati}ev kamen, Kokino Selo, R. Makedonija 6. Tri kru{ke, Klinovac 7. Dve mogili, Pelince, R. Makedonija 8. Bawka–kaldrma, Miratovac 9. Pribovce, Lopate, R. Makedonija Map 1. Early Bronze Age sites in the upper course of the South Morava and in Gornja P~inja Karta 1. Lokaliteti ranog bronzanog doba u gorwem toku Ju`ne Morave i u Gorwoj P~iwi relationships with the neighbouring populations.4 After these explorations, which gave rise to new indicative information about this period, a need to write a study that would integrate all the available data arose, with its focus on the new, unpublished sites and finds, allowing a comprehensive insight into the problems of the origins and development of the Early Bronze Age culture in the south of the central Balkans. southern, about 20 m apart. No burial traces have been discovered between the two parts. The northern part is poorly preserved, since the graves were dug to a very small relative depth (less than 0.2 m in places), and the upper part of the necropolis was destroyed by agricultural and other ground works. In the southern part of the necropolis (Fig. 1), with a surface area of around 100 m² which was much better preserved, cremated human remains were interred in shallow pits, approximately 0.5 m in diameter, into which, in most cases, one or more round stones or large pebbles were placed. In some graves the remains of the deceased were covered with a vessel, most usually a cup or a bowl. In all the graves offerings were noted, such as one or more SITES AND FINDS 1. Ranutovac, Meani{te site, Vranje5 The site of Meani{te is situated around 5 km north of Vranje, on the southern outskirts of the village of Ranutovac (Map 1/1). From a geological perspective, the site is positioned partially in the alluvial plain (the eastern part of the Iron Age settlement and the Early Bronze Age necropolis), and partially in the area with moderate washout and gullies (the western part of the Iron Age settlement). The necropolis is at a distance of around 600 m west of the current South Morava river bed. It consisted of two parts, the northern and the 4 Explorations were conducted by teams from the Institute of Archaeology in Belgrade and the Department of Archaeology of the Faculty of Philosophy in Belgrade, during the period of August 2011 – July 2013. 5 The site was explored by a team from the Institute of Archaeology in Belgrade (under the leadership of A. Bulatovi}) in the period of April – July 2012. The publication detailing the results of the site explorations is being prepared. 58 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 Fig. 1. Necropolis in Ranutovac, southern part, view from the south. Sl. 1. Nekropola u Ranutovcu, ju`ni deo, pogled sa juga whole vessels (Figs. 2, 3) or fragments of vessels, whilst in grave 2, in the northern part of the necropolis, along with the vessels there was a perforated axe made of polished grey stone of a compact structure. The graves were then covered with a circular construction 1–2 m in diameter, which was made of broken stone, whilst some graves were encircled with a stone ring 1.5–3 m in diameter. According to the quantity of stone from the destroyed grave constructions and the appearance of the few preserved whole ones, it is assumed that the northern (destroyed) part of the necropolis had similar stone constructions above the graves, whilst the remains of the deceased had the same ritual treatment as in the southern part. 13 graves in the northern part of the necropolis and 10 in the southern part were discovered and explored. The only difference between these two parts of the necropolis, from the aspect of grave ritual, was that in the southern part of the necropolis, burnt vessels were not noted, whilst in the northern part, several burnt vessels were discovered. That is to say, in grave 3 in the northern part of the necropolis, vitrified vessels (two bowls, a dual vessel, an oven model and an askos) were discovered with the remains of a deceased male of between 20 and 29 years of age which had, apparently, been burnt on a funeral pyre at a very high temperature together with the deceased and then, with the remains of the deceased, interred in the grave.6 Near the grave, on its south-eastern side, an area of scorched earth of an irregular shape, almost 2 m long, around 1 m wide and 6 For the information regarding the temperature of burning of the deceased, as well as for the anthropological analyses, I thank H. Miladinovi} and D. Bizjak who performed a complete anthropological analysis. 59 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 Fig. 2 and 3. Necropolis in Ranutovac, northern part, grave 7 and 10 Sl. 2 i 3. Nekropola u Ranutovcu, severni deo, grob 7 i 10 1–4 cm thick, onto which the still hot remains from the pyre were laid before being placed in the pit, was noted. Other differences in the ritual, the grave constructions and the manner of burial in these two parts of the necropolis, were not discerned. However, certain differences in the stylistic-typological characteristics of the ceramics were recorded. In the southern part of the necropolis (grave 21) was, for example, noted a beaker of a sharper profile, whilst that kind of profile was not evident in vessels in the northern part. In addition, in the southern part of the necropolis, on several cups and bowls (grave 17) were noted plastic triangular extensions on the rim, which were also not present on the vessels in graves in the northern part of the necropolis. A much larger number of vessels from both parts of the necropolis, however, have similar stylistic-typological characteristics. The most prevalent type of vessels in the necropolis is that of the pear-shaped beakers with two handles, with a ribbon-like cross section that exceed the rim (Pl. I/5–7, 12–14), followed by hemispherical or, more rarely, biconical cups (Pl. I/2) also with a handle, with a ribbon-like cross section that exceeds the rim (Pl. I/3, 11) and bowls. The bowls are of a hemispherical or S profile (more rarely conical), with ribbonshaped handles that exceed the rim, decorated with small button-like motifs with an indentation in the middle on the belly (Pl. I/1) or with triangular plastic extensions on the rim (Pl. I/10). Along with the globular amphorae with a cylindrical neck and four ribbon-shaped handles on the belly (Pl. II/15), which appear relatively often, a vessel was noted with four tunnel-like handles on the belly, standing on a tall bell-shaped hollow foot, with four elongated oval perforations (Pl. I/9), as well as a dual vessel that had been burnt at a high temperature (Pl. I/4). Besides the plastic decorations, an ornament resembling an incised hanging triangle combined with small circular pricks was noted on one fragment (Pl. I/17). Very indicative are the finds of baked clay which represent stylised oven models.7 Four oven models have been discovered in total, three of which have a similar shape, with a vertical cylinder and a horizontal part with a small handle on the back of it, and whose front end has the shape of a shovel (Pl. I/8), whilst one example was modelled into a bird shape, with small circular perforations on its body (Pl. II/16). The vessels from the necropolis mostly have a polished exterior either brown, ochre or dark brown in colour. 2. Donji Neradovac, Golo Rebro site, Vranje8 We are actually only discussing one fragment of a sharp profiled belly, with a triangular plastic handle (Pl. II/18), which was noted in a layer of river sediment above the Roman cultural layer.9 Clearly, the fragment reached this place in the runoff from the nearby hill, or it drifted there in the South Morava, whose current river 7 Bulatovi} 2013, 1–13, fig. 1–4. The site was explored by a team from the Institute of Archaeology in Belgrade (under the leadership of V. Filipovi}) in September 2011. 9 Filipovi}, Jeremi}, Bulatovi} 2012, 136–137. 8 60 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 bed is around 500 m south-east of the site (Map 1/2). The fragment is grey in colour and with a smooth finish. 3. Pavlovac, Kova~ke Njive site, Vranje10 The village of Pavlovac is situated around 7 km south of Vranje (Map 1/3), whilst the site of Kova~ke Njive comprises the south-eastern outskirts of Donji Pavlovac. The site is located outside the alluvial plain, on a terrain with moderate washout and gullies, around 250 m north of the current South Morava river bed. At this site were noted the remains of Middle and Late Neolithic settlements, along with Early Bronze Age finds, as well as pits from the Iron Age and the Middle Ages. The Early Bronze Age finds were discovered in a pit from the Iron Age with a more or less circular base, approximately 1.5 m in diameter and around 1.2 m deep. It would appear that we are dealing here with the remains of the cultural layer from the Early Bronze Age, which was destroyed by this pit. The pit was registered in the northernmost square of the explored sector, thus it is assumed that the cultural layer (settlement?) from the Early Bronze Age is located north of the explored part of the site, in the area which was not threatened by the building of the motorway. The finds from the pit comprise a fragment of a shallow conical vessel with semicircular plastic extensions on the rim (Pl. II/19), a fragment of a pear-shaped beaker with an arched handle of a ribbon-like cross section (Pl. II/Fig. 22), fragments of deeper vessels with plastic extensions on the rim (Pl. II/20, 21), as well as rough textured vessels of larger dimensions, decorated with imprints made with a fingertip on the fresh surface of the vessel (Pl. II/23), and rectangular handles with a split in the middle (Pl. II/24). The ceramics were mostly dark brown or dark in colour, with a polished surface. Fig. 4. Ritual pit in Davidovac Sl. 4. Ritualna jama u Davidovcu as well as a larger fragment of an amphora (almost a half) were found. The belly of the amphora is elongated, with one vertically and one horizontally placed handle of ribbon-like cross section, and a small and slightly everted rim (Pl. III/29). The drum is modelled into an hourglass shape, with four oval perforations on each cone, which were probably used for attaching and tightening the skin (Pl. III/30). The height of the drum is 16.5 cm, with almost the same diameter at both open ends. The dimensions of the perforations on the lower cone are 4 x 2 cm, whilst on the upper cone they are slightly smaller at 3 x 1.5 cm. The bowls have a hemispherical shape with a short, oblique, everted rim or a rim with a slight S profile, and handles with a ribbonlike cross section which exceed the level of the rim (Pl. II/25–28). One of the bowls is decorated with a plastic ornament on the rim, which resembles two joined hornlike extensions (Pl. II/25). The bowls are baked and have a black-ochre colour, the amphora is grey and dark brown in colour, whilst the drum is black. The ceramics have semi polished or polished surfaces. Although the majority of the area of this site was explored, aside 4. Davidovac, Gradi{te site, Bujanovac11 Davidovac is situated around 10 km south-west of Vranje, on the widest part of the Vranje–Bujanovac valley (Map 1/4). The Gradi{te site is located in the northwestern outskirts of the village, on top of the hill to the west of the motorway, as well as on the southern, northern and eastern slopes of the hill. This is a terrain that passes from the alluvial plain to the plateau, and is ideal for building a settlement. On the south-western edge of the site, not far from the Roman thermae, a shallow, circular, Early Bronze Age pit was explored (Fig. 4) which was approximately 0.2 m deep and 0.7 m in diameter, in which a drum (tarabuka) made of baked clay, three larger and two smaller fragments of different bowls, 10 The site was explored by a team from the Institute of Archaeology in Belgrade and the Department of Archaeology of the Faculty of Philosophy in Belgrade (under the leadership of J. Vukovi} and A. Bulatovi}) in the period October – December 2011. 11 The site was explored by a team from the Institute of Archaeology in Belgrade in the period of August – December 2011. The leaders were S. Petkovi} (Antiquity) and A. Bulatovi} (prehistory). 61 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 from this pit there were no other finds from this period. It is assumed that the settlement which this pit, of a probably ritual character, belongs to (at least based on its content and context), is situated west of the explored part of the site, on the vast plateau outside the endangered zone. 5. Kokino Selo, Taticev kamen site, Kumanovo The site is located on a dominant mountain elevation, at an altitude of about 1000 m, on the left side of the Kumano–Arbana{ko road, around 35 km northeast of Kumanovo (Map 1/5).12 The site has been previously published, but the finds presented in this paper have not yet been published. They are significant because they illustrate some characteristic stylistic elements of ceramics from the Early Bronze Age in this territory.13 The site comprises two distinct areas, both sacral complexes – the megalithic observatory and a ritual space.14 The megalithic observatory is situated on the top of the site and consists of a lower flattened plateau on a rock, with four stone chairs facing east, and some higher rocks on the eastern horizon, approximately 20 m from the plateau, on which some triangular and rectangular notches are marked for tracking the movement of the sun. The ritual space is located on the northern slope of the site, below the peak and is approximately 80 m long and 50 m wide. Pits were noted there, sometimes with circular stone constructions over and around them, in which were stored offerings such as whole vessels or fragments of vessels (Fig. 5), stone tools, parts of anthropomorphic and zoomorphic figurines, fragments of house daub and other items. Since the terrain drops very steeply towards the north, it was retained with several layers of dry stone wall. It was noted that the rituals related to the sacrificial pits were in use over a longer period, from the Early to the Late Bronze Age, and that the oldest pits were located in the eastern part of the ritual space. The most numerous finds from the ritual space are the ceramic vessels. Certainly, the most common types of vessels from the Early Bronze Age are the beakers with two handles, mostly of a ribbon-like cross section, and bowls with rounded shoulders and inverted rims. On rare examples of beakers the handles exceed the level of the rim, but more numerous are examples with handles at the same level as the rim or slightly exceeding it (Pl. III/38).15 The bowls with inverted rims often have them decorated with plastic extensions of different shapes (rectangular, hornlike or trapezoidal), by far the most numerous of which are the extensions of triangular Fig. 5. Shrine in Kokino Selo, ritual pit with offerings of beakers and a funnel, detail. Sl. 5. Svetili{te u Kokinom Selu, obredna jama sa darovima u vidu pehara i levka, detaq shape (Pl. III/32, 33, 35).16 These bowls often have one or two ribbon-shaped handles on the rim (Pl. III/31), but there are examples with horizontally or obliquely placed handles of an oval cross-section which exceed the level of the rim (Pl. III/34). Besides the aforementioned vessels, globular amphorae with a cylindrical neck, hemispherical vessels on a hollow conical foot, bowls with an S profile and others were also noted at this site. The ornaments are rare and consist of plastic decorations such as button-like motifs with an indent in the middle, plastic vertical ribs, horseshoe shaped handles and the already mentioned plastic extensions on the rims of the vessels. Decorations such as incised triangles, zigzag or wavy lines, which usually follow the small circular indentations, are scarce (Pl. III/36).17 12 Stankovski 2002, 29–48; Bulatovi}, Stankovski 2012, 85–95 and quoted literature. 13 The site of Taticev kamen has been explored since 2001, under the leadership of J. Stankovski, to whom I am grateful for the provided material. 14 For further details regarding the observatory and the ritual space see: Bulatovi}, Stankovski 2012, 269–277. 15 Bulatovi}, Stankovski 2012, T. LIX/4, LX/15, 16, LXI/20, LXII/36. 16 Bulatovi}, Stankovski 2012, T. LX/9–11, LXI/18, 21, 22, 24–28. 17 Bulatovi}, Stankovski 2012, T. LIX/1, 6, 7, LX/13, 14, LXI/20, 23, LXII/32–44. 62 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 culture, as well as with the sites in western Bulgaria, rather than with the Armenochori culture in the south,22 whilst on the other hand, at the site in Lopate, not far from Kumanovo, ceramics exclusively from the Armenochori culture were noted.23 The only certain thing is that, at least for a brief period, these cultures were concurrent, given that, at the sites in Kokino Selo and Pelince, the finds with visible influences from the Armenochori culture were noted.24 Unfortunately, since we still do not have at our disposal any absolute dates, the question of the exact mutual relationship between the communities from these cultures remains open. The necropolis discovered in Ranutovac, as well as the finds from the other sites presented in this paper have, however, served to supplement our current understanding, at least in the sense of the interpretation of the material and spiritual culture in this region in the Early Bronze Age, which will be further discussed later in this paper. The most indicative type of vessel from this period, characteristic of both cultures, is a pear-shaped beaker with two handles. It was noted, however, that these beakers in Armenochori culture are mostly of a slimmer shape, and that the handles considerably exceed the level of the rim of the vessel, whilst in Bubanj–Hum III culture they are mainly stocky, with handles predominantly at the same level as the rim, or only slightly above it.25 The fact that a large number of these vessels of a recognisable appearance was noted in a wide area of the central Balkans in the Early Bronze Age, led some authors to define a specific culture named after Apart from the vessels, two oven models, of a simpler form than the examples from the necropolis in Ranutovac, were also noted as offerings in the ritual space. *** The sites that were discovered during the protective archaeological campaign, conducted along the E 75 motorway, especially the necropolis in Ranutovac, provided important information about the life of the Early Bronze Age communities in the territory of southeastern Serbia. First of all, the location of the sites of Kova~ke Njive in Pavlovac and Gradi{te in Davidovac confirmed that, in the Early Bronze Age, in Southern Pomoravlje, the settlements were formed mainly on the terraces of larger rivers,18 assuming the cultural layer noted in Pavlovac and the pit in Davidovac actually belong to the settlements. From a geological perspective, these settlements were formed on the edge of the alluvial plain, in places where they pass into a terrain of moderate washout and gullies, an area where the configuration of the terrain gently rises. They appear, at least based on current information, exclusively on the left bank of the South Morava.19 Certainly, one of the most significant results obtained throughout the exploration of these sites was the large number of different types of vessels, which significantly added to the knowledge of the material culture of the Early Bronze Age communities in this territory. Previous researchers were understandably cautious throughout the cultural determination of the Early Bronze Age communities around the upper and middle course of the South Morava, since finds with characteristic features were lacking. Based on the small number of existing finds, it was assumed that the Bubanj–Hum III culture stretched to the Skopje Valley in the south, where it came into contact with the Armenochori culture.20 Just a few years ago, findings were published which indicate that this territory belonged to a border area where the influences of Bubanj–Hum III culture21 from the north and Armenochori culture from the south were combined, which more or less confirmed these assumptions. However, what remained were the problems of delimitation and the chronological relationship between these two cultures, which were again made topical with explorations of the Early Bronze Age sites in north-eastern Macedonia. Specifically, ceramic production from the sites in Kokino Selo and Pelince pointed to much stronger cultural links with the north, the Bubanj–Hum III 18 Bulatovi}, Stankovski 2012, 197–198. A possible reason for this is the motorway which passes along the left bank of the river, whilst the right bank remains unexplored. However, it is more likely that the left bank was more suitable for living, since it faces east and, as such, has a more favourable position regarding the cardinal points, which was an important parameter in choosing a place for a settlement in prehistory. 20 The finds from Armenochori and neighbouring sites were singled out and classified into the Early Bronze Age by W. Heurtley (1939, 85), however, the Armenochori culture was ultimately defined by M. Gara{anin (1958, 122–123). About the prevalence of the culture see: Gara{anin 1983, 720. Macedonian authors for this culture in the territory of Pelagonia and southern Macedonia use the name Karamani or Pelagonian cultural group (Kitanoski 1978, 51; Mitrevski 2003, 44). 21 Bulatovi} 2007, 35. 22 Mitrevski 2003, 45; Alexandrov 2007, 229; Bulatovi} 2011, 8. 23 Sanev 1999, 130, 133–134. 24 Bulatovi}, Stankovski 2012, 341. 25 Bulatovi} 2011, 4. 19 63 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) them – the culture (horizon) of two handled beakers.26 It is interesting that, at the necropolis in Ranutovac, even though it is located more than 50 km north of the northernmost site of the Armenochori culture (Lopate near Kumanovo), finds characteristic of the Armenochori culture were noted in the largest number. These are mainly beakers with handles that exceed the level of the rim (Pl. I/5, 12, 14), although the shape of some of these beakers actually represents a variant of both the Armenochori and Bubanj–Hum III type of beakers (Pl. I/6–7, 13). The finds also included a dual vessel (Pl. I/4), a globular amphora with a cylindrical neck (Pl. II/15), cups with a handle that greatly exceeds the level of the rim (Pl. I/2, 3, 11), a vessel on a tall foot (Pl. I/9), an askos and others. The aforementioned type of beaker with handles that exceed the level of the rim appear in a wide area from eastern Albania (Maliq III, Tren IIIa),27 to Thrace (Ezero I–II, Junacite III, Razkopanica V),28 and from northern Greece (Armenochori, Kastanas – from layer 26, Xeropigado, Archontiko, Sitagroi Vb),29 through Macedonia (Bukri, Kravari, Suvodol, Karamani, Lopate),30 to Southern Pomoravlje (Ranutovac), also sporadically in the homeland of the Bubanj–Hum III cultural group (Bubanj),31 and in the south of Greece.32 Similar beakers, with handles placed on the vessel’s shoulder, have already appeared in a larger number from the settlement Troy II.33 Along with these beakers, in the southern part of the necropolis in Ranutovac, were noted globular amphorae with a cylindrical neck, also characteristic of the Early Bronze Age in the territory of Troy (starting from the settlement of Troy II),34 through Thrace and northern Greece, to Macedonia.35 An askos was found in the southern part of the necropolis, and these vessels are characteristic of Early Bronze Age Thrace and the northern Aegean,36 although they are also known from previous periods.37 A classic shape of askos, as is the case of the example from Ranutovac, appears in Troy only from settlement IV, which chronologically coincides with the EH III period, that is the EBA III period, to which settlements in south-eastern Serbia belong.38 As well as in Troy, askoi of the Early Bronze Age (EH II/III) were noted in the previously mentioned territory that comprises Thrace (Bereketskata Mogila, Djadovo – II stratum, Karanovo VII),39 northern Greece (Agios Mamas)40 and eastern Albania (Maliq IIIa).41 The dual vessel (Pl. I/4), noted in grave 1 in the northern part of the necropolis in Ranutovac, however, is not characteristic of the material culture of the communities in this territory. These vessels appear here only sporadically, and mostly with STARINAR LXIV/2014 ceramic inventory characteristic of this territory (Razkopanica VI, the oldest stratum at Donja Brnjica, Nova~ka ]uprija and Star Karaorman).42 These kinds of vessels, however, very often occur in the Middle and Late Bronze Age in the northern and central part of the Balkans. The vessel that stands out, by virtue of its appearance, is most certainly the one on a tall hollow bellshaped foot (drum?), found in grave 6 in the northern part of the necropolis (Pl. I/9). The vessels on tall feet, but without perforations, were noted at the sites of Tsoungiza and Zygouries,43 whilst those on shorter feet, with similar perforations to those on the vessel from Ranutovac, were also noted at Zygouries and Troy I,44 which points to a southern origin of the tall bell-shaped feet.45 The finds that unambiguously connect the necropolis in Ranutovac with the communities that lived in the south, in the territory of the Armenochori culture, are the stylised oven models (Pl. I/8, Pl. II/16). These models were noted solely in the territory of northern 26 Stoji} 1996, 248–250; Nikolova 1999, 233–234. Prendi, Bunguri 2008, T. VII, XXXIV/3, XLVII/6. 28 Le÷akov 1992, 18, 44, obr. 4/m, 17/e; Nikolova 1999, 227, fig. 10. 2; Detev 1981, fig. 37. 29 Heurtley 1939, cat. no. 318–349; Aslanis 1985, Taf. V/15; Maniatis, Ziota 2011, fig. 8; Papaefthymiou-Papanthimou, PilaliPapasteriou 1997, 152, fig. 4; Rutter 1982, fig. 1/1; Renfrew et al. 1986, fig. 13. 20, 13. 20/25. 30 Simoska, Sanev 1976, kat. br. 239, 242; Simoska 1984, sl. 8, 13; Simoska 1984a, sl. 4; Sanev 1999, Pl. VI. 31 Gara{anin, \uri} 1983, kat. br. 162, 164. This type of beaker, together with a dual vessel, was noted in the oldest stratum at D. Brnjica (Srejovi} 2002, 169, sl. 25). 32 Rutter 1982, fig. 1/1; Rutter 2008, fig. 15. 33 Blegen et al. 1950, type A43, fig. 370a. 34 Blegen et al. 1950, fig. 389/35. 485, 395/35. 521. 35 Le÷akov 1992, obr. 17/i, 26/a1; Aslanis 1985, T. 111/9; Blegen 1928, fig. 95–97; Bulatovi}, Stankovski 2012, T. LIX/1. 36 Blegen et al. 1951, fig. 154b/D29. 37 Georgiev et al. 1979, obr. 174/`, z. 38 Kokowski 2010, 66. 39 Le÷akov 1992, obr. 11/a, b, 18/`, 26/i. 40 Heurtley 1939, cat. no. 191, 192; Aslanis 1985, Taf. 110/11, 119/8. 41 Prendi, Bunguri 2008, T. XXII/7. 42 Detev 1981, fig. 37; Srejovi} 2002, 169, sl. 25; Krsti} et al. 1986, T. XVI/1–3; Nacev 2009, kat. br. 65. 43 Pullen 2011, fig. 6. 75–6. 76; Blegen 1928, fig. 116. 44 Blegen et al. 1950, fig. 262/27; Blegen 1928, fig. 117. 45 Regarding the possibility that this object actually represents a drum will be discussed further in the text. 27 64 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 and Ezero (strata II–III), 54 and the earliest examples date back to Troy I.55 The bowls with horizontal handles were discovered at Sitagroi (stratum Vb),56 Agios Mamas (pit D28), Kritsana (stratum IV), Kastanas (stratum 24),57 Karanovo VII,58 as well as many other sites throughout this territory. Another interesting find is a ceramic, hourglass shaped, bottomless object, found in the pit in Davidovac which, according to analogies with other prehistoric sites, represents a drum (tarabuka). Similar items are known in prehistoric cultures all over Europe59 and, based on the shape and dimensions of these drums, the vessel on a tall, hollow, bell-shaped foot from the necropolis in Ranutovac could also be interpreted as a drum (Pl. I/9), except that the skin on the example from Ranutovac was pulled tight by tunnel-like handles,60 and on the example from Davidovac using the oval perforations.61 The spiritual life of the communities of the Early Bronze Age in this area is already well-known, primarily due to the ritual places in Pelince and Kokino, as well as the observatory in Kokino. The custom of storing offerings, in the form of whole or parts of vessels, or objects of stone, bone, ceramics and the like, either in the ground or on a rock with circular stone constructions around and above the pits, as noted in Kokino (Fig. 5) and Pelince has, for now, no analogy in the immediate vicinity. A similar ritual place, where the ritual pits with gifts were surrounded by a large trench, was discovered in Kazanlak, near Stara Zagora.62 The Greece, eastern Albania and Macedonia, and, chronologically, they are tied to the Early Bronze Age and the beginning of the Middle Bronze Age.46 In Ranutovac, as many as four such examples were found which, considering also the stylistic-typological features of other ceramics from the necropolis, confirms that the necropolis belongs to the Armenochori cultural circle and excludes the possibility that they are only culturally influenced or imported from the south. The ceramics from the other sites, on the whole, also resemble the ceramics of the Early Bronze Age (EBA III) from the territory of northern Greece, although they also have elements of the Bubanj–Hum III culture (Davidovac, Pavlovac). The vessel with the globular belly and large ribbon-shaped, obliquely placed handles, from the pit in Davidovac, is reminiscent of those from Kastanas (strata 22 and 23).47 The bowls with a slight S profile, or with an inverted rim with plastic decorations were noted at all the sites presented in this paper, with the exception of Golo Rebro. These decorations are one of the main characteristics of the Early Bronze Age in Southern Pomoravlje and, judging by the analogies, also in north-eastern Macedonia (Kokino, Pelince) and western Bulgaria,48 with sporadic appearances in Pelagonia (Bukri) as well.49 According to some authors, these decorations survived from the previous period (Bubanj–Hum II, or Glina IV in western Bulgaria),50 and they are of an autochthonous origin.51 There is, however, also the possibility that they originated from the Aegean Early Bronze Age, since these types of decorations (identical to the one on the bowl from Pavlovac – Pl. III/35) were frequently present on the bowls from the settlement of Troy I, although some examples from Troy were decorated with engraved representations of the human face.52 This second assumption is less likely since the chronological period between Troy I and the Early Bronze Age (EBA III) in the interior of the Balkans is quite long, and within that period similar decorations on ceramics, which could possibly link these two cultures, have not been seen, meaning they obviously originate from the north and east, from the Bubanj–Hum II culture, or from the culture of the Early Bronze Age (EBA III) of western Bulgaria. The bowls with an inverted rim and short ribbon-shaped handles that exceed the level of the rim (Pl. III/31) or with either horizontally or obliquely placed handles on the rim (Pl. III/34) are, on the other hand, a typical occurrence of the Early Bronze Age in Thrace and the northern Aegean. The bowls with vertical handles were noted at Kastanas from stratum 2653 46 For detail regarding the purpose, chronology, distribution and cultural affiliation of these oven models see: Bulatovi} 2013, 1–13, with quoted literature. 47 Aslanis 1985, Taf. 40/8, 57/16. 48 Alexandrov 1998, fig. 3/8, fig. 5. 49 Simoska 1984, sl. 6. 50 Alexandrov 2007, 228–229, Pl. VIII. 51 Bulatovi} 2011, 8–9. 52 Blegen et al. 1950, fig. 253/18. 53 Aslanis 1985, Taf. 8/18. 54 Le÷akov 1992, obr. 4/a. 55 Blegen et al. 1950, fig. 266/1. 56 Renfrew et al. 1986, fig. 13. 27/16, 19. 57 Aslanis 1985, Taf. 12/12, 106/4, 111/3, 112/3. 58 Le÷akov 1992, obr. 26/o. 59 Gergov 2011, 269–273; Wyatt 2008, 1–22; Aiano 2006, 31–42. 60 Gergov 2011, obr. 4; Aiano 2006, fig. 1–8; Wyatt 2008, fig. 1. 61 Gergov 2011, obr. 1, 2. 62 Nikolov 2008, 23–32. 65 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 Sites of the Bubanj–Hum III – Pelince – Pernik culture Lokaliteti Bubaw–Hum III – Pelince – Pernik kulture Sites of the Armenochori culture Lokaliteti Armenohori kulture Sites with elements of both cultures Lokaliteti sa elementima obe kulture Map 2. More significant sites of Bubanj–Hum III – Pelince – Pernik culture and Armenochori culture Karta 2. Va`niji lokaliteti Bubaw–Hum III – Pelince – Pernik kulture i Armenohori kulture finds from these pits belong to the Early Bronze Age III of Thrace,63 which chronologically corresponds to the sites in Kokino Selo and Pelince. Pits from the Bronze Age containing different offerings are characteristic of shrines in the region of the Southern Rhodopes, but they also appear in western Bulgaria,64 which could point to the direction through which this custom arrived in the southern parts of the central Balkans as well. Some similarities with these rituals have been observed in Cretan mountain shrines, from the beginning of the MMII period, in which a circular space was formed using river stones, where offerings were deposited, similar to those from Kokino Selo and Pelince.65 The shallow pit discovered in Davidovac probably belongs to that sacral idea of storing offerings in the ground, into the bosom of the Great Mother, the goddess of fertility and protector of crops, whereby several, almost whole vessels and one drum were deposited, although around this pit a stone construction was not noted (Fig. 4). These customs which, according to the analogies of finds and the distribution of the sites, belong to the Aegean–Thracian cultural circle, later expanded into the interior region of the Balkans.66 The necropolises from the Early Bronze Age with exclusively cremated deceased, as in Ranutovac, 63 64 65 66 66 Nikolov 2008, 32. Kostova 2006, 2. Nowicki 1994, 34. Pekovi}, Jevti} 2007, 125–140. Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 beakers with two handles were found, similar to the examples from Kokino Selo, Ranutovac and others, was dated to the periods 2275–2139; 2199–2135 and 2196–2050 BC (23rd–21st century BC). At the same site, oven models similar to those in Ranutovac were also discovered, which connect these two sites from the perspective of material culture as well.74 Maliq – strata IIIa–IIIb, as well as Kastanas – strata 26–22b (the most numerous analogies with the ceramics from our sites were noted in stratum 22b) in which was identified a large number of stylistic-typological characteristics identical or similar to the ceramics from the sites of the upper course of the South Morava and from Gornja P~inja, were also dated to the period of the final quarter of the 3rd and the beginning of the 2nd millennium BC.75 To approximately the same period we can date the presented finds from Davidovac, Neradovac and Pavlovac, whilst it should be noted that the mutual chronological relation between these sites cannot be precisely determined without absolute dates. The dates for the Bubanj–Hum III culture, whose cultural elements, together with the elements of the Early Bronze Age of western Bulgaria,76 dominate on the ceramics from the sites in Pelince and Kokino Selo, appear to be slightly younger, at least in the far north of its territory. Thus, the pit from the Early Bronze Age at Nova~ka ]uprija, with finds characteristic of the Bubanj–Hum III culture, was determined by calibrated dates to 2160–1850 BC,77 whilst the oldest stratum have, so far, not been noted, either in these areas, or in any distant regions.67 In this period the deceased were usually inhumed, very rarely cremated, and only then in the biritual necropolises in which inhumation dominated. At the Xeropigado necropolis, in the north-west of Greece, for example, where more than 200 graves with inhumation have been discovered, only 12 were cremations, or around 5%.68 Necropolises with only inhumed deceased were noted in Thrace. In Albania, tumuli with inhumed deceased are prevalent, whilst graves with cremated deceased are extremely rare and appear solely in the north of Albania in the Early Bronze Age, whereas in the Middle Bronze Age they also appear in the south of Albania.69 In the area of the central Balkans, in the Early Bronze Age, the custom of solely cremation burials has not been noted, but there are numerous necropolises with biritual burials. In the area of the Beloti}–Bela Crkva culture, several graves with cremated deceased were noted,70 and at the necropolises under the tumuli in the West Morava valley, both rituals were more or less equally present.71 The closest territory in which, around this time, the almost exclusive ritual of cremation burials was noted is Oltenia.72 However, due to the large distance between the regions, it cannot be linked with this custom in the far south-east of Serbia. For now, the necropolis in Ranutovac, where exclusively cremated deceased were noted, remains the only one of its kind in the territory of the central Balkans, and it is possible that the reason for this sole ritual is social, micro regional, sacral, economic or some other in nature, and it does not necessarily have to be linked with the cultural affiliation of the communities that practiced it. Nevertheless, it has to be mentioned that the spiritual concept of storing offerings in shallow pits with circular stone constructions at the necropolis of Ranutovac closely resembles the rituals noted in nearby Kokino Selo, Pelince and Davidovac, as well as the other sacred places which belong to the Aegean–Thracian sacral area. The necropolis in Ranutovac can, at this time, only be chronologically determined by relative dates. Closest to this necropolis, according to both the burial custom (both inhumation and cremation) as well as the grave offerings (beakers with two handles), is the necropolis of Xeropigado, in the north-west of Greece, which has been chronologically determined using absolute dates. This necropolis is dated to a period of about seven centuries, starting at the end of the 25th century BC.73 The house at Archontiko, in which 12 67 In central Greece and in the Peloponnese, for example, in the Early Bronze Age (EH I–III) necropolises with burnt deceased were not noted (Weiberg 2007, 188), nor in Thrace (Todorova 2003, 295–296) or Muntenia (Schuster 2003, 109–119). 68 Maniatis, Ziota 2011, 461. 69 Prendi, Bunguri 2008, 225. 70 Gara{anin 1973, 361–367. 71 Dmitrovi} 2009, 115. 72 Schuster 2003, 138. 73 Maniatis, Ziota 2011, 461–478. 74 Papleuqumion–Papanqimon, Pilalh–Papasterion 1998, 87. 75 Prendi, Bunguri 2008, 274; Aslanis 1985, 318, Abb. 130. 76 For this culture that covered the territory of western Bulgaria, Southern Pomoravlje, eastern Serbia and north-eastern Macedonia, the proposed name was Bubanj III – Pernik culture (Nikolova 1999, 233), that is Bubanj–Hum III – Pelince II–III – Pernik culture (Bulatovi} 2011, 11). S. Alexandrov also considers that the territory of south-western Bulgaria in the Early Bronze Age (EBA III) belonged to the cultural area of the central Balkans (Alexandrov 1998, 231). 77 Krsti} et al. 1986, 34. 67 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) at Ljuljaci is dated to 1950–1900 BC.78 Stratum III at Junacite, where beakers of the Bubanj–Hum III type were noted along with other finds, is dated to 2300–2000 BC.79 As can been seen, all the known dates for both cultures are derived from different sites, which are, additionally, quite far apart from each other. Therefore, based on these dates, their chronological relation cannot be defined. The only certain thing is that the territory of the upper course of the South Morava and north-eastern Macedonia were, at least for a short while, a contact zone between the Bubanj–Hum III and Armenochori cultures (Map 2), whilst their more precise chronological relation can only be determined with a series of absolute dates, from several indicative sites. According to some elements noted on the ceramics from absolutely dated sites of the Early and Middle Bronze Age in the central Balkans, such as the vessels with horizontally placed handles with horn-like extensions, beakers with two handles which exceed the level of the rim (Ljuljaci – second horizon),80 or dual vessels (Nova~ka ]uprija),81 it can be determined that the period in which the Armenochori culture spread towards the north and actually met the Bubanj–Hum III culture in the region of Kumanovo and in the upper course of the South Morava, was somewhere between the end of the 3rd and the beginning of the 2nd millennium BC. STARINAR LXIV/2014 Greece, and eastern Albania in the south, to northern Macedonia in the north. The results of the explorations of the new sites and the finds presented in this paper confirmed the assumptions that the Armenochori culture included, at least in one period of its existence, the territory of Southern Pomoravlje as well. Based on the available data, it cannot be determined, with any precision, the time at which this group started its expansion towards the north, that is to say when it spread to Southern Pomoravlje. However, according to the available data and absolute dates, it is generally assumed that it happened at sometime towards the end of the 3rd or the beginning of the 2nd millennium BC. Based on the topography of the Early Bronze Age sites in the territory of north-eastern Macedonia and the upper course of the South Morava, it can be said that the Early Bronze Age was a relatively peaceful period, as the settlements were set in lowland parts of the territory, and hill fort settlements were absent. In the mountainous region a larger, possibly even regional shrine (Kokino) was formed, where stylistic-typological elements of both cultures were noted, whilst smaller ritual places (Davidovac, Pelince) were also organised in the lowland parts of the territory, probably near settlements. Such topographical features of the sites, considering the presence of finds from both cultures in some of them, point to the fact that, even if there were certain cultural or ethnic changes in this period, such as cultural penetrations or migrations from the south, they were not violent, but gradual and led to the peaceful coexistence of the communities of both cultures. These two, obviously related cultures, with pottery inventory of very similar stylistic-typological characteristics, probably arose from similar cultural roots. However, their cultural-chronological relation, unfortunately, cannot be more precisely defined without a series of absolute dates and further research. Within the cultures from the end of the Early and into the Middle Bronze Age in the central Balkans, however, the elements of the Armenochori culture are recognisable, which would indicate that, apart from the autochthonous Bubanj–Hum III culture, this culture also left its mark on the genesis and the development of the Middle Bronze Age in the Balkans. CONCLUSION Based on all previously presented data, as well as the data presented in this paper about the material, and especially the spiritual culture of the communities from this territory, if we take a wider view of the Early Bronze Age Balkans, it is possible to discern a certain cultural unity in the territory of the northern Aegean, Thrace, northern Greece and the Central Balkans, which can be defined as a uniform cultural zone. This cultural unity is the result of the development of the autochthonous cultures, but also the interaction of the cultures from the Aegean circle with cultures from the interior of the Balkan Peninsula throughout the previous periods. Within this uniform cultural zone, among others, can also be defined two almost concurrent cultural manifestations, the Bubanj–Hum III – Pelince – Pernik culture (complex), which comprises western Bulgaria, Southern Pomoravlje, eastern Serbia and north-eastern Macedonia, and the Armenochori (Pelagonian) culture, whose territory stretches from northern and western 78 79 80 81 68 Bogdanovi} 1986, 70. Nikolova 1999, 226. Bogdanovi} 1986, kat. br. 50, 51. Krsti} et al. 1986, T. XVI/1–3. Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 BIBLIOGRAPHY: Bulatovi}, Kapuran 2013 – A. Bulatovi}, A. Kapuran, Praistorijski lokaliteti na trasi autoputa E 75 – prethodni izve{taj, Leskova~ki zbornik LIII, Leskovac, 70–79. Detev 1981 – P. Detev, Le tell Razkopanica, Cultures préhistoriques en Bulgarie, Izvesðiý na Arheoloái~eskiý insðiðuð XXXVI, Sofiý, 141–188. Dmitrovi} 2009 – K. Dmitrovi}, Bronzano doba u ~a~anskom kraju, neobjavqen magistarski rad odbrawen na Filozofskom fakultetu u Beogradu. Gara{anin 1958 – M. Gara{anin, Neolithikum und Bronzezeit in Serbien und Makedonien, Bericht der Romisch-Germanischen Kommission des Deutschen 39, 1–130. Gara{anin 1973 – M. Gara{anin, Praistorija na tlu Srbije, Srpska kwi`evna zadruga, Beograd. Gara{anin 1983 – M. Gara{anin, Grupa Bubanj– Hum III, u: A. Benac (ur.), Praistorija jugoslavenskih zemalja IV, ANUBIH, Sarajevo, 719–722. Gara{anin, Ajdi}, Dejanovi} 1971 – M. Gara{anin, R. Ajdi}, D. Dejanovi}, Praistorijske kulture Pomoravqa i isto~ne Srbije, katalog izlo`be, Narodni muzej, Ni{. Gara{anin, \uri} 1983 – M. Gara{anin, N. \uri}, Bubaw i Velika Humska ^uka, katalog izlo`be, Narodni muzej, Ni{. Georgiev et al. 1979 – G. Georgiev, N. Ý. Merpert, R. V. Katin~arov, D. G. Dimitrov (eds.), Ezero, rannobronzovoðo seli÷e, Bølgarskata Akademiý na naukite, Sofiý. Gergov 2011 – V. Gergov, Kerami~ki tarambuki ot praistori~eskoto seli÷e Teli{–Redutite, u: Ý. Boýxiev, S. Terziiska-Ignatova (ur.), Zlaðnoðo íeðo hilâoleðie, NIAM–BAS, Sofiý, 269–273. Heurtley 1939 – W. A Heurtley, Prehistoric Macedonia, Cambridge. Kitanoski 1978 – B. Kitanoski, Ranobronzenodopski grobovi od Varo{ kaj Prilep, MAA 4, 33–55. Kokowski 2010 – A. Kokowski, Troja, Sen Henryka Schliemanna, Muzeum im. Stanislawa Staszica – Instytut Archeologii Uniwersytetu MCS w Lublinie, Lublin. Kostova 2006 – K. Kostova, Sve÷eni mesða i riðualni sðrukðuri oð bronzovaða eíoha v Drevna Trakiý, avtoreferat. (http://www. scribd. com/doc/ 53922116/K–Kostova-avtoreferat – pristupqeno 4. 2. 2014.) Krsti} et al. 1986 – D. Krsti}, A. Bankoff, M. Vukmanovi}, F. Winter, Praistorijski lokalitet Aiano 2006 – L. Aiano, Pots and drums: an acoustic study of Neolithic pottery drums, Eurorea 3, EXARC, Eindhoven, 31–42. Alexandrov 1995 – S. Alexandrov, The Early Bronze Age in Western Bulgaria: periodization and Cultural Definition, in: D. W. Bailey and I. Panayotov with S. Alexandrov (eds.), Prehistoric Bulgaria, Monographs in World Archaeology No. 22, Madison Wisconsin, 253–270. Alexandrov 1998 – S. Alexandrov, Pottery of the Early Bronze Age in Western Bulgaria, in: M. Stefanovich, H. Todorova, H. Hauptmann (eds.) J. Harvey Gaul in Memoriam, Sofia, 223–233. Alexandrov 2007 – S. Alexandrov, Bronze Age materials from Bagacina (North–West Bulgaria), in: M. Stefanovich, C. Angelova (eds.) Prae, in honorem Henrieta Todorova, Sofia, 103–129. Aslanis 1985 – I. Aslanis, Kastanas, die Fruhbronzezeitlichen Funde und Befunde, Berlin. Blegen 1928 – C. Blegen, Zygouries, a prehistoric settlement in the valley of Cleonae, The American School of Classical Studies at Athens, Harvard University press. Blegen et al. 1950 – C. Blegen, J. Caskey, M. Rawson, J. Sperling, Troy, general introduction the first and second settlements, vol. I. part 2: plates, New Jersey. Blegen et al. 1951 – C. Blegen, J. Caskey, M. Rawson, Troy, the third, fourth, and fifth settlements, vol. II, part 2: plates, New Jersey. Bogdanovi} 1986 – M. Bogdanovi} Quqaci, naseqe protovatinske i vatinske kulture, Narodni muzej, Kragujevac. Bulatovi} 2007 – A. Bulatovi}, Vrawe – kulturna stratigrafija praistorijskih lokaliteta u Vrawskoj regiji, Arheolo{ki institut, Narodni muzej, Beograd–Vrawe. Bulatovi} 2011 – A. Bulatovi}, Relations between Cultural Groups in the Early Bronze Age in Southeastern Serbia, Western Bulgaria and North-eastern Macedonia, Archaeologica Bulgarica, XV, 2, Sofia, 1–13. Bulatovi} 2013 – A. Bulatovi}, Oven models from Early Bronze Age Settlements in Central and Southern Parts of the Balkan Peninsula, Archaeologia Bulgarica XVII, 1, Sofia, 1–13. Bulatovi}, Stankovski 2012 – A. Bulatovi}, J. Stankovski, Bronzano doba u basenu Ju`ne Morave i u dolini P~iwe/Bronzeno vreme vo basenot na Ju`na Morava i dolinata na P~iwa, Arheolo{ki institut – N.U. Muzej, Beograd–Kumanovo. 69 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 Village in Northeast Greece, vol. 1, The Institute of Archaeology, The University of California, Los Angeles. Rutter 1982 – B. J. Rutter, A group of distinctive pattern-decorated Early Helladic III pottery from Lerna and its implications, The Journal of the American School of Classical Studies at Athens, vol. 51, 459–488. Rutter 2008 – J. B. Rutter, Anatolian Roots of Early III Drinking Behaviour, in: H. Erkanal, H. Hauptmann, V. Sahogly, R. Tuncel (eds.), The Aegean in the Neolithic, Chalcolithic and the Early Bronze Age, Ankara University, Ankara, 461–482. Sanev 1999 – V. Sanev, Pottery in Favour of Designating the Cultural Borders of the Early Bronze Age in Macedonia, in: Macedonia and the Neighbouring Regions from 3rd to 1st milenium b. c., papers presented at the International Symposium in Struga 1997, Museum of Macedonia, Skopje, 127–136. Schuster 2003 – S. Schuster, Zur Bestattungsweise in Südrumäniem in der Bronzezeit, u: R. Vasi} (ed.) Sahranjivanje u bronzano i gvozdeno doba, Simpozijum odr`an u ^a~ku 2002. godine, Narodni muzej u ^a~ku i Arheolo{ki institut u Beogradu, ^a~ak, 109–130. Simoska, Sanev 1976 – D. Simoska, V. Sanev (ur.) Praistorija vo centralna Pelagonija, katalog izlo`be, Naroden muzej, Bitola. Simoska 1984 – D. Simoska, Visok rid, praistorijska povekeslojna naselba (preliminaren izve{taj), Zbornik na trudovi 4–5, Bitola, 5–20. Simoska 1984a – D. Simoska, Arheolo{ki kompleks Suvodol, lokalitet Anovi – II, Zbornik na trudovi 4–5, Bitola, 21–32. Srejovi} 2002 – D. Srejovi}, Iliri i Tra~ani, prir. V. Jovi}, Srpska kwi`evna zadruga, Beograd. Stankovski 2002 – J. Stankovski, Tatikev Kamen – megalitska opservatorija i svetili{te, Muzejski glasnik 7–9, Kumanovo, 29–48. Stoji} 1996 – M. Stoji}, Le Bassin de la Morava a l’age de bronze et a la periode de transition de l’age de bronze a celui de fer, The Yougoslav Danube basin and the neighbouring regions in the 2 nd millennium b. c., Beograd–Vr{ac, 247–256. Stoji}, Joci} 2006 – M. Stoji}, M. Joci}, Ni{ – kulturna stratigrafija praistorijskih lokaliteta u ni{koj regiji, Arheolo{ki institut – Narodni muzej, Beograd–Ni{ 2006. Todorova 2003 – H. Todorova, Prehistory of Bulgaria, in: D. V. Grammenos (ed.) Recent Research in the Prehistory of the Balkans, Archaeological Institute of Northern Greece. Thessaloniki, 257–328. Nova~ka }uprija, Zbornik Narodnog muzeja XII – 1, Beograd, 17–63. Le÷akov 1992 – K. Le÷akov, Izsledovaniý vørhu bronzovata epoha v Trakiý I. sravnitelna stratigrafiý na seli÷nite mogili ot rannata bronzova epoha v ügoizto~na Bølgariý, Goâi{nik na sofiçskiý universiðeð „Sv. Klimenð Ohriâski“ ðom 84–85, Sofiý, 5–119. Maniatis, Ziota 2011 – Y. Maniatis, Ch. Ziota, Systematic 14C dating of a unique Early and Middle Bronze Age Cemetery at Xeropigado Koiladas, West Macedonia, Greece, Radiocarbon 53/3, 461–478. Mitrevski 2003 – D. Mitrevski, Prehistory in Republic of Macedonia – F.Y.R.O.M, in: D. V. Grammenos (ed.) Recent research in the Prehistory of the Balkans, Thessaloniki, 13–72. Nacev 2009 – T. Nacev, Triðe keramiðki – rezulðaði od za{ðiðniðe arheolo{ki isðra{uvawa Zleðovica 2007–2009, N.U. Zavod za za{tita na spomenicite na kulturata i muzej, [tip. Nikolova 1999 – L. Nikolova, The Balkans in Later Prehistory, Periodization, Chronology and Cultural Development in the Final Copper and Early Bronze Age (Fourth and Third Millenia BC), BAR International Series 791. Nowicki 1994 – K. Nowicki, Some Remarks on the Pre-and Protopalatial Peak Sanctuaries in Crete, Aegean Archaeology I, Warsaw, 31–48. Papaefthymiou-Papanthimou, Pilali-Papasteriou 1997 – A. Papaefthymiou-Papanthimou, A. PilaliPapasteriou, Excavation at Arhondiko, 1993, resume, To arcaiologiko ergo sth Makedonia kai sth Qrakh 7, 152. Papleuqumion-Papanqimon, Pilalh-Papasterion 1998 – A. Papleuqumion-Papanqi mon, A. Pilalh-Papasterion, Anaskafh Arcontikon 1994 (tomšaj) – To arcaiologiko ergo sth Makedonia kai sth Qrakh 8, 83–90. Pekovi}, Jevti} 2007 – M. Pekovi}, M. Jevti}, Za{titna iskopavawa nalazi{ta Mihajlov ponor na Miro~u, Glasnik Srpskog arheolo{kog dru{tva 23, Beograd, 125–140. Prendi, Bunguri 2008 – F. Prendi, A. Bunguri, The Early Bronze Age in Albania. The Albanological studying center – The Institute of Archaeology, Prishtinë. Pullen 2011 – D. J. Pullen, The Early Bronze Age village on the Tsoungiza hill, The American School of Classical Studies at Athens, Princeton. Renfrew et al. 1986 – C. Renfrew, M. Gimbutas, E. S. Elster (eds.), Excavations at Sitagroi, A Prehistoric 70 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) Filipovi}, Jeremi}, Bulatovi} 2012 – V. Filipovi}, G. Jeremi}, A. Bulatovi}, Preliminarni rezultati arheolo{kih iskopavawa na lokalitetu Golo rebro u selu Dowi Neradovac kod Vrawa, Glasnik SAD 28, Beograd, 133–152. Wayatt 2008 – S. Wyatt, The Classification of the Clay Drums of the Southern Trichterbecher Culture, Rezime: STARINAR LXIV/2014 Journal of Neolithic Archaeology, http://www.jna.uni-kiel. de/index.php/jna/article/view/22/22 (pristupqeno 3. 2. 2014.) Weiberg 2007 – E. Weiberg, Thinking the Bronze Age, Life and Death in Early Helladic Greece, Upsala universitet, Upsala. ALEKSANDAR P. BULATOVI], Arheolo{ki institut, Beograd NOVI NALAZI KAO PRILOG PROU^AVAWU RANOG BRONZANOG DOBA U JU@NOM DELU CENTRALNOG BALKANA Kqu~ne re~i. – Rano bronzano doba, ju`no Pomoravqe, Gorwa P~iwa, Armenohori kultura, Bubaw–Hum III kultura. Svi lokaliteti nalaze se na terasi na levoj obali Ju`ne Morave, na rubu aluvijalne ravni. Stilsko-tipolo{ke karakteristike keramike iz Ranutovca i Doweg Neradovca odgovaraju Armenohori kulturi, koja je u rano bronzano doba (EBA III) obuhvatala teritoriju severne Gr~ke, isto~ne Albanije i Pelagonije, dok nalazi iz Pavlovca, Davidovca i Kokinog Sela imaju karakteristike i Armenohori kulture, ali i Bubaw–Hum III kulture, kao i kulture ranog bronzanog doba zapadne Bugarske (Pernik kultura). Sakralni obi~aj pohrawivawa darova u zemqu, u vidu celih posuda ili wihovih delova, kao i drugih predmeta, sa kru`nim kamenim konstrukcijama oko jame i iznad we, konstatovan u Kokinu (sl. 5) i Pelincu, ali i u Davidovcu (ali bez kamene konstrukcije), svakako pripada Egejskotra~kom sakralnom arealu, a veoma sli~an duhovni koncept pohrawivawa darova u plitke jame okru`ene i pokrivene kru`nim kamenim konstrukcijama prisutan je i na nekropoli u Ranutovcu. Na osnovu svih ranije prezentovanih podataka, kao i podataka predstavqenih u ovom radu o materijalnoj, a naro~ito duhovnoj kulturi zajednica ranog bronzanog doba na ovoj teritoriji, na Balkanu je, {ire posmatrano, mogu}e uo~iti izvesno kulturno jedinstvo na teritoriji severne Egeje, Trakije, severne Gr~ke i centralnog Balkana, koje se mo`e definisati kao jedinstvena kulturna zona. Ovo kulturno jedinstvo je rezultat razvoja autohtonih kultura, ali i me|usobnih uticaja kultura iz Egejskog kruga i kultura iz unutra{wosti Balkanskog poluostrva tokom prethodnih perioda. Unutar te jedinstvene kulturne zone mogu se definisati, izme|u ostalih, i dve pribli`no istovremene kulturne manifestacije: Bubaw – Hum III – Pelince – Pernik kultura (kompleks), koja obuhvata zapadnu Bugarsku, ju`no U ovom radu su predstavqeni rezultati istra`ivawa lokaliteta ranog bronzanog doba koji su evidentirani na trasi autoputa E 75, u okolini Vrawa, kao i novi nalazi sa lokaliteta Tati}ev kamen u Kokinom Selu kod Kumanova, u Republici Makedoniji. Re~ je o rezultatima istra`ivawa nekropole spaqenih pokojnika u Ranutovcu, zatim ritualne jame u Davidovcu, kao i nalazima iz Doweg Neradovca, Pavlovca i Kokinog Sela (karta 1). Nekropola u Ranutovcu, koja se nalazi na oko 5 km severno od Vrawa, sastojala se iz dva dela – ju`nog i severnog, na me|usobnoj udaqenosti od oko 20 m. Ju`ni deo nekropole je bio mnogo boqe o~uvan i u wemu su konstatovani grobovi koje su ~inile plitke jame, sa slobodno pohrawenim ostacima spaqenog pokojnika, koji su u nekim slu~ajevima bili pokriveni posudom (sl. 1–3). U grobovima su konstatovani prilozi u vidu celih posuda ili ve}ih fragmenata posuda (T. I/1–17), a oko jama i iznad wih je formirana kru`na kamena konstrukcija pre~nika 1–2 m. U severnom delu nekropole otkriveno je i istra`eno 13 grobova, a u ju`nom delu 10 grobova. Osim posuda, u grobovima su na|eni stilizovani modeli pe}i (T. I/8, T. II/16) i jedna kamena perforirana sekira. U Davidovcu, udaqenom 10 km ju`no od Vrawa, prilikom za{titnih iskopavawa 2011. godine otkrivena je jedna kru`na jama, pre~nika oko 0,7 m i dubine oko 0,2 m (sl. 4), u kojoj su na|eni jedan bubaw od pe~ene zemqe (T. III/30), zatim tri ve}a i dva mawa fragmenta razli~itih zdela, kao i ve}i fragment amfore (T. II/25–28, T. III/29). U Pavlovcu, na oko 7 km ju`no od Vrawa, na lokalitetu Kova~ke wive, otkriveno je nekoliko nalaza koji pripadaju ranom bronzanom dobu (T. II/19–24), a jedan fragment keramike iz tog perioda otkriven je i na lokalitetu Golo rebro u Dowem Neradovcu, na oko 3 km ju`no od Vrawa (T. II/18). 71 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 je ova grupa zapo~ela ekspanziju prema severu, odnosno kada se pro{irila do ju`nog Pomoravqa, ali se, prema dostupnim podacima i apsolutnim datumima, pretpostavqa da se to okvirno dogodilo krajem III ili po~etkom II milenijuma pre n. e. Uticaj Armenohori kulture je prepoznatqiv i u kulturama sredweg bronzanog doba na centralnom Balkanu, ~ime je ova kultura ostavila trag i u genezi i razvoju bronzanog doba na Balkanskom poluostrvu. Pomoravqe, isto~nu Srbiju i severoisto~nu Makedoniju, i Armenohori (Pelagonijsku) kulturu, ~ija se teritorija prostire od severne i zapadne Gr~ke, i isto~ne Albanije, na jugu, do severne Makedonije, na severu (karta 2). Rezultati istra`ivawa novih lokaliteta i nalazi prezentovani u ovom radu potvrdili su pretpostavke da Armenohori kultura obuhvata, bar u jednom periodu svoje egzistencije, i teritoriju ju`nog Pomoravqa. Na osnovu dostupnih podataka ne mo`e se ta~no odrediti vreme kada 72 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) 1 STARINAR LXIV/2014 2 3 4 5 6 7 9 8 10 11 12 13 14 Plate I – Finds from the necropolis in Ranutovac, northern part (1, 3, 6, 8 – grave 7; 2 – grave 2; 4, 5 – grave 1; 7 – grave 5; 9 – grave 6); southern part (10–12 – grave 17; 13 – grave 20; 14 – grave 21) Tabla I – Nalazi iz nekropole u Ranutovcu, severni deo (1, 3, 6, 8 – grob 7; 2 – grob 2; 4, 5 – grob 1; 7 – grob 5; 9 – grob 6); ju`ni deo (10–12 – grob 17; 13 – grob 20; 14 – grob 21) 73 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 17 15 16 19 20 18 21 22 23 24 25 26 27 28 Plate II – Finds from the southern part of the necropolis in Ranutovac (15 – grave 15; 16 – grave 20); fragment from the northern part of the necropolis (17 – grave 1); find from Donji Neradovac, Golo Rebro site (18); finds from Pavlovac, Kova~ke Njive site (19–24); Davidovac, Gradi{te site, pit finds (25–28) Tabla II – Nalazi iz ju`nog dela nekropole u Ranutovcu (15 – grob 15; 16 – grob 20); fragment iz severnog dela nekropole (17 – grob 1); nalaz iz Doweg Neradovca, lokalitet Golo rebro (18) nalazi iz Pavlovca, lokalitet Kova~ke wive (19–24); Davidovac, lokalitet Gradi{te, nalazi iz jame (25–28) 74 Aleksandar P. BULATOVI], New finds as a contribution to the study of the Early Bronze Age… (57–75) STARINAR LXIV/2014 29 30 31 33 34 32 35 36 37 38 Plate III – Davidovac, Gradi{te site, pit finds (29, 30); finds from the shrine at Tatikev Kamen site, Kokino Selo, Republic of Macedonia (31–38) Tabla III – Davidovac, lokalitet Gradi{te, nalazi iz jame (29, 30); nalazi iz svetili{ta na lokalitetu Tati}ev kamen, Kokino Selo, Republika Makedonija (31–38) 75 IVANA POPOVI] Institute of Archaeology, Belgrade NEW DISCOVERIES OF MARBLE SCULPTURES IN THE SIRMIUM IMPERIAL PALACE UDK: 904:730"652"(497.113) ; 902.2(497.113)"2012/2013" DOI: 10.2298/STA1464077P e-mail: [email protected] Received: February 19, 2014 Accepted: July 21, 2014 Original research article Abstract. – In the course of archaeological excavations carried out in 2012 and 2013, in the northwestern section of the palatial complex in Sirmium (locality 85), many fragments of porphyry and marble sculptures were discovered. Worth mentioning among the marble sculptures is a female head with a lunular diadem that had, most probably, been made during the Antonine period. The head was used as spolia incorporated in the medieval wall. It was a fragment of a statue of some goddess, possibly Juno, Minerva or the deified empress Faustina the Younger, and erected in the area of the palatial complex during the Late Antique period. Key words. – Roman period, Sirmium, imperial palace, sculptures, marble. A structure (fig. 1a) were discovered. Next to the square were recorded remains of two rooms whose floors were paved with mosaics with geometric motifs. These mosaics were damaged in a later period by the intrusion of the postholes used as the substructure of the Gepidean houses which were built at the time of disintegration of the palatial complex during the 5th and 6th centuries. Three contemporary burials, most probably of the Gepidean population, were also discovered in the immediate vicinity of these dwelling structures.1 The cathedral church of medieval Dmitrovica was built in the medieval period on top of the remains of antique rchaeological excavations of the northwestern section of the palatial complex in Sirmium (locality 85) have been in progress since 2002, with an interruption from 2007 to 2010, resulting from a lack of funds to continue investigations. Investigations revealed a large structure (31 x 39 m) with substantial walls reinforced on the inside with pilasters, that was most probably the granary (horreum), had been built in this area in the 4th century. Sometime later the annexes were added on the north and south side of the eastern wall. The remains of a fountain were recorded in the northern section of the annex. A square connecting this section of the complex with structures in its western area (locality 37) was built along the eastern annex wall in the second half of the 4th century. In the course of excavations conducted in 1968 and 1969, remains of a luxurious Late Roman 1 Jeremi}, Popovi} 2004, 284–288; Jeremi} 2009, 470–491, Fig. 9, 17–20, 30; Miladinovi}-Radmilovi} 2011, 427–432, Fig. 213. * The article results from the project: Romanization, urbanization and transformation of urban centres of civil, military and residential character in the Roman provinces on territory of Serbia (no 177007) funded by the Ministry of Education, Science and Technological development of the Republic of Serbia. 77 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) STARINAR LXIV/2014 Fig. 1a. Sirmium – investigated localities within the palatial complex (plan B. Popovi}) Sl. 1a. Sirmijum – istra`ivani lokaliteti u sklopu palatijalnog kompleksa (plan: B. Popovi}) architecture in the western section of locality 85.The fresco-painted crypt of the church used the wall of a Late Roman structure of economic character.2 The multi-layered necropolis used in the period between the 12th and 16th centuries was established around the church.3 Medieval structures and communications were also recorded in the eastern section of the site. During the course of archaeological excavations carried out in the autumn of 2012, medieval and antique layers at locality 85 were levelled off 4, and during the campaign in the autumn of 2013, medieval structures and communications were removed (fig. 1b). Many fragments of carved porphyry and marble belonging to Roman period sculptures were found on these occasions. Of these carvings, a fragmented marble female head and fragments of a porphyry male head were preserved. Unfortunately, most of these sculptural fragments were not discovered in situ, while some of them were used as spolia in medieval structures or communications. Nevertheless, this material, together with sculptural works discovered during excavation of this area in the preceding years, offers important elements for the comprehension of the decorative concept applied in this section of the palatial complex. Among the few fragments of marble structures discovered at locality 85 in 2012, only the partially preserved head of a young woman (C–7/2012: h. 15.5 cm; w. 15 cm) offers clear data for analysis. This head had been used as spolia in the Middle Ages, as it was found incorporated in wall XXXIII (fig. 2) built above the street, which was dated by monetary finds to the 13th century.5 At the time of its discovery the head was covered with a thick layer of mortar and looked like some architectural element spherical in shape (fig. 3 a–d). It turned out, after cleaning and conservation, that it is the marble head of a woman whose lower sections of the face are missing and the nose is damaged.6 The eyes 2 Jeremi} 2006, 164–167. Miladinovi}-Radmilovi} 2011, 486–494; 523–541. 4 Popovi}, Pop-Lazi}, Popovi}, Vujadinovi} 2012, 38–41. 5 According to preliminary analysis conducted by Pavle Popovi}, the museum counsellor from the Museum of Srem, at the level of medieval street where discovered copper coins of Hugarian king Istvan IV (1162–1163) and silver Croatiani coins from the period between 1235 and 1349 (from the reign of Bela IV to Miklos Cehi). 6 Popovi} 2013a, 113, sl. 46; Popovi} 2013b, 113, fig. 46. 3 78 STARINAR LXIV/2014 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) Fig. 1b. Sirmium – locality 85, area investigated in 2012 and 2013 (plan B. Popovi}) Sl. 1b. Sirmijum – lokalitet 85, prostor istra`ivan 2012. i 2013. godine (plan: B. Popovi}) that she had a chignon at the back of her head. Slightly wavy hair gathered at the nape represents one of the characteristic hairstyles of empresses from the time of the Antonines, primarily of Faustina the Younger, wife of Marcus Aurelius,7 and that could indicate the date of origin of the head of this goddess. In favour of such chronological determination speaks the fact that private portraits and other sculptural works during the Antonine period had been modelled after the sculptures of rulers and their wives and that their characteristic is an optical effect of contrast between the voluminously sculptured hair and the smooth, incarnate hair.8 The head of the goddess from Sirmium also reveals such characteristics as thick wavy hair above the forehead and face, which is without significant vibrations, and slightly curly locks which are divided by channels, although much shallower and less pronounced than those used are of an almond shape with lightly engraved corneas and crescent shaped pupils. Her hair is parted along the middle of the head and is depicted by relatively deep, slightly meander-like channels, creating a light–dark contrast. It was probably gathered in a chignon, but the back of the head is also damaged, so that segment of the hair style is missing. On the head is a diadem of lunular shape, damaged on the upper arched section (fig. 4 a–c). Near the top in central section is shallow, symmetrical, circular impression preserved in the lower semicircular segment (fig. 5). Judging by the lunular diadem, which is an attribute of some female deities, the head represents a goddess, possibly Juno or Minerva, although, because of the damage and lack of attributes, a precise identification is not possible. The execution of hair indicates the work of Asia Minor sculptors and the general style of creation links it to the sculptures from the period around the middle or second half of the 2nd century. Despite the fact that the hairstyle of the depicted goddess is not completely preserved, and considering that the hair is falling downward from the top of the head, it seems 7 8 79 Wegner 1938, 324–325, Abb.4. Cambi 2005, 95–96. Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) STARINAR LXIV/2014 Fig. 2. Medieval wall XXXIII at locality 85 and head of the goddess extracted from the place where it had been built-in (photo S. Pop-Lazi}) Sl. 2. Sredwovekovni zid XXXIII na lokalitetu 85 i glava bogiwe izvu~ena sa mesta gde je bila uzidana (foto: S. Pop-Lazi}) for depicting the hair by masters from the sculptural workshops in Aphrodisias in Asia Minor, sometime later during the Severan epoch. Artisans from those workshops travelled throughout the Empire, carving nicely modelled sculptures. Judging by the execution of the hair on the top of the much damaged head of the woman with a lunular diadem from Divo{ near Sremska a b c d Fig. 3 a–d. Head of the goddess after being extracted from the medieval wall (photo S. Pop-Lazi}) Sl. 3 a–d. Glava bogiwe posle va|ewa iz sredwovekovnog zida (foto: S. Pop-Lazi}) 80 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) STARINAR LXIV/2014 a b c Fig. 4 a–c. head of the goddess after cleaning and conservation (photo N. Bori}) Fig. 5. Lunular diadem on the goddess’ head (photo N. Bori}) Sl. 4 a–c. Glava bogiwe posle ~i{}ewa i konzervacije (foto N. Bori}) Sl. 5. Lunulasta dijadema na glavi bogiwe (foto: N. Bori}) Mitrovica,9 this sculpture could also have been the work of masters from the Asia Minor workshops, made during the Antonine period. The head of the goddess, Minerva or Juno, also with a lunular diadem, found at ^itluk near Sinj (Aequum), is also the work of artisans from Aphrodisias.10 In Aquileia, from the Antonine to the Late Roman period, were displayed medallions with, most probably, representations of 12 deities, made according to the traditions of the works of masters from this stone-carving centre in Asia Minor.11 The head from Sirmium, as well as the one from nearby Divo{, are stylistically close to the Antonine period and that raises some questions related to the time, circumstances and place of their display. As the Sirmium head was incorporated into the medieval wall, its original architectural context could not be established. It could be assumed, with some certainty, that this sculpture was, in the Late Roman period, 9 Popovi} 2012, 108–109, kat. 34. Cambi 2005, 123, sl. 180. 11 Mian 2013, 326–231, cat. 34–39. 10 81 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) displayed in the northwestern area of the palatial complex, on top of which were constructed dwelling structures and communications in the medieval times. If this assumption is correct, it would mean that the sculpture of the goddess dating from the earlier period, i.e. from the second half of the 2nd century, existed within the zone of the imperial palace, most probably built at the end of the 3rd to the beginning of the 4th century and adapted on many occasions until the final third of that century. Such a phenomenon has been already encountered in the very same area since the head of Venus, discovered in 2003 as an element of the ornamental system of the fountain in the north annex of the 4th century structure, was most probably made at the end of 1st century.12 This phenomenon was explained as the wish of Constantine to show cultural continuity and confirm his legitimacy as the new Augustus, by displaying a statue of the goddess Venus, ancestress of the Julio–Claudian dynasty.13 For similar reasons the mentioned medallions with the deities in Aquileia, originally placed on some arch or gateway, were transferred to the imperial palace and displayed there.14 However, if the assumption that the Sirmium head was made during the Antonine period is correct, it also opens some new options for the interpretation of its later reuse. It is, however, well-known that Marcus Aurelius used Sirmium as the base for military operations in Pannonia during the wars against Quadi and Marcomanni at the end of the seventh decade of the 2nd century.15 The stay of this emperor and his court in Sirmium is also recorded in written sources, which mention that his palace – bas/leia16 was in this city. The location of this imperial residence is unknown and, although layers from earlier epochs have been encountered under the Late Roman imperial palace,17 there is no evidence that there was also the imperial residence of Marcus Aurelius. However, during the stay of that emperor, the city had been decorated with various sculptures including those depicting female deities. Also, we must not overlook the fact that Marcus Aurelius abandoned the earlier tradition of the adoption of future emperors, and proclaimed his son Commodus as heir apparent. Because of that, the role of empress, the mother of the future emperor, gained in importance and that also remained so in the ensuing period, during the reign of the Severi, who also established their dynasty based on blood relationships. A large number of cameos in the Danube basin with representations of women with hairstyles characteristic of empresses from the dynasties of Antonines and Severi could be STARINAR LXIV/2014 Fig. 6. Fragment of marble sculpture (upper arm and shoulder?) discovered at locality 85 in 2012 (photo S. Pop-Lazi}) Sl. 6. Fragment mermerne skulpture (mi{ica i rame ruke?) otkrivene 2012. godine na lokalitetu 85 (foto: S. Pop-Lazi}) explained as being a consequence of that practice.18 Many empresses from these dynasties had the rank of Augusta, and Faustina the Younger (ca. 130–176), wife of Marcus Aurelius, got the title Mater Castrorum in 174.19 Empresses from that period were sometimes depicted on coins minted on their behalf with symbols of the goddesses – with lunular diadem on their heads. Faustina the Younger had also been depicted in such a way on various series of coins,20 and the temples in Aquileia and Zadar (Iader) were dedicated to her cult. In those temples the priestesses performed certain rituals and the name of one of them, Cosutia, was confirmed in the inscriptions from these two cities.21 The Cult of Faustina the Younger and her husband remained popular in the ensuing centuries and even Constantine 12 Popovi} 2008, 153–159; Popovi} 2009, 707–709; Popovi} 2012, 19–21, 78–82, kat. 9. 13 Popovi} 2008, 159. 14 Mian 2013, 226. 15 Mirkovi} 1971, 33; Mirkovi} 2006, 61. 16 Philostr., V. Marci 168–169. 17 Mirkovi} 1971, 36; Mirkovi} 2006, 65. 18 Popovi} 2010, 217. 19 RIC III, 206. 20 Among many specimens with representation of Faustina the Younger with diadem on the head we are distinquishing the denarius from the find in Sotine (Cornacum), cf. RIC III, No 713. 21 Cambi 2005, 95, nap. 327, 329. 82 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) STARINAR LXIV/2014 a b c d Fig. 7. Fragment of marble sculpture/composition discovered at locality 85 in 2012: a) front side; b) back side; c) left lateral side; d) right lateral side (photo S. Maksi}) Sl. 7. Fragment mermerne skulpture/kompozicije otkrivene 2012. godine na lokalitetu 85: a) predwa strana; b) zadwa strana; s) leva bo~na strana; d) desna bo~na strana (foto: S. Maksi}) 83 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) STARINAR LXIV/2014 two connected cylindrical elements of varied thickness (fig. 7a), carved out of one piece of marble (fig. 7b). On the lateral side of the left, thicker part (R 10.41˜11.70 cm) of the composition is a spherical protrusion resembling a knot in a piece of wood (fig. 7c), while on the lateral side of the right element (R 7.20 cm) is some irregularly shaped damage (fig. 7d). On a separate, semi circularly ending part, which is connects the two cylindrical elements at the front, is a ribbed ornament consisting of a central relief band with laterally slanting bands radiating from it (fig. 7a). It is possible that, in such manner, a leaf with its venation was depicted. The preserved fragment of marble sculpture represents, most probably, a segment of a tree branching off into the upper zone. If a leaf (vine leaf?) is depicted, it was probably an element (support?) from a composition showing a scene from the cult of Dionysus, like the similar fragment from the hoard of sculptures from Mediana.26 The new finds of marble sculptures together with the fragmented marble capitals, architectural decorations, heads of goddesses and the segment of the upper arm and shoulder, discovered in 2003 and 2005,27 confirm that the area to the east of the economy building at locality 85 was an integral part of the residential zone of the palatial complex. considered Marcus Aurelius to be one of five good emperors and wanted to present himself as his legitimate heir.22 Regarding these attempts, and as an element of imperial propaganda, Constantine’s wife Fausta was depicted on the cameos from Bela Palanka (Remesiana)23 and on the coins minted after the year 320 with a hairstyle resembling that favoured by Faustina the Younger, i.e., with slightly wavy hair gathered in a chignon at the nape.24 Fausta, nobilissima femina, also got the title of Augusta in 324. Taking into account that Constantine’s court was in Sirmium from the beginning of 320 until the beginning of 324 and that the emperor himself stayed there on many occasions for quite long periods of time,25 it is logical to assume that the area of the imperial palace was representative and that a statue of the goddess dating from the Antonine period, possible of Faustina the Younger herself was displayed there as a reflection of imperial propaganda. The fragments of other sculptural works made of marble have also been discovered in addition to this head, at locality 85 during excavations in 2012. These were the segment of the upper arm and shoulder (C–63/ 2012: l. 11.2 cm, w. 5.6 cm) (fig. 6) and a fragment of a figural composition, (C–57/2012: l. 15 cm, w. 21 cm) which, unfortunately, could not be identified. These are 22 Hanestad 1988, 325. Popovi} 2010, 211, no 39, 40 (with quoted literature). 24 RIC VII, 571, No 8, Pl. 18. 12. 25 Hannestad 1988, 319–327, figs. 195–197. 26 Jovanovi} 1975, 58–59, Sl. 10. 27 Popovi} 2008, 153–162, Fig. 2, 4, 6; Popovi} 2009, 707–712, Fig. 1–2; Popovi} 2012, 19–21, 78–82, 110–112, 119, kat. 9, 36, 43. 23 84 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) STARINAR LXIV/2014 SOURCES AND ABREVIATIONS: Philostr. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Philostratus, Vita Marci, in: The Lives of the Sophists, transleted by W. C. Wright, London–New York 1922. H. Mattingly, E. A. Sydenham, Antoninus Pius to Commodus, The Roman Imperial Coinage, Vol. III, London 1968. P. Bruun, Constantine and Licinius. A. D. 313–337, The Roman Imperial Coinage Vol. VII, ed. C. H. V. Sutherland, D. Litt, R. A. G. Carson, London 1966. RIC III . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RIC VII . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BIBLIOGRAPHY: Mirkovi} 2006 – M. Mirkovi}, Sirmium. Istorija rimskog grada od I do kraja VI veka, Sremska Mitrovica 2006. Popovi} 2008 – I. Popovi}, Marble Sculptures from the Imperial Palace in Sirmium (Rezime: Mermerne skulpture iz carske palate u Sirmijumu), Starinar (n. s.) LVI (2006), 2008, 153–166. Popovi} 2009 – I. Popovi}, Les sculptures du palais impérial de Sirmium (fouilles 2003–2005), in: Les ateliers de sculpture réginoaux: techniques, styles et iconographie, Actes du Xe Colloque international sur l’art provincial romain, ed. V. Gaggadis-Robin, A. Hermary, M. Reddé, C. Sintes, Arles et Aix-en-Provence 2009, 707–712. Popovi} 2010 – I. Popovi}, Roman cameos with female busts from Middle and Lower Danube, Pallas. Revue d’études antiques. Glyptique romaine, 2010, 203–224, Pl. IX–XIII. Popovi} 2012 – I. Popovi}, Sirmium – Mermerne skulpture / Sirmium – Marble Sculptures. Beograd – Sremska Mitrovica 2012. Popovi} 2013 a – I. Popovi}, Sirmijum – carska rezidencija, panonska metropola i hri{}anska „glava Ilirika“, u: Konstantin Veliki i Milanski edikt 313. Ra|awe hri{}anstva u rimskim provincijama na tlu Srbije, ur. I. Popovi}, B. Bori}Bre{kovi}, Beograd 2013, 102–117. Popovi} 2013 b – I. Popovi}, Sirmium – Imperial Residence, Pannonian Metropolis and Christian “Head of Illyricum”, in: Constantine the Great and the Edict of Milan 313. The Birth of Cristianity in the Roman Provinces on the Soil of Serbia, eds. I. Popovi}, B. Bori}Bre{kovi}, Beograd 2013, 102–117. Cambi 2005 – N. Cambi, Kiparstvo rimske Dalmacije (Summary: The Sculpture of the Roman Province of Dalmatia), Split 2005. Hannestad 1988 – N. Hannestad, Roman Art and Imperial Policy, Aarhus 1988. Jeremi} 2009 – M. Jeremi}, The Sirmium Imperial Palace Complex, in Light of the Recent Archaeological Investigations, in: Diocletian, Tetrarchy and Diocletian’s Palace on the 1700 th Anniversary of Existence (International Conference, Split, September 2005), eds. N. Cambi, J. Belamari}. T. Marasovi}, Split, 471–499. Jeremi}, Popovi} 2004 – M. Jeremi}, I. Popovi}, Arheolo{ka istra`ivawa Sirmijuma u Sremskoj Mitrovici na lokalitetima 79 i 85 (u periodu od 2000. do 2003. godine), Starinar (n. s.) LIII–LIV, 2004, 281–288. Jovanovi} 1975 – A. Jovanovi}, Neki aspekti problema skupnog nalaza skulptura sa Medijane kod Ni{a (Résumé: Certains aspects du probleme de la trouvaille collective de sculptures à Mediana près de Naissus), Starinar XXIV–XXV (1973–1974) 1975, 57–65. Mian 2013. – G. Mian, Clipei con busti di divinità, cat. nr. 34–39, in: Constantino e Teodoro. Aquileia nel IV cecolo, eds. C. Tiussi, L. Villa, M. Novello, Aquileia 2013, 226–231. Miladinovi}-Radmilovi} 2011 – N. Miladinovi}Radmilovi}, Sirmium. Necropolis, Beograd–Sremska Mitrovica 2011. Mirkovi} 1971 – M. Mirkovi}, Sirmium-its History from the I Century A. D. to 582 A. D. The Inscriptions from Sirmium and its Territory, in: Sirmium I, V. Popovi}, ed., Beograd, 1971, 5–94. 85 Ivana POPOVI], New discoveries of marble sculptures in the Sirmium imperial palace (77–86) Popovi}, Pop-Lazi}, Popovi}, Vujadinovi} 2012 – I. Popovi}, S. Pop-Lazi}, B. Popovi}, V. Vujadinovi}, Arheolo{ka iskopavanja u Sremskoj Mitrovici 2012. godine – Lokalitet 85, Arheologija u Srbiji. Projekti Rezime: STARINAR LXIV/2014 Arheolo{kog instituta u 2001. godini (priredile V. Biki}, S. Golubovi}, D. Antonovi}), Beograd 2012, 38–41. Wegner 1938 – M. Wegner, Datierung römischer Haartrachten, Archäologischer Anzeiger 53, 198, 276–328. IVANA POPOVI], Arheolo{ki institut, Beograd NOVI NALAZI MERMERNIH SKULPTURA U CARSKOJ PALATI U SIRMIJUMU Kqu~ne re~i. – rimski period, Sirmijum, carska palata, skulpture, mermer. cija ne mo`e precizno utvrditi. Tretman kose ukazuje na rad maloazijskih klesara, a op{ti stil izrade je povezuje sa skulpturama iz perioda oko sredine ili druge polovine II veka. Blago talasasta kosa, skupqena iznad vrata, predstavqa jednu od karakteristi~nih frizura carica iz doba Antonina, pre svega Faustine Mla|e, `ene Marka Aurelija, {to bi moglo da upu}uje na vreme nastanka glave bogiwe. S obzirom da je Konstantin smatrao Marka Aurelija za jednog od pet dobrih careva, `ele}i da se predstavi wegovim legitimnim naslednikom, na nekim serijama novca i kameja je wegova `ena Fausta prikazivana sa frizurom karakteristi~nom za caricu Faustinu Mla|u. Statua bogiwe nastala tokom antoninskog perioda, mo`da same deifikovane carice, verovatno je bila postavqena na prostoru palatijalnog kompleksa u Sirmijumu kao segment Konstantinove propagandne politike. Delovi drugih mermernih skulptura (sl. 6, 7) otkriveni 2012. godine na lokalitetu 85 ne pru`aju mogu}nost sigurnog definisawa skulpturalnih kompozicija kojima su pripadali, ali ukazuju na reprezentativnost tog prostora, koji su krasila brojna vajarska dela izvedena u mermeru. Tokom arheolo{kih iskopavawa, sprovedenih 2012. i 2013. godine na prostoru severozapadnog dela palatijalnog kompleksa u Sirmijumu (lokalitet 85) (sl. 1a, 1b), otkriveno je vi{e fragmenata skulptura od porfira i mermera. Od nekoliko delova mermernih skulptura otkrivenih 2012. godine na lokalitetu 85, izvesne podatke za analizu pru`a samo delimi~no o~uvana glava mlade `ene, kori{}ena tokom sredweg veka kao spolija, budu}i da je bila ugra|ena u zid (sl. 2) podignut iznad ulice, koja je monetarnim nalazima datovana u XIII vek. U trenutku nalaza glava je bila oblo`ena debelim slojem maltera, pa je izgledala kao neki gra|evinski element u obliku kugle (sl. 3 a–d). Posle ~i{}ewa i konzervacije uvidelo se da je re~ o mermernoj glavi `ene, kojoj nedostaju dowe partije lica i kojoj je nos obijen. Na glavi se nalazi lunulasta dijadema, o{te}ena na gorwem, lu~nom delu (sl. 4 a–c). Pri vrhu, na centralnom delu, nalazi se plitko, pravilno kru`no udubqewe, sa~uvano u dowem, polukru`nom delu (sl. 5). Sude}i po lunulastoj dijademi, obele`ju nekih `enskih bo`anstava, glava predstavqa bogiwu (mogu}e je Junonu ili Minervu), mada se zbog o{te}ewa i nedostatka atributa identifika- 86 SOFIJA PETKOVI], Institute of Archaeology, Belgrade NATA[A MILADINOVI]-RADMILOVI], Institute of Archaeology, Belgrade MILITARY GRAVES FROM THE LATE ROMAN NECROPOLIS AT SLOG IN RAVNA (TIMACUM MINUS) UDK: 904:726.8"652"(497.11) DOI: 10.2298/STA1464087P e-mail: [email protected] Received: February 24, 2014 Accepted: July 23, 2014 Original research article Abstract. – The necropolises of the Roman fortification and settlement Timacum Minus, in the village of Ravna, near Knja`evac, were partially explored by systematical and rescue archaeological excavations. The most extensively explored was the part of the Late Roman necropolis on the eastern slope of the Slog hill, about 400m west of the fortification, where 80 graves from this period have been investigated. The analysis of the human osteological material, and the archaeological finds from the aforementioned necropolis, confirmed 17 military graves, containing adult male individuals with traces of injuries, stress markers and pathological changes, characteristic of a military population, as well as military equipment and weapons. At the time of the formation of the Late Roman necropolis at the site of Slog, during the second half of the 4th and the first half of the 5th century, the garrison of the Timacum Minus fortification consisted of an equestrian unit of pseudocomitatenses Timacenses, a part of the auxiliary formation that secured the forts and roads in the Timok region. Among the graves from the three phases of the Late Roman necropolis, similarities as well as certain differences are apparent, indicating changes in the structure of the civilian and military population of Timacum Minus. Key words. – Late Roman period, necropolis, Timacum Minus, Roman army, Chernyakhov culture, sharp force traumas, blunt force injuries, injuries inflicted by an arrow. T as far back as the end of the 1st century AD, but it was abandoned after the Hun invasion of the Danube Basin in the middle of the 5th century (Fig. 1).3 Around the fortified administrative, mining-metallurgical and craft centre, an urban settlement with public he Roman fortification and settlement on the left bank of the Beli Timok, in the vicinity of the village of Ravna, 10 km north of Knja`evac, was, using systematic archaeological explorations, identified as Timacum Minus (Map 1).1 There are several hypotheses regarding the nature of this Roman site, however, the most credible assumption is that it was a fortified administrative centre of the Upper Moesia mining region Territoria metallorum, which comprised the north-eastern part of the Upper Moesia province, later the provinces of Dacia Ripensis and Dacia Mediterranea.2 The fortification was first established 1 Petrovi} 1995, 21–37; Petrovi}, Jovanovi} 1997, 18–19. Du{ani} 1980, 32–34; Petrovi} 1996, 195 et sequ; Du{ani} 1996, 224, Note 39. 3 Petkovi}, Jovanovi} 2000, 277–280; Petkovi} et al. 2005, 13–17; Iliji} 2009, 11–13. 2 * This article is the result of the project: Romanisation, urbanisation and transformation of urban centres of civil, military and residential character in Roman provinces in the territory of Serbia (no. 177007), funded by the Ministry of Education, Science and Technological Development of the Republic of Serbia. 87 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Karta 1. Polo`aj arheolo{kog nalazi{ta Timacum Minus na karti Republike Srbije between the fortification in the east, the Ropinski brook in the north, the Slog hill in the west and the Seli{ki (Zubanov) brook in the south (Plan 1). On the other hand, parts of the Late Roman necropolis were explored on the right bank of the Ropinski brook, to the north-west of the fortification and on the eastern slope of the Slog hill. In the course of the three campaigns of rescue archaeological excavations, prompted by the reconstruction of the local Ravna – Debelica road in 1994–1996, on the left side of the road, on the eastern slope of the Slog hill, a section of the multilayered Late Roman and Early Medieval necropolis was explored. 140 graves were discovered, out of which 72 with 75 buried individuals are from the Late Roman period (Plan 2). It was concluded that it was a necropolis formed in three successive phases in the period from the middle of the 4th to the middle of the 5th century: phase I – from 350 to 380, phase II – from 380 to 410 and phase III – from 410 to 450.6 The results of the rescue archaeological excavations of the Slog necropolis were published in 2005 in a separate monograph.7 On that occasion, the results of the anthropological analysis, carried out by Dr Zsuzsana Zoffmann, the anthropologist of the National Museum in Budapest,8 were also published. However, since 2011 a project of processing, analysing and publishing all archaeological data and finds gathered during the systematical and rescue archaeological excavations at the site of Ravna – Timacum Minus9 has been in progress. New data, baths, villas, workshops and temples was formed over time. Some of these structures were confirmed by archaeological excavations, whilst the others were acknowledged indirectly, through the discovered sculptures and epigraphic monuments. Based on the latter, which were built into the walls and buildings during reconstruction in the Late Roman period, from the middle of the 3rd to the first half of the 5th century, the existence of a large necropolis in the 2nd and 3rd centuries was ascertained.4 Unfortunately, for the mentioned reason, most of the gravestones had been dislocated, thus, very little is known about this necropolis. More recent non-destructive archaeological exploration, which used geophysical methods of measuring,5 along with the data gathered from locals, who work the fields and gardens west of the Roman fortification, locate the necropolis of the cremated deceased from the 2nd–3rd century to the stretch of [irina situated 4 Petrovi} 1995, 13–16, 40; Petkovi}, Iliji} 2013, 64–65, Table 1. 5 Iliji} 2011. 6 Petkovi} et al. 2005, 145. 7 Petkovi} et al. 2005. 8 Petkovi} et al. 2005, 279–323. 9 The Project Preparation of the monograph of the Roman fortification and the town of Timacum Minus in Ravna near Knja`evac, co-funded by the Ministry of Culture and Information of the Republic of Serbia and the Municipality of Knja`evac, was carried out in 2010–2012 by the Institute of Archaeology in Belgrade and the Homeland Museum of Knja`evac, with an expert team consisting of: Dr Sofija Petkovi}, senior research associate of the Institute of Archaeology – the head of the project, Dr Nata{a Miladinovi}-Radmilovi}, research associate of the Institute of Archaeology – anthropologist, Bojana Iliji} MA, senior curator of the Homeland Museum of Knja`evac – archaeologist, Nikola Radinovi} MA, Marija Jovi} MA and Milica Miti} – archaeologists, Sa{a Milutinovi}-Lete}i, photographer and Tomislav @ivkovi}, conservator of the Homeland Museum of Knja`evac, Dragica Bizjak – senior undergraduate of Archaeology and Nikola Ivankovi} – senior undergraduate of Architecture. Map 1. Position of the archaeological site Timacum Minus on the map of the Republic of Serbia 88 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Fig. 1. Aerial photography of the archaeological site Timacum Minus, taken from the south-east Sl. 1. Aero-snimak arheolo{kog nalazi{ta Timacum Minus, sa jugoistoka with the aforementioned characteristics were analysed according to the phases of burials at the Late Roman necropolis of Slog in order to acquire, by means of acquired from the revision of the human osteological material and by analyses carried out using modern methodological procedures, led us to reconsider the premises and conclusions that relate to the population of the Late Roman Ravna.10 This paper will try to supplement the data about the military component of the Timacum Minus population, which undoubtedly played a significant role in the socio-economic picture of this settlement. Within the explored Late Roman graves from the Slog necropolis in Ravna, the burials that contained archaeological finds of military equipment and weapons were singled out (Plan 2). By a detailed anthropological analysis of the osteological finds from these graves, we primarily attempted to determine, as much as was possible given the degree of bone preservation, the presence of ante-mortem and perimortem traumas which occurred, not as a consequence of accidents, but as a result of intentional violence,11 as well as the presence of characteristic markers of occupational stress12 and pathological changes,13 which would indicate that the deceased were in active military service. The graves 10 Petkovi} et al. 2005, 143–144. The injuries caused by intentional violence include fractures of the upper and lower jaw, tooth fractures, and fractures of the nose and cheekbones, blows inflicted by hard and sharp objects (stone or hilt of a sword or knife), blows inflicted by hard and blunt objects, injuries caused by sharp or blunt blade, injuries inflicted by projectiles (arrows, stones from a slingshot), etc. 12 The examination of markers of occupational stress, typical for cavalry and infantry, is very intricate work. In horsemen, for example, there is a whole series of muscle and ligament attachment points in the region of the shoulder and elbow joints, the lower back, the sacroiliac region and in the region of the legs (most frequently the femur). In addition, what should be taken into consideration are the deformities caused by using certain types of weapons (swords, bows, spears and shields), as well as from wearing military equipment (helmets, amour, vests etc.). 13 Pathological changes caused by frequently poor living conditions (unvaried diet, poor hygiene conditions, exposure to cold and damp, long gruelling marches, often in full battle dress and general exhaustion due to irregular sleep, stress, infectious diseases, etc.). 11 89 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plan 1. General plan of the Timacum Minus fortification, civil settlement and the Late Roman Slog necropolis Plan 1. Generalni plan utvr|ewa Timacum Minus, civilnog naseqa i kasnoanti~ke nekropole Slog METHODOLOGICAL FRAMEWORK comparing data, chronologically relevant results which inform on the development of the military garrison of the Timacum Minus fortification, as well as similarities and differences within its function throughout the second half of the 4th and the first half of the 5th century. Subsequently, the correlation between these results and the data acquired by the analysis of the stratigraphy of the cultural layers and the construction phases of the fortification was formulated. The preservation degree of the examined skeletons is given in the form of a descriptive scheme comprising five categories, as proposed by Miki}:14 I – complete, well preserved skeleton; II – incomplete, well preserved skeleton; III – moderately preserved skeleton;15 IV – partially preserved skeletal remains16 and V – poorly preserved skeletal remains.17 HUMAN OSTEOLOGICAL MATERIAL 14 Miki} 1978, 9. Moderate preservation refers to the situation when a complete skeleton is present in the grave, but the bones are quite brittle and break in the course of excavation. 16 Partial preservation refers to the situation when only parts of a skeleton are present in the grave, which are very brittle and cannot be easily lifted, packed and transported. 17 Poor preservation refers to the situation when skeletal remains can only be registered in traces and are practically impossible to lift completely. 15 The anthropological analysis included 17 individuals (Tables 1, 5 and 10). Based on the archaeological research, the graves were divided into three phases. Phase I (350–380) includes the following graves: 44, 69A, 79, 91, 99 and 141; phase II (380–410) includes graves 15, 25, 26, 27A, 28, 76, 97 and 123; phase III (410–450) includes graves 100, 102 and 108. 90 STARINAR LXIV/2014 0 10 m PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) Plan 2. Late Roman phase of the Slog necropolis in Ravna Plan 2. Kasnoanti~ka faza nekropole Slog u Ravni 91 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 sutures (Vallois’ scheme);25 changes on maxillary and mandibular teeth (changes were compared on the occlusal surface of the dental material with the numerical classification of the attrition of the upper (occlusal) surface of molars in relation to age, as defined by Brothwell26 and changes on the occlusal surface of the dental material with the numerical classification of the attrition of the upper surface of all teeth in relation to age, as defined by Lovejoy);27 morphological changes in the sternal ends of the ribs (metamorphoses of the depth, joint cavities, shape, edges and ridge configuration were examined, together with the overall state of the bone, based on nine (0–8) phases of progression, covering an age range from 18 to over 70);28 morphological changes on the medial end of the clavicle (morphological changes were observed on the medial end of the clavicle, as documented by Scheuer and Black,29 who established five (1–5) phases of progression, covering an age range from 14 to 29); morphological changes on the joint surface of the pubic symphysis (Todd’s method was used here, in which the metamorphosis of the surface of the pubic symphysis is divided into ten chronological phases during ageing, starting with the age of 18, up to 50 and over);30 sacroiliac region (the age of adult individuals was determined based on the model defined by Lovejoy and his associates,31 whereby they classified the changes within this region into eight phases, from late adolescence to the phase of old age, devoting most attention to the examination of the position, marginal lipping and bone porosity of this region). 26 epigenetic variations on the cranial and 11 on the postcranial part of the skeleton were observed.32 For determining sex on skeletal material of juvenile and adult individuals, the combination of morphological and metrical methods was opted for. Special attention was given to the morphological elements of the skull (glabella, planum nuchale, processus mastoideus, processus zygomaticus, arcus supercilialis, protuberantia occipitalis externa, os zygomaticum, tubera frontale et parietale, the inclination of the os frontale, margo supraorbitalis and the shape of the orbitae) and the pelvis (sulcus praearicularis, incisura ischiadica s. ischialis major, arcus pubis s. pubicus et angulus subpubicus, arc compose, the appearance of the os coxae, corpus ossis ischii, foramen obturatum, crista iliaca, fossa iliaca, pelvis major, pelvis minor; subpubic region: ventral arc, subpubic concavity and the medial appearance of the ischio-pubic branch). The methodology was adopted from a group of European anthropologists18 and Buikstra and Ubelaker.19 The morphological elements of the mandible were also analysed (the general appearance of the mandible (corpus mandibulae, ramus mandibulae and angulus mandibulae), mentum, angulus mandibule and margo inferior), according to the criteria established by Ferembach and his associates,20 and metrical elements relevant for sex determination on the skeletons.21 Based on the obtained metrical elements, calculated indices were presented in tables separately for each grave (Tables 3, 7 and 12). The mesiodistal and vestibulolingual diameters were measured on the teeth, in the manner recommended by Hillson.22 The differences in tooth size, obtained via these measurements, were monitored mainly on the canines, and when they were missing in the osteological material, other teeth such as molars, premolars and incisors were used.23 Whilst analysing other bones of the postcranial skeleton, both morphological and metrical elements were also examined. The morphological elements that drew our attention were the degrees of development: tuberositas deltoideae, tuberositas radii and margo interosseus (of radius), tuberositas ulnae and margo interosseus (of ulna), linea aspera and tuberositas tibiae. Bone appearance, body curvature and facies auricularis were morphological elements observed in the sacrum.24 Metrical elements have a decisive role in sex determination on the skeletal remains of the postcranial skeleton, hence, special attention was paid to them. Based on the acquired metrical elements, the separately calculated indices for the left and the right side of the body were presented in tables for each separate grave (Tables 4, 8, 9 and 13). The individual ages of the deceased were determined based on: the degree of obliteration of the cranial 18 Ferembach, Schwidetzky and Stloukal 1980, 519–527. Buikstra and Ubelaker 1994, 15–21. 20 Ferembach, Schwidetsky and Stloukal 1980, 523–525. 21 Ferembach, Schwidetsky and Stloukal 1980, 523–525; Bass 1995, 84, 85. 22 Hillson 1990, 240–242; idem. 1996, 80–82. 23 Garn, Lewis and Kerewsky 1965. 24 Miki} 1978, 18, 19; Bass 1995, 114. 25 Vallois 1937. 26 Brothwell 1981, 72. 27 Lovejoy 1985. 28 Iºcan, Loth and Wright 1984a; idem. 1984b; idem. 1985. 29 Scheuer and Black, 2000. 30 Todd 1920, 285–334; idem. 1921a; idem. 1921b. 31 Lovejoy et al. 1985. 32 Hauser and De Stefano 1989; \uri}-Sreji} 1995, 238–260. 19 92 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the tendon attachment point of the left calcaneus (tendo calcaneus (Achillis)). The stature of juvenile and adult individuals was calculated based on formulae established by Trotter and Gleser (Tables 2, 6 and 11).33 Grave 69A Sondage F II/1995. CATALOGUE OF GRAVES WITH ANTHROPOLOGICAL ANALYSIS 34 Rectangular burial pit, dug into the subsoil (greenyellowish marl), dimensions: 2.30 x 0.80 m, with bottom at a relative depth of 1.80 m. Orientation: west-east. Skeleton supine with legs extended and hands on pelvis. PHASE I OF THE NECROPOLIS (350–380) (Pl. I) Grave 44 Sondage F I/1994. Freely dug burial pit, with bottom at a relative depth of 1.80 m. The find of an iron nail in the vicinity of the left foot indicates the existence of a wooden casket. Orientation: west-east, with a 10° deviation towards the north. Skeleton supine with legs extended. Finds: 1. a conical glass goblet near the pelvis (Pl. II, 5),42 2. a bronze crossbow fibula near the right forearm (Pl. II, 6).43 Note: It was damaged by the subsequent digging of the Late Roman graves 53 and 58. Finds (Pl. III, 1): 1. small beads of glass paste near the left knee, 2. a bronze crossbow fibula near the left lower leg (Pl. II, 1),35 3. a glass bottle near the left lower leg,36 4. a bronze belt buckle with an oval plate and a frame shaped like the Latin letter B, near the left lower leg (Pl. II, 2),37 5–6. bronze shoe buckles with rectangular plates and frames shaped like the Latin letter D, near the left and right feet (Pl. II, 3),38 7–8. leaf-shaped, bronze, shoe strap-ends near the left and right feet (Pl. II, 4),39 9. two bronze coins of Constantius II, minted in the period 337–348, next to the left foot,40 10. an iron nail, below the left foot. ANTHROPOLOGICAL ANALYSIS:44 In the grave, skeletal remains of an adult individual, of unknown sex, aged 30+ were discovered.45 33 Trotter and Gleser 1952. Although the monograph from 2005 (Petkovi} et al. 2005, 25–46, 49–126) contains catalogues of all explored graves and finds, a military graves catalogue is presented and considered in the article due to certain additions and corrections which mainly relate to the analysis of anthropological and archaeological material. 35 Petkovi} 2010, 264 –266, sl. 93, kat. 1537 – type 34/D 2 is dated to the period of the Valentinian Dynasty and Theodosius I (363–395). 36 The glass bottle had been damaged to such an extent that the reconstruction of its shape was not possible. 37 Petkovi} et al. 2005, 91, Cat. 2, Fig. 18b, Pl. 10, 3 – is dated to the second half of the 4th century. 38 Petkovi} et al. 2005, 97, Cat. 1, Pl. VI, G. 44, 5–6, Pl. 5, 3 – is dated to the second half of the 3rd and the 4th century. 39 Petkovi} et al. 2005, 97, Cat. 2, Pl. VI, G. 44, 7–8, Pl. 5, 3 – is dated to the 4th century. 40 Petkovi} et al. 2005, 116, Cat. 7, 9, Pl. 5, 3. 41 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 42 Ru`i} 1994, 51, tip VII/12b, T. XXXVIII, 6, T. XL, 3 – is dated to the second half of the 4th and the first half of the 5th century. 43 Petkovi} 2010, 266–269, sl. 97, T. LXVIII, 1, kat. 1635 – type 34/E 1 is dated to the period of the reign of Theodosius I and Arcadius (380–408). 44 In the same grave, the partially preserved postcranial skeletal remains of an adult female (?) individual, of unknown age (Grave 69) were also discovered. 34 Note: The grave is located below the Late Roman graves 26 and 27, which partially damaged it. The human osteological material and coin finds from grave 26 possibly belong to this grave. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of a male (?) juvenile/ adult individual, aged 20–24 (dental age), were discovered.41 The only paleopathological change detected is the dislocation of the left ankle joint. Dental analysis of tooth 22 indicated moderately pronounced calculus and abrasion of the 2nd degree (exposed dentin). 93 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) INDIVIDUAL AGE juvenilis I (15-18 years) juvenilis II (19-22 years) adultus I (23-30 years) adultus II (31-40 years) maturus I (41-50 years) maturus II (51-60 years) senilis I (61-70 years) senilis II (71 and more) unknown age TOTAL NUMBER OF INDIVIDUALS MALE 1 1 - FEMALE 1 - - - 2 5 STARINAR LXIV/2014 UNDETERMINED SEX 1 - - 1 TOTAL 1 1 1 - 1 2 6 Table 1. Sex and age structure of individuals Tabela 1. Polna i starosna struktura individua ment points of both femurs (all muscle attachment points are pronounced along the lineae asperae). Markers of occupational stress in the form of hypertrophy (cortical defect) are also visible on the ligament attachment points of both femurs (lig. cruciatum posterius, lig. cruciatum anterius). On the left tibia, two so-called “squatting facets” are noticeable. Paleopathological changes were not noticed. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the ligament attachment point of the left femur (lig. capitis femoris). Grave 79 Sondage F I (V)/1995. Grave 91 Sondage F I (VI)/1995. Freely dug burial pit, with bottom at a relative depth of 1.00 m. Orientation: west-east. Skeleton supine with legs extended. A destroyed grave construction of stone and bricks, with the bottom of the burial pit at a relative depth of 1.25 m. 10 iron nails, arranged above the head, near the right upper leg and below the feet, testify to the existence of a wooden casket. Orientation: west-east, with a 20° deviation towards the north. Skeleton supine with legs extended and arms straight, placed alongside the body. Finds: 1–2. a consecration coin of Constantine I from 341–346 and a coin of Constantius II, minted between 341–348, near the right foot,46 3. two fragments of bronze plating, 4. an atypical fragment of a glass vessel. Note: The upper part of the skeleton was destroyed by grave robbers. The grave is situated below the Late Roman graves 38 and 32. No finds. Note: The grave was damaged by grave robbers. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, of unknown age, were discovered.47 The only paleopathological change detected is the dislocation of the right ankle joint. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attach- 45 Degree of bone preservation: category IV (partially preserved postcranial skeleton). 46 Petkovi} et al. 2005, 115–116, Cat. 6, 10, Pl. 11, 1, 4. 47 Degree of bone preservation: category IV (partially preserved postcranial skeleton). 94 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STATURE (cm) humerus radius ulna femur tibia fibula medium GRAVE 91 176 ¯ 4 176 ¯ 4 STARINAR LXIV/2014 diameter), 38 (occlusal, 0.5 and 0.2 cm in diameter), 44 (distal, 0.1 cm in diameter), 45 (so-called gross-gross caries) and 46 (so-called gross-gross caries). Buccal cysts are present on teeth 35 (0.65 cm in diameter) and 46 (0.90 cm in diameter). Occlusion: tête-à-tête. On the postcranial part of the skeleton, the epigenetic characteristic facies articularis superior atlantis bipartite was detected on the atlas. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle and ligament attachment points of the mandible (m. genioglassus, m. geniohyoideus), manubrium (m. sternocleidomastoideus, m. pectoralis major, lig. sternoclaviculare anterius) (Plate IV/2), ribs (Mm. levatores costarum), both scapulae (m. biceps brachii – Caput longum, m. triceps brachii – Caput longum, m. teres minor, m. infraspinatus, m. subscapularis), right clavicle (m. deltoideus), right humerus (m. supraspinatus, m. subscapularis, m. latissimus dorsi, m. teres major, m. pectoralis major, m. deltoideus, m. coracobrachialis), left humerus (m. deltoideus, m. coracobrachialis) and right femur (all muscle attachment points are pronounced along the lineae asperae, m. gastrocnemius – Caput mediale, m. adductor magnus, m. plantaris, m. gastrocnemius – Caput laterale). Other observations: On the tuberosity of the right tibia, brown coloured traces of metal oxidation are visible. The same traces are visible on the anterior side of the lower end of the right femur. GRAVE 141 168 ¯ 5 165 ¯ 5 168 ¯ 4 167 ¯ 5 Table 2. Stature Tabela 2. Telesna visina ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 50+ were discovered.48 The perceived paleopathological change is a trace of an injury in the form of a cut on the left side of the mentum (Pl. IV, 1). The cut could have been inflicted by a right-handed person with a short dagger or sword in a close frontal clash. Other noticeable paleopathological changes are a fused fissure in the lower half of the body of the right tibia, on the anterior side, osteoarthritis on the glenoid cavities of both scapulae and mild traces of cribra orbitalia on the right orbit roof. Dental analysis indicated the presence of the following teeth in the jaws: 11, 12, 13, 14, 15, 21, 22, 23, 24, 27, 31, 32, 33, 34, 35, 36 (root), 37, 38, 41, 42, 43, 44, 45 (root) and 46 (root). Teeth 16, 25, 26 (?), 28, 47 (?), 48 (?) were lost ante-mortem, and tooth 28 (?) postmortem. Abrasion of the 1st degree (in enamel) was noted on tooth 27, of the 2nd degree (exposed dentin) on teeth 22 and 44, of the 3rd degree (to the bottom of the fissure) on teeth 11 ( 4th), 12 ( 4th), 15, 21 ( 4th), 35, 38, 42 ( 4th) and 43, and of the 4th degree (pulp exposure) on teeth 13, 14, 23, 24, 31, 32, 33, 34, 37 and 41. Parodontopathy is very pronounced on the maxilla, and moderately on the mandible. Hypoplasia is moderately pronounced on the maxilla, and on the mandible it ranges from mild to moderate. The present jaw and dental arch anomalies are the distal rotation of teeth 34 (30°) and 35 (30°), as well as the dislocation of tooth 27 in place of tooth 26, lost ante-mortem. Caries was noted on the following teeth: 14 (distal, 0.5 cm in diameter), 15 (distal and mesial, so-called gross caries), 23 (distal, carious stain, 0.5 cm in diameter), 24 (mesial, 0.5 cm in diameter), 34 (mesial, 0.2 cm in diameter), 35 (distal, so-called gross caries), 36 (so-called gross-gross caries), 37 (mesial, between the crown and the root, 0.85 cm in Grave 99 Sondage F II (VII)/1996. Rectangular burial pit dug into the subsoil (greenyellowish marl), dimensions: 2.10 x 0.75 m, with bottom at a relative depth of 2.00 m. Orientation: west-east. Skeleton dislocated. Finds: 1. an amphora-shaped bronze belt-end among the dislocated bones (Pl. III, 2),49 2. fragments of bronze plating among the dislocated bones. 48 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 49 Petkovi} et al. 2005, 94, Cat. 1, Fig. 19 a, Pl. XI, G. 99, 1 – is dated to the second half of the 4th and the first half of the 5th century. 95 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) CRANIAL SKELETON Mandible Mandibular Index Mandibular Robustness Index Mandibular Branch Index Fronto–mandibular Index GRAVE 91 GRAVE 99 86.96 - 27.78 30.78 44.52 47.14 90.38 - STARINAR LXIV/2014 attachment points of the right clavicle (lig. conoideum) and left clavicle (lig. trapezoideum, lig. conoideum). Grave 14151 Sondage I/2012. Freely dug burial pit. The dimensions and relative depth cannot be determined since the grave was damaged by erosion and grave robbers. Two iron nails next to the left upper arm and one below the left leg, testify to the existence of a wooden casket. Orientation: southwest-northeast. Skeleton dislocated, with head and right side of the body missing. Table 3. Indices on the cranial skeleton Tabela 3. Indeksi na kranijalnom skeletu Finds: 1. rectangular silver belt plating, decorated longitudinally with an incised “waving spiral”, next to the left pelvic bone (Pl. II, 7),52 2. a bronze coin of Constantine I, minted 326–328, next to the left foot,53 3–4. 3 iron nails next to the left upper arm. Note: The grave was damaged by grave robbers. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 30–35, were discovered.50 The perceived paleopathological changes are a trace of an injury in the form of a shallow depression on the right femur (3.7 x 1 cm in size) and the appearance of osteosarcoma (?) – the bone is very decomposed and it is difficult to determine, with any certainty, precisely what type of tumour this is. Dental analysis indicated the presence of the following teeth in the jaws: 23, 24, 25, 26, 35, 36, 37, 38, 42, 43, 45, 46, 47 and 48. Teeth 21, 22, 27, 28, 31, 32, 33, 34, 41 and 44 were lost post-mortem. Abrasion of the 1st degree (in enamel) was noted on teeth 24, 25 ( 2nd), 35, 38, 45 and 48, and of the 2nd degree (exposed dentin) on teeth 23, 26, 36, 37, 42, 43, 46 and 47. Parodontopathy (maxilla and mandible) and calculus (mandible) are mildly pronounced. The present jaw and dental arch anomaly is the mesial rotation of tooth 24 (30°). Caries was not perceived. Occlusion could not be determined. The noticeable epigenetic characteristics on the norma occipitalis are ossa suturae lambdoideae (five on the right side, of the following dimensions: 1.1 x 0.8 cm, 0.8 x 1.5 cm, 0.75 x 2 cm, 0.9 x 1.9 cm and 1.8 x 2.5 cm; the occipital bone was damaged post-mortem on its left side, however, it appears that the ossa suturae lambdoideae exist there as well). Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the right humerus (m. supraspinatus). Markers of occupational stress in the form of hypertrophy (cortical defect) are also visible on the ligament Note: The grave was destroyed by the erosion of the eastern slope of the Slog hill and by grave robbers. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male (?) individual, of unknown age, were discovered.54 50 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 51 Excluding grave 141, which was explored during the site survey in 2012, in the course of the 2013 rescue excavations, 16 more graves were discovered at the Slog necropolis, 7 of which were Late Roman (G. 143, G. 149–152, G. 154 and G. 157) – Petkovi}, Radinovi}, Iliji} 2014, 33–38. In 2014, more 31 graves were excavated (G. 159–188), among them at least 8 from the Late Roman period. 52 Rectangular silver plating of a belt set, with faceted narrower sides, decorated longitudinally with an incised “waving spiral”, was attached by two rivets onto the leather strap (dim: 2.5 x 0.8 cm). Generally, it can be dated to the second half of the 4th and the first half of the 5th century, and more specifically to the last third of the 4th and the first decade of the 5th century (Bullinger 1969, 47, Typ Furfooz, 58 –59, Typ Abbeville, Abb. 16, 1, Abb. 36, 2, Abb. 42, 1, T. XXX, 2, T. LVI, T. LX, 1; Sommer 1984, 62–68, Gruppe 2–3, T. 46, 15.16, T. 58, 7–9; Nagy 2005, 450–456, T. Abb. 13, 7–18, Abb. 26–27). 53 Obv: CONSTAN TINVSAVG, rev: PROVIDEN TIAEAVGG, a gate fortified with two turrets and a star between them, no door, mint-mark of Thessalonica SMTS (RIC VII, 153). 54 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 96 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) POSTCRANIAL SKELETON Humerus Robusticity Index Cross–Section Index Radiohumeral Index Radius The Length–Thickness Index Cross–Section Index Femur Robusticity Index Pilastric Index Platymeric Index Tibia Platycnemic Index STARINAR LXIV/2014 GRAVE 69A GRAVE 79 GRAVE 91 GRAVE 99 GRAVE 141 - - 75 84.78 - 72.34 - 21.94 89.13 72.58 - - - - 8.41 5.68 - 106.90 110.34 - 12.34 103.51 - - 13.45 105.26 80.70 69.23 mesocn. - 70.13 eurycn. 71.43 eurycn. 68 mesocn. - - 68.83 mesocn. Table 4. Indices on the postcranial skeleton Tabela 4. Indeksi na postkranijalnom skeletu Other observations: On the 1st metacarpal bone, a trace of green coloured metal oxidation is visible. This kind of trace, considering that it is on a hand bone, could have been made by a part of the belt set. The perceived paleopathological change is myositis ossificans traumatica (2.5 x 8 cm in size), in the region of the m. tibialis posterior (on the middle of the body of the left tibia, on the posterior side) (Pl. IV, 4–7). The epigenetic characteristic observed on the postcranial part of the skeleton is perforatio fossae olecrani (0.3 cm in diameter) on the left humerus. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachments points of the left scapula (m. triceps brachii – Caput longum, m. subscapularis, m. infraspinatus, m. teres minor, m. teres major, m. deltoideus), left humerus (m. subscapularis, m. pectoralis major, m. latissimus dorsi, m. teres major, m. deltoideus, m. coracobrachialis), left radius (m. biceps brachii), left ulna (m. supinator, m. brachialis, m. flexor digitorum superficialis, m. pronator teres, m. flexor carpi ulnaris, m. extensor carpi ulnaris) and left femur (m. adductor magnus, m. gastrocnemius – Caput mediale). Markers of occupational stress in the form of hypertrophy (cortical defect) are also visible on the ligament attachment point of the left femur (lig. capitis femoris). PHASE II OF THE NECROPOLIS (380–410) (Pl. V) Grave 15 Sondage F I/1994. Rectangular burial pit dug into the subsoil (greenyellowish marl), dimensions: 1.80 x 0.80 m, above which there was a construction of broken stone and tegulae that was destroyed and dislocated next to the grave. Bottom at a relative depth of 1.20 m. Orientation: west-east, with a deviation of 13° towards the north. Skeleton dislocated. No finds. Note: The grave was robbed which resulted in the destruction of its stone construction. 97 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) INDIVIDUAL AGE juvenilis I (15-18 years) juvenilis II (19-22 years) adultus I (23-30 years) adultus II (31-40 years) maturus I (41-50 years) maturus II (51-60 years) senilis I (61-70 years) senilis II (71 and more) unknown age TOTAL NUBER OF JUVENILES AND ADULTS TOTAL NUMBER OF INDIVIDUALS MALE 1 4 1 - FEMALE 1 1 - - - STARINAR LXIV/2014 UNDETERMINED SEX - - TOTAL 1 4 1 - 1 1 - - - - - 8 - - 8 8 - - 8 Table 5. Sex and age structure of individuals Tabela 5. Polna i starosna struktura individua 33, 34, 35, 37, 42, 43, 44, 45 and 47, and of the 4th degree (pulp exposure) on teeth 22, 23, 24, 25, 26, 36, 41 and 46. On maxillary teeth, hypoplasia is mildly pronounced and parodontopathy is very pronounced. On the mandibular teeth, hypoplasia and parodontopathy are moderately pronounced, and calculus mildly to moderately pronounced. Occlusion: tête-à-tête. The noticeable epigenetic characteristic on the norma lateralis is os incisurae parietalis (on the left side 0.75 x 1.20 cm). On the postcranial part of the skeleton, facies articularis superior atlantis bipartite (on the left side) and trochanter tertius (on the left femur) were observed. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the ribs (Mm. levatores costarum), both clavicles (m. trapezius, m. deltoideus, m. pectoralis major, m. sternocleidomastoideus, m. subclavius), left humerus (m. deltoideus, m. coracobrachialis, m. brachialis), left ulna (m. flexor digitorum superficialis, m. pronator teres, m. brachialis, m. supinator), left radius (m. flexor digitorum superficialis), right and left tibia (m. quadriceps femoris, m. solei) and both femurs (m. gastrocnemius – Caput mediale, m. adductor magnus, m. plantaris, m. gastrocnemius – Caput laterale, m. popliteus, m. biceps femoris – Caput breve, m. adductor brevis, m. vastus intermedius, m. vastus medialis, m. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 35–45, were discovered.55 The perceived paleopathological changes are traces of injuries in the form of two cuts on the frontal bone (Pl. VIII, 1–3). Both of them were probably inflicted by the sharp blade of a sword. One is in the region above the right orbit. Although it is a serious injury (the separated part of the bone is 2 x 1.7 cm in size, and the cut is 0.4 cm deep), traces of healing are clearly visible in the picture (Pl. VIII, 3). The second injury is above the left orbit, also with traces of healing. The injuries were inflicted in a direct clash by a right-handed opponent. Other noticeable paleopathological changes are a fused fissure on the malleolus medialis of the right tibia, spondylolysis (on the atlas), osteoarthritis (on the right glenoid cavity and vertebrae L1, L2, L4 and L5, on the olecranon of the left ulna, the 1st right metacarpal bone, the pelvic bones, the right patella and on the lower tibia ends), spondylarthrosis (of the second degree on C4, C6 and C7), osteoporosis (on the promontorium, and the thoracic and lumbar vertebrae) and osteochondritis dissecans (near the upper end on the anterior side of the left femur, 0.5 cm in diameter). Dental analysis indicated the presence of the following teeth in the jaws: 12, 13, 16, 17, 22, 23, 24, 25, 26, 32, 33, 34, 35, 36, 37, 41, 42, 43, 44, 45, 46, 47 and 48. Teeth 11, 14, 15, 21, 31 and 38 were lost postmortem. Abrasion of the 1st degree (in enamel) was noted on tooth 48, of the 3rd degree (to the bottom of the fissure) on teeth 12, 13 ( 4th), 16 ( 4th), 17, 32, 55 Degree of bone preservation: category III (moderately preserved cranial and postcranial skeleton). 98 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STATURE (cm) humerus radius ulna femur tibia fibula medium GRAVE 15 170 ¯ 5 167 ¯ 4 168 ¯ 4 168 ¯ 4 GRAVE 25 165 ¯ 4 165 ¯ 4 GRAVE 26 166 ¯ 4 166 ¯ 4 GRAVE 76 181 ¯ 5 178 ¯ 5 183 ¯ 5 181 ¯ 5 STARINAR LXIV/2014 GRAVE 123 171 ¯ 4 171 ¯ 4 Table 6. Stature Tabela 6. Telesna visina ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 20–25, were discovered.57 The perceived paleopathological changes are spina bifida occulta on the sacrum, osteoarthritis (on both calcanei) and a bony growth (osteoma?), right of the mentum (about 1 cm in diameter). Dental analysis indicated the presence of the following teeth in the jaws: 12, 13, 14, 15, 16, 17, 21, 23, 24 (root), 25, 26, 27, 31, 32, 33, 34, 37, 38, 41, 42, 43, 44, 45 and 48. Teeth 36, 46 and 47 were lost ante-mortem, and tooth 35 post-mortem. Abrasion of the 1st degree (in enamel) was noted on teeth 15, 16, 25, 26, 27 and 37, of the 2nd degree (exposed dentin) on teeth 12 ( 3rd), 17, 23, 32, 34, 42, 44 and 45, and of the 3rd degree (to the bottom of the fissure) on teeth 13, 14, 21 ( 4th), 31, 33, 41 and 43. Parodontopathy is very pronounced on the maxilla. On the mandible, parodontopathy ranges from moderately to very pronounced. Calculus is moderately pronounced on the mandibular teeth. Hypoplasia is moderately pronounced on the maxillary and mandibular teeth. The present jaw and dental arch anomaly is the mesial rotation of teeth 31 (30°) and 41 (30°). A cyst is visible on the buccal side of tooth 47 (1 cm in diameter). Occlusion could not be determined. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the ribs (Mm. levatores costarum), both humeri (m. deltoideus, m. coracobrachialis, m. adductor magnus, m. vastus lateralis). Markers of occupational stress in the form of hypertrophy (cortical defect) are also visible on the ligament attachment points of both clavicles (lig. trapezoideum, lig. conoideum, lig. costoclaviculare), manubrium (lig. sternoclaviculare posterius) and both femurs (lig. capitis femoris, lig. cruciatum posterius, lig. cruciatum anterius). Both clavicles exhibit a flattening of the body (direction: superior-inferior) (Pl. VIII, 4). The incisura clavicularis on the manubrium is perceptibly extended to the left side (Pl. VIII, 5–6). The olecranon of the left ulna is slightly separated, and the margo interosseus is mildly pronounced. The margo interosseus of the left radius is also pronounced. Grave 25 Sondage F I/1994. Grave construction made of river pebbles, preserved on both sides of the head, below the feet and along the right side of the skeleton. The bottom of the pit at a relative depth of 1.10 m. Orientation: west-east, with a 3° deviation towards the south. Skeleton supine with legs extended, arms bent at elbows and hands on pelvis. Finds (Pl. VII, 1): 1. 32 decorative (?) iron rivets with pyramidal heads, from shoes or belt, 2. a frame of an iron buckle in the form of the Latin letter B, near the feet (Pl. VI, 1).56 56 Petkovi} et al. 2005, 91, Cat. 3, Fig. 18 c, Pl. III, G. 25, 1 – is dated to the last quarter of the 4th and the first half of the 5th century. 57 Degree of bone preservation: category III (moderately preserved cranial and postcranial skeleton). Note: The grave construction was destroyed by the digging of the medieval grave 6. 99 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) brachialis), right ulna (m. brachialis) and right radius (m. extensor digiti minimi, m. extensor indicis, m. extensor pollicis longus, m. extensor carpi radialis brevis, m. extensor carpi radialis longus, m. abductor pollicis longus, m. extensor pollicis brevis). Markers of occupational stress in the form of hypertrophy (cortical defect), 1.8 x 0.9 cm and 1.4 x 0.70 cm in size, are visible on the ligament attachment points of both clavicles (lig. costoclaviculare) (Pl. IX, 1). STARINAR LXIV/2014 Dental analysis indicated the presence of the following teeth in the jaws: 11, 12, 13, 14, 15, 16, 17, 21, 22, 23, 24, 25, 31 (?), 32, 33, 34, 38, 42, 44, root (35 or 45) and 48 (root). Abrasion of the 1st degree (in enamel) was noted on teeth 16, 17, 24, 25, 33 and 34 ( 2nd), of the 2nd degree (exposed dentin) on teeth 14 ( 3rd), 15, 22 and 23, of the 3rd degree (to the bottom of the fissure) on teeth 13, 21, 31 (?), 32, 42 and 44, and of the 4th degree (pulp exposure), on teeth 11 and 12. Parodontopathy is very pronounced on the maxilla and mandible. Hypoplasia and calculus are mildly pronounced. Decolourisation of the enamel is visible on the labial side of teeth 11 and 12. The present jaw and dental arch anomalies are the mesial rotation of teeth 11 and 21 (30°) and the distal rotation of tooth 15 (30°). Caries was noted on teeth 12 (distal, 0.3 cm in diameter and mesial, 0.1 cm in diameter), 13 (mesial, 0.4 cm in diameter), 15 (distal, 0.6 cm in diameter), 16 (lingual, so-called gross caries and buccal, 0.5 cm in diameter), 17 (buccal, 0.5 cm in diameter), 23 (mesial, caries (?), 0.2 cm in diameter), 24 (mesial, 0.4 cm in diameter), 25 (mesial, between the crown and the root, 0.5 cm in diameter), 37 (occlusal/ mesial, 0.7 cm in diameter), 38 (distal, between the crown and the root, 0.65 cm in diameter) and 48 (socalled gross-gross caries). A cyst exists on the buccal side of tooth 24 (0.4 cm in diameter). Occlusion could not be determined. Grave 26 Sondage F I/1994. Destroyed grave construction of river pebbles above the rectangular burial pit, with bottom at a relative depth of 1.00 m. Finds of nails, below the legs and along the left side of the skeleton, indicate the existence of a wooden casket. Orientation: west-east, with a 4° deviation towards the south. Skeleton supine with legs extended, arms bent at elbows and hands on pelvis. Finds (Pl. VII, 2): 1. a bronze consecration coin of Constantine I from 341–346 and two bronze coins of Constans, minted 341–348, to the left of the skull,58 2. a bronze buckle in the region of the pelvis,59 3. an amphora-shaped bronze belt-end in the region of the pelvis,60 4. a small, olive green glazed ceramic vessel with three handles, next to the right foot (Pl. VI, 2).61 Grave 27A Sondage F I/1994. Destroyed rectangular construction of river pebbles above a burial pit, dimensions: 1.80 x 0.90 m, with bottom at a relative depth of 1.70 m. Orientation: west-east, with a 7° deviation towards the south. Skeleton dislocated. Note: It is situated above the Late Roman grave 44. The coin finds from this grave might belong to the inventory of grave 44. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 20–40, were discovered.62 The perceived paleopathological change is a trace of an injury on the left processus zygomaticus, 2 x 1 cm in size (Pl. IX, 2). The injury was most probably inflicted in a direct clash by a right-handed individual with a blunt object. Other noticeable paleopathological changes are osteoarthritis (on calcanei), defect in the joint of the right ankle and spondyloarthrosis of the second degree on C3 and C4. The noticeable epigenetic characteristics on the norma lateralis are two foramen zygomaticofaciale (on the right zygomatic bone). 58 Petkovi} et al. 2005, 115, 117, Cat. 5, 16 –17. The shape of the bronze belt buckle could not be determined, since it was destroyed in conservation process. 60 Petkovi} et al. 2005, 93 –94, Cat. 2, Pl. IV, G. 26, 3 – is dated to the second half of the 4th and the 1st half of the 5th century. 61 Petkovi} et al. 2005, 111, Cat. 3, Fig. 21, type III/1, Pl. 9, 5; Cvijeti}anin 2006, 87, KAG 146; Jeremi} 2009, 93, Type II/9, Fig. 51, Cat. 249 –250, 252 – is dated to the 3rd and 4th century. 62 Degree of bone preservation: category III (moderately preserved cranial and postcranial skeleton). 63 Petkovi} et al. 2005, 89–90, 92, Fig. 18d, Pl. 10, 4, Cat. 6 – is dated to the last quarter of the 4th and the beginning of 5th century. 59 100 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 2. a bronze loop among the dislocated bones, 3. a conical shank of a leaf-shaped iron arrowhead among the dislocated bones. Finds (Pl. VII, 3): 1. a bronze buckle with a rectangular plate and a circular frame among the dislocated bones (Pl. VI, 3),63 Table 7. Indices on the cranial skeleton Tabela 7. Indeksi na kranijalnom skeletu CRANIAL SKELETON Primary cranial measures Cranial Index GRAVE 15 GRAVE 25 GRAVE 28 GRAVE 76 GRAVE 123 81.46 brachycranic - 82.02 brachycranic 75.92 mesocranic - Cranial Lenght-Height Index 76.40 hypsicranic - - - - Cranial Breadth-Height Index 93.79 metriocranic - - - - Mean Height Index 84.21 high - - - - Cranial Module 15.30 - - - - Mean Porion–Height Index 74.30 high - - 73.81 high - 5.88 Index of Flatness of the Cranial Base Fronto–Parietal Index Facial Skeleton Upper facial Index - - - - 66.9 metriometopic - 69.86 metriometopic 64.14 stenometopic - - - 49.28 euryenic - - - - 48.08 leptorrhinic - - - - 79.49 chamaeconchic - - - - 103.92 dolichuranic - - - - 78.35 leptostaphilinic - - 88 - - 94.02 - 41.27 52.38 - 33.90 42.86 55.56 40.15 - 46.21 - 85.09 - - - - The Nose Nasal Index The Orbits Orbital Index Maxilla Maxilloalveolar Index The Palate Palatal Index Mandible Mandibular Idex Mandibular Robustness Index Mandibular Branch Index Fronto–mandibular Index 101 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Dental analysis indicated the presence of the following teeth in the jaws: 12, 13, 14, 15, 18, 25, 28, 32, 33, 43 and 48. Teeth 16, 17, 26 and 27 were lost antemortem, and teeth 11, 21, 22, 23, 24, 31, 34, 35, 36, 37 and 38, post-mortem. Abrasion of the 1st degree (in enamel) was detected on teeth 18, 28 ( 2nd) and 48, of the 2nd degree (exposed dentin) on teeth 32, 33 and 43, of the 3rd degree (to the bottom of the fissure) on teeth 13, 14 and 15, and of the 4th degree (pulp exposure) on teeth 12 and 25. Parodontopathy is moderately pronounced on the maxilla. Hypoplasia is mildly pronounced on the maxillary and mandibular teeth. The present jaw and dental arch anomaly is the distal rotation of teeth 14 (30°) and 24 (30°). Caries was noted on tooth 48 (occlusal, 0.15 cm in diameter). A periapical cystic cavity is visible on the buccal side of tooth 24 (0.65 cm in diameter). Occlusion could not be determined. Other observations: Batricephalia is present on the occipital bone. Note: The stone construction of the grave was destroyed by grave robbers. It is situated above the Late Roman grave 44. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male (?) individual, aged 30–35, were discovered.64 Paleopathological changes were not noticed. Dental analysis indicated the presence of the following teeth in the jaws: 11, 14, 15, 16, 17, 18, 21, 23, 24, 25, 26, 27, 28, 32, 33, 34, 36, 37, 38, 43, 44, 45, 46, 47 and 48. Abrasion of the 1st degree (in enamel) was noted on teeth 17, 18, 27 ( 2nd), 28, 34, 37, 38, 44 ( 2nd), 47 and 48, and of the 2nd degree (exposed dentin) on teeth 11, 14, 15, 16, 21, 23, 24, 25, 26, 32, 33, 36, 43, 45 and 46. Hypoplasia is mildly pronounced on the maxillary and mandibular teeth. Caries was noted on teeth 16 (buccal, 0.45 cm in diameter), 18 (occlusal, in the form of a dot) and 14 (distal, 0.3 cm in diameter). Occlusion could not be determined. Grave 76 Sondage III/1995. Grave 28 Sondage F III/1994. Destroyed grave construction of river pebbles above a burial pit, with bottom at a relative depth of 1.75 m. Orientation: west-east, with a 2° deviation towards the south. Skeleton supine with legs extended, arms bent at elbows, and hands on pelvis. The skull was found in the region of the stomach. Freely dug burial pit, with bottom at a relative depth of 1.00 m. Orientation: west-east, with a 40° deviation towards the south. Skeleton supine with legs extended and arms straight, placed alongside the body. Finds: 1. a conical glass goblet next to the right foot (Pl. VI, 4).67 Finds: 1. a frame of an iron buckle in the shape of the Latin letter B, on the right pelvic bone (Pl. VII, 4).65 ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 40–50, were discovered.68 Note: The grave was robbed and on that occasion the grave construction was destroyed. ANTHROPOLOGICAL FINDS: In the grave, skeletal remains of an adult male individual, aged 30–40 (dental age 30–35), were discovered.66 The only paleopathological change detected is scurvy (changes are visible on the roofs of the orbits, zygomatic bones, the palatine bone, temporal bones and on bones of the skull base). The noticeable epigenetic characteristics on the norma frontalis are openings and notches in the supraorbital region and a partially fused metopic suture (sutura metopica). 64 Degree of bone preservation: category V (poorly preserved postcranial skeleton). In the same grave the remains of the poorly preserved dental material (category V) of a child, aged 10 years ¯ 30 months, of unknown sex, were discovered (Grave 27B). 65 See note no. 55. 66 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 67 Ru`i} 1994, 50–52, tip VII/12 b, T. XXXVIII, 12 – is dated to the second half of the 4th and the first half of the 5th century. 68 Degree of bone preservation: category III (moderately preserved cranial and postcranial skeleton). 102 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) The perceived paleopathological change is scurvy (changes are visible on the roofs of the orbits, the palatine bone, the mandibular condyles and on the wings of the sphenoid bone).69 Other noticeable paleopathological changes are a fused fracture of the head of the right femur (Pl. IX, 3), subperiosteal hematoma (in the lower half of the body of the left fibula on the facies medialis), superiosteal hematoma (on the medial side of the right femur throughout the whole length of the body), anomalies in the skeletal development (bipartite right patella) and dislocation of the disk in thoracic vertebrae. Dental analysis indicated the presence of the following teeth in the jaws: 11, 13, 14, 16 (root), 17, 18, 21, 22, 23, 24, 26 (root), 31, 32, 33, 34, 35, 36, 37, 38, 41, 42, 43, 44, 45, 46, 47 and 48. Teeth 12, 15 and 25 were lost post-mortem. Abrasion of the 1st degree (in enamel) was noted on teeth 18, 24 and 34, of the 2nd degree (exposed dentin) on teeth 13, 21, 22, 32, 33, 38, 44 and 48, of the 3rd degree (to the bottom of the fissure) on teeth 11 ( 4th), 14, 17, 23 ( 4th), 35, 37 ( 4th), 42, 43, 45 and 47 ( 4th), and of the 4th degree (pulp exposure) on teeth 31, 36, 41 and 46. On the maxillary teeth, hypoplasia is mildly to moderately pronounced, and parodontopathy moderately to very pronounced. On the mandibular teeth, hypoplasia is mildly pronounced, and parodontopathy mildly to moderately pronounced. The present jaw and dental arch anomalies are the mesial rotation of teeth 33 (30°) and 43 (45°), as well as the retrusion of teeth 22, 32, 42 and 45 towards the lingual side. Caries was noted on teeth 16 (so-called gross-gross caries) and 26 (so-called gross-gross caries). Periapical cystic cavities are visible on the buccal side of teeth 16 (0.6 cm in diameter) and 26 (0.8 cm in diameter). Fenestration (0.4 x 0.7 cm in size) is present on the buccal side of tooth 25. Occlusion: tête-à-tête. The detected epigenetic characteristic on the postcranial part of the skeleton is trochanter tertius on the left femur. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the ribs (Mm. levatores costarum), right and left clavicle (m. deltoideus), right and left scapula (m. deltoideus, m. biceps brachii – Caput longum, m. biceps brachii – Caput breve, m. triceps brachii – Caput longum, m. subscapularis, m. infraspinatus, m. teres minor, m. teres major), right humerus (m. subscapularis, m. latissimus dorsi, m. teres major, m. pectoralis major, m deltoideus, m. coracobrachialis, m. brachialis, m. brachioradialis, m. extensor carpi radialis longus, POSTCRANIAL SKELETON Scapula Scapular Indeks Clavicle Claviculohumeral Index Robustness Index STARINAR LXIV/2014 GRAVE 25 GRAVE 76 - 70.97 - 28 - 43.66 43.18 22.58 24.34 Table 8. Indices on the postcranial skeleton Tabela 8. Indeksi na postkranijalnom skeletu m. extensor carpi radialis brevis, m. pronator teres, m. extensor digitorum, m. extensor digiti minimi, m. extensor carpi ulnaris, m. supinator, m. flexor carpi radialis, m. palmaris longus, m. flexor carpi ulnaris, m. flexor digitorum superficialis), left humerus (m. latissimus dorsi, m. teres major, m. pectoralis major, m deltoideus, m. coracobrachialis, m. brachialis, m. brachioradialis, m. extensor carpi radialis longus, m. extensor carpi radialis brevis, m. pronator teres, m. extensor digitorum, m. extensor digiti minimi, m. extensor carpi ulnaris, m. supinator, m. flexor carpi radialis, m. palmaris longus, m. flexor carpi ulnaris, m. flexor digitorum superficialis), right and left ulna (m. flexor digitorum superficialis, m. pronator teres, m. brachialis, m. supinator, m. pronator quadratus (on the left ulna it created a sulcus: 0.5 cm in width, 3.6 cm in length and 0.03 cm in depth), m. triceps brachii, m. flexor carpi ulnaris), right radius (m. biceps brachii, m. abductor pollicis longus, m. extensor pollicis brevis, m. pronator teres, m. extensor digiti minimi, m. extensor digitorum, m. extensor indicis, m. extensor pollicis longus, m. extensor carpi radialis brevis, m. extensor carpi radialis longus, m. abductor pollicis longus), left radius (m. biceps brachii, m. abductor pollicis longus, m. extensor pollicis brevis, m. pronator teres), right and left femur (m. iliopsoas, m. adductor magnus, m. gluteus maximus, m. vastus lateralis, m. pectineus, m. vastus medialis, m. adductor brevis, m. vastus intermedius, m. adductor longus, m. biceps femoris – Caput breve). The olecranons of both 69 Scurvy appears as a consequence of the deficiency of vitamin C in the diet. It is common in times of drought or war, when the diet is unvaried and fruit and vegetable intake is reduced. 103 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) ulnas are slightly separated. The corpus sterni is slightly curved to the left side. The spina and acromion of both scapulae are deformed. The corpus costae of seven fragmented ribs have a more pronounced curvature than usual (near the facies articularis tuberculi costae). STARINAR LXIV/2014 2. a leaf-shaped iron arrowhead in the region of the right pelvic bone,71 3. an animal bone (sheep or goat’s pelvic bone) in the region of the chest. Note: The grave was probably robbed, since the leg bones were dislocated. Grave 97 Sondage F II (VII)/1995. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged around 40, were discovered.72 The only paleopathological change detected is osteoarthritis (on the tuberosity of the left tibia, on both tali and both calcanei). Dental analysis indicated the presence of the following teeth in the jaws: 12, 13, 21, 24, 25, 41, 43, 46 and 47. Abrasion of the 1st degree (in enamel) was noted Remains of a stone construction around the skull. The bottom of the burial pit at a relative depth of 2.00 m. Orientation: west-east. Skeleton supine with legs extended. Finds (Pl. VII, 5): 1. a shank of an iron spear to the right of the skull,70 Table 9. Indices on the postcranial skeleton Tabela 9. Indeksi na postkranijalnom skeletu POSTCRANIAL SKELETON Humerus GRAVE 15 GRAVE 25 GRAVE 26 GRAVE 27A GRAVE 28 GRAVE 76 GRAVE 97 GRAVE 123 17.18 Robusticity 17.90 Index 76.09 76.19 77.78 92.10 Cross–Section 71.70 75.61 82.93 Index 73.24 Radiohumeral 73.86 Index Radius 16.94 The Length– 16.19 Thickness Index 4.84 Cross–Section 4.66 Index 14.11 The Length– Breadth Index Ulna 16.36 14.17 Caliber 17.76 Index Femur 13.93 13.13 Robusticity 14.81 12.72 Index 120.37 100 93.55 107.02 95.08 Pilastric 122.22 91.07 85.71 105.45 100 Index 93.33 eurym. 76.93 platy. 69.33 platym. 87.50 eurym. 77.61 platy. Platymeric 103.23 sten. 70.59 platy. 75.64 platym. 80.88 platym. 87.69 eury. Index Tibia 22.60 20.41 The Length– 20.98 Breadth Index 80.30 euryc. 65.15 meso. 66.67 mesoc. Platycnemic 80.60 euryc. 68.57 mesoc. 74.63 eurycn. Index 104 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 48, of the 3rd degree (to the bottom of the fissure) on teeth 11, 12, 16, 21, 22 and 26, and of the 4th degree (pulp exposure) on tooth 28. On the maxillary teeth, hypoplasia is mildly to moderately pronounced and parodontopathy moderately to very pronounced. On the mandibular teeth, both hypoplasia and parodontopathy are moderately pronounced. Calculus is mildly pronounced on both maxillary and mandibular teeth. The present jaw and dental arch anomaly is the mesial rotation of teeth 31 (30°) and 41 (30°). Decolourisation of the enamel is visible on teeth 16 (buccal), 21 (labial), 23 (labial) and 24 (labial). Caries was noted on teeth: 16 (distal, 0.3 cm in diameter), 22 (distal, carious stain, 0.25 cm in diameter), 23 (mesial, carious stain, 0.2 cm in diameter), 24 (occlusal, 0.2 cm in diameter), 26 (buccal, carious stain, 0.35 cm in diameter), 37 (distal, so-called gross caries), 45 (distal, carious stain, 0.25 cm in diameter) and 47 (mesial, 0.35 cm in diameter). A cyst, 2.1 x 1.4 cm in size, was noted buccally on teeth 17 and 18. On the buccal side of the crowns of teeth 15 and 25, vertical furrows are visible (0.4 cm in length, 0.35 cm in width and 0.05–0.1 cm in depth). Occlusion could not be determined. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the right and left clavicle (m. subclavius), left scapula (m. triceps brachii – Caput longum), right and left humerus (m. subscapularis, m.triceps brachii – Caput mediale), right and left ulna (m. brachialis, m. triceps brachii, m. flexor digitorum profundus, m. flexor carpi ulnaris, m. anconeus), right radius (m. supinator, m. flexor digitorum superficialis), left radius (m. biceps brachii, m. supinator, m. flexor digitorum superficialis), right and left femur (all muscle attachment points are pronounced along the linea asperae). Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the ligament on teeth 12, 13 ( 2nd), 24, 25 and 43, of the 2nd degree (exposed dentin) on tooth 47, and of the 3rd degree (to the bottom of the fissure) on teeth 21, 41 and 46. Hypoplasia is moderately pronounced on the maxillary and mandibular teeth. Occlusion could not be determined. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the right humerus (m. deltoideus, m. brachialis, m. coracobrachialis), left humerus (m. deltoideus) and both femurs (m. gastrocnemius – Caput mediale). Grave 123 Sondage F I (VII)/1996. Freely dug burial pit, with bottom at a relative depth of 1.20 m. Orientation: west-east, with a 7° deviation towards the north. Skeleton supine with legs extended, arms bent at elbows and hands crossed on the stomach. The bones of the lower leg and the feet are missing. Finds (Pl. VII, 6): 1. an iron tri-bladed arrowhead between the ribs on the right side of the chest cavity (Pl. VI, 5),73 2. a bronze buckle with an oval plate and an oval frame in the region of the pelvis.74 Note: It is positioned below the Late Roman grave 102, which damaged it. The iron arrowhead does not represent a grave offering, and the place where it was found indicates the possible cause of death of the deceased. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 45–65, were discovered.75 The perceived paleopathological changes are traces of the healing of an injury in the region of the right asterion, spondyloarthrosis (of the 1st degree on T11 and T12, and the 3rd degree on T9 and T10), Schmorl’s node (on T10) and a defect in the sacroiliac joints. Dental analysis indicated the presence of the following teeth in the jaws: 11, 12, 13, 14, 15, 16, 21, 22, 23, 24, 25, 26, 28, 31, 32, 33, 34, 35 (root?), 36 (root?), 37, 41, 42, 43, 45, 47 and 48. Teeth 17, 18, 44 and 46 were lost ante-mortem, and tooth 38 post-mortem. Abrasion of the 2nd degree (exposed dentin) was noted on teeth 13, 14, 15, 23 ( 3rd), 24 and 25, of the 2nd–3rd degree on teeth 31, 32, 33, 34, 35, 37, 41, 42, 43 and 70 Petkovi} et al. 2005, 100 –101, Cat. 2, Pl. X, G. 97, 2. Petkovi} et al. 2005, 100 –101, Cat. 4, Pl. X, G. 97, 1 – arrow type characteristic of the 3rd and the 4th century. 72 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 73 Petkovi} et al. 2005, 100 –101, Cat. 3, Pl. XI, G. 123, 1 – arrow type characteristic of the end of the 4th and the 5th century. 74 Petkovi} et al. 2005, 88, Fig. 18 a, Cat. 1, Pl. XI, G. 123, 2 – is dated to the second half of the 4th and the beginning of the 5th century. 75 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 71 105 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 the mandible, parodontopathy is very pronounced and calculus moderately pronounced. Hypoplasia is very pronounced on both the maxillary and mandibular teeth. A jaw and dental arch anomaly of hypodontia of teeth 38 and 48 is present. Caries was noted on teeth: 11 (distal, 0.3 cm in diameter) and 14 (distal, between the crown and the root, 0.5 cm in diameter). Occlusion could not be determined. The perceived epigenetic characteristic on the norma occipitalis is a mildly pronounced linea nuchae suprema. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the ribs (Mm. levatores costarum), right humerus (m. deltoideus, m. brachioradialis, m. brachialis, m. coracobrachialis, m. pronator teres, m. triceps brachii – Caput laterale), left humerus (m. supraspinatus, m. subscapularis, m. latissimus dorsi, m. teres major, m. pectoralis major, m. deltoideus, m. coracobrachialis, m. brachioradialis, m. extensor carpi radialis longus, m. extensor carpi radialis brevis) and left ulna (m. supinator, m. brachialis, m. pronator teres, m. flexor digitorum superficialis, m. pronator quadratus, m. triceps brachii, m. flexor carpi ulanaris, m. extensor carpi ulnaris). Markers of occupational stress in the form of hypertrophy (cortical defect) are also visible on the ligament attachment points of the left clavicle (lig. trapezoideum, lig. conoideum) and right femur (lig. capitis femoris). attachment point of the left femur (lig. capitis femoris), and on the ligaments of both clavicles (lig. trapezoideum, lig. conoideum, lig. costoclaviculare). Special observations: In the course of the excavation of the skeleton, among the ribs, on the right side, an arrowhead was found. A tri-bladed iron arrowhead in the chest cavity undoubtedly speaks to the cause of death of this person. Unfortunately, on the partially preserved rib fragments, no traces of injury were detected. PHASE III OF THE NECROPOLIS (410–450) (Pl. X) Grave 100 Sondage F I (VII)/1996. Destroyed grave construction of stone and brick joined with mortar, with bottom at a relative depth of 0.70 m. Orientation: west-east. Skeleton dislocated. Finds: 1. an atypical fragment of a glass goblet among the dislocated bones, 2. two atypical fragments of a ceramic lamp among the dislocated bones. Note: The grave was destroyed by grave robbers. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 40–50, were discovered.76 The perceived paleopathological changes are an irregularly fused fracture of the right fibula in the upper quarter of its body (Pl. XI, 1–2), a trace of a subperiosteal hematoma near the upper end of the right tibia (on the lateral side), osteoarthritis (on the head of the right femur, on the tuberosity of the right tibia and on both calcanei), irregularly fused fracture of the neck of the right femur and a fracture of one rib, accompanied by osteomyelitis. Dental analysis indicated the presence of the following teeth in the jaws: 11, 13, 14, 25, 31, 32, 33, 34, 36, 37, 41, 42, 43, 44 and 47. Teeth 35, 45 and 46 were lost post-mortem. Abrasion of the 1st degree (in enamel) was noted on tooth 47 ( 2nd), of the 2nd degree (exposed dentin) on teeth 31, 32, 34 and 44, of the 3rd degree (to the bottom of the fissure) on teeth 11, 13, 25 ( 4th), 33, 37 ( 4th), 41 ( 4th), 42 and 43, and of the 4th degree (pulp exposure) on teeth 14 and 36. On Grave 102 Sondage F I (VII)/1996. Destroyed rectangular grave construction of river pebbles and bricks, with bottom at a relative depth of 1.00 m. Orientation: west-east, with a 5° deviation towards the north. Skeleton supine with legs extended, upper part of skeleton dislocated. Finds: 1. an iron buckle with a circular frame in the region of the pelvis (Pl. VII, 7).77 76 Degree of bone preservation: category IV (partially preserved cranial and postcranial skeleton). 77 Petkovi} et al. 2005, Fig. 18 e, 90, 92, Cat. 7 – is dated to the second half of the 4th and the first half of the 5th century. 106 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) INDIVIDUAL AGE juvenilis I ( 15-18 years ) juvenilis II ( 19-22 years ) adultus I ( 23-30 years ) adultus II ( 31-40 years ) maturus I ( 41-50 years ) maturus II ( 51-60 years ) senilis I ( 61-70 years ) senilis II ( 71 and more ) unknown age TOTAL NUMBER OF INDIVIDUALS MALE 1 - FEMALE - 2 - - - 3 - STARINAR LXIV/2014 UNDETERMINED SEX - - - - - - TOTAL 1 2 - 3 Table 10. Sex and age structure of individuals Tabela 10. Polna i starosna struktura individua on tooth 34, of the 3rd degree (to the bottom of the fissure) on teeth 33 ( 4th) and 43 ( 4th), and of the 4th degree (pulp exposure) on teeth 13, 14, 22, 23, 25, 31, 42 and 44. Parodontopathy is very pronounced on the maxilla and mandible. Caries was noted on tooth 28 (mesial, between the crown and the root, 0.3 cm in diameter). A periapical cystic cavity, as a result of caries, is visible labially on tooth 22 (0.5 cm in diameter) and buccally on tooth 27 (0.7 cm in diameter). Occlusion: tête-à-tête. The epigenetic characteristics noticeable on the norma frontalis are openings and notches in the supraorbital region, on the norma verticalis – foramina parietalia (one on the right and one on the left), on the norma occipitalis – linea nuchae suprema (very pronounced), on the norma lateralis – two foramen zygomaticofaciale (on the left zygomatic bone) and sutura squamomastoidea (on both mastoid processes). On the postcranial part of the skeleton, perforatio fossae olecrani on the left humerus was also noted. Markers of occupational stress in the form of hypertrophy (cortical defect) are visible on the muscle attachment points of the ribs (Mm. levatores costarum), left scapula (m. triceps brachii – Caput longum, m. Note: The grave construction was destroyed by grave robbers. It is situated above the Late Roman grave 123. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of an adult male individual, aged 46, were discovered.78 The perceived paleopathological changes are traces of several injuries on the frontal bone (Pl. XI, 3) and several fused fractures on the postcranial part of the skeleton (these are fused fractures of the ribs (Pl. XI, 4), the right ulna and the malleolus lateralis of the right fibula). The injuries on the frontal bone, in the form of shallow perforated circular depressions, are present in two almost parallel lines (Pl. XI, 3).79 Although serious, these injuries on the frontal bone, most probably inflicted by a blunt object, which had on it several spikes (cestus?), show traces of healing. Other noticeable paleopathological changes are osteoarthritis (on the left glenoid cavity, on the 1st metacarpal bone, on the pubic and ischiadicus parts of both pelvic bones, on both patellas, and on the ends of the femurs and tibias), Schmorl’s nodes (on five thoracic and one lumbar vertebrae), osteoporosis (on vertebrae L1–L5) and osteochondritis dissecans (on both os capitate). Dental analysis indicated the presence of the following teeth in the jaws: 13 (root), 14 (root), 22 (root), 23 (root), 25 (root), 28, 31 (root), 33, 34, 42, 43 and 44. Teeth 15, 21 (?), 24, 26, 27 (one root), 45 and 46 were lost ante-mortem, and teeth 11, 12, 32, 35 and 41, postmortem. Abrasion of the 1st degree (in enamel) was noted on tooth 28, of the 2nd degree (exposed dentin) 78 Degree of bone preservation: category III (moderately preserved cranial and postcranial skeleton). 79 The frontal bone around the first two depressions was damaged post-mortem. 107 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) stature (cm) humerus radius ulna femur tibia fibula medium GRAVE 100 171 ¯ 4 171 ¯ 4 GRAVE 102 183 ¯ 5 172 ¯ 4 176 ¯ 4 177 ¯ 4 GRAVE 108 171 ¯ 5 170 ¯ 5 173 ¯ 5 170 ¯ 4 171 ¯ 5 CRANIAL SKELETON Primary cranial measures Cranial Index Cranial Lenght-Height Index Cranial Breadth-Height Index Mean Height Index Approximate Cranial Size Mean Porion–Height Index Index of Flatness of the Cranial Base Fronto–Parietal Index Prognatic Index Facial Skeleton Total Facial Index Upper Facial Index The Nose Nasal Index The Orbits Orbital Index Maxilla Maxilloalveolar Index The Palate Palatal Index Mandible Mandibular Robustness Index Mandibular Branch Index Fronto–mandibular Index Table 11. Stature Tabela 11. Telesna visina subscapularis, m. infraspinatus, m. teres minor, m. teres major), left humerus (m. brachioradialis, m. brachialis, m. extensor carpi radialis longus, m. extensor carpi radialis brevis, m. extensor digitorum, m. extensor digiti minimi, m. extensor carpi ulnaris, m. supinator, m. pronator teres, m. flexor carpi radialis, m. palmaris longus, m. flexor carpi ulnaris, m. flexor digitorum superficialis, m. triceps brachii – Caput mediale, m. flexor carpi ulnaris, m. anconeus), right and left ulna (m. flexor digitorum superficialis, m. pronator teres, m. brachialis, m. supinator, m. flexor digitorum profundus, m. pronator quadratus, m. triceps brachii, m. anconeus, m. flexor digitorum profundus, m. flexor carpi ulnaris, m. extensor carpi ulnaris, m. abductor pollicis longus, m. extensor pollicis brevis, m. extensor pollicis longus, m. extensor indicis) (Pl. XI, 5), right radius (m. brachioradialis, m. pronator quadratus, m. extensor digitorum, m. extensor indicis, m. extensor pollicis longus, m. extensor carpi radialis brevis, m. extensor carpi radialis longus, m. abductor pollicis longus, m. extensor pollicis brevis), left radius (m. pronator teres, m. abductor pollicis longus, m. extensor pollicis brevis, m. supinator, m. biceps brachii, m. flexor digitorum superficialis), right femur (m. iliopsoas, m. adductus magnus, m. gluteus maximus (and all others along the lineae asperae), m. gastrocnemius – Caput mediale, m. adductor magnus, m. plantaris, m. gastrocnemius – Caput laterale, m. popliteus) (Pl. XI, 6–7) and left femur (m. piriformis, m. gluteus minimus, m. vastus lateralis, m. iliopsoas, m. adductus magnus, m. quadriceps femoris, m. gluteus maximus (and all others along the lineae asperae), m. gastrocnemius – Caput mediale, m. adductor magnus, m. plantaris, m. gastrocnemius – Caput laterale, m. popliteus). Markers of occupational stress in the form of hypertrophy (cortical STARINAR LXIV/2014 GRAVE 100 GRAVE 108 80.59 brachycranic 72.35 orthocranic 89.78 tapeinocranic 80.13 medium - 14.33 72.96 high 5.69 low 69.34 metriometopic 91.21 ortognatic - 92.91 leptoprosopic 55.90 leptenic - 43.14 leptorrhinic - 86.84 mesoconchic 89.74 mesoconchic - 121.43 brachyuranic - 75.55 leptostaphylnic 35 40.32 44.93 42.74 97.43 mesomandibular - - Table 12. Indices on the cranial skeleton Tabela 12. Indeksi na kranijalnom skeletu 108 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) POSTCRANIAL SKELETON Humerus Robusticity Index Cross–Section Index Radiohumeral Index Radius The Length–Thickness Index Cross–Section Index The Length–Breadth Index Ulna Caliber Index Femur Robusticity Index Pilastric Index Platymeric Index Tibia The Length–Breadth Index Platycnemic Index Fibula The Length–Breadth Index STARINAR LXIV/2014 GRAVE 100 GRAVE 102 GRAVE 108 83.33 86.96 - 78.18 - 17.19 17.98 85 87.18 75.71 - - 16.89 4.89 14.44 15.22 - 18 - 17.57 - - 14.50 15.22 127.59 108.57 69.23 platymeric 68.83 platymeric 11.36 11.67 104.08 104 73.01 platymeric 85.18 eurymeric 18.48 76.12 eurycnemic 67.65 mesocnemic 21.28 73.96 eurycnemic 79.66 eurycnemic 85.45 eurycnemic 11.85 - - - Table 13. Indices on the postcranial skeleton Tabela 13. Indeksi na postkranijalnom skeletu Grave 108 Sondage F I (VII)/1995. defect) are also visible on the ligament attachment points of both femurs (lig. iliofemorale, lig. pubofemorale, lig. cruciatum posterius, lig. cruciatum anterius). The olecranons of both ulnas are slightly separated. Other observations: Brown traces made by metal oxidation are visible on the ribs, left scapula, lower end of the left humerus, lower end of the left tibia, ends of the left femur and on the right pelvic wing. The position of the stains made by metal (iron) oxidation indicates that the deceased might have been buried with weapons (these are most likely a spearhead at shoulder height, a long sword along the left arm and left leg, and a short dagger in the region of the right pelvic bone). Rectangular grave construction, a cist of tegulae laid edgewise, dimensions: 1.80 x 0.50 m. Bottom tiled with horizontally laid bricks, at a relative depth of 2.00 m. Orientation: west-east. Skeleton dislocated. No finds. Note: The upper part of the grave construction, the lid of the cist, which most likely consisted of horizon- 109 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 The noticeable epigenetic characteristic on the norma occipitalis is os sutura lambdoidea (one on the right side, 1.1 x 0.5 cm in size). On the postcranial part of the skeleton, foramen processus transversi bipartitum on C5 was also observed. Markers of occupational stress in the form of the hypertrophy (cortical defect) are visible on the muscle attachment points of the right and left clavicle (m. deltoideus), right and left scapula (m. triceps brachii – Caput longum), right and left humerus (m. deltoideus, m. brachialis), right and left ulna (m. brachialis) (Pl. XII, 4), both femurs (m. adductor magnus). The clavicles are slightly deformed in the direction from superior to inferior (it is possible that this, as well as the occurrence of Schmorl’s nodes, is a consequence of carrying heavy loads).83 The olecranons of both ulnas are slightly separated. Additional facets are visible on the lower ends of the right and left 1st metatarsal bone. tally laid tegulae, was destroyed by grave robbers. The grave is situated above the Late Roman grave 128. ANTHROPOLOGICAL ANALYSIS: In the grave, skeletal remains of a juvenile male individual, aged about 20, were discovered.80 The perceived paleopathological changes are traces of several injuries. Most obvious of these is a tribladed arrow inflicted injury (right facies orbitalis) (0.7 x 0.7 x 0.7 cm in size) with traces of healing (Pl. XII, 1–3).81 Unfortunately, the eye could not have been saved. Injuries also worth mentioning are those in the form of shallow depressions on the right processus zygomaticus and on the mandible (left of the mentum) (Pl. XII, 2–3). Other noticeable paleopathological changes are anomalies in the skeletal development (foramen sternale, a cleft below the dens on the body of the axis), osteoarthritis (on the olecranons of both ulnas and on the heel bones), Schmorl’s nodes (on T6–T12 and on L1 and L2 (Pl. XII, 5), 1–2.5 cm in length and 0.5–1 cm in width), dislocation of the right foot,82 a slight collapse of the body of the thoracic and lumbar vertebrae, cribra orbitalia (particularly on the roof of the left orbit), cribra humera (near the upper ends of both humeri; the surface area of the affected parts of the bones is 1.5 x 1.5 cm and 1.2 x 1 cm), cribra femora (near the upper ends of both femurs; the surface area of the affected parts of the bones is 2.5 x 1.5 cm and 1.5 x 3 cm), osteoporosis (on vertebra L5), infectious osteomyelitis (on the right talus and the right calcaneus) and osteochondritis dissecans (on the right and left 3rd metatarsal bone, 1 x 1 cm in size). Dental analysis indicated the presence of the following teeth in the jaws: 13, 15, 16, 17, 22, 23, 24, 25, 26, 27, 32, 33, 36, 37, 38, 42, 43, 44, 45, 46, 47 and 48. Teeth 11, 12, 14, 21, 28, 31, 34, 35 and 41 were lost post-mortem. Abrasion of the 1st degree (in enamel) was noted on teeth 15, 17, 22, 23, 32, 33, 37, 38, 43, 44, 45, 47 and 48, of the 2nd degree (exposed dentin) on teeth 16, 42 and 46, and of the 3rd degree (to the bottom of the fissure) on teeth 26 and 36. Hypoplasia, calculus and parodontopathy (moderately pronounced on the maxilla) were mildly pronounced. Decolourisation of the enamel is visible on the labial and buccal side on the maxillary teeth. The present jaw and dental arch anomalies are hypodontia of tooth 18, the distal rotation of tooth 32 (30°), the labial curvature of tooth 24 and lingual of teeth 37, 38, 44 and 48. Caries was noted on tooth 17 (occlusal, in the form of a dot) and 27 (occlusal, in the form of a dot). Occlusion: tête-à-tête. DISCUSSION Burying the deceased with belt sets and fibulae, which are not only items of clothing but also military equipment, is a common occurrence in Roman necropolises throughout the Empire. Military belts and certain types of fibulae were an indication of military rank and/or a specific unit, and at the same time they revealed the social status of the person who wore them.84 Therefore, it was preferable for the deceased to be buried in a “uniform”, with items of military equipment which defined him as a member of the military elite. This was of particular importance in the border provin- 80 Degree of bone preservation: category II (incomplete, well preserved cranial and postcranial skeleton). 81 The entry angle of the arrow undoubtedly shows that, when wounded, this youth was probably on a higher elevation in relation to the opponent, or maybe on a horse. (Pl. XII, 1–3). 82 There are almost no individuals in these 17 graves who did not have some kind of problem with ankle joints (osteoarthritis, sprains or dislocations). Ankle joints in wartime circumstances obviously suffered a great deal of pressure. What contributed to it most was the carrying/wearing of heavy military equipment, pivoting and swift changes of direction in battle, jumping over obstacles, jumping from heights, dismounting from a horse, etc. 83 The military equipment of Roman soldiers was extremely heavy. Just the wooden containers with leather straps, similar to modern backpacks, weighed 20–40 kg. 84 Bullinger 1969, 68–70; Petkovi} 2011a, 121–124; Quast 2012, 254–256. 110 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 iron belt buckles with a frame in the shape of the Latin letter B and a prong that exceeds the frame (Pl. VI, 1, Pl. VII, 1, 4), a bronze belt buckle of the “Crimean type”, with a rectangular plate, circular frame and a prong with a zoomorphic thickening on the tip, which exceeds the frame (Pl. VI, 3, Pl. VII, 3) and a fluted glass goblet of a conical shape (Pl. VI, 4), characteristic of the late phase of the Chernyakhov – Sântana de Mureº culture.91 Furthermore, in grave 97, in the region of the chest of the buried individual, an animal bone, part of the pelvis of a goat or a sheep, was found. This was a remnant of a posthumous feast, given as an offering to the deceased. The placement of meat into the grave is not a part of the Roman burial ritual, whilst posthumous feasts are characteristic of the funerary cult of the late phase of the Chernyakhov – Sântana de Mureº culture.92 It is therefore fair to assume that the buried men were newcomers, most probably recruited from within the ranks of the Danube foederati. In addition, the type of the weaponry points to the Gothic foederati from the Lower Danube, the majority of which made up the ces. On the other hand, the graves in which weapons were placed along with the deceased are rare in the necropolises of the Roman period but they do, however, appear in all provinces of the Empire.85 The genesis and significance of these graves differ from one region of the Roman Empire to the next. In the region of the central Balkans, three typologically and chronologically differentiated groups of Roman graves with weapons are found: 1. graves with cremated deceased, of the Mala Kopa{nica – Sase type, dated to the 2nd and the first half of the 3rd century, in which the inclusion of weapons and military equipment can be explained by a strong autochthonous tradition and weak Romanisation,86 2. graves with cremated and inhumed deceased – masonry tombs, tumuli and sarcophagi of members of the provincial military elite, dated to the last decades of the 3rd and the beginning of the 4th century, the Tetrarchy period, whereby weapons and military equipment are included as insignia of the status of the buried individual, as a consequence of the merging of autochthonous tradition and Roman culture,87 3. graves with inhumed deceased, in freely dug pits or in cists made of tegulae, dated to the second half of the 4th and the first half of the 5th century, in which burial with weapons is a consequence of the acceptance of “barbarian” funereal customs into the Late Roman culture.88 What have not been considered here are graves with weapons which can undoubtedly be ascribed to barbarian populations in the territory of the Empire, such as the eastern Germanic foederati, noted in separate necropolises of the 5th century at Singidunum and Viminacium.89 The burial with weapons in grave 97 belongs to phase II of the Slog necropolis, that is to the period after the Battle of Hadrianopolis, in the last decades of the 4th and first decade of the 5th century, when the fortification’s garrison consisted of an equestrian unit of pseudocomitatenses Timacenses, an auxiliary unit of light cavalry of the “Alanic type” (Pl. V, G. 97, Pl. VII, 5).90 Devastated grave 27 also belongs to the same phase of the necropolis, in which dislocated bones of an adult male individual and a child of about 10 years of age were discovered, along with parts of a bronze belt set and the conical shank of a leaf-shaped arrowhead (Pl. V, G. 27, Pl. VI, 3, Pl. VII, 3). The influence of barbarians recruited to the aforementioned cavalry unit is not only reflected in the material culture of this period of the fortification but also in the grave finds, such as 85 Diaconscu 1999, 204–206, 215–216; Quast 2012, 247–256, Fig. 8; Tejral 2012, 118–120, Fig. 3. 86 Such graves were noted at the eponymous necropolis of the cremated deceased near Leskovac, at the Vi{e Grobalja necropolis at Viminacium and at the necropolis in Davidovac, near Vranje. (Zotovi} 1968, 24; Fidanovski, Cvijeti}anin 2005, 91, 93, T. III, 2, T. IV, 5, graves XIV and XVII; Zotovi}, Jordovi} 1990, 84, 93, 103, T. LXXXV, 7, T. CXXXII, 4, T. CLXXIV, 4, G1–58, G1–137, G1–222). 87 Burials of this group were noted in Silistra (Durostorum) in Bulgaria, Taranes in Macedonia, at Gamzigrad (Romuliana) and [arkamen in Serbia, and Hârºova (Carsium) in Romania (Diaconescu 1999, 205–206, Abb. 2–3; Ivanovski 1987; Petkovi} 2009, 253 et sequ., Figs. 4–12;Tomovi} et al. 2005, 44–45, 170–172,G. 6, Cat. 90). 88 We assume that there is a larger number of burials of this type at Roman necropolises in Serbia although, apart from G. 27 and G. 97 from the site of Slog in Ravna (Timacum Minus), we are only familiar with one grave from ]uprija (Horreum Margi) and one from the necropolis at Viminacium – Vi{e Grobalja (Pileti} 1969, 21–22, sl. 28, sl. 38; Kora}, Golubovi} 2009, 365, G–443). 89 Ivani{evi}, Kazanski 2009, 126–129. 90 Petrovi} 1995, 34; Petkovi} 1999, 225–228. 91 Tejral 1999, 242–248, Abb. 18, 21–22; Kazanski 2012, 383–386, Fig. 2 – graves which contain similar finds, a belt buckle and a conical glass goblet, as well as weapons, most commonly leafshaped spearheads, long swords, faceted conical shield umbos, and occasionally horse tack, appear in the last third of the 4th and at the beginning of the 5th century north-east of the Danube Limes, as well as in the Lower Danube and on the northern coast of the Black Sea. 92 Magomedov 2000, 59–63. 111 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 almost parallel lines, and fractures of the ribs, right ulna and right fibula (grave 102)97 and a transfixing injury caused by a tri-bladed arrow, and injuries to the processus zygomaticus and mentum (grave 108). These graves are without finds, except for the iron belt buckle with a circular frame from grave 102 (Pl. VII, 6). However, what should be taken into account is that all three graves were destroyed by grave robbers, the skeletons were dislocated, and that the most valuable parts of the military equipment, such as the fibulae and the belt sets were probably stolen from the pits. On the other hand, in most graves of adults and children of both sexes in this phase, there were no finds, indicating two facts – the population of Timacum Minus were Christian and they were poor. What bears testament to this fact are the cultural layers in the fortification from the period of the first half of the 5th century, during which time an almost rural settlement emerged within its walls, comprising structures of light materials and crudely and poorly adapted older buildings.98 The deceased buried in grave 108 might also have belonged to the non-Roman population, considering that, at the approximate age of 20 he had already been seriously wounded and had ailments caused by military life. Generally speaking, Roman citizens did not join the army before they turned 17 or 18 years of age, although for the recruited “barbarians” this rule did not apply. Thus, this young warrior most probably started service at the age of 14 or 15, resulting in pathological changes on the spine. The impression given is that Timacum Minus once again experienced changes to its population at the beginning of the 5th century, considering the graves from phase III, containing no offerings, and with a west-east orientation almost without exception, which testifies to complete Christianisation. infantry, but also the equestrian units formed under the influence of Sarmatian tribes, primarily the Alans.93 In any case, the Alano-Hunnic steppe component of the auxiliary units of the Roman light cavalry, in the eastern part of the Empire in the time of Theodosius I, can also be observed at the Late Roman Slog necropolis through the finds of the tri-bladed arrowheads (Pl. VI, 5).94 Along with the numerous recorded injuries (cuts on the head and the fused fissure of the right tibia (grave 15), injuries to the processus zygomaticus (grave 26) and injuries to the head in the region of the asterion (grave 123)), what is surely worth mentioning are the cases of scurvy observed in the individuals buried in graves 28 and 76, which would indicate that they moved over to the Timok valley, or that they resided over a longer period of time outside of this fertile area, which could provide fruit and vegetables rich in vitamins. Although the grave finds from phase I of the Slog necropolis do not essentially differ from those from graves of the subsequent phase (Pl. II–III), the component of the regular Roman army is more prominent, which is reflected in the finds of crossbow fibulae (Pl. II, 1, 6) and bronze belt sets (Pl. III, 1–2). Generally, it is possible that they are members of the same military unit, pseudocomitatenses Timacenses auxiliarii, or officers of the II Aurelia Dardanorum equestrian cohort, which unquestionably made up the Timacum Minus garrison from the middle of the 2nd to the end of the 3rd or the beginning of the 4th century, possibly even to the last quarter of the 4th century, that is to the Battle of Hadrianopolis in 378.95 However, what should not be ruled out are the equestrian units of Numeri Dalmatorum, testified to by stamps on bricks from the 4th phase of the fortification reconstruction, during the period of the reign of Valentinian I and Valens (364–375).96 That they were really warriors is also confirmed by, among other things, a cut inflicted by a sharp blade, most likely a sword or knife, on the mentum and a fissure of the right tibia (grave 91), an injury on the right femur (grave 99) and a myositis ossificans traumatica noted on the left tibia (grave 141). In phase III of the Slog necropolis, three graves were noted, in which adult male individuals were buried who, in the course of their lifetime, acquired injuries in battle and/or characteristic pathological changes on the bones caused by military life, fractures of the right fibula, the neck of the right femur and the ribs, and a subperiosteal hematoma on the right tibia (grave 100), injuries on the frontal bone, in the form of shallow perforated circular depressions present in two 93 Kazanski 2013, 494–502, 504–506. Kazanski 2013, 513. 95 Petkovi} et al. 2005, 13, Note 3. 96 Petrovi} 1995, 34 – stamps NVMDIIII and NVMDIIIII were noted on the bricks. 97 This specific individual, in addition to a large number of injuries along with evidence of an athletic built on the skeleton, displayed the highest levels of occupational stress in the entire Slog necropolis, thus his “barbarian” origins can be assumed. 98 Petkovi}, Jovanovi} 2000, 279–280; Petkovi}, Iliji} 2013, 53–54, 65–68 – this horizon of life at Timacum Minus was explored in the sector of the southern gate of the fortification and in the large building in its interior, considered by previous researchers to be a horreum. 94 112 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Fig. 2. Medical instruments discovered at the archaeological site Timacum Minus, now in the archaeological exhibition of the Homeland Museum of Knja`evac Sl. 2. Medicinski instrumenti otkriveni na arheolo{kom nalazi{tu Timacum Minus, sada u arheolo{koj postavci Zavi~ajnog muzeja u Kwa`evcu CONCLUSION sporadically, so one can speak of the barbarisation of the Timacum Minus population rather than of the stationing of military units comprising the barbarians, the Goths, the Alans or the Huns. On the other hand, the socioeconomic standing of the population buried at the Slog necropolis is very uniform and could be described as very modest. However, the individuals buried in the analysed military graves generally lived well into adulthood, despite frequent wounding and very serious injuries, as well as evidence of a gruelling military life. It is not possible to determine whether these are retired soldiers, veterans, or active warriors but, based on the individual of approximately 20 years of age buried in Based on the analysis of the military graves from the Slog necropolis, several conclusions regarding the population of the Late Roman settlement and fortification Timacum Minus can be drawn.99 The population was composed of Romanised natives, with a prominent autochthonous Dardanian-Tribalian component (Timachi) and newcomers, who settled in several waves from the last quarter of the 3rd to the first half of the 5th century.100 The newcomers always came from the left bank of the Danube, either from the abandoned Romanised Dacia or from the territory in Thrace where the foederati originated, which is testified to by pronounced material elements of the Chernyakhov – Sântana de Mureº culture. This population was integrated, as demonstrated by the distribution of the graves in all three phases of the Late Roman Slog necropolis (Plan 2). The graves were, most likely, grouped according to family ties rather than ethnic background or social status. This conclusion is based on the fact that the elements of a “barbarian” material culture appear 99 A small scale archaeological exploration of the civil settlement revealed horizons of life from the 2nd to the end of the 4th century, which leads to the conclusion that it was moved, at the end of the 4th and the beginning of the 5th century, most likely to the interior of the fortification (Petrovi}, Jovanovi} 1997, 21–24; Petkovi}, Iliji} 2012). 100 Petkovi}, Iliji} 2013, 63–65. 113 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) grave 108, who already had traces of serious injuries and pathological changes caused by the carrying of heavy loads and excessive walking, it is evident that they entered into military service at a very early age.101 In the 2nd–3rd century, Timacum Minus was an attractive place for veterans to settle due to the natural resources in the area, the fertile soil and mining potential, testified to by the epigraphs on gravestones and votive monuments.102 The warriors buried in the Late Roman Slog necropolis belonged to an equestrian unit, which is confirmed by the characteristic stress markers on the osteological material, particularly in the cases of the deceased buried in graves 15, 76, 91, 100, 102, 108, 123 and 141. Most likely these were light cavalry archers (graves 15, 25, 76, 91, 99, 100, 102, 108, 123 and 141) who could engage similar barbarian units. This is evident in both the injuries of the individuals from graves 15, 26, 76, 91, 99, 100, 102, 108 and 123, and the weapons placed in graves 27, 97 and 102. In the case of the deceased from graves 27 and 97, these are tri-bladed and leafshaped arrowheads (Pl.VI, 5, Pl. VII, 3, 5–6). Regardless of the relative poverty of the population of the Late STARINAR LXIV/2014 Roman Timacum Minus, the military population buried at the Slog necropolis had, despite all the hardships of military life, very good and efficient medical care, given that the majority of the serious injuries were treated in such a way that they left no major consequences on the individual’s health and life (Pl. IV, 1, Pl. VIII, 1–3, Pl. XI, 1–3, Pl. XII, 1–3). This indicates that the Timacum Minus fortification had one or more military surgeons. A medical service in this fortification in the previous period is evidenced by the gravestone of the medicus of the II Aurelius cohort of Dardanians, Titus Aelius Martialis,103 as well as a votive monument devoted to Aesculapius, a patron of doctors and healers, and several surgical instruments found in Ravna (Fig. 2).104 The analysis of the military graves from the Slog necropolis confirmed the results of the previous research of the Late Roman period in eastern Serbia, in the Roman province of Dacia Ripensis, as well as the hypothesis about the garrisons of light cavalry in the fortifications of the Timok region. Characteristics of the Late Roman material culture of this region are militarisation, barbarisation, Christianisation and ruralisation, which announce the onset of the Early Medieval period.105 101 Similar changes on bones caused by the carrying of heavy loads were noted in the individual from grave 15. 102 Petkovi}, Iliji} 2013, 63, Table 1. 103 Petrovi} 1995, 97, No. 49 – D(is) M(anibus) | T. Ae(lio) | Mar|tia[l(i)] med(ico) | co[h(ortis)] II Aur(eliae) | Dar(danorum)… 104 Petrovi} 1995, 63, No. 1; Jovanovi} 1989, 92–93. 105 Petkovi} et al. 2005, 143–145; Petkovi} 2011 b, 122–126. 114 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 BIBLIOGRAPHY: Garn, Lewis and Kerewsky 1965 – S. Garn, A. Lewis and R. Kerewsky, X – linked inheritance of teeth size. Journal of Dental Research 44, 1965 439–441. Hauser and De Stefano1989 – G. Hauser and G. F. De Stefano, Epigenetic Variants of Human Skull. Stuttgart: E. Schweizerbart’sche Verlagsbuchhandlung 1989. Hillson 1990 – S. Hillson, Teeth. Cambridge: Cambridge University Press 1990. Hillson 1996– S. Hillson, Dental Anthropology. Cambridge: Cambridge University Press 1996. Iliji} 2009 – B. Iliji}, Rimsko utvr|enje Timacum Minus, Knja`evac: Zavi~ajni muzej Knja`evac 2009. Iliji} 2011 – B. Iliji}, Primena geofizi~kih metoda na arheolo{kom nalazi{tu Timacum Minus, Razvitak, godina L, broj 239–240/2011, 50 godina, 2011, 78–86. Iºcan, Loth and Wright 1984a – M. Y. Iºcan, S. R. Loth and R. K. Wright, Metamorphosis at the sternal rib end: A new method to estimate age at death in males. American Journal of Physical Anthropology 65, 1984, 147–156. Iºcan, Loth and Wright 1984b – M. Y. Iºcan, S. R. Loth and R. K. Wright, Age estimation from the rib by phase analysis: White males. Journal of Forensic Sciences 29, 1984, 1094–1104. Iºcan, Loth and Wright 1985 – M. Y. Iºcan, S. R. Loth and R. K. Wright, Age estimation from the rib by phase analysis: White females. Journal of Forensic Sciences 30, 1985, 853–863. Ivani{evi}, Kazanski 2009 – V. Ivani{evi}, M. Kazanski, Nouvelle nècropole des Grandes migrations de Singidunum, Starinar LVII/2007 (2009), 115–137. Ivanovski 1987 – M. Ivanovski, The Grave of a Warrior from the Period of Licinius I found at Tarane{, Archaeologia Iugoslavica 24, 1987, 81–90. Jeremi} 2009– G. Jeremi}, Saldum. Roman and Early Byzantine Fortification, Belgrade: Archaeological Institute 2009. Jovanovi} 1989 – S. Jovanovi}, Rimski medicinski spomenici iz tvr|ave Timacim Minus kod sela Ravne, Razvitak, god. XXIX, br. 1, 1989 (Zaje~ar), 90–93. Kazanski 2012 – M. Kazanski, Radaigaise et la fin de la civilization ^ernjahov, in: The Pontic-Danubian Realm in the Period of the Great Migration, ed: V. Ivani{evi}, M. Kazanski, Collège de France – CNRS, Centre de recherché d’histoire et civilization de Byzance, Arheolo{ki institut, Paris–Beograd 2012, 381–403. Bass 1995 – W. M. Bass, Human Osteology, A Laboratory and Field Manual. Columbia: Missouri Archaeological Society 1995. Brothwell 1981– D. R. Brothwell, Digging up bones. London: British Museum (Natural History) and Oxford: Oxford University Press 1981. Buikstra and Ubelaker 1994 – J. E., Buikstra and D. H., Ubelaker, Standards for data collection from human skeletal remains. Arkansas Archeological Survey Research Series, No 44. Fayettville, Arkansas: Arkansas Archeological Survey 1994. Bullinger 1969 – H. Bullinger, Spätantike Gürtelbeschläge, Brügge: Dissertationes Archaeologiae Gadenses 12, 1969. Cvijeti}anin 2006 – T. Cvijeti}anin, Kasnoanti~ka gler|osana keramika. Gle|osana keramika Prve Mezije, Priobalne Dakije, Sredozemne Dakije i Dardanije, Beograd: Narodni muzej 2006. Diaconescu 1999 – A. Diaconescu, Ornamenta dignities. Gradabzeichenung Symbole des Sozialen Status be idem Lokalen Eliten von Dakien nach dem Aurelianischen Rückzug, Acta musei Napocensis 36, 1, 1999, 203–241. Du{ani} 1980 – S. Du{ani}, Organizacija rimskog rudarstva u Noriku, Panoniji, Dalmaciji i Gorwoj Meziji, Istorijski Glasnik 1–2, 1980, 7–55. Du{ani} 1996 – S. Du{ani}, Late Roman Mining in Illyricum: Historical Observations, in: Ancient Mining and Metallurgy in Southeast Europe, International Symposium, Donji Milanovac, May 20–25, 1990, ed: P. Petrovi}, S. \ur|ekanovi}, B. Jovanovi}, Belgrade, Bor: Archaeological Institute, Museum of Mining and Metallurgy 1996, 219 – 225. \uri}-Sreji} 1995 – M. \uri}-Sreji}, Uvod u fizi~ku antropologiju drevnih populacija. Beograd: Zavod za uxbenike i nastavna sredstva 1995. Ferembach, Schwidetzky and Stloukal 1980 – D. Ferembach, I. Schwidetzky and M. Stloukal, Recommendations for age and sex diagnosis of skeletons. Journal of Human Evolution 7, 1980, 517–549. Fidanovski, Cvijeti}anin 2005 – S. Fidanovski i T. Cvijeti}anin, Rimska vila i nekropola na lokalitetu Kamenitica – Moravi{te u Maloj Kopa{nici kod Grdelice (Roman Villa and Necropolis at the Site Kamenitica–Moravi{te in Mala Kopa{nica near Grdelica), u: E 75, Arheolo{ka istra`ivawa, sveska 1/2004 (Archaeological Research, Volume 1/2004), ur: M. Brmboli}, Beograd 2005, 50–121. 115 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Petkovi}, Iliji} 2012 – S. A. Petkovi}, B. Iliji}, Prilog prou~avawu rimskog naseqa na lokalitetu Timacum Minus kod Kwa`evca, Glasnik SAD 28, 2012, 153–178. Petkovi}, Iliji} 2013 – S. Petkovi}, B. Iliji}, Votive Altar of Lucius Petronius Timachus, Starinar LXIII/2013, 53–72. Petkovi}, Jovanovi} 2000 – S. Petkovi}, S. Jovanovi}, Arheolo{ka iskopavawa rimskog utvr|ewa Timacum Minus kod sela Ravna, op{tina Kwa`evac tokom 1997–1998. godine – sektor ju`ne kapije, Starinar L/2000, 275–280. Petkovi}, Radinovi}, Iliji} 2014 – S. Petkovi}, N. Radinovi}, B. Iliji}, Arheolo{ka istra`ivawa nekropole Slog u Ravni kod Kwa`evca 2013. godine, Razvitak, god. LII, br. 243–244, Zaje~ar 2014, 33–38. Petrovi} 1995 – P. Petrovi}, Inscriptions de la Mésie Superieure III – 2. Timacum Minus et la Vallé du Timok, Beograd: Centre d’étudest épigraphiques et numismatiques de la Faculté de philosophie de l’Université de Beograd 1995. Petrovi} 1996 – P. Petrovi}, Der römische Bergbau in Ravna. Einige Archäologische Notizen, in: Ancient Mining and Metallurgy in Southeast Europe, International Symposium, Donji Milanovac, May 20–25, 1990, eds: P. Petrovi}, S. \ur|ekanovi}, B. Jovanovi}, Belgrade, Bor: Archaeological Institute, Museum of Mining and Metallurgy, 1996, 195–202. Petrovi}, Jovanovi} 1997 – P. Petrovi}, S. Jovanovi}, Kulturno blago kwa`eva~kog kraja. Arheologija, Beograd–Kwa`evac: Arheolo{ki institut, Zavi~ajni muzej Kwa`evac. Pileti} 1969 – D. Pileti}, Rimaki kastrum ]uprija – Horreum Margi, Vesnik Vojnog muzeja 15 (Beograd), 1997, 9–57. Quast 2012 – D. Quast, Martial writers – Intellectual warriors: Remarks on a group of Late Antique male graves, in: The Pontic-Danubian Realm in the Period if the Great Migration, eds: V. Ivani{evi}, M. Kazanski, Collège de France – CNRS, Centre de recherché d’histoire et civilization de Byzance, Arheolo{ku institut, Paris– Beograd 2012, 243–262. RIC VII – P. M. Brunn, Constantine and Licinius (313–337), Roman Imperial Coinage, Vol. VII, London 1966: Spink and son, Ltd. Ru`i} 1994 – M. Ru`i}, Rimsko staklo u Srbiji, Beograd: Centar za arheolo{ka istra`ivanja Filozofskog fakulteta 1994. Kazanski 2013– M. Kazanski, in: War and Warfare in Late Antiquity, ed: A. Saramtis, N. Christie, Brill– Leiden–Boston 2013, 493–522. Kora}, Golubovi} 2009 – M. Kora} and S. Golubovi}, Viminacium 2. Vi{e grobalja, Beograd: Arheolo{ki institut 2009. Lovejoy 1985 – C. O. Lovejoy, Dental Wear in the Libben Population: Its Functional Patterns and Role in the Determination of Adult Skeletal Age at Death. American Journal of Physical Anthropology 68, 1985, 47–56. Lovejoy et al. 1985 – C. O. Lovejoy et al., Chronological matamorphosis of the auricular surface of the ilium: A new method for the determination of skeletal age at death. American Journal of Physical Anthropology 68, 1985, 15–28. Magomedov 2000 – B. Magomedov, The tradition of ritual feasts as a part of burial customs in Chernyakhov –Sintana culture, Archaeologia Bulgarica IV, 2000 – 2, 59–64. Miki} 1978 – @. Miki}, O antropolo{koj metodologiji terenske obrade skeletnih nalaza. Godi{njak Centra za balkanolo{ka ispitivanja ANUBiH 16/14: 3–44, 1978, 201–242. Nagy 2005 – M. Nagyi, Zwei Spätrömerzeitlicher Waffengräber am Westenrand der canabae von Aquincum, Acta archaeologica ASH 56 (2005), 403–486. Petkovi} 1999 – S. Petkovi}, Meaning and Provenance of Horses’ Protomes Decoration on the Roman Antler Combs, Starinar XLIX /1998 (1999), 213–228. Petkovi} 2009 – S. Petkovi}, Late Roman Necropolis of Romuliana. Area South of the Fortified Palace (Research 2005–2006), Starinar LVII/2007 (2009), 251–275. Petkovi} 2010 – S. Petkovi}, Rimske fibule u Srbiji od I do V veka n.e., Beograd: Arheolo{ki institut 2010. Petkovi} 2011a – S. Petkovi}, The Crossbow Fibulae from Gamzigrad (Romuliana), Starinar LX/2010 (2011), 111–136. Petkovi} 2011b – S. Petkovi}, Gamzigrad – Romuliana in der Zeit nach dem Kaiserlichen Palast, in: Bruckneudorf und Gamzigrad. Spätantike Pälaste und Gro¼villen im Donau-Balkan Raum, KVF Band 15, Hrgs: G. von Bülow, H. Zabehlicky, Römisch-Germanische Komission des Deutschen Archäologische Instituts, Österraichisches Archäologische Institut, Bonn 2011, 113–128. Petkovi} et al. 2005 – S. Petkovi}, M. Ru`i}, S. Jovanovi}, M. Vuksan, Zs. Zoffmann, Roman and Medieval Necropolis in Ravna near Knja`evac, Beograd: Archaeological Institute 2005. 116 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Todd 1921a – T. W. Todd, Age changes in the pubic bone: II The pubis of the male Negro–white hybrid; III The pubis of the white female; IV The pubis of the female Negro–white hybrid. American Journal of Physical Anthropology 4, 1921, 1–70. Todd 1921b – T. W. Todd, Age changes in the pubic bone: VI The interpretation of vriations in the symphyseal area. American Journal of Physical Anthropology 4, 1921, 407–424. Tomovi} et al. 2005 – M. Tomovi}, I. Popovi}, B. Bori}-Bre{kovi}, T. Cvijeti}anin, S. Stefanovi}, A. V. Popovi}, [arkamen (Eastern Serbia): A Tetrarchic Imperial Palace. The Memorial Complex, Belgrade: Archaeological Institute 2005. Trotter and Gleser 1952 – M. Trotter and G. C. Gleser, Estimation of stature from long bones of American whites and Negroes, American Journal of Physical Anthropology 10, 1952, 463–514. Vallois 1937 – H. W. Vallois, La Durre de la vie chez l’Homme fossile. L’Anthropologie 47, 1937, 499–532. Zotovi} 1968 – Q. Zotovi}, Nekropole spaqenih pokojnika na teritoriji Gorwe Mezije, Leskova~ki zbornik VIII, 1968, 19–30. Zotovi}, Jordovi} 1990 – Q. Zotovi} i ^. Jordovi}, Viminacium 1. Nekropola Vi{e grobaqa, Beograd 1990. Scheuer and Black 2000 – L. Scheuer and S. Black, Developmental Juvenile Osteology. London: Academic Press 2000. Sommer 1984 – M. Sommer, Die Gürtel und Gürtelbeschläge des 4.und 5. Jahrhunderts in Römischen Reich, Bonner Hefte zur Vorgeschichte 22, Bonn: Institut für Vor– und Frühgeschichte der Rheinischen Friedrich-Wilhelms-Universitat 1984. Tejral 1999 – J. Terjal, Die spätantiken militärischen Eliten beiderseits der norisch-pannonischen Grenze aus der Sicht der Grabfunde, in: Germanen Beidersseits des spätantiken Limes, Hrsg: T. Fischer, G. Precht, J. Tejral, Archäologische Imstitut der Universität zu Köln, Archeologickú Ústav AV^, Köln–Brno 1999, 217–292. Tejral 2012 – J. Tejral, Cultural or ethnic changes? Continuity and discontinuity on the Middle Danube ca A.D. 500, in: in: The Pontic-Danubian Realm in the Period if the Great Migration, eds: V. Ivani{evi}, M. Kazanski, Collège de France – CNRS, Centre de recherché d’histoire et civilization de Byzance, Arheolo{ku institut, Paris–Beograd 2012, 115–188. Todd 1920 – T. W. Todd, Age changes in the pubic bone: I The male white pubis. American Journal of Physical Anthropology 3, 1920, 285–334. ILLUSTRATION AUTHORS: photographs – Dr Sofija Petkovi}, Dr Nata{a Miladinovi}-Radmilovi}, Neboj{a Bori}, Pero Pra{talo MSc, Sa{a Milutinovi}-Lete}i; plans and drawings of finds – Dr Aleksandar Kapuran, map and illustration postproduction – Miro Radmilovi} 117 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) Rezime: STARINAR LXIV/2014 SOFIJA PETKOVI], Arheolo{ki institut, Beograd NATA[A MILADINOVI]-RADMILOVI], Arheolo{ki institut, Beograd VOJNI^KI GROBOVI SA KASNOANTI^KE NEKROPOLE SLOG U RAVNI (TIMACUM MINUS) Kqu~ne re~i. – kasna antika, nekropola, Timacum Minus, rimska vojska, ~erwahovska kultura, povrede nanete o{trim se~ivom, tupim predmetima i strelom. Rimsko utvr|ewe i naseqe na levoj obali Belog Timoka, u ataru sela Ravne, na 10 km severno od Kwa`evca, identifikovano je sistematskim arheolo{kim istra`ivawima kao Timacum Minus (karta 1). O karakteru ovog rimskog nalazi{ta postoji vi{e hipoteza, ali je najverodostojnija pretpostavka da se radi o utvr|enom administrativnom centru gorwomezijske rudne oblasti Territoria metallorum, koja je zauzimala severoisto~ni deo provincije Gorwe Mezije, odnosno kasnije provincije Priobalnu i Sredozemnu Dakiju (Dacia Ripensis, Dacia Mediterranea). Utvr|ewe je nastalo jo{ krajem I veka n. e., a napu{teno je posle hunske invazije u Podunavqe, sredinom V veka (sl. 1). Oko utvr|enog administrativnog, rudarsko-metalur{kog i zanatskog sedi{ta vremenom se formiralo urbano naseqe sa javnim kupatilima, gradskim vilama, radionicama i hramovima. Neki od tih objekata su konstatovani arheolo{kim iskopavawima, dok se o drugima zna posredno preko otkrivenih skulptura i epigrafskih spomenika. Na osnovu ovih potowih, uzidanih u bedeme i gra|evine tokom rekonstrukcija u kasnoanti~kom periodu, od sredine III do prve polovine V veka, utvr|eno je i postojawe velike nekropole u II i III veku. Na`alost, iz pomenutog razloga, ve}ina nadgrobnih spomenika je dislocirana, tako da se o ovoj nekropoli malo zna. Novija nedestruktivna arheolo{ka istra`ivawa, koja su koristila metode geofizi~kih merewa, uz podatke prikupqene od me{tana koji obra|uju wive i ba{te zapadno od rimskog utvr|ewa, lociraju nekropolu kremiranih pokojnika iz II–III veka na potesu [irina, koji se nalazi izme|u utvr|ewa na istoku, Ropinskog potoka na severu, brda Slog na zapadu i Seli{kog (Zubanovog) potoka na jugu (plan 1). S druge strane, delovi kasnoanti~ke nekropole istra`eni su na desnoj obali Ropinskog potoka, severozapadno od utvr|ewa i na isto~noj padini brda Slog. Tokom tri kampawe za{titnih arheolo{kih iskopavawa, uslovqenih rekonstrukcijom lokalnog puta Ravna–Debelica tokom 1994–1996. godine, sa leve strane puta, na isto~noj padini brda Slog istra`en je deo vi{eslojne kasnoanti~ke i ranosredwovekovne nekropole. Otkriveno je 140 grobova, od ~ega 72 sa 75 sahrawenih individua iz kasnoanti~kog perioda (plan 2). Konstatovano je da se radi o nekropoli formiranoj u tri sukcesivne faze u periodu od sredine IV do sredine V veka: I faza – od 350. do 380. godine, II faza – od 380. do 410. godine i III faza – od 410. do 450. godine. Rezultati za{titnih arheolo{kih iskopavawa nekropole Slog publikovani su 2005. godine u posebnoj monografiji.1 Tom prilikom objavqeni su i rezultati antropolo{ke analize, koju je izvr{ila dr @u`ana Cofman, antropolog Nacionalnog muzeja u Budimpe{ti. Me|utim, od 2011. godine je u toku projekat obrade, analize i publikovawa svih arheolo{kih podataka i nalaza prikupqenih tokom sistematskih i za{titnih arheolo{kih istra`ivawa nalazi{ta Ravna – Timacum Minus. Novi podaci, dobijeni revizijom antropolo{kog osteolo{kog materijala i analizama sprovedenim savremenim metodolo{kim postupcima, naveli su nas da ponovo razmotrimo premise i zakqu~ke koji se odnose na populaciju kasnoanti~ke Ravne. U ovom radu smo dopunili podatke o vojni~koj komponenti stanovni{tva Timakum Minusa, koja je, bez sumwe, igrala zna~ajnu ulogu u socioekonomskoj slici ovoga naseqa. Me|u istra`enim kasnoanti~kim grobovima sa nekropole Slog u Ravni izdvojili smo sahrane koje su sadr`ale arheolo{ke nalaze vojne opreme i oru`je (plan 2, T. I–III, T. V–VII, T. X). Detaqnom antropolo{kom analizom osteolo{kih nalaza iz ovih grobova prevashodno smo poku{ali da ustanovimo, koliko je to stepen o~uvanosti kostiju dozvoqavao, prisustvo antemortalnih i perimortalnih povreda koje su nastale ne kao posledica nesre}nih slu~ajeva, ve} kao posledica namernog nasiqa, zatim karakteristi~nih markera okupacionog stresa, i patolo{kih promena, koji bi ukazivali da su pokojnici bili u aktivnoj vojnoj slu`bi (T. IV, T. VIII–IX, T. XI–XII). Na osnovu analize vojni~kih grobova sa nekropole Slog mo`emo izneti nekoliko zakqu~aka o populaciji kasnoanti~kog naseqa i utvr|ewa Timacum Minus. Stanovni{tvo je bilo sastavqeno od romanizovanih starosedelaca, sa izra`enom autohtonom dardansko-tribalskom komponentom (Timachi) i prido{lica, koje su se doselile u vi{e talasa od posledwe ~etvrtine III do prve polovine V veka. Do{qaci su uvek pristizali sa leve obale Dunava, bilo iz napu{tene romanizovane Dakije ili sa federatske teritorije u Trakiji, o ~emu svedo~e izra`eni elementi materijalne kulture tipa ^erwahov – Sintana de Mure{. Ova populacija je bila integrisana, {to se vidi po distribuciji grobova na kasnoanti~koj nekropoli Slog u sve tri faze (plan 2). Grobovi su, najverovatnije, grupisani prema porodi~nim vezama, a ne po etni~koj pripadnosti ili na osnovu socijalnog statusa. Ovo zakqu~ujemo na osnovu ~iwenice da se elementi „varvarske“ materijalne kulture javqaju sporadi~no, pa se pre mo`e govoriti o varvarizaciji stanovni{tva Timakum Minusa nego o stacionirawu vojnih jedinica sastavqenih od varvara, Gota, Alana ili Huna. S druge strane, socioekonomski polo`aj stanovni{tva sahrawenog na nekropoli Slog je veoma ujedna~en i mo`e se opisati kao veoma skroman. Ipak, mu{karci sahraweni u analizira- 118 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 pokojnika iz grobova 27 i 97, to su trobride i listolike strelice (T. VI, 5, T. VII, 3, 5–6). Bez obzira na relativno siroma{tvo stanovnika kasnoanti~kog Timakum Minusa, vojni~ka populacija sahrawena na nekropoli Slog je, pored svih te{ko}a ratni~kog `ivota, imala veoma dobru i efikasnu lekarsku za{titu, jer je ve}ina te{kih povreda sanirana bez ve}ih posledica po zdravqe i `ivot (T. IV, 1; T. VIII, 1–3; T. XI, 1–3; T. XII, 1–3). Ovo pokazuje da je utvr|ewe Timacum Minus imalo jednog ili vi{e vojnih hirurga. O slu`bovawu lekara u ovom utvr|ewu u prethodnom periodu svedo~i nadgrobni spomenik lekara II Aurelijeve kohorte Dardanaca, Tita Elija Martijala. Tako|e, iz Ravne poti~e votivni spomenik Eskulapu, koji je za{titnik lekara i vidara, kao i vi{e medicinskih hirur{kih instrumenata (sl. 2). Analiza vojni~kih grobova sa nekropole Slog potvrdila je rezultate ranijih istra`ivawa kasnoanti~kog perioda u isto~noj Srbiji, odnosno u rimskoj provinciji Priobalnoj Dakiji (Dacia Ripensis), kao i hipotezu o posadama lake kowice u utvr|ewima Timo~ke zone. Karakteristike kasnoanti~ke materijalne kulture ove oblasti su militarizacija, varvarizacija, hristijanizacija i ruralizacija, {to predstavqa uvod u ranosredwovekovni period. nim vojni~kim grobovima su ve}inom do`iveli zrelo doba, uprkos ~estom rawavawu i veoma te{kim povredama, kao i tragovima napornog vojni~kog poziva. Nije mogu}e utvrditi da li se radi o umirovqenim vojnicima, veteranima, ili o aktivnim ratnicima, ali se na osnovu individue od oko 20 godina, sahrawene u grobu 108, koja je ve} imala tragove te{kih rawavawa i patolo{ke promene izazvane no{ewem tereta i preteranim pe{a~ewem, vidi da su veoma rano stupali u vojnu slu`bu. U II–III veku Timacum Minus je bio privla~no mesto za naseqavawe veterana, zbog prirodnih resursa u okolini, plodnog zemqi{ta i rudnih bogatstava, o ~emu svedo~e epigrafski natpisi na nadgrobnim i votivnim spomenicima. Ratnici sahraweni na kasnoanti~koj nekropoli Slog pripadali su kowi~koj jedinici, o ~emu svedo~e karakteristi~ni markeri stresa na osteolo{kom materijalu posebno u slu~aju pokojnika sahrawenih u grobovima 15, 76, 91, 100, 102, 108, 123 i 141. Najverovatnije se radi o lakoj kowici strelaca (grobovi 15, 25, 76, 91, 99, 100, 102, 108, 123 i 141), koja je mogla da parira sli~nim varvarskim odredima. To se vidi, s jedne strane, prema povredama individua iz grobova 15, 26, 76, 91, 99, 100, 102, 108 i 123, a, s druge, prema oru`ju polo`enom u grobove 27, 97 i 102. U slu~aju 1 119 Petkovi} et al. 2005. PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate I – Graves from phase I of the Late Roman Slog necropolis (350–380) Tabla I – Grobovi I faze kasnoanti~ke nekropole Slog (350–380. godine) 120 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate II – Grave finds from phase I of the Late Roman Slog necropolis (350–380): 1) bronze crossbow fibula from grave 44; 2) bronze belt buckle from grave 44; 3) bronze shoe buckle from grave 44; 4) bronze shoe strap-end from grave 44; 5) conical glass goblet from grave 69A; 6) bronze crossbow fibula from grave 69A; 7) silver plating of a belt set from grave 141 Tabla II – Nalazi iz grobova I faze kasnoanti~ke nekropole Slog (350–380. godine): 1) krstoobrazna bronzana fibula iz groba 44; 2) bronzana pojasna kop~a iz groba 44; 3) bronzana kop~a obu}e iz groba 44; 4) bronzani jezi~ak obu}e iz groba 44; 5) stakleni koni~ni pehar iz groba 69A; 6) krstoobrazna bronzana fibula iz groba 69A; 7) srebrna oplata pojasne garniture iz groba 141 121 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate III – Grave finds from phase I of the Late Roman Slog necropolis (350–380): 1) finds from grave 44; 2) bronze belt-end of a belt set from grave 99 Tabla III – Nalazi iz grobova I faze kasnoanti~ke nekropole Slog (350–380. godine): 1) nalazi iz groba 44; 2) bronzani jezi~ak pojasne garniture iz groba 99 122 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate IV – 1) grave 91 – cut on the left side of mentum; 2) grave 91 – pronounced muscle attachment points on manubrium; 3–6) grave 141 – myositis ossificans traumatica on left tibia Tabla IV – 1) grob 91 – posekotina sa leve strane mentuma; 2) grob 91 – istaknuta mi{i}na hvati{ta na manubrijumu; 3–6) grob 141 – myositis ossificans traumatica na levoj tibiji 123 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate V – Graves from phase II of the Late Roman Slog necropolis (380–410) Tabla V – Grobovi II faze kasnoanti~ke nekropole Slog (380–410. godine) 124 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate VI – Grave finds from phase II of the Late Roman Slog necropolis (380–410): 1) iron belt buckle from grave 25; 2) small vessel with three handles from grave 26; 3) bronze belt buckle from grave 27; 4) conical glass goblet from grave 76; 5) tri-bladed iron arrowhead from grave 123 Tabla VI – Nalazi iz grobova II faze kasnoanti~ke nekropole Slog (380–410. godine): 1) gvozdena pojasna kop~a iz groba 25; 2) lon~i} sa tri dr{ke iz groba 26; 3) bronzana pojasna kop~a iz groba 27; 4) stakleni koni~ni pehar iz groba 76; 5) trobrida gvozdena strelica iz groba 123 125 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate VII – Grave finds from phase II (380–410) and phase III (410–450) of the Late Roman Slog necropolis: 1) iron buckle from grave 25; 2) finds from grave 26; 3) finds from grave 27; 4) iron buckle from grave 28; 5) finds from grave 97; 6) finds from grave 123; 7) iron buckle from grave 102. Tabla VII – Nalazi iz grobova II (380–410. godine) i III faze (410–450. godine) kasnoanti~ke nekropole Slog: 1) gvozdena kop~a iz groba 25; 2) nalazi iz groba 26; 3) nalazi iz groba 27; 4) gvozdena kop~a iz groba 28; 5) nalazi iz groba 97; 6) nalazi iz groba 123; 7) gvozdena kop~a iz groba 102 126 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate VIII – 1, 2, 3) grave 15 – injuries on frontal bone inflicted by a sharp blade; 4) grave 15 – flattened body of the clavicle (direction: superior-inferior); 5, 6) grave 15 – incisura clavicularis on the manubrium extended to the left. Tabla VIII – 1, 2, 3) grob 15 – povrede na frontalnoj kosti nanete o{trim se~ivom; 4) grob 15 – spqo{tenost tela klavikule (pravac: superior–inferior); 5, 6) grob 15 – razvu~ena incisura clavicularis na manubrijumu u levu stranu 127 STARINAR LXIV/2014 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) 1 2 3 Plate IX – 1) grave 25 – pronounced ligament attachment points on clavicles (lig. costoclaviculare); 2) grave 26 – trace of injury on the left processus zygomaticus inflicted by a blunt object; 3) grave 76 – fused fracture of the head of right femur Tabla IX – 1) grob 25 – izra`ena hvati{ta ligamenata na klavikulama (lig. costoclaviculare); 2) grob 26 – trag povrede nanete tupim predmetom na levom processus zygomaticus-u; 3) grob 76 – srasla fraktura glave desnog femura Plate X – Graves from phase III of the Late Roman Slog necropolis (410–450). Tabla X – Grobovi III faze kasnoanti~ke nekropole Slog (410–450. godine) 128 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate XI – 1, 2) grave 100 – irregularly fused fracture of right fibula in the upper quarter of its body; 3) grave 102 – injuries on frontal bone; 4) grave 102 – irregularly fused rib fractures; 5) grave 102 – pronounced muscle attachment points on right ulna; 6, 7) grave 102 – pronounced muscle attachment points on right femur Tabla XI – 1, 2) grob 100 – nepravilno srastao prelom desne fibule u gorwoj ~etvrtini tela; 3) grob 102 – nepravilno srastao prelom rebara; 4) grob 102 – povrede na frontalnoj kosti; 5) grob 102 – istaknuta mi{i}na hvati{ta na desnoj ulni; 6, 7) grob 102 – istaknuta mi{i}na hvati{ta na desnom femuru 129 PETKOVI], MILADINOVI]-RADMILOVI], Military graves from the Late Roman necropolis… (87–130) STARINAR LXIV/2014 Plate XII – 1, 2, 3) grave 108 – injury inflicted by a tri-bladed arrow with traces of healing; 4) grave 108 – pronounced muscle attachment points on ulnas; 5) grave 108 – Schmorl’s nodes on one thoracic and two lumbar vertebrae Tabla XII – 1, 2, 3) grob 108 – povreda naneta trobridom strelom, sa tragovima zaceqewa; 4) grob 108 – istaknuta mi{i}na hvati{ta na ulnama; 5) grob 108 – [morlovi defekti na jednom grudnom pr{qenu i na dva slabinska pr{qena 130 VUJADIN IVANI[EVI], Institut archéologique, Belgrade MICHEL KAZANSKI, Centre Nationale de la Recherche Scientifique, Paris ILLYRICUM DU NORD ET LES BARBARES À L'ÉPOQUE DES GRANDES MIGRATIONS (Ve–VIe siècle) UDK: 904(363)"03/05" ; 904(497)"03/05" DOI: 10.2298/STA1464131K e-mail: [email protected] Recu : 19 février 2014 Accepté : 25 juillet 2014 Article scientifique original Sommaire. – Les auteurs disscutent des importants changements historiques et sociaux sur la frontière danubienne qui ont fait suite à l’anéantissement de l’armée impériale à Andrinople en 378. Cet échec ne fit qu’ouvrir toute grande la porte à l’installation des Barbares dans les provinces frontalières de l’Empire. Leur arrivée, par vagues successives, devait avoir l’effet d’un travail de sape, démantelant les capacités défensives de l’Empire sur ses frontières septentrionales, tout en provoquant le repli des populations locales plus à l’intérieur de l’Illyricum et dans les provinces balkaniques méridionales. Mots clés. – Illyricum, Balkans, Antiquité tardive, Grande migration, Germains, Huns, Slaves. LES INVASIONS BARBARES DE LA FIN DU IVe – DÉBUT DU Ve SIÈCLE Bien que la frontière septentrionale de l’Illyricum du Nord soit située sur le limes danubien, ce n’est que vers la fin du IVe et le début du Ve s. que les premiers Barbares s’implantent sur le territoire de la diocèse de Dacie. Les incursions des Sarmates aux cours de la première moitié du IVe s. n’ont pas démantelé le système défensif, qui a été, il faut le souligner, renforcé par la reconstruction des anciennes forteresses du limes, la construction des fortins sur la rive gauche du Danube, tel que Constantia (Kovin), Translederata (Sapaja), Drobeta (Turnu Severin), etc. Le système défensif a été d’avantage assuré par la construction des grands ouvrages en terre sur le territoire barbare en Sarmatie. Deux lignes de cette fortification débutent, et cela n’est pas un hasard, du limes danubien, une de Constantia et l’autre de Translederata, et elles s’étendent vers le nord. Ces nouvelles limites coupaient l’accès à l’ennemi venant de l’ouest et du nord, en premier lieu les Goths qui se sont répandues après l’abandon de la Dacie par l’empereur Aurélien en 271.1 Les importants changements historiques et sociaux sur la frontière danubienne ont fait suite à l’anéantissement de l’armée impériale à Andrinople en 378. Cet échec ne fit qu’ouvrir toute grande la porte à l’installation des Barbares dans les provinces frontalières de l’Empire. Leur arrivée, par vagues successives, devait avoir l’effet d’un travail de sape, démantelant les capacités défensives de l’Empire sur ses frontières septentrionales, tout en provoquant le repli des populations locales plus à l’intérieur de l’Illyricum et dans les provinces balkaniques méridionales (Fig. 1). 1 Soproni 1969, 117–135; Vaday 2001, 249–276; Bugarski, Ivani{evi~ 2012, 482–511. * Cette étude est effectuée dans le cadre du projet: Processus d’urbanisation et de développement de la société médiévale (n°177021), financé par le Ministère de l’éducation, de la science et de développement technologique de la République de Serbie. 131 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 Fig. 1. Carte de l'Illyricum du Nord à l'époque des Grandes Migrations : y grande nécropole ; ™ trouvailles isolées y tombe isolée ou petite nécropole ; Sl. 1. Karta Severnog Ilirika u vreme seobe naroda: y velika nekropola; ™ pojedina~ni nalazi y usamqen grob ili mala nekropola; Les Barbares, dans leur avancé vers les Balkans n’épargnèrent pas le secteur du limes danubien s’étendant entre Singidunum (Belgrade) et Aquae (Prahovo), où leur passage se solda par la destruction d’un grand nombre de forteresses. A Singidunum la fouille d’une aire située sur la rive du Danube a livré une couche de destruction et d’incendie datée par 227 monnaies, dont les émissions finales datent de 375–378. Une couche d’incendie semblable renfermant plus de 600 pièces de monnaies, a été constatée dans la forteresse de Gratiana (Saldum). Il s’agit de plusieurs dépôts monétaires dispersés à l’intérieur du camp. Cette époque vit également la destruction des fortifications de Campsa (Ravna), où la couche d’incendie est datée par les monnaies de Valentinien Ier et Valens, de Taliata (Donji Milanovac) et d’autres camps jusqu’à Diana (Karata{) et Pontes. En plus des couches de destruction cette violente irruption des Barbares se voit également attestée par les dépôts monétaires. A proximité même de Singidunum on enregistre trois dépôts de bronze ayant pour terminus ante quem 375–378, alors que le site de Donja Dubona, entre Aureus Mons (Seona) et Vinceia (Smederevo), a livré un petit dépôt de 5 solidi avec dernière émission en 382/383. Le site de Naku~ani, sur l’emplacement d’une villa rustica, a livré un second trésor démembré de solidi se terminant par un solidus de Gratien datée de 378–383. L’horizon de destruction et les dépôts monétaires précédemment mentionnés peuvent être mis en relation avec les incursions des Barbares que plus rien n’arrêtait dans leur avancé sur l’Illyricum après la défaite d’Andrinople en 378.2 132 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 sur la rive gauche du Danube.10 Ces ceinturons représentent de bons témoins de la culture militaire « romano-germanique » qui se développa surtout dans la Gaule du Nord.11 Elle fut diffusée le long du limes danubien au Noricum et en Pannonie,12 quelques exemplaires arrivent même jusqu’au Bas-Danube. La tombe de Pontes fut interprétée par Vladislav Popovi} comme celle d’un barbare au service de l’Empire romain selon la position du ceinturon dans la tombe sous les pieds. A l’appui de sa thèse il site l’exemple de la tombe 67 de Csakvar.13 Il convient également de mentionner ici de nombreuses trouvailles sporadiques – fibules et peignes en os- qui peuvent être rattachées à la culture de ^ernjahov – Sîntana de Mureº, appartenant aux Goths et à leurs alliés germaniques et non-germaniques. Il a depuis longtemps été remarqué que ces objets sont nombreux dans la zone le long de la frontière danubienne, et plus précisément entre Singidunum et Aquae, où ils apparaissent au sein des couches de la basse antiquité fouillées dans de nombreux ouvrages défensifs.14 Notons la trouvaille de sept peignes du type Thomas III dans une des tours de Romuliana (Gamzigrad).15 Ce même processus d’infiltration peut également être constaté plus à l’ouest, puis au nord, le long du limes danubien bordant respectivement la Pannonie II et la province de Valeria. On peut citer, comme une des preuves archéologiques, la découvertes des peignes en os germaniques, du type Thomas I, à Sirmium.16 Plus à l’intérieur du pays de telles traces sont rares et n’apparaissent que sporadiquement. A proximité de Naissus (Ni{) la nécropole de Mediana a livré une fibule en tôle de bronze, et le site de Ni{avac (Timacum Après la vague d’irruptions, s’étant traduite par les destructions des fortifications, le limes fut restauré durant le règne de Théodose Ier. Les raids de pillage et de dévastations lancés par les Barbares laissèrent alors la place à un processus d’implantation faisant suite à l’enrôlement des Germains dans l’armée romaine au cours des dernières décennies du IVe s. Cette infiltration progressive des Barbares, peut être suivie à travers les tombes isolées et le matériel archéologique sporadique datés de la fin du IVe et de la première moitié du Ve s., mis au jour sur le territoire nord de l’Illyricum. Cet horizon est ainsi attesté par un petit nombre de tombes dégagées dans les nécropoles de Viminacium.3 Il s’agit en l’occurrence des sépultures 134 et 1607 de la nécropole de Vi{e grobalja (Viminacium II), qui ont été datées, sur la base de leur mobilier, de la fin du IVe et de la première décennie du Ve (phase D1 de la chronologie du Barbaricum européen – 360/370–400/410 et D2 – 380/400–440–450). Toujours en Mésie I nous pouvons également attribuer à cet horizon les tombes mises au jour dans la nécropole aménagée à l’intérieur du castrum de Singidunum (nécropole II)4 ainsi que les tombes les plus anciennes de la nécropole de Singidunum III. Il s’agit des sépultures 19a, 28, 59, 78 et, peut-être, 42 et 99, qui, sur la base d’analogies, ont livré des objets datables de la phase D1 – D2.5 Plus à l’est, le long du limes danubien, dans la province de Dacie ripuaire, deux tombes, à Vajuga, et à Pontes, appartient à la même époque.6 En ce qui concerne Vajuga, une petite nécropole auprès de la fortification du Bas-Empire y a été fouillée. Deux fibules ont été mises au jour parmi les autres objets caractéristiques, dans la sépulture 18 appartenant à une fille (Fig. 2). Elles sont, selon Vladislav Popovi}, proches de celle de la tombe féminine d’Untersiebenbrunn. J. Tejral a soutenu, lui aussi l’attribution de la tombe de Vajuga au même horizon d’Untersiebenbrunn, c’est à dire à la phase D2, 380/400–440–450.7 La position de deux fibules sur les épaules avec la tête retournée vers le bas suggère qu’il s’agit, selon V. Popovi}, d’une jeune Gothe. En effet, le costume à deux fibules sur la partie supérieur du corps est caractéristique des Germains orientaux à l’époque des Grandes Migrations.8 A la même époque, mais à un différent contexte, appartient la tombe 7 de Pontes avec un ceinturon dit militaire, formé de deux plaques en bronze, à décor en taille biseautée (Fig. 3). Les objets apparents proviennent des fortins proches de Smorna (Boljetin) et de Campsa (Ravna), mais également de Singidunum9 ainsi que des forteresses de Dierna, de Drobeta et de Hinova 2 Ivani{evi} 1999, 95–96; Vasi} 2008, 102–103; Jeremi} 2009, 201–221. 3 Ivani{evi} et alii 2006. 4 Bjelajac, Ivani{evi} 1993, fig. 3. 5 Ivani{evi}, Kazanski 2002, 124. 6 Popovi} 1987, 129–137. 7 Tejral 2011, 190, 308. 8 cf. en particulier Bierbrauer 1975, 71–78. 9 Pop-Lazi} 2013, 70, fig. 22. 10 Bondoc 2009, 55–70, fig. 85, 131 et 158–160. 11 Nombreux exemples: Böhme 1974; Sommer 1984. 12 Voir la carte de diffusion: Tejral 2011, Abb. 212. 13 Popovi} 1987, 129–137. 14 Ivani{evi} 1999, 97–107. 15 Petkovi} 2006, 363–366. 16 Popovi} 1987, pl. 4.1, 3. 133 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 2 STARINAR LXIV/2014 3 4 5 Fig. 2. Vajuga – sépulture 18 (Popovi} 1987) Sl. 2. Vajuga, grob 18 (Popovi} 1987) Maius) une autre fibule du même modèle.17 Ces fibules spécifiques sont aussi bien représentées dans les fortins du limes danubien. Un exemplaire provient de Translederata, et deux de Burgenae (Novi Banovci) et ils peuvent être datées de la phase D1 – 370/80–400/10.18 A Viminacium cet horizon se distingue toutefois des autres, par la présence d’arme dans le mobilier de la tombe 1607. Nous pouvons à juste titre supposer que cette population barbare précoce peut être mise en relation avec les premiers fédérés engagés pour assurer la défense des frontières de l’Empire. Il convient également de noter que l’occupant de la sépulture 1607 avait un crâne artificiellement déformé,19 de sorte qu’il apparaît avoir été un des tout premiers adeptes de ce 17 Ivani{evi} 2013. Ivani{evi} 1999, 97–107, fig. 1.1; Dimitrijevi} et alii. 1962, 82–83, fig. 9–10. 19 Miki} 1994, 193. 18 134 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 1 2 3 5 4 Fig. 3. Pontes – sépulture 7 (Popovi} 1987) Sl. 3. Pontes, grob 7 (Popovi} 1987) de Théodose II dans la défense des provinces balkaniques de l’Empire, notamment entre les années 410 et 425. – mentionnons l’édit de Théodose II au préfet de l’Illyricum Herculius ou on prescrits l’obligation pour toutes les personnes, quoique soit leurs rangs, de participer à la construction des remparts.21 nouvel usage qui s’est développé parmi les Barbares du Danube, sous l’influence des Alains et des Sarmates à l’époque de la domination des Huns.20 Ces deux différentes manifestations de civilisation matérielle, étrangères à celle de l’Illyricum du Nord, sont très probablement diffusées par les militaires. Les fédérés dont la présence est incontestable, ainsi que les unités romaines venues de l’Occident et comportant peut être aussi des Germains, ont été engagés dans la défense du limes. Ceci peut être mis en rapport avec les efforts 20 21 135 Kazanski 1980; Anke 1998, 126–128. Cod. Theod. 11.17, 4. IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 STARINAR LXIV/2014 2 Fig. 4. Markovac – sépulture dispersée (Popovi} 2001) Sl. 4. Markovac, uni{ten grob (Popovi} 2001) LES HUNS nople de ne pas respecter les clauses du traité conclu en 434 et, prétextant le comportement de l’évêque de Margum, attaquent la frontière septentrionale de l’Empire. Ils démolirent nombre de forteresses et de villes sur la rive droite du Danube, et parmi celles-ci Singidunum, Margum, Viminacium, Naissus et bien d’autres.26 Il ne faut pas exclure non plus la possibilité que les Barbares eux-mêmes se sont réfugiés sur le territoire de l’Empire devant la menace hunnique. A l’appui de cette opinion citons les clauses du traité conclu entre Byzance et les Huns en 434, qui entre autres choses, prescrivaient que l’État byzantin livrera tous les fugitifs barbares aux Huns. A en juger par les sources disponibles, la grande avancée des Huns en 441 constitua un véritable tournant dans l’histoire des provinces frontalières du nord de l’Illyricum et, avec elles, des villes du Danube. Ceci est notamment le cas pour Viminacium, alors investi d’un rôle clé dans la défense de la frontière, qui fut alors pris Le nouveau danger est représenté par les Huns. Ces derniers s’installent au début du Ve s. en Moldavie et Munténie, à proximité du limes Scythicus et limes Mysiacus.22 Les Huns ont pillé la Thrace en 408, sous la conduite de Uldis, ils ont même occupé temporairement Castra Martis dans la province de Dacie ripuaire.23 La frontière danubienne dans les provinces de Scythie, Mésie II, Dacie ripuaire et Mésie I était pour l’Empire un important point de défense contre la menace barbare. Ainsi près d’un tiers des garnisons, recensées dans la Notitia Dignitatum, 104 en tout, ont été déployées le long de cette partie de frontière. Ce même document mentionne Singidunum comme résidence du préfet de la légion IV Flavia: praefectus legionis quartae Flaviae, Singiduno.24 Le limes fut consolidé à cette période comme le prouve le fait que les Romains ont maintenu la forteresse de Constantia sur la rive gauche, contre Margum.25 Les Huns devinrent vite les maîtres de la plaine pannonienne et s’imposèrent aux autres peuples barbares qui occupaient la région. La présence des Huns au confins de l’Illyricum du Nord et leur puissance croissante se traduisit d’abord par les disputes et ensuite par les hostilités. Les Huns accusent en 441 Constanti- 22 23 24 25 26 136 Ciupercã, Mãgureanu 2008; Ciobanu, Constantinescu 2008. Zosimus, V 22. Notitia Dignitatum, or., XLI.30. Priscus, frag. 2.21–23. Priscus, frag. 2. IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) et dévasté.27 Les nécropoles mises au jour, notamment à Singidunum28 et à Viminacium,29 reflètent tout particulièrement les changements d’ordre démographiques survenus en Mésie I en 441 et en Dacie ripuaire en 443, au lendemain de la percée des Huns et de la chute du limes. A cette occasion, s’avancant à l’intérieur de l’Illyricum, les Huns atteignirent même Naissus qu’ils enlevèrent. La gravité de la crise à laquelle fut alors confronté l’Empire est attestée par Priscus qui rapporte qu’en 448 Attila alla jusqu’à exiger le retrait de toute la population romaine vivant entre Singidunum et Novae, sur une profondeur de cinq jours de marche soutenue en direction de Naissus. Toutefois, oubliant cette même exigence, Attila se contenta l’année suivante de proposer de fixer la frontière sur le Danube.30 Quelques trésors monétaires amassés le long du limes danubien témoignent des dévastations hunniques. On note à Viminacium la découverte d’un dépôt de solidi de Théodose II, ainsi qu’un autre trésor de solidi de Théodose II trouvé dans les environs d’Aquae. A Pontes, le sol d’une maison incendiée a livré 5 pièces d’or de Théodose II. En plus des trésors de monnaies d’or on doit noter la découverte de nombreux dépôts de monnaies de bronze à Viminacium qui a été mise en relation avec l’incursion des Huns en 441, ainsi que deux dépôts d’Horreum Margi datés de 442/43.31 Cependant les objets pouvant être attribués aux Huns sont rares. Une de ces découvertes provenant de l’Illyricum du Nord qui pourrait peut être liée aux Huns est un torque d’or et une plaque-boucle d’argent provenant de Markovac, lieu située auprès de la rivière de Margus (Morava), celle-ci représente la grande diagonale balkanique et la route vers le sud32 (Fig. 4). Cette trouvaille pouvait faire partie du mobilier d’une tombe détruite. Le torque et la plaque-boucle ressemblent à ceux provenant de la tombe de KeszthelyZiegelfabrik.33 La nécropole détruite de Singidunum IV – Faubourg Ouest, près du Danube, un peu plus au Sud-Ouest par rapport à la nécropole de Singidunum I – Ville basse, a livré au moins une tombe de l’époque hunnique, datée de la période D2/D3 (430/440–460/470) : sépulture 2. 2006.34 C’est une inhumation de guerrier, probablement d’un chef militaire, dans une fosse rectangulaire, la tête au Sud-Ouest. Elle était accompagnée notamment d’une fibule en arbalète, de tradition germanique, typique de la région danubienne et d’une plaque-boucle, caractéristique de la partie initiale de la période D2/D3. La panoplie de ce guerrier se compose d’un bouclier à umbo conique, d’une épée à garde de fer, datable de la STARINAR LXIV/2014 première moitié du Ve s., d’un poignard, d’une lance, d’un arc renforcé par des plaques en os et d’un carquois contenant des flèches à trois ailettes, du type steppiques, des flèches losangiques à tige et celles à courte pointe de section triangulaire (Fig. 5). En outre la tombe a livré des monnaies de Valens et, probablement, d’Honorius. Plus à l’Ouest, au delà de la Sava, en Pannonie l’époque « hunnique » est marquée par plusieurs découvertes significatives. C’est la tombe de chef militaire à Ne{tin, ayant livré une longue épée et plaque-boucle à décor cloisonné, typique de la période D2,35 ainsi que deux tombes de Vranja, dont une féminine, contenait des parures typiques de l’époque hunnique et l’autre masculine, qui contenait un arc renforcé des plaques en os36 (Fig. 6). Selon J. Tejral, ces deux sépultures appartiennent à la phase D2 tardive, ou au début de la phase D2/D3, c’est à dire 420/430–450, ce qui correspond, toujours selon J. Tejral, à l’horizon de « l’empire » hunnique sur le Danube moyen.37 Citons également une fibule en tôle d’argent provenant de Rakovac.38 D’après sa taille de longueur de 11,1 cm, elle est comparable avec celles du type Makó – Maklár – Csongrád – Werboczigasse, appartenant, selon J. Tejral à la phase D2, c’est-à-dire 380/400–440/450 environ39 (Fig. 7.1). Une deuxième fibule, cette fois-ci en arbalète, du type Almgren 158, a été mise au jour à Rakovac.40 D’après les proportions elle est proche de celles de Vranja et de Singidunum IV. Deux fibules en tôle d’argent, de 21,1 et 22,1 cm de longueur, du type Smolin, proviennent d’une tombe découverte en 1883 à Ilok41 (Fig. 7.2). Leur relativement grande taille suggère leur attribution plutôt à la 27 Priscus, frag. 2 et 8. Bjelajac, Ivani{evi} 1993; Ivani{evi}, Kazanski 2002; Ivani{evi}, Kazanski 2009. 29 Ivani{evi} et alii 2006. 30 Priscus frag. ib, 7, 14. 31 Vasi} 1981, 123–161; Vasi} 1990; Ivani{evi} 1999, 97–107. 32 Popovi} 2001, 89–90, fig. 25. 33 cf. Germanen 1987, 181, V.50. 34 Ivani{evi}, Kazanski 2009, 117–139. 35 Vinski 1957, pl. 19.75 et pl. 23.88. 36 Dautova-Ru{evljan 1981; Dautova-Ru{evljan 1998. 37 Tejral 2007, 78–81; Tejral 2011, 355, 367, 368. 38 Vinski 1957, 31, Pl. 20.73. 39 Tejral 2007, 76, Abb. 3–5, 8, 9. 40 Vinski 1978, Taf. 5.4. 41 Brun{mid 1905, 210, fig. 31; Vinski 1978, Taf. 1, 2; Germanen 1987, 224, V,10.a. 28 137 STARINAR LXIV/2014 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 2 4 3 5 6 7 8 10 9 Fig. 5. Singidunum IV – sépulture 2.2006 (Ivani{evi}, Kazanski 2009) Sl. 5. Singidunum IV, grob 2.2006 (Ivani{evi}, Kazanski 2009) 138 11 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 STARINAR LXIV/2014 3 2 5 4 7 6 8 9 11 13 12 14 10 15 16 Fig. 6. Vranja – sépulture féminine (Dautova-Ru{evljanin 1981) Sl. 6. Vrawa, `enski grob (Dautova-Ru{evljanin 1981) 139 17 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 période D2/D3. A Zemun, (Taurunum) en 1868 on a mis au jour une tombe, contenant deux grandes fibules digitées, portant un décor en taille biseautée, dérivées sans doute du groupe Artánd-Záhony,42 une plaqueboucle, un collier en perles en tôle d’or et un pendentif en carcasse métallique43 (Fig. 8). On pense, que cette sépulture appartient à l’époque post-hunnique et, probablement marque l’installation des Ostrogoths dans la région de Sirmium après 454.44 Cependant J. Tejral l’a attribué à la période D2/D3.45 APRÈS ATTILA 1 La mort d’Attila en 453 et la défaite des Huns en 454/455 sur la Nedao furent suivies de la dislocation de l’empire hunnique et de vastes mouvements des peuples barbares. Certaines tribus, principalement germaniques, s’installent sur le territoire de l’Empire, y compris dans les importants centres romains sis sur la frontière nord, tels que Singidunum et Viminacium. Leur présence à Singidunum est notamment attestée par la découverte au pied du castrum romain d’une nécropole (Singidunum III) à forte composante germanique, où un certain nombre des sépultures est 2 Fig. 7. Fibules en tôle d’argent provenant de la région de Sirmium : 1) Rakovac ; 2) Ilok (Germanen 1987) Sl. 7. Srebrne lu~ne fibule iz oblasti Sirmijuma: 1) Rakovac; 2) Ilok (Germanen 1987) 4 5 1 3 Fig. 8. Zemun – sépulture (Bierbrauer 1975) Sl. 8. Zemun, grob (Bierbrauer 1975) 140 2 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 4 5 2 9 6 1 7 8 3 11 0 12 13 10 3 cm 14 Fig. 9. Singidunum I : 1–3) sépulture 8 ; 4–6) sépulture 16 ; 7) sépulture 4 ; 8) sépulture 10 ; 9, 10) sépulture 9 ; 11, 12) sépulture 7 ; 13, 14) sépulture 14 (Bjelajac, Ivani{evi} 1993) Sl. 9. Singidunum I: 1–3) grob 8; 4–6) grob 16; 7) grob 4; 8) grob 10; 9, 10) grob 9; 11, 12) grob 7; 13, 14) grob 14 (Bjelajac, Ivani{evi} 1993) datée de la période D2/D3.46 Les tombes de la période D2/D3 apparaissent aussi à Singidunum I – Ville basse,47 (Fig. 9) et à Singidunum II – Castrum48 (Fig. 10). Les tombes isolées de Belgrade, exhumé en 1911 sur un site inconnu,49 et de Karaburma50 appartiennent à la même époque. Ce processus d’installation barbare est attesté également à Viminacium par les nécropoles de Viminacium I – Burdelj et de Viminacium II – Vi{e grobalja – secteur sud-est,51 dont une partie considérable des inhumations appartient à la phase D2/D352 (Fig. 11). La culture matérielle, attestée dans ces nécropoles, serait principalement attribuable à des Germains orientaux. Cette présence de vastes nécropoles barbares sur le sol des deux principales villes frontalières – Viminacium et Singidunum – illustrent bien les importants changements démographiques survenus à la suite de l’établissement d’un nouveau rapport de forces et d’une nouvelle situation sur le plan politique. L’analyse de la culture matérielle dénote cependant une importante différence quand à la durée même des nécropoles de Singidunum, d’une part, et de Viminacium, d’autre part. En l’occurrence, celle de Singidunum perdure de toute évidence durant la période suivante, à savoir les périodes D3 (450–470/480) et E (470/480–510), alors que cet horizon est moins important à Viminacium. Si on y ajoute d’autres trouvailles provenant du territoire de Singidunum, telles une tombe dégagée à Ostru`nica53 et 42 cf. Tejral 2007, 90; Bierbrauer 2008, 126, 127. Brun{mid 1905, 214, fig. 33; Bierbrauer 1975, 127–129, Taf. 86.1–5; Germanen 1987, 230, 231, V, 21; Prohazka 2007. 44 Kiss 1979, n° 5; Prohazka 2007. 45 Tejral 2007, 90, 92. 46 Ivani{evi}, Kazanski 2002. 47 Sépultures 7 et 14: Bjelajac, Ivani{evi} 1993, fig. 8. 48 Sépultures 14 et 15: Bjelajac, Ivani{evi} 1993, fig. 10. 49 Tejral 1988, fig. 34.4–10. 50 Bjelajac, Ivani{evi} 1993, fig. 6. 51 Ivani{evi} et alii 2006. 52 Voir Tejral 1988; 1997 et 2005. 53 Tati}-Duri} 1958; Bierbrauer 1975, Taf. 78. 43 141 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 10 7 4 STARINAR LXIV/2014 8 0 3 cm 2 5 3 6 9 11 Fig. 10. Singidunum II : 1–3) sépulture 14 ; 4) sépulture 10 ; 5–11) sépulture 15 (Bjelajac, Ivani{evi} 1993) Sl. 10. Singidunum II: 1–3) grob 14; 4) grob 10; 5–11) grob 15 (Bjelajac, Ivani{evi} 1993) LES OSTROGOTHS EN ILLYRICUM une seconde, à Rospi ]uprija, avec une épée-spatha et une boucle appartenant, d’après son décor cloisonné, à l’horizon Apahida–Blu~ina–Tournai,54 il s’avère que cet horizon atteste à Singidunum une forte implantation de Germains. Cette ville est le point stratégique – clé qui contrôlait les voies de communication en direction de la Pannonie et de la Mésie I. La présence de Germains à Singidunum et dans ses environs est également signalée par les sources historiques. Nous savons ainsi que c’est à proximité de Singidunum que séjournaient les Ostrogoths avec lesquels l’empereur Marcien (450–457) conclut un foedus. De même, des données quelques peu confuses fournies par Jordanès laissent penser que les Goths occupèrent un certain temps Singidunum. Certaines tombes de Singidunum III en effet contiennent des objets typiques des Ostrogoths55 (Fig. 12). L’instabilité persistante de la situation militaire devait conduire à de nouveaux mouvements parmi les barbares. Ainsi, les Goths passèrent en 475 en Mésie II, avec l’approbation de l’empereur Zénon (474–475; 476–491), alors que les Gépides s’installaient à leur place en Pannonie. De nouveaux grands changements survinrent en 488 lorsque, sous la conduite de Théodoric, les Ostrogoths prirent la route de l’Italie où ils devaient s’assurer une base solide. De là, s’avancant au nord, ils étendirent bientôt les frontières de leur Etat jusqu’aux rives du Danube, de la Save et de la Drina. Ils réoccupèrent ainsi la région de Sirmium en 504, qu’ils rattachèrent au royaume ostrogothique d’Italie qui perdura jusqu’en 535. Cette même année 504, la partie septentrionale de la Mésie I vit la pénétration d’un allié des Goths, le prince gépide Mundo. Il semble toutefois, au vu du traité conclu entre Byzance et les Ostrogoths, que la Mésie I et la partie orientale de la Pannonie II, y compris la ville de Bassianae, revenaient à l’Empire, alors que les Ostrogoths conservaient la plus grande partie de cette province avec Sirmium.56 54 55 56 142 Bjelajac, Ivani{evi} 1993, 136, fig. 11.1. Ivani{evi~, Kazanski 2002, fig. 8, 11. Mirkovi} 2006. IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 STARINAR LXIV/2014 2 6 4 3 5 10 9 8 7 12 13 11 14 17 22 15 18 23 19 24 16 20 25 Fig. 11. Viminacium I et Viminacium II (secteur sud-est) : 1–12) sépulture 39 ; 13–21) sépulture 1193 ; 22–26) sépulture 1758 (Ivani{evi} et alii 2006) Sl. 11. Viminacium I i Viminacium II (jugoisto~ni sektor): 1–12) grob 39; 13–21) grob 1193; 22–26) grob 1758 (Ivani{evi} et alii 2006) 143 21 26 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 3 1 2 4 5 Fig. 12. Singidunum III – sépulture 1 (Ivani{evi}, Kazanski 2002) Sl. 12. Singidunum III, grob 1 (Ivani{evi}, Kazanski 2002) 1 2 Fig. 13. Ma~vanska Mitrovica, sépulture 149 (Ercegovi}-Pavlovi} 1980) Sl. 13. Ma~vanska Mitrovica, grob 149 (Ercegovi}-Pavlovi} 1980) 144 3 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 On peut attribuer à l’époque ostrogothique quelques découvertes à l’Ouest de la Sava, comme la tombe 149 de Ma~vanska Mitrovica,57 avec deux fibules – l’une de tradition gépide, à pied losangique et à tête rectangulaire et l’autre de tradition balkanique occidentale, à peid attaché – ainsi qu’une plaque-boucle à plaque losangique du type Ma~vanska Mitrovica – Noºlak58 (Fig. 13). On peut également citer une fibule digitée, du type Ar~ar–Histria, provenant de Rovine,59 ainsi que des grandes fibules digitées, originaires de Sirmium, site 3, tombe 1.1959.60 Certaines trouvailles de cet horizon sont enregistrées d’ailleurs en Mésie I. En premier lieu il faut mentionner l’applique en or, portant un décor cloisonné, provenait de Translederata, qui a été mise en relation par Zdenko Vinski avec les trouvailles apparente de l’Italie gothique.61 LES GÉPIDES Fig. 14. Batajnica, sépulture de chef militaire – casque (Vogt 2006) Quant aux Gépides, leurs sites archéologiques attestés dans la région de Sirmium par plusieurs découvertes. C’est la tombe de chef militaire à Batajnica, contenant un casque du type Baldenheim, une épée, un bouclier à umbo, une lance, des mors de cheval62 (Fig. 14). L’épée, découverte à Rakovac,63 provient sans doute d’une autre tombe du même type (Fig. 18.1). Une plaque-boucle, ornée de tête d’aigle provenant de Sirmium 64 est considérée comme caractéristique des Gépides65 (Fig. 15). Une des rares nécropoles gépides au nord du Danube a été fouillée à Constantia (Kovin), fortin construit contra Margum au IVe s. Notons parmi les objets découverts encore une plaque-boucle, ornée de tête d’aigle.66 La culture matérielle gépide peut être suivie à Sirmium même. Les fouilles des dernières années ont révélé de nombreux restes de cabanes construite à l’intérieure du soi-disant palais impérial, ainsi que des tombes.67 D’autre part il est plus difficile d’identifier culturellement les tombes découvertes à Sirmium, autour du mausolée octogonal.68 D’une part ces tombes ont livré un récipient, habituel des contextes funéraires gépides, mais d’autre part, leur mobilier réduit, essentiellement des boucles de ceinture et des peignes, est plus caractéristique de la population romaine, qui sans doute restait à Sirmium. La nécropole de Kormadin–Jakovo, dans la région de Srem69 est d’habitude considérée comme gépide. Ce cimetière apparaît durant la période de D3/MD2 (450–470/480) de la chronologie danubienne, voire D2/ Sl. 14. Batajnica, grob vojnog zapovednika – {lem (Vogt 2006) D3/MD1 (430–460), à en juger d’après les découvertes dans la tombe 15.1956–1957 des fibules du type Viminacium.70 Une plaque-boucle à plaque réniforme ayant des parallèles à Vy{kov,71 appartient à la période E/MD3 (470/480–510). La nécropole existe jusqu’à la période MD4 (510–540/550) au moins, comme le témoignent les découvertes des boucles à ardillon à 57 58 59 60 Ercegovi}-Pavlovi} 1980, 15, 39, pl. 18 et 31. Kiss 1984, 58–60. Germanen 1987, 231, V.19. Bierbrauer 1975, Taf. 78.1; Miladinovi}-Radmilovi} 2011, 421. 61 Vinski 1964, 157–178. Vinski 1957, 3–27, Pl. 1–9. 63 Vinski 1957, 34, Pl. 22.80; Germanen 1987, 230, V, 18.b. 64 Vinski 1957, 33, sl. 94. 65 Bóna 1976, 28–30; Nagy 2002. 66 Milinkovi} 2005, 208–212; Ivani{evi}, Bugarski 2008, 45, fig. 8; Kazanski 2013, 115–130. 67 Jeremi} 2002, 43–58; Pejovi}, Lu~i} 2011, 389–413. 68 Brukner 1995. 69 Dimitrijevi} 1960; Dimitrijevi} 1964. 70 Ivani{evi} et alii 2006, 17. 71 cf. Tejral 2005, fig. 6.B.3. 62 145 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 Fig. 15. Diffusion des plaques-boucles à tête d’aigle (Bóna 1976) Sl. 15. Rasprostrawenost pojasnih kop~i sa glavom orla (Bóna 1976) le cadre de la reconstruction du système défensif.74 De fait, on peut difficilement concevoir que cette restauration d’une frontière sûre, incluant également le relèvement de forteresses intérieures, soit l’œuvre d’une seule époque et d’un seul souverain. A l’appui de cette thèse citons l’exemple de Ratiaria pour laquelle Procope mentionne qu’elle a été restaurée pendant le règne de Justinien Ier, ce qui est à rejeter, conformément l’inscription Anastasiana Ratiaria semper floreat trouvée dans le site, qui indique clairement que la reconstruction fut commencé durant le règne de Anastase Ier.75 C’est avec l’approbation de l’empereur Anastase Ier que les Hérules s’installent vers 512 sur le territoire de l’Empire.76 Sous Justinien Ier ce peuple s’installa dans base scutiforme et rivets scutiformes, voire plus tard. En effet, la plaque-boucle dite méditerranéenne du type Legoux–Périn–Vallet 161, découverte dans une sépulture de Kormadin–Jakovo, est datable de la période MA2 tardive de chronologie mérovingienne (520/530– 560/570) et surtout de celle de MA3 (560/570–600/ 610)72 (Fig. 16). Les plaques-boucles de même type sont également attestées dans la nécropole de Sirmium.73 LA RESTAURATION DU LIMES AU VIe SIÈCLE ET LES HÉRULES La restauration du limes danubien commenca sous le règne d’Anastase Ier. Selon Procope (De aedificiis) c’est l’ouvre de Justinien, qui a relevé les fortifications détruites et en a édifié de nouvelles. Les vastes fouilles archéologiques effectués le long du Danube sont venues confirmer cette restauration du limes danubien, tout en ramenant à la lumière du jour de nombreuses forteresses inconnues des sources écrites, érigées ou relevées dans 72 73 74 75 76 146 Legoux, Périn, Vallet 2004, n° 161. Germanen 1987, 230–231, V, 24. Kondi} 1984, 147–155. Velkov 1985, 883–891. Mirkovi} 1981, 103–104. IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 2 5 6 STARINAR LXIV/2014 3 7 8 4 9 Fig. 16. Nécropole de Kormadin–Jakovo – objets caractéristiques (Dimitrijevi} 1964) Sl. 16. Nekropola Kormadin–Jakovo – karakteristi~ni prilozi (Dimitrijevi} 1964) les alentours de Singidunum qu’il a tenu jusque vers 550.77 C’est dans les environs de cette ville que Narsès leva en 545 des mercenaires hérules pour les amener en Italie.78 Les Hérules participaient en tant que fédérés, aux expéditions de l’armée byzantine sur les nombreux fronts de l’Empire – contre les Vandales, les Goths et la Perse – ainsi qu’à la répression de la sédition dite Nika à Constantinople.79 Le nombre des Hérules était considérable. Sous Justinien Ier, l’armée hérule représentait plus de 4500 combattants,80 ce qui permet d’estimer leur nombre en 20000–25000 personnes81. On peut donc supposer que tout ce monde occupait, à part Singidunum,82 d’autres forteresses proches du Danube.83 A Svetinja, au nord-ouest de la ville romaine de Viminacium, ont été repérés les restes d’un ouvrage défensif érigé, très vraisemblablement, après 542. Il s’agit en l’occurrence d’un rempart d’une longueur de 100 à 110 m, flanqué de deux tours rectangulaires à ses extrémités. Outre cela, on a constaté dans l’espace fermé par ce mur, la présence d’une série d’habitations qui abritaient, à en juger par les trouvailles de 77 Procope, BG II.15.30, III.33.13. Procope BG III, 13, 33. 79 Procope BG II, 4, 95; 19, 133; III, 1, 173–174; 13, 205; 27, 240–241; 39, 262; IV, 26, 295–296, 30–31, 309; Procope BV I, 11, 29; Agathias, Historia, I, 11, 337; II, 7, 361. 80 Procope, BG III.34.40. 81 Partant de la notion d’une famille biologique moyenne de 4 à 5 personnes. 82 Ivani{evi}, Kazanski 2010, 147–157; Sarantis 2011, 361–402. 83 Le recours aux Barbares pour assurer la défense du Danube a été imposée par le déficit démographique provoqué par la peste qui a frappé l’Empire en 541–543, auquel s’ajoutait l’engagement de l’armée sur un grand nombre de fronts à travers tout l’Empire (Teall 1965, 319–322). Il faut noter aussi que Justin, le magister militum per Illyricum a été engagé dans les campagnes d’Italie, ainsi que Vitalius et finalement Jean, ce qui indique que les commandants, ainsi que leurs armées ont été absents de 537 jusqu’aux années 550 de leur territoire – les Balkans (PLRE III, 748–749, 1380–1381 et 652–662). A part les Hérules l’Empire a engagé dans son armée d’autres Barbares de la frontière danubienne. Ainsi les Huns, les Antes et les Sclavènes sont présents dans l’armée de Bélisaire (voir pour plus des détails: Kazanski 2009). Narsès, quand à lui, disposait, à part 3000 Hérules, de 2500 cavaliers et 3000 « écuyers » lombardes, de 400 Gépides, ainsi que des ressortissants d’autres peuples, dont les effectifs ne sont pas mentionnés (Procope BG IV, 26, 5–17). 78 147 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 STARINAR LXIV/2014 2 3 Fig. 17. Viminacium II (secteur nord-ouest) – objets de tradition scandinave : 1) sépulture 138 ; 2) sépulture 133 ; 3) Kasidol (Ivani{evi} et alii 2006) Sl. 17. Viminacium II (severozapadni sektor) – predmeti skandinavske tradicije: 1) grob 138; 2) grob 133; 3) Kasidol (Ivani{evi} et alii 2006) céramique et d’autres objets, une population formée de fédérés. Cet habitat est daté des années 570–580.84 Enfin quelques restes d’une couche paléo-byzantine ont également été identifiés à Seli{te, près de l’aire occupée par la ville romaine. C’est à celle-ci que pourraient être rattachées les nécropoles du VIe s. à Burdelj (Viminacium I) et à Vi{e grobalja (Viminacium II),85 et, plus particulièrement, le secteur nord-ouest de cette seconde nécropole. Celui-ci peut être très précisément daté entre la fin de la phase MD 4 et le début de la phase MD 6 de la chronologie du Barbaricum danubien, soit de 530 à 560.86 D’après le mobilier la nécropole de Viminacium II – Vi{e grobalja – le secteur nord-ouest, peut être attribuée à des fédérés chargés de la défense des frontières de l’Empire sous Justinien Ier. Ce mobilier dénote une influence de la culture matérielle des Germains septentrionaux et occidentaux. Ainsi, le port des fibules, pour certaines, en haut de la poitrine et au niveau du bassin (Viminacium, tombes 112, 132, 133, 144), et pour d’autres, au niveau du bassin et des jambes (tombe 143) est caractéristique des Germains occidentaux.87 D’autre part trois couteaux de tisserand découvertes à Viminacium, (tombes 118, 143 et 2083)88 sont bien connus chez les Fig. 18. Epées à chape d’éntre en bande métallique : 1) Rakovac (Germanen 1987) ; 2) Viminacium, tombe 1876 (Ivani{evi} et alii 2006) Sl. 18. Ma~ i metalni tokovi korica: 1) Rakovac (Germanen 1987); 2) Viminacium, grob 1876 (Ivani{evi} et alii 2006) 84 1 Popovi} 1988, 34–35. Ivani{evi} et alii 2006. 86 Tejral 2005. 87 Par ex. Clauss 1987 : Abb. 12.2; 15; 25.1; 29; 33; Hansen 2004: Abb. 59. 88 Ivani{evi} et alii 2006, 43. 2 85 148 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 2 3 4 5 6 7 9 10 STARINAR LXIV/2014 8 11 12 Fig. 19. Ulpiana–Justiniana Secunda – sépulture (Milinkovi} 2003) Sl. 19. Ulpiana–Justiniana Secunda, grob (Milinkovi} 2003) Lombards du Danube.89 Typiques de ce peuple,90 ils sont également attestés en Angleterre du Sud, en Allemagne entre le Rhin Moyen et l’Elbe moyenne, par ex. chez les Thuringiens, plus rarement en Gaule, en Italie et en Pannonie.91 Deux fibules à tête rectangulaire et pied à motifs zoomorphes dégradés (tombes 133 et 138) sont également typique avant tout pour les Germains occidentaux et septentrionaux – Lombards, thuringiens, Francs, alamans, Scandinaves, Anglo-Saxons etc92 (Fig. 17.1–2). Une bande métallique de fourreau porte un décor en tresse provenant de la tombe 187693 (Fig. 18.2).94 Les chapes d’entrée en bande métallique sont attestées surtout en Gaule du Nord et de l’Est, en Germanie, en Angleterre, épisodiquement en Scandinavie, en Espagne, en Finlande et rarement sur le Danube moyen.95 Les umbo des tombes 115 et 2093 possèdent des rivets plaqués d’argent,96 caractéristiques de l’Occident mérovingien, ainsi que des Lombards et des Thuringiens.97 Tous ces traits caractérisent une culture matérielle des Germains occidentaux et septentrionaux, tels les Lombards, voire les Alamans et les Thuringiens. La présence des crânes portant des traces de déformation artificielle dans certaines tombes de cette phase98 89 Werner 1962, Taf. 1.10, 16.7, 55.9; Bóna, Horváth 2009, Taf. 20.2.2, 22.18.1, 29.6.2, 50.9, 138.1,2; Tejral et alii 2011, Taf. 42.28.2, 51.13. 90 Werner 1962, 82,84. 91 La carte de diffusion : Werner 1962, Taf. 68.2, Fundliste 2. 92 Ivani{evi} et alii 2006, 15. 93 Ivani{evi} et alii 2006, 38. 94 Elle rappelle celle de Rakovac, découverte en 1909 (Vinski 1957, pl. 22.80, Germanen 1987, 230, V,18.d), cette dernière porte cependant un décor linaire et appartient donc au type Entringen– Sindelfingen, caractéristique de l’Alemagne du Sud-Ouest (Menghin 1983, Abb. 50). 95 Menghin 1983, Abb. 11 et 12. 96 Ivani{evi} et alii 2006, 42. 97 Werner 1962, Fundliste 1. 98 Miki} 1994, 193. 149 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 STARINAR LXIV/2014 2 Fig. 20. Udovice (Popovi} 2008) Sl. 20. Udovice (Popovi} 2008) de tradition lombarde (Fig. 19). Les fibules digitées, à décor géométrique et à un arque sur les doigts de la tête, mais à pied de forme u différente, sont attestées chez les Lombards en Pannonie, donc avant l’exode de 568.110 En Italie les fibules semblables, du type BF4a, sont datées de 570–590.111 ne s’oppose pas à ce rapprochement. Cette pratique est rare, certes, chez les Lombards au Nord du Danube,99 mais attestée davantage chez les Thuringiens et les Alamans à l’époque post-hunnique.100 Or, à la même époque, les crânes déformés sont bien connus chez les Germains orientaux du Danube101 et leur présence à Viminacium pourrait alors témoigner de l’intégration des Germains danubiens. Parallèlement on note aussi une influence byzantine sur le costume de cette nouvelle population, qui portait notamment des fibules à pied attaché de tradition balkanique.102 Un horizon semblable des tombes a été identifié dans la nécropole de Singidunum aux environs duquel s’étaient installés des Hérules103 Ainsi, on peut émettre la hypothèse de l’appartenance aux Hérules des tombes de Viminacium.104 Cela expliquerait la découverte de quelques objets de la tradition scandinave dans la région de Viminacium, car on connaît les liens de ce peuple avec la Scandinavie.105 Ce sont par ex. la fibule de la tradition scandinave, provenant de Kasiodol, au sud de Viminacium106 (Fig. 17.3). Elle remonte aux prototypes scandinaves du milieu du Ve – premier quart du VIe s.107 En outre, on attribue aux Germains septentrionaux ou occidentaux (Hérules, Thuringiens…) la célèbre tombe féminine trouvée à Gra~anica (Ulpiana – Justinana Secunda dans la province de Dardanie).108 Elle a livré deux fibules de tradition sud-scandinave, à tête rectangulaire, datables de 550–600109 et une fibule digitée, 99 Un cas dans la nécropole de Lu`ice nous est signalé par J. Tejral, la publication de la nécropole: Tejral et alii 2011, 225–312. 100 Werner 1962, 116, 117, Fundliste 8; Anke 1998, 130, 131. 101 Anke 1998, 129, 130. 102 Uenze 1992, 146–154. 103 Procope, BG II.15.30; III.33.13; Ivani{evi}, Kazanski 2002, 124. 104 Ivani{evi} et alii 2006, 133, 134; Ivani{evi}, Kazanski 2010. 105 Procope, BG II.15. 106 Ivani{evi} et alii 2006, 15,16. 107 (Type Sjovold A1b, attesté notamment à Kvale, en Norvège et à Lundbjers/Suderbjärs, dans l’île de Gotland, voir : Sjóvold 1993 : pl. 3 ; Näsman 1984: Taf. 15.1). Les fibules comparables sont attestées à Augusta (Kaleto), en Bulgarie, ainsi qu’dans la nécropole gépide de Szolnok-Szanda, dans le bassin de la Tisza (Haralambieva 1984, obr. 6; Bona, Nagy 2002,Taf. 38.73.3). 108 Vierck 1981, 68; Popovi} 1984, 174; Milinkovi} 2003, 169–177. 109 cf. par ex. Hóilund Nielsen 1987, fig. 10. E2. 110 Par ex. Werner 1962, Taf. 1.1,2, 9.1,2. 111 Jórgensen 1992, fig. 5. 150 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) A cette même période peuvent être assigne deux tubuli avec solidi mis à jour sur le limes danubien à Udovice près de Grocka (Ad sextum miliarem ?), non loin de Singidunum. Le premier tubulus comporte deux solidi, dont un de Valentinien III frappé à Rome ou Ravenne en 425–455, et l’autre de Severus issue à Ravenne en 461–465. Le second tubulus comprend aussi deux solidi émis à Ravenne, un de Valentinien III de 425–455 et un d’Honorius frappé en 421. (Fig. 20) Cette trouvaille est selon Ivana Popovi} originaire de l’Italie du Nord et elle date de la période d’occupation barbare du limes de l’Illyricum du Nord.112 D’autre part Svante Fischer indique que ces tubuli ont été fabriqués en Scandinavie du Sud après 475. Ils témoignent, selon l’auteur, de la présence des mercenaires barbares, entre 475 et 500, sur le limes danubien.113 Cependant, comme l’a justement souligné I. Popovi}, en s’appuyant sur les archéologiques scandinaves, les colliers de ce type ont la datation très large, du début du IVe au début du VIIe s.114 Ainsi, le fameux torque en or provenant d’Alleberg et contenant des éléments comparables est attribué par V. Holmqvist à la première moitié du VIe s., et un autre torque du même type, découvert à Möne) – au VIe s.115 Ainsi, on peut admettre que cette découverte peut être mise en liaison avec le séjour des Hérules sur le Danube sous Justinien. D’autre part J. Tejral a remarqué les parallèles entre la céramique de certaines tombes de Singidunum (57, 61, 85, 106)116 et celle de l’époque pré-lombarde au Nord du Danube.117 Cette zone est habituellement attribuée aux Hérules avant leur défaite par les Lombards.118 Cependant une différence notable existe de la culture matérielle entre les tombes de Viminacium et celles de la phase pré-lombarde (phase MD 3, ou la période E – 470/480–510) au Nord du Danube.119 Ceci nous rappelle à la prudence quant à l’attribution hérule des tombes de Viminacium.120 De toute facon, il est bien possible que les Hérules, lors de leur migration, ont entrainé avec eux d’autres groupes germaniques et non germaniques, y compris très probablement ceux des Germains d’Elbe (Lombards, Thuringiens), qui étaient leurs voisins dans le pays nord-danubien.121 Nous rattachons donc la population du VIe s. enterrée dans les nécropoles de Viminacium II – Vi{e grobalja – secteur nord-est et à Singidunum III (phase tardive) à des Germains occidentaux et septentrionaux, sans vouloir considérer ici plus avant le problème de leur appartenance ethnique, compte tenu de l’évident mélange de culture matérielle. Notons cependant que ces nécropoles ont été attribuées aux Gépides,122 STARINAR LXIV/2014 quoique la culture matérielle des Gépides se distingue nettement de celle des Barbares de Viminacium. Or, ce peuple qui occupait de vastes territoires aux portes de l’Empire, a pratiquement été en conflit permanent avec les Byzantins.123 Toute une série d’informations relatives aux événements militaires touchant l’Illyricum, le font précisément apparaître au nombre des principaux adversaires de l’Empire sur cette partie de la frontière sous Justinien Ier. Qui plus est, entre 535 et 567, les Gépides ont tenu une partie de la Pannonie avec la ville de Sirmium, et ce contre la volonté de Constantinople. Pour ces raisons, nous serions enclins à penser qu’aux fins de défendre ses frontières l’Empire veillait à recruter des peuples germaniques hostiles aux Gépides. D’ailleurs un des piliers de la politique militaire de l’empereur Justinien était précisément, comme le rapporte Procope, d’opposer les barbares les uns aux autres.124 Les régions sises aux confins de l’Illyricum n’en restaient pas moins sous la menace constante des incursions barbares. Une des plus sérieuses, survenue en 548–549, aboutit à un affrontement opposant Germains du Danube, Gépides et Hérules rebelles, d’une part, à des Lombards, Romains et Hérules fédérés, d’autre part. En dépit de la victoire de ses troupes, l’Empire fut d’ailleurs contraint, en raison du danger de nouvelles incursions des Gépides, de laisser une 112 Popovi} 2001, 55–58, fig. 9; Popovi} 2008. Fischer 2008, 81–88 ; Voir aussi : Ivani{evi}, Kazanski 2010, 154–157, fig. 7. 114 Popovi} 2008 : 78. 115 Holmqvist 1965 : 168, 169. 116 La tombe 85, d’après les grosses perles en ambre est datable plutôt de la période D2/D3 ou D3–E1. Celles 57, 61 et 106 ne contiennent pas d’autre mobilier qui se prête à la datation. Cependant d’après les parallèles provenant de Viminacium (tombes 136 et 143), les cruches provenant des tombes 61 et 106 appartiennent plutôt au VIe s. La petite cruche de la tombe 61 est proche, quant à elle, à celle de l’époque pré-lombarde, provenant de [aratice (Tejral 2005, fig. 8.C.6). 117 Tejral 2005, 135, fig. 8.A–C; Tejral 2007, 102, Abb. 26. 118 Par ex. Wolfram 1990, 273, carte 5. 119 Nombreuses fibules digitées dans des tombes féminines, peu d’armes dans des tombes masculines : voir Tejral 2005, fig. 5 et 6. 120 Ivani{evi}, Kazanski 2010, 147–157. 121 Tejral et alii 2011, 64–66. 122 Zotovi}, 1994, 183–190 ; Milinkovi} 1998, 242–244 ; Mirkovi} 1998, 123–125. 123 Voir en dernier lieu Kazanski 2013. 124 Procope Hist. arc. 11. 113 151 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) partie de l’armée sur le Danube au lieu de l’envoyer en Italie.125 STARINAR LXIV/2014 précisément à cette époque, ont été recrutés en grand nombre par l’empereur Tibère II. Par la suite, nous savons que ces mêmes Germains ont suivi Maurice lors de sa campagne sur le lointain front perse.134 Ajoutons qu’en 582 Maurice a été le dernier empereur connu à recevoir le titre de comte des fédérés.135 Viminacium était à cette époque une importante place stratégique face au Barbaricum, ainsi qu’à la province de Pannonie, actuelle région de Srem. Ceci est notamment attesté par l’existence d’une fabrica ou fabricenses identifiées dans une inscription fragmentaire non publiée, trouvée à Viminacium et datée du règne de Tibère II. On y reconnaît à la dernière ligne, parfaitement lisible, les termes IMP.TIB.FABRI.136 Toute l’importance stratégique de Viminacium devait d’ailleurs s’affirmer quelque temps plus tard, en 600, lors de la dernière tentative de Byzance visant à endiguer la pénétration des Avars dans la vallée du Danube moyen. Cette place servira alors de base pour les opérations lancées en profondeur en territoire ennemi. Nous pouvons également rattacher à cet horizon, attestant la présence de barbares dans la partie nord de l’Illyricum, toute une série de trouvailles provenant du LES SLAVES, LES AVARS ET LA FIN DU LIMES La situation sur la frontière danubienne s’aggrave avec l’apparition des nouveaux ennemis, les Sclavènes et les Antes. Selon Procope dès le début du règne de Justinien les Huns, les Sclavènes et les Antes chaque année font des incursions sur l’Illyricum et sur la Thrace.126 En 548–552 les invasions sclavènes se produisent en Illyricum pratiquement chaque année, de plus en plus dévastatrices.127 La carte des trésors monétaires montre une situation de stresse militaire dans la partie orientale d’Illyricum.128 On peut noter deux périodes dans les guerres des Slaves contre l’Empire. La première englobe les années 527–578, quand les Sclavènes se contentent de faire des incursions sur le territoire byzantin, sans essayer de rester dans les Balkans. La deuxième période des guerres danubiennes va de 579/580 à 626. Durant cette période, les Sclavènes restent définitivement sur le territoire de la péninsule balkanique129. Les cartes de répartition des trésors monétaires montre que pratiquement toute la péninsule balkanique est exposée au danger.130 A cette époque les Sclavènes agissent soit en tant que les alliés subordonnés des Avars, soit, très souvent, à leur propre compte.131 Les Barbares continuèrent leur garde sur la frontière du nord de l’Illyricum sous le règne de Tibère II et de Maurice, comme l’ont montré les travaux de fouilles sur l’ouvrage défensif byzantin sis à Svetinja, dans l’agglomération de Viminacium. Ceux-ci ont mis au jour, à proximité même, les traces d’un habitat de fédérés comptant plusieurs habitations qui a été attribué aux Gépides par M. Popovi}. D’après ce même auteur, il s’agirait des éléments gépides qui, sous la conduite d’Usdibade, sont passés sur le territoire de l’Empire après la défaite de Gépides face au Lombards en 567.132 Les trouvailles mêmes, constituées de tessons provenant de 14 pots dont douze ornés de motifs estampillés et deux avec décoration lustrée, peuvent toutefois aussi bien s’inscrire dans l’héritage gépide que dans celui des Lombards. Cette céramique est en effet caractéristique du bassin des Carpates, mais aussi de l’Italie du Nord, où se sont installés les Lombards en 568.133 La garnison de fédérés cantonnés à Svetinja, dont la présence peut être datée des années 570–580, pourrait donc tout aussi bien être mise en relation avec les Lombards qui, 125 (Procope BG III, 35.) Si le recours aux Barbares restait principalement destiné à la lutte contre d’autres peuples hostiles, les auxiliaires germaniques de l’Empire apparaissent, à l’occasion, dans la résolution de certains conflits internes du monde barbare. On sait ainsi qu’en 551, répondant à un appel des Lombards en conflit contre les Gépides, Byzance leur envoya un détachement de forces mixtes commandées par les fils du général byzantin Germain, Justin et Justinien, Aratius Kamsarakan, l’ex-roi Hérule Svartua et Amalafrede, fils d’un neveu de Théodoric et beau-frère du roi lombard Avdui. Finalement, seul Amalafrede se porta au devant des Lombards avec lesquels il défit les Gépides, alors que les autres chefs, sur ordre de l’empereur, restèrent à Ulpiana, dans la province de Dardanie, compte tenu qu’une révolte était survenue dans la ville en raison de querelles religieuses (Procope BG IV, 25). 126 Procope, Hist. arc. 18.20. 127 Procope, BG III.29.1, 2, 38.1–23, IV.25.1–5. 128 Morrisson et alii 2006, carte 3. 129 Cependant les premiers traces archéologiques incontestables de la présence des Slaves dans la péninsule balkanique remontent à présent au VIIe s. (par ex. Vida, Völling 2000) ; voir aussi : Popovi} 1975, 1978 et 1980. 130 Morrisson et alii 2006, cartes 5–8. 131 Kazanski 1999, 65–76. 132 Popovi} 1988, 21–23, fig. 17. 133 Werner 1962, 51, fig. 4, 54–57, 176–178, pl. 18–19; von Hessen 1968. 134 Haldon 1979, 22, n° 8. 135 Zuckerman 2004, 168. 136 Mirkovi} 1998, 124–127. 152 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 2 3 STARINAR LXIV/2014 4 Fig. 21. Viminacium III – sépulture 23 (Ivani{evi} et alii 2006) Sl. 21. Viminacium III, grob 23 (Ivani{evi} et alii 2006) limes, mais aussi de sites se trouvant plus à l’intérieur du territoire byzantin. Sur le limes c’est par ex. la tombe isolée de Bole~ica, en aval du castrum d’Octavum près de Singidunum, dont proviennent une boucle de fer ovale et un peigne à une rangée de dents, qui présente de nombreuses analogies précisément avec le matériel de la nécropole Viminacium III – secteur nord-ouest. On peut également évoquer les trouvailles de céramique estampillée provenant des forteresses de Taliata (Veliki Gradac) et de Transdierna (Tekija) (Simoni 1978, 215–216, 226, n. 108, Pl. III 1). Il faut noter aussi quatre tombes mises à jour à Margum,137 ainsi que deux sépultures isolées découvertes à Kamenovo, près de Petrovac na Mlavi, au sud de Viminacium, sur l’axe principal conduisant à Naissus. Ces deux sépultures de Kamenovo, dont une de guerrier, possèdent de nombreux parallèles dans le matériel de la nécropole la plus récente de Viminacium II – Vi{e grobalja. K. Simoni les a rattachées aux Gépides ayant trouvé refuge sur le territoire de l’Empire après leur défaite en 567.138 En ce qui concerne les territoires situés plus à l’intérieur de l’Empire, nous mentionnerons les découvertes dans le massif du Juhor,139 à Cari~in Grad (Justiniana Prima)140 ainsi que à Jelica, où l’on a mis au jour de la céramique, un étui de peigne, une agrafe avec plaque cordiforme et une fibule circulaire avec protomes. Ces découvertes, ainsi que d’autres objets, mis au jour en Serbie, sont rattachées par M. Milinkovi} aux Gépides.141 Cependant toutes ces trouvailles cependant présentent à nouveau, dans une large mesure, des traites typiques également de la culture des Lombards et d’autres Germains.142 L’ouvrage défensif de Svetinja a été endommagé, et l’habitat voisin incendié, au cours de l’été 584, lors de la grande percée des Avars, comme l’ont montré les trouvailles numismatiques.143 Cette avancée des Avars, lorsque, selon Théophylactes Simokattès, sont également tombées Singidunum, et Avgusta, a mis un terme à la longue résistance de Viminacium en tant que ville frontalière de l’Empire.144 On a constaté dans les ruines de Svetinja un second niveau, reflétant une nouvelle installation ne renfermant plus aucune trace de céramique estampillée caractéristique, bien que certaines trouvailles, telles un étui de peigne et un manipule de bouclier sont typiques de la culture matérielle des Germains145. La fortification de Svetinja a perdu sa fonction durant la dernière décennie du VIe s., vraisemblablement à l’occasion d’une incursion des Avars en 593, ou 596.146 137 Cunjak 1992, 34–40 ; Bugarski, Ivani{evi} 2013. Simoni 1978, 209–214. 139 Brmboli} 1986, 204, fig. 207. 140 Popovi} 1984, 160–178; Ivani{evi} 2012, 57–69. 141 Milinkovi} 2000, 352–359. 142 D’ailleurs ces territoires intérieurs étaient alors peu sécurisés. Ainsi en 579–580 l’ambassadeur avar a été tué en Illyricum par les Sclavènes, qui étaient en train de dévaster cette région (Ménandre, fr. 64). 143 Popovi} 1988, 34–35. 144 Théophylacte Simocattès, Hist, I, 3, 4. 145 Les manipules de bouclier comparables ont été également mis au jour au Proche-Oruent, dans le contexte purement byzantin (Quast 2012, 357, Abb. 5.2). 146 Popovi} 1988, 26–35, fig. 20.5 et 24. 138 153 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) 1 STARINAR LXIV/2014 les rares tombes, comme celle déjà citée d’Ulpiana – Justinana Secunda.149 Dans le même contexte on peut qualifier les rares trouvailles « barbares » de Cari~in Grad (Justiniana Prima). Il s’agit de plaque-boucle d’origine germanique, de bouton d’umbo, de rares fragments de céramique ornée de motifs estampillés, mais aussi de deux fibules dites slaves. Une de ces fibules150 est typique des trésors dits du type Martynovka et de la civilisation de Penkovka, cette dernière est attribuable aux Antes.151 (Fig. 22.1) L’autre152 ait partie du type danubien de Kiskorös–Sarmizegetusa, diffusé dans le bassin des Carpates, sur le Haut–Dniestr, dans la région du Dniepr moyen, en Crimée et en Prusse orientale153 et parfois attesté sur les sites slaves de la civilisation de Prague154 (Fig. 22.2). La fibule de Cari~in Grad rappelle le plus les pièces provenant de l’habitat slave de Dem’janov et de la nécropole avare de Kiskorös.155 Ces pièces représentent qu’une petite partie d’objets parmi des centaines d’autres d’origine byzantine.156 Ces fouilles démontrent que l’intérieur de l’Illyricum du Nord fut protégé par l’armée byzantine dont témoignent de divers éléments de ceinturon, mais avant tout les nombreux fragments de casque d’officier, dit de Baldenheim.157 La chute du limes danubien dans les premières décennies du VIIe s. marqua également la fin de l’administration byzantine sur le vaste territoire de l’Illyricum. Il est donc question de la fin d’un long processus au cours duquel les Barbares, dans notre cas principalement des Germains, jouèrent alternativement le rôle d’agresseurs, de colons, de peuples souverains et de mercenaires. Et finalement, l’éclipse de Byzance, contrainte de quitter la scène historique dans les Balkans au début du VIIe s., vit également leur disparition lorsqu’ils durent laisser la place aux Slaves, aux Avars et, plus tard, aux Bulgares. 2 Fig. 22. Fibules digitées de Cari~in Grad (Institut archéologique de Belgrade) Sl. 22. Lu~ne fibule iz Cari~inog grada (Arheolo{ki institut Beograd) Nonobstant l’absence des sources écrites le confirmant clairement, le maintien de Barbares sur le territoire de Viminacium est également attesté par le dernier horizon des inhumations, datées de la fin du VIe et du début du VIIe s. Celui-ci a été constaté à Vi{e grobalja – secteur nord-ouest (Viminacium II) et surtout à Lanci (Viminacium III)147 (Fig. 21). Ajoutons que ce même horizon apparaît également dans la nécropole de Singidunum III.148 Ayant été rapidement repris par les Byzantins, Viminacium devait se trouver une dernière fois au premier plan de la scène historique lors de leur campagne contre les Avars en 600. La ville constitua alors la principale base et place forte à partir de laquelle l’armée byzantine lancait ses attaques contre les Avars et leurs alliés slaves et germaniques, tenant la rive gauche du Danube. Plusieurs opérations des forces byzantines, sous le commandement de Priscus, se soldèrent par des succès de sorte que les Avars, vaincus sur la Tisza, furent repoussés plus au nord. L’existence des barbares sur le limes est bien documenté, tandis que leur présence à l’intérieur de l’Illyricum du Nord est épisodique comme le démontre 147 148 149 150 151 152 153 154 155 156 157 154 Ivani{evi} et alii 2006. Ivani{evi}, Kazanski 2002, 124–127. Milinkovi} 2003, 143–178. Popovi} 1984, fig. 188. Voir en dernier lieu Kazanski 2013a. Popovi} 1984, fig. 186. Gavrituhin 1991, 128–130, Pl. 1.A; Curta 2006. Dem’janov: Gavrituhin 2005, 430, ris. 21.4. Gavrituhin 1991, 128, 129. Ivani{evi} 2012, 57–69. Bavant 2008, 327–353. IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 SOURCES: Agathias, Historia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Agathias, Historiarum Libri Quinque, éd. R. Keydell. Berlin–New York, 1967. Codex Theodosianus, éd. Th. Mommsen, P. Meyer, Berlin, 1905. The history of Menander the Guardsman, éd. R.C. Blokley, Liverpool, 1985. Notitia Dignitatum, éd. O. Seeck, Berlin, 1876, Francfort, 1962. Prisci fragmenta, éd. L. Drindorf, HGM I. Leipzig, 1970. Procopii Caesarensis De Bello Gothico (De Bellis libri V–VIII), éd. J. Haury. Leipzig, 1905. Procopii Caesarensis De Bello Vandalico (De Bellis libri III–IV), éd. J. Haury. Leipzig, 1905. Procopii Caesarensis De Aedificiis, éd. J. Haury. Leipzig, 1913. Procopii Caesarensis Historia Arcana, éd. J. Haury. Leipzig, 1906. Théophylacte Simocatta, The History of Theophylact Simocatta, éd. M. et M. Whitby. Oxford, 1986. Zosimus historicus, éd. F. Paschoud, Paris, 1971–1989. Cod. Theod. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ménandre . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Notitia Dignitatum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Priscus frag. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Procope BG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Procope BV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Procope De aedif. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Procope Hist. arc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Théophylacte Simocatta, Hist. . . . . . . . . . . . . . . . . . . . Zosimus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BIBLIOGRAPHIE: Bóna, Nagy 2002 – I. Bóna, M. Nagy, Gepidische Gräberfelder am Tiszagebiet I, Budapest 2002. Bóna, Horváth 2009 – I. Bóna, J. B. Horváth, Langobardische Gräberfelder in West-Ungarn, Budapest 2009. Bondoc 2009 – D. Bondoc, The Roman Rule to the North of the Lower Danube, Cluj – Napoca 2009. Brmboli} 1986 – M. Brmboli}, Ranovizantijsko utvr|ewe na Juhoru. Zbornik Narodnog muzeja u Beogradu 12.1, 1986, 190–210. Brukner 1995 – O. Brukner, Mauzolej – oktogonalna gra|evina. In : Arheolo{ko istra`ivawa du` autoputa kroz Srem, Novi Sad 1995, 175–180. Brun{mid 1905 – J. Brun{mid, Starine ranijega srednjega vijeka iz Hrvatske i Slavonije, Vjesnik Hrvatskoga arheolo{koga dru{tva, 8, 1905, 208–220. Bugarski, Ivani{evi} 2012 – I. Bugarski, V. Ivani{evi~, Pograni~úe Rimskoç imperii i varvarov: sistema oboronù Imperii ot Kuciý do Lederatù: Lesnaý i lesostepnaý zonù Vosto~noç Evropù v ûpohi rimskih vliýniç i Velikogo pereseleniý Anke 1998 – B. Anke, Studien zur reiternomadischen Kultur des 4. Bis 5. Jahrhunders, Weissbach 1998. Bavant 2008 – B. Bavant, Fragments de casques de type Baldenheim trouvés à Cari~in Grad, Mélanges de l’École francaise de Rome – Moyen Âge 120, 2008, 327–353. Bierbrauer 1975 – V. Bierbrauer, Die ostgotischen Grab- und Schatzfunde in Italien, Spoleto 1975. Bierbrauer 2008 – V. Bierbrauer, Ethnos und Mobilität im 5. Jahrhundert aus archäoloigscher Sicht: Vom Kaukasus bis nach Niederösterreich, München 2008. Bjelajac, Ivani{evi} 1993 – Lj. Bjelajac, V. Ivani{evi}, Les témoignages archéologiques des Grandes Invasions à Singidunum, Starinar 42, 1991, 123–139. Böhme 1974 – H.-W. Böhme, Germanische Grabfunde des 4. bis 5. Jahrhunderts zwichen unterer Elbe und Loire, München 1974. Bóna 1976 – I. Bóna, A l’aube du Moyen Age. Gépides et Lombards dans le bassin des Carpates, Budapest 1976. 155 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 Moskva, 1991, 127–141. Gavrituhin 2005 – I. O. Gavrituhin, Kompleksù pra`skoç kulúturù s datiruü÷imi ve÷ami. In: Archeologia o poczatkach Sÿowian, eds. P. Kaczanowski, M. Parczewski, Kraków 2005, 403–461. Germanen 1987 – Germanen, Hunnen und Awaren. Schätze der Völkerwaznderungszeit, Nuremberg 1987. Haldon 1979 – J. Haldon, Recruitment and Conscription in the Byzantine Army c. 550–950., Wien 1979. Hansen 2005 – C. M. Hansen, Frauengräber im Thüringerreich (Basler Hefte zur Archäologie 2), Bâle 2005. Haralambieva 1984 – A. Haralambieva, Døgovidni fibuli ot epohata na Velikototpereselenie na narodite, Arheologiý 1984/1, 45–52. von Hessen 1968 – O. von Hessen, Die langobardische Keramik aus Italien, Wiesbaden 1968. Hóilund Nielsen 1987 – K. Hóilund Nielsen, Zur Chronologie der Jüngeren Germanischen Eisenzeit auf Bornholm, Acta Archaeologica 56, 1987, 47–86. Holmqvist 1965 – W. Holmqvist, L’art des Germains depuis le Ve s. In: Eggers H. J., Will E., Joffroy R., Homkvist V, Les Celtes et les Germains à l’époque paðenne, Paris 1965, 152–221. Ivani{evi} 1999 – V. Ivani{evi}, Le début de l’époque des Grandes Migrations dans l’Illyricum du Nord. In: L’Occident romain et l’Europe centrale à l’époque des Grandes Migrations, éds. J. Tejral, C. Pilet, M. Kazanski, Brno 1999, 95–108. Ivani{evi} 2012 – V. Ivani{evi}, Barbarian Settlements in the Interior of Illyricum: The Case of Cari~in Grad, In: The Pontic-Danubian Realm in the Period of the Great Migration, éds. V. Ivani{evi} and M. Kazanski, Paris–Beograd 2012, 57–69. Ivani{evi} 2013 – V. Ivani{evi}, Petite fibule ansée de tradition germanique provenant de Ni{avac, In : From Central Balkans to Danube – Archeology and Heritage, éds. V. Petrovi} et V. Filipovi}, Beograd 2013, sous presse. Ivani{evi}, Bugarski 2008 – V. Ivani{evi}, I. Bugarski, Western Banat during the Great Migration Period, In: The Turbulent Epoch II, éds. B. Niezabitowska-Wisniewska, M. Juscinski, P. Luczkiewicz, S. Sadowski, Lublin 2008, 39–61. Ivani{evi} et alii 2006 – V. Ivani{evi}, M. Kazanski, A. Mastykova, Les nécropoles de Viminacium à l’époque des Grandes Migrations, Paris 2006. Ivani{evi}, Kazanski 2002 – V. Ivani{evi}, M. Kazanski, La nécropole de l’époque des Grandes Migra- narodov, red. A. M. Voroncov, I. O. Gavrituhin, Tula 2012, 482–511. Bugarski, Ivani{evi} 2013 – I. Bugarski, V. Ivani{evi}, Migration Period Finds from Margum: A Possible Interpretation, Nyíregyházi Jósa András Múzeumi Évkönyve LV, 2013, (sous presse). Ciobanu, Constantinescu 2008 – D. Ciobanu, Constantinescu E.-M., Die Hunnen im Nordosten Munteniens. In: Hunnen zwischen Asien und Europe, Langenweissbach 2008, 131–141. Ciupercã, Mãgureanu 2008 – B. Ciupercã, A. Mãgureanu, Huns and others peuples – archaeological evidence in present-day Romania. In: Hunnen zwischen Asien und Europe, Langenweissbach 2008, 119–130. Clauss 1987 – G. Clauss, Die Tragsitte von Bügelfibeln. Eine Untersuchung zur Frauentracht im Frühen Mittelalter, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 34, 1987, 491–602. Cunjak 1992 – M. Cuwak, Prilog prou~avawu prisustva Gepida u Podunavqu, Glasnik Srpskog arheolo{kog dru{tva 8, 1992, 34–40. Curta 2006 – F. Curta, A Contribution to the Study of Bow Fibulae of Werner’s Class IG, Arheologia Molodvei 29, 2006, 93–123. Dautova-Ru{evljanin 1981 – V. Dautova-Ru{evljanin, Ein germanisches Grab auf dem Fundort Vranja bei Hrtkovci in Syrmien, Archaeologia Jugoslavica 20–21, 1981, 146–153. Dautova-Ru{evljanin 1998 – B. Dautova-Ru{evqanin, Kasnoanti~ki grob na lokalitetu Vraw u Hrtkovcima, Rad Muzeja Vojvodine 40, 1998, 97–99. Dimitrijevi} 1960 – D. Dimitrijevi}, Gepidska nekropola „Kormadin“ kod Jakova, Rad Vojvo|anskih muzeja 9, 1960, 5–50. Dimitrijevi} 1964 – D. Dimitrijevi}, Le cimetière gépide à « Kormadin » près de Jakovo en Srem, Serbie, Inventaria Archaeologica VI, fasc. 7, Bonn 1964. Dimitrijevi} et alii. 1962 – D. Dimitrijevi}, J. Kova~evi}, Z. Vinski, Seoba naroda – Arheolo{ki nalazi jugoslovenskog Podunavlja, Zemun 1962. Ercegovi}-Pavlovi} 1980 – S. Ercegovi}-Pavlovi}, Les nécropoles romaines et médiévales de Ma~vanska Mitrovica (Sirmium XII), Belgrade 1980. Fischer 2008 – S. Fischer, The Udovice Solidus Pendants – Late-5th Century Evidence of South Scandinavian Mercenaries in the Balkans, Fornvännen 103, 2008, 81–88. Gavrituhin 1991 – I. O. Gavrituhin, Palú~atùe fibulù pra`skih pamýtnikov Podnestrovúý. In: Drevnosti Severnogo Kavkaza i Pri~ernomorúý, 156 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 Legoux, Périn, Vallet 2004 – R. Legoux, P. Périn, F. Vallet, Chronologie normalisée du mobilier funéraire mérovingien entre Manche et Lorraine, Saint-Germainen-Laye 2004. Miki} 1994 – Z. Miki}, Erste Ergebnisse Anthropälogischer Untersuchung des Germanenfriedhofes von Viminacium/Serbien, Starinar 43–44, 1994, 191–199. Milinkovi} 1998 – M. Milinkovi}, Germanska plemena na Balkanu. Arheolo{ki nalazi iz vremena seobe naroda, Beograd 1998 (Thèse doctorale non publiée). Milinkovi} 2000 – M. Milinkovi}, Ulpijana kod Gra~anice na Kosovu i Gradina na Jelici kod ^a~ka u svetlu akulturacionih procesa u Iliriku VI v. In : Tre}a jugoslovenska konferencia vizantologa, Kru{evac 2000, 343–360. Milinkovi} 2003 – M. Milinkovi}, O tzv. `enskom germanskom grobu iz Ulpiane. In: Spomenica Jovana Kova~evi}a, Beograd 2003, 143–178. Milinkovi} 2005 – M. Milinkovi}, Serbien. Reallexikon der Germanischen Altertumskunde, H. Beck, D. Geuenich, H. Steuer (dir.), 28, Berlin 2005, 197–218. Mirkovi} 1981 – M. Mirkovi}, Centralne balkanske oblasti u doba poznog carstva. In: Istorija srpskog naroda, éd. S. ]irkovi}, Beograd 1981, 89–105. Mirkovi} 1986 – M. Mirkovi}, Inscription de la Mésie Superieure II, Viminacium et Margum, Belgrade 1986. Mirkovi} 1998 – M. Mirkovi}, The Legionary Camps at Singidunum and Viminacium in the Defensive System in the fourth-fifth and sixth centuries: Romans and Barbarians. In: The Roman Frontier at the Lower danube 4th–6th centuries, Studia Danubiana I, Bucharest, 1998, 117–130. Mirkovi} 2006 – M. Mirkovi}, Sirmium – Istorija rimskog grada od I do kraja VI veka, Sremska Mitrovica 2006. Miladinovi}-Radmilovi} 2011 – N. Miladinovi}Radmilovi}, Sirmium. Necropolis, Beograd – Sremska Mitrovica 2011. Morrisson et alii 2006 – C. Morrisson, V. Popovi}, V. Ivani{evi}, Les Trésors monétaires byzantins des Balkans et d’Asie Mineure (491–713), Paris 2006. Nagy 2002 – M. Nagy, Die Gepidischen Adlerschnallen und ihre Beziehungen, Budapest Régiségei XXXVI, 2002, 363–392. Näsman 1984 – U. Näsman, Zwei Relieffibeln von der Insel Öland, Prähistorische Zeitschrift 59/1, 1984, 48–80. tions à Singidunum. In: Singidunum 3, Belgrade 2002, 101–157. Ivani{evi}, Kazanski 2009 – V. Ivani{evi}, M. Kazanski, Nouvelle nécropole des Grandes Migrations de Singidunum, Starinar LVIII, 2009, 117–139. Ivani{evi}, Kazanski 2010 – V. Ivani{evi~, M. Kazanskiç, Gerulù Üstiniana v Severnome Illirikume i ih arheologi~eskie sledù, Startum plus 5, 2010, 147–157. Jankovi} 1989 – M. Jankovi}, Arheolo{ki nalaz iz okoline Beograda, grob VI veka, Godi{wak grada Beograda 36, 1989, 5–15. Jovanovi} 1996 – A. Jovanovi}, The Problem of the Location of Lederata. In: Roman Limes on the Middle and Lower Danube, éd. P. Petrovi}, Belgrade 1996, 69–72. Jeremi} 2002 – M. Jeremi}, Graditeljstvo Sirmijuma u V i VI veku, Saop{tenja 34, 2002, 43–58. Jeremi} 2009 – G. Jeremi}, Saldum, Roman and Early Byzantine Fortification, Belgrade 2009. Jórgensen 1992 – L. Jórgensen, A.D. 568 – A Chronological Analysis of Lombard Graves in Italy. In: Chronological Studies of Anglo-Saxon England, Lombard Italy and Vendel Periode Sweden, ed. L. Jórgensen, Copenhagen 1992, 94–122. Kazanski 1980 – M. Kazanski, A propos de l’apparition de la coutume de la déformation crânienne artificielle chez les tribus germaniques de la Gaule. Bulletin de liaison de l’Association francaise d’Archéologie mérovingienne, 1980, 85–88. Kazanski 1999 – M. Kazanski, Les Slaves. Les origines, Ier–VIIe siècle après J.-C., Paris 1999. Kazanski 2009 – M. Kazanski, La cavalerie slave à l’époque de Justinien, Archaeologia Baltica 11, 2009, 229–239. Kazanski 2013 – M. Kazanski, Les Gépides et la Crimée, Starinar 63, 2013, 115–130. Kazanski 2013a – M. Kazanski, The Middle Dnieper area in the seventh century: an archaeological survey, Travaux et Mémoires 17, 2013, p. 769–864. Kiss 1979 – A. Kiss, Ein Versuch die Funde und das Siedlungsgebiet der Ostgotyrn in Pannonien zwischen 456–471 zu bestimmen, Acta Archaeologica Academiae Scientarum Hungaricae 31, 1979, 329–339. Kiss 1984 – A. Kiss, Über eine silbervergoldete gepidische Schnalle aus dem 5. Jahrhundert von Ungarn, Folia Archaeologica 35, 1984, 57–76. Kondi} 1984 – V. Kondi}, Les formes des fortifications protobyzantines dans la région des Portes de Fer. In : Villes et peuplement dans l’Illyricum protobyzantin, Rome 1984, 147–155. 157 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) PLRE III – The Prosopography of the Later Roman Empire III, éd. J. R. Martindale, Cambridge 1992. Pejovi}, Lu~i} 2011 – Z. Pejovi}, B. Lu~i}, Nekropola iz perioda seobe naroda sa lokaliteta 1a Sirmijuma, Zbornik Narodnog muzeja – serija: Arheologija 20.1, 2011, 389–413. Petkovi} 2006 – S. Petkovi}, Unilateral antler combs from Romuliana, Starinar 56, 2006, 363–366. Pop-Lazi} 2013 – S. Pop-Lazi}, Pregled rimske vojske u Srbiji tokom kasne antike, In: Konstantin Veliki i Milanski edikt 313, éd. I. Popovi}, Beograd 2013, 60–73. Popovi} 1975 – V. Popovi}, Les témoins archéologiques des invasions avaro-slaves dans l’Illyricum byzantin, Mélanges de l’École francaise de Rome. Antiquité 87, 1975, p. 445–504. Popovi} 1978 – V. Popovi}, La descente des Koutrigours, des Slaves et Avars vers la mer Égée: le témoignage de l’archéologie, Académie des Inscriptions et Belles-Lettres. Comptes rendus, 1978, p. 596–648. Popovi} 1980 – V. Popovi}, Aux origines de la slavisation des Balkans : la constitution des premières sklavinies macédoniennes vers la fin du VIe siècle, Académie des Inscriptions et Belles-Lettres. Comptes rendus, 1978, p. 230–257. Popovi} 1984 – V. Popovi}, Un étui de peigne en os de type »mérovingien« et les objets d’origine ethnique étrangère à Cari~in Grad, Cari~in Grad I, Belgrade– Rome 1984, 160–178. Popovi} 1987 – V. Popovi}, Die süddanubischen Provinzen in der Spätantike vom Ende des 4. bis zur Mitte des 5. Jahrhunderts. In: Die Völker Südeuropas im 6. bis 8. Jahrhunderts, Berlin–Munich 1987, 95–140. Popovi} 1988 – M. Popovi}, Svetiwa, novi podaci o ranovizantijskom Viminacijumu, Starinar 38, 1988, 1–35. Popovi} 2001 – I. Popovi}, Kasnoanti~ki i ranovizantijski nakit od zlata u Narodnom muzeju u Beogradu, Beograd 2001. Popovi} 2008 – I. Popovi}, Solidi with Filgereed Tubular Suspension Loops from Udovice in Serbia. Fornvännen, 103, p. 73–80. Prohászka 2007 – P. Prohászka, Ein »altes« ostgotisches Frauengrad aus Beograd–Zemun, Folia Archaeologica 3, 2007, 179–192. Quast 2012 – D. Quast, Einige alte und neue Waffenfundeaus dem frühbyzantinischen Reich. In: Thesaurus Avarorum, ed. T. Vida, Budapest 2012, 351–370. Sarantis 2011 – A. Sarantis, The Justinianic Herules: from allied barbarians to Roman provincials, In: STARINAR LXIV/2014 Neglected Barbarians, F. Curta ed., Brepols, Turnhout 2011, 361–402. Simoni 1978 – K. Simoni, Dva priloga istra`ivanju germanskih nalaza seobe naroda u Jugoslaviju, Vjesnik Arheolo{kog muzeja u Zagrebu 10–11, 1978, 209–233. Sjóvold 1993 – T. Sjóvold, The Scandinavian relief brooches of the Migration Period. An attempt at a new classification (Norske oldfunn 15), Oslo 1993. Sommer 1984 – M. Sommer, Die Gürtel und Gürtelbeschlage des 4. und 5. Jahrhunderts im römischen Reich, Bonn 1984. Soproni 1969 – S. Soproni, Limes Sarmatiae. Móra Ferenc Múzeum Ëvkönyve 2, 117–135 Tati}-\uri} 1958 – M. Tati}-\uri}, Gotski grob iz Ostru`nice, Zbornik radova Narodnog muzeja I, 1958, 161–185. Teall 1965 – J. Teall, The Barbarians in Justinian’s Armies, Speculum, 40, 1965, 294–322. Tejral 1988 – J. Tejral, Zur Chronologie der frühen Völkerwanderungszeit im mittleren Donauraum, Archaeologia Austriaca 72, 1988, 223–304. Tejral 1997 – J. Tejral, Neue Aspekte der frühvölkerwanderungszeitlichen Chronologie im Mitteldonauraum. In: Neue Beiträge zur Erforschnug der Spätantike im mittleren Donauraum, éd. J. Tejral, H. Friesinger, M. Kazanski, Brno 1997, 321–392. Tejral 2005 – J. Tejral, Zur Unterscheidung des vorlangobardischen und elbgermanisch-lanogbardischen Nachlasses. In: Die Langobarden. Herrschaft und Identität, Vienne 2005, 103–200. Tejral 2007 – J. Tejral, Das Hunnenreich und die Identitätsfragen der barbarischen »gentes« im Mitteldonauraum aus der Sicht de Archäologie. In: Barbaren im Wandel, ed. J. Tejral, Brno 2007, 55–119. Tejral 2011 – J. Tejral, Einhemische und Fremde. Das norddanubische Gebiet zur Zeit der Völkerwanderung, Brno 2011. Tejral et alii 2011 – J. Tejral, S. Stuchlík, M. ^i`már, Z. Klanica, S. Klanicová, Langobardische Gräberfelder in Mähren I, Brno 2011. Uenze 1992 – S. Uenze, Die spätantiken Befestigungen von Sadovec (Bulgarien), Munich 1992. Vaday 2001 – A. Vaday, Militia inermis, militia armata. Bemerkungen zur Frage des Limes Sarmatiae, Slovenská Archeológia 49, 249–276. Vasi} 1906 – M. Vasi}, Starosrpska nalazi{ta u Srbiji, VI Kostolac, Starinar I, 1906, 56–66. Vasi} 1981 – M. Vasi}, Ostava rimskog bronzanog novca IV i V veka iz Viminacijuma, Starinar XXXI, 1981, 123–161. 158 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) STARINAR LXIV/2014 Vinski 1964 – Z. Vinski, Okov Teodorikova vremena s ostrva Sapaja na Dunavu, Zbornik Narodnog muzeja IV, 157–178. Vinski 1978 – Z. Vinski, Archäologische Spuren ostgotischer Anwesenheit im heutigen Bereich Jugoslawiens. In: Problemi seoba naroda u Karpatskoj kotlini, Novi Sad 1978, 33–47. Werner 1962 – J. Werner, Die Langobarden in Pannonien, München 1962. Wolfram 1990 – H. Wolfram, Histoire des Goths, Paris 1990. Zotovi} 1994 – Lj. Zotovi}, Die Gepidische Nekropole bei Viminacium, Starinar 43–44, 1994, 183–190. Zuckerman 2004 – C. Zuckerman, L’armée. In: Le Monde byzantin I – L’empire romain d’Orient (330–641), éd. C. Morrisson, Paris 2004, 143–180. Vasi} 1990 – M. Vasi}, Nalazi rimskog bronzanog novca IV i V veka iz municipijuma Horreum Margi, Beograd 1990. Vasi} 2008 – M. Vasi}, Zlatni i srebrni novac kasne antike 284–450. godine iz zbirke Narodnog muzeja u Beogradu, Beograd 2008. Velkov 1985 – V. Velkov, Frühbyzantinische Inschriften aus Dacia Ripensis, Byzantina 13, 1985, 883–891 Vida, Völling 2000 – T. Vida, T. Völling, Das slawische Brandgräberfeld von Olympia, Rahden/Westf. 2000. Vierck 1981 – H. Vierck, Imitatio imperii und interpretatio Germanica vor der Wikingerzeit. In: Les pays du Nord et Byzance (Scandinavie et Byzance), Upsalla 1981, 64–113. Vinski 1957 – Z. Vinski, Arheolo{ki spomenici velike seobe naroda u Srijemu (Situla 2), Ljubljana 1957. 159 IVANI[EVI], KAZANSKI, Illyricum du Nord et les Barbares à l’époque des Grandes Migrations (131–160) Rezime: STARINAR LXIV/2014 VUJADIN IVANI[EVI], Arheolo{ki institut, Beograd MI[EL KAZANSKI, Nacionalni centar za nau~no istra`ivawe, Pariz SEVERNI ILIRIK I VARVARI U VREME SEOBE NARODA (4–6. VEK) Kqu~ne re~i. – Ilirik, Balkan, kasna antika, velike seobe, Germani, Huni, Sloveni. na grob vojnog zapovednika iz Batajnice, ali i na ostatke koliba i grobova iz Sirmijuma (sl. 14). Poznata nekropola iz Jakovo-Kormadina u Sremu se naj~e{}e pripisuje Gepidima (sl. 15–16). Uprkos zna~ajnim politi~kim promenama i obnovi carske vlasti u Severnom Iliriku po~etkom 6. veka, va`nu ulogu nastavqaju da igraju Varvari – jedni kao federati, a drugi kao neprijateqi Carstva. Justinijan I naseqava Herule u blizini Singidunuma, gde se oni zadr`avaju do 550. godine. Tom periodu pripadaju nekropole Viminacijum I – Burdeq i Viminacijum II – Vi{e Grobaqa – severozapadni sektor, koje se mogu pripisati federatima. Prilozi u grobovima odra`avaju uticaje materijalne kulture severnih i zapadnih Germana (sl. 17–18). Sli~an horizont je konstatovan i u Singidunumu, u ~ijoj okolini su bili naseqeni Heruli. U prilog prisustvu Herula govorio bi i niz predmeta u nekropolama jasne skandinavske tradicije. I poznati grob iz Ulpijane mo`e se delom vezati za severne i zapadne Germane (Heruli i Tirin`ani), dok bi se tubulusi iz Udovica mogli pripisati severnim Germanima (sl. 19–20). Situacija na prostoru Severnog Ilirika se iznova mewa kada se na dunavskoj granici pojavquju Sklaveni i Anti, a kasnije i Avari. Granicu i daqe ~uvaju federati – Germani – kako pokazuje analiza naseobine na Svetiwi pored samog Viminacijuma. Garnizon federata na Svetiwi, ~ije se prisustvo datuje oko 570–580. godine i kasnije, mo`e se dovesti u vezu sa Langobardima regrutovanim u velikom broju od strane cara Tiberija. Ovom horizontu pripadaju i nalazi iz Talijate, Transdijerne, Kamenova i dr., kao i grobovi iz nekropola u Viminacijumu II – severozapadni sektor, Viminacijumu III – Lanci i u Singidunumu III (sl. 21). Iz unutra{wosti su to arheolo{ki predmeti, mahom kerami~ke posude, otkriveni na Juhoru, Jelici i Cari~inom gradu. Sa ovog posledweg nalazi{ta posebno se izdvajaju nalazi dve „slovenske“ fibule. Jedna od fibula predstavqa tipi~an nalaz u ostavama tipa Martinovka, koje se vezuju za Ante, i kulturu Penkovka. Druga fibula, dunavskog tipa Kiskoros-Sarmizegetusa, javqa se tako|e i u kontekstu kulture Prag (sl. 22). Pad dunavskog limesa po~etkom 7. veka predstavqao je kraj vizantijske uprave nad Severnim Ilirikom. Re~ je o jednom dugom istorijskom procesu u kome su Varvari, u prvom redu Germani, a kasnije Sloveni i Avari, igrali va`nu ulogu u oblikovawu politi~kih i dru{tvenih odnosa u podunavskim provincijama Carstva. Autori daju pregled naseqavawa Varvara na prostoru Severnog Ilirika od kraja 4. do kraja 6. veka (sl. 1). Velike istorijske i dru{tvene promene nastupaju nakon poraza carske vojske kod Hadrijanopoqa 378. godine, ~ime je otvoren put za upade i naseqavawe u prvom redu Germana, a kasnije i drugih Varvara. Prvi talas podrazumevao je naseqavawe grani~nih provincija, {to je dovelo do povla~ewa autohtonog stanovni{tva dubqe u unutra{wost. Taj proces se na prostoru Severnog Ilirika jasno prati na osnovu brojnih sporadi~nih grobova otkrivenih du` dunavske granice na nekropolama u Viminacijumu, Vajugi i Pontesu (sl. 2–3). Ovde vredi pomenuti i brojne nalaze, u prvom redu fibule i ~e{qeve, koji pripadaju kulturi ^erwahov – Sintana de Mure{, ~iji su glavni nosioci bili Goti i wihovi saveznici. Osim isto~nih Germana, prisutni su i nosioci zapadnog rimsko-germanskog kulturnog kruga. Velike promene na prostoru Severnog Ilirika nastupaju s dolaskom Huna i wihovim prodorom duboko u unutra{wost balkanskih provincija. U tim naletima postradali su gradovi na dunavskoj granici (Singidunum, Margum, Viminacijum), kao i oni u unutra{wosti (Naisus i dr.). O ovom naletu svedo~e ostave novca, posebno solida Teodosija II, i drugi retki nalazi, kao {to je zlatni torkves iz Markovca (sl. 4). Hunskom periodu pripadaju i grobovi iz nekropola Singidunum IV i iz Vrawa kod Hrtkovaca (sl. 5–6). Smrt Atile 453. godine i poraz Huna 454/455. godine na reci Nedao dove{}e do sloma Hunskog saveza i otvori}e {irom vrata naseqavawu Varvara na podru~je Severnog Ilirika, posebno podunavskih provincija i gradova, poput Singidunuma i Viminacijuma. O ovom talasu naseqavawa svedo~e brojni grobovi u nekropolama Singidunum I–III i Viminacijum I i II – jugoisto~ni sektor (sl. 9–12). Materijalna kultura se najve}im delom mo`e pripisati isto~nim Germanima (sl. 7–8). Postojawe velikih varvarskih nekropola na podru~ju dva glavna grani~na grada – Singidunuma i Viminacijuma – jasno ilustruje velike demografske promene koje se javqaju kao posledica novih dru{tvenih i politi~kih prilika. Nove promene nastaju 488. godine kada Ostrogoti, sa Teodorikom na ~elu, odlaze u Italiju, koju osvajaju da bi, naknadno, svoje granice pro{irili sve do Sirmijuma. Ovom horizontu pripada nekoliko nalaza sa tog podru~ja, iz Ma~vanske Mitrovice, Rovina i samog Sirmijuma, kao i iz Translederate (sl. 13). Nove prostore osvajaju i Gepidi, ~iji se tragovi mogu pratiti u arheolo{kom materijalu. To se posebno odnosi 160 MIROSLAV VUJOVI] Odeqewe za arheologiju, Filozofski fakultet, Beograd RANOVIZANTIJSKI KANTAR IZ BEOGRADA e-mail: [email protected] UDK: 904:681.26(497.11)"04/05" ; 069.51:902/904(497.11) DOI: 10.2298/STA1464161V Primqeno: 17. februar 2014. Prihva}eno: 25. jul 2014. Originalni nau~ni ~lanak Apstrakt. – Rad je posve}en kantaru od bakarne legure na|enom u Beogradu 1929. godine u okviru slu~ajnog nalaza grupe predmeta, koji }e se, sticajem okolnosti, na}i u zbirkama Narodnog muzeja i Muzeja grada Beograda. Delovi kantara, koji su ~inili poluga sa dvostrukom mernom skalom i punktiranim natpisom, lan~ani sistem sa kukama i teg u obliku biste vizantijske carice, bivaju razdvojeni, a podaci o wihovom zajedni~kom poreklu zanemareni. Poluga kantara je prvobitno zavedena u okviru Tursko-austrijske zbirke Muzeja grada Beograda, da bi kasnije ispravno bila opredeqena u ranovizantijski period. Teg je zasebno publikovan i datovan u kraj V veka, dok se lanac sa kukama prvi put publikuje ovom prilikom. Autor iznosi osnovne stilsko-hronolo{ke, epigrafske i metri~ke osobenosti ovog nalaza. Posebna pa`wa posve}ena je ~itawu natpisa koji do sada nije bio detaqnije tuma~en. Kantar je datovan u posledwe decenije V i prvu polovinu VI veka. Kqu~ne re~i. – Beograd, Singidunum, kantar, teg, epigrafika, metrologija, ranovizantijski period, merni instrumenti, ikonom. P rvobitni ciq ovog rada bio je detaqno publikovawe, nau~nim krugovima nedovoqno poznate, poluge ranovizantijskog kantara koji se ~uva u Muzeju grada Beograda, sa posebnim osvrtom na wegove metri~ke odlike i tuma~ewe natpisa punktiranog na kra}em kraku. Me|utim, nakon uvida u muzejsku dokumentaciju Muzeja grada Beograda i Narodnog muzeja u Beogradu, pri~a o kantaru iz Singidunuma dobila je potpuno nov tok. Kako }e se ispostaviti, potvr|ena je pretpostavka da poluga kantara iz Muzeja grada Beograda predstavqa deo istog skupnog nalaza sa dobro poznatim tegom u obliku poprsja vizantijske carice i lan~anim sistemom sa kukama za ka~ewe te- reta koji se ~uvaju u Narodnom muzeju u Beogradu. Detaqnijim pregledom muzejskih inventarnih kwiga, u najve}oj meri je rasvetqen neobi~an splet okolnosti koji je, pre vi{e od osamdeset godina, doveo do razdvajawa sastavnih delova beogradskog kantara. *** Kantari, kao i druga sredstva za merewe ranovizantijske epohe poput terazija, merica i raznorodnih tegova, predstavqaju ne tako ~este nalaze na tlu Srbije. Posebno su retki i zna~ajni oni koji poti~u sa sistematski istra`enih arheolo{kih * Ovaj rad je nastao anga`ovawem autora 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. 161 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 Sl. 1. [ematski prikaz ranovizantijskog kantara tipa Konstantinopoq (precrtano iz: N. Franken 1993) Fig. 1. Schematic representation of the Early Byzantine steelyard type Constantinople (redrawn from: N. Franken 1993) lokaliteta.1 Izvestan broj tih predmeta dospeo je u muzejske zbirke putem otkupa, pa o uslovima i mestu wihovog nalaza naj~e{}e nema odgovaraju}ih podataka. Wihova distribucija i wihove stilske i metri~ke osobenosti su izuzetno va`ni pokazateqi istorijskih, dru{tvenih i ekonomskih odnosa. Kantar (statera2, stat»ra, campana, kampanÒj3) jeste vrsta vage, sa nejednakim kracima, koja funkcioni{e po zakonu poluge (ravnote`e).4 Kantari su bili slo`enije konstrukcije ali univerzalnije namene od terazija ili jednakokrakih vaga (libra, trutina, zugÒj) i kori{}eni su za merewe ve}ih te`ina, odnosno kabastijeg tereta. Poznati su jo{ i kao „brze vage”, zbog ~ega su posebno bili omiqeni me|u uli~nim prodavcima, u bakalnicama i na pijacama, gde se mogu i danas videti. Wihovu funkciju i mehani~ka svojstva detaqno je izlo`io Vitruvije (Vitr. 10. 3. 4), pomiwu}i ih kao staterae, kao i Isidor Seviqski (Isid. orig. 16, 25, 6), koji ovu vrstu vage navodi pod terminom campana.5 Kantari su naj~e{}e izra|ivani od bakarne legure (bronza, mesing), mada su u upotrebi bili i primerci od gvo`|a, drveta ili kombinacije metala i drveta. Svaki kantar sastoji se od poluge sa du`im krakom (scapus), na kome je merna skala, i kra}im krakom, 1 Tati}-\uri} 1962 (Beograd); Lalovi} 1983, 169, kat. br. 348 (Gamzigrad); @ivi} 2003, Cat. 328–344 (Gamzigrad); Simi} 1987, 16–18, sl. 2 (Rogatac); Bavant 1990, 245, Pl. XLIV/ 299, 300, 301 (Cari~in grad); Popovi} 1999, 116. cat. 160, fig. 67 (Ras); Ivani{evi}, [pehar 2005 141–142, Fig. 4/1, Fig. 5 (^e~an, Streoc); Mitrovi} 2010, 43–44, 55, 68, kat. 46, 47, 103, 111, 123–130, 158 (okolina Vrawa); Milinkovi} 2010, sl. 39–41, T. XIV/5, 60–61; 229–131 (Jelica). Za nalaze iz ranovizantijskih utvr|ewa sa prostora \erdapa: Jovanovi} 1984, 328 (Hajdu~ka Vodenica); Jeremi} 2009, 189–190, Fig. 89/ cat. 626–628 (Saldum), [pehar 2010, T X, 78–81. 2 Daremberg, Saglio 1877, 1225 (statera). 3 ODB I, 1947 (kampanÒj). 4 Kada je kantar uravnote`en, proizvod mase merenog tereta i du`inske vrednosti izme|u fulkruma i ta~ke oslonca tereta na kra}em kraku poluge jednaka je proizvodu mase kontratega i du`inske vrednosti izme|u fulkruma i pozicije kontratega na mernoj skali du`eg kraka. 5 Wei{ 2001, 200. 162 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) prsja od bakarne legure, koje je naknadno ispuwavano olovom kako bi se postigla odgovaraju}a masa saglasna fizi~kim karakteristikama poluge odre|enog kantara. Ta poprsja nekad prikazuju obi~ne smrtnike, negroide ili predstavnike nekih egzoti~nih zanimawa, poput atleta, i imaju uglavnom dekorativni karakter. U rimskom periodu su znatno ~e{}i bili tegovi u vidu likova anti~kih bo`anstava, a posebno onih pod ~ijom jurisdikcijom su se nalazili trgovina i merewe (Atena, Minerva). Budu}i da je od ispravnosti ovog tega zavisila ta~nost merewa i po{tena trgovina, prikaz odgovaraju}eg bo`anstva trebalo je da potvr|uje ili obezbedi verodostojnost merewa. Tokom kasnoanti~ke i ranovizantijske epohe ustali}e se upotreba tegova u formi poprsja careva ili carica, ~iji su likovi, poput onih na novcu, ozna~avali i potvr|ivali oficijelnost mere.12 Osim toga {to su bile garancija vrednosti, ove predstave su ujedno podse}ale na kazne za kr{ewe pravila i ne~asnu upotrebu vladarskog lika kao zakonodavca. U Teodosijevom kodeksu je nekoliko zakona posve}eno upravo pravnom regulisawu kori{}ewa odgovaraju}ih mera i tegova (mensurae et pondera), odnosno zloupotreba tegova mase ve}e od predvi|ene me|u poreskim slu`benicima.13 U Kwizi Prefekata, istina znatno kasnijoj (IX vek), navedeni su carigradski piqari (saldamaroi) koji su posedovali tegove lak{e od zvani~nih mera.14 Upotreba kantara bila je propisana za merewe upravo u piqarnicama koje su prodavale doma}u, naj~e{}e prehrambenu robu za potrebe lokalnog `ivqa.15 U oba slu~aja su vlasnici tih neispravnih mera zgrtali na kome je nazna~ena najmawe jedna fiksna ta~ka oslonca kantara (fulcrum) u vidu kuke (ansa) o koju je kantar ve{an. Kod rimskih i vizantijskih kantara (Sl. 1) postoje obi~no dva fulkruma – jedan na gorwoj a drugi na dowoj strani kra}eg kraka poluge, kao i dve naspramno postavqene merne skale ozna~ene na skapusu.6 Takav kantar je imao „debqu“ i „tawu“ meru, koje su omogu}avale merewe tereta mawih ili ve}ih te`ina u zavisnosti od trenutne potrebe. Vrstu mere odre|ivala je udaqenost fulkruma od centra ravnote`e, te bi fulkrum bli`i kraju kra}eg kraka bio za te`e terete a onaj bli`i centru ravnote`e za lak{e. Postoje, me|utim, i kantari koji, osim „debqe“ i „tawe“ mere, imaju jo{ jednu – „sredwu“ meru, kojoj je odgovarala posebna merna skala na skapusu.7 Pored poluge kantar je imao i lan~ani sistem sa kukama ili tasom za ka~ewe tereta. Ukoliko je kantar imao tas, obi~no su postojala tri lanca za ka~ewe. U zavisnosti od robe koja je merena, sistem za suspenziju tereta je mogao imati tri lanca, koja su osloncima postavqenim pod 120º omogu}avali stabilnost tasa, ili dva lanca sa kukama za ve{awe kabastog tereta poput ve}ih komada mesa, robe u vre}ama ili balama.8 Na lan~anim sistemima sa kukama mogla je na kantarima biti merena i roba transportovana u amforama.9 Pojedini primerci kantara mogli su imati i kombinovani sistem sa tasom i kukom.10 Lanci su bili naj~e{}e spojeni sa lu~nom pre|icom i s potkovi~asto oblikovanim nosa~em koji se oslawao na posebno predvi|eno le`i{te na kraju kra}eg kraka kantara. Tragovi pohabanosti, usled stalnog kori{}ewa i trewa metalnih delova kantara, jasno ukazuju da se to le`i{te, odnosno ta~ka oslonca merenog tereta, nalazilo u `lebu urezanom ispred ukrasnog dugmeta na kraju kra}eg kraka i fulkruma.11 Veoma zna~ajan deo kantara je pomi~ni kontrateg (aequipondium). On je ka~en o skapus pomo}u kuke sa naro~ito oblikovanim, spqo{tenim lukom koji omogu}ava lak{e klizawe tega i wegovo precizno pozicionirawe du` merne skale, a samim tim i ta~no odre|ivawe te`inske vrednosti merene robe. Tegovi kantara su naj~e{}e izra|ivani od olova, nekad prevu~enog tawim limom od bakarne legure. Neki od tih tegova su bili liveni i mogli su da imaju jednostavniji ili slo`eniji oblik. Najve}i broj je koni~nog, bikoni~nog, loptastog ili ovoidnog oblika. Ima i onih koji su figuralno modelovani u vidu {upqeg antropomorfnog po- 6 Pitarakis 2012, 407–410. Za tehni~ke karakteristike i klasifikaciju kantara vidi: Jenemann 1989; Franken 1993; Jenemann 1995. 8 Hill 1955, 51–55. 9 Sams 1982, 212–219; Van Alen 1996, 205. 10 Ross 1962, 63–64, T XLIII. 11 Ta~na upotreba ovog dela kantara ~esto nije odgovaraju}e nazna~ena na ilustracijama publikovanih primeraka. Tako ovaj lanac za ka~ewe tereta ~esto „zaluta“ i biva oka~en na skapus ili kuku fulkruma (Anti~ka bronza u Jugoslaviji 1969, sl. 338), {to nikako ne odgovara wegovoj funkcionalnoj nameni. 12 Franken 1994; Pitarakis 2012, Fig. 16.15. 13 Ch.T. XI. 8.3 (409). 14 TÕ ^parcikÕn bibl/on, 13, 5. 15 U ovim trgovinama su se, za razliku od prodavnica uvozne robe, poput za~ina, sapuna, voska i mirisa, merene na finijim jednokrakim vagama, mogli kupiti usoqeno ili su{eno meso 7 163 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 ri~kim oznakama te`ine (T. II). Wihovo ba`darewe izvedeno je u skladu sa anti~kim duodecimalnim sistemom mera koji je kori{}en u rimskom i vizantijskom periodu. To se dobro vidi na „tawoj“ odnosno lak{oj skali skapusa, gde su izdvojeni podeoci gravirani na poluskupine od {est kra}ih ureza, dok je svaki {esti podelak du`i. Po~etni i svaki dvanaesti podelak obele`eni su punktiranom oznakom za litru, koja podse}a na obrnuto slovo Y.23 Merna skala sa desne, „tawe“ strane (T. II/1) po~iwe slovom A (broj 1) i zavr{ava se sa E, koje bi trebalo da ozna~ava broj 5, ali ovde je zapravo obele`en broj 15 jer mu na skali prethodi broj 10 obele`en gr~kim slovom I. Visina slova (numeri~kih oznaka) iznosi oko 1 cm. Slovo A je izvedeno sa prelomqenom popre~nom crtom a E s jednakim hori- nezakonitu zaradu koja je mogla biti zakonski sankcionisana.16 Sve do IV veka, nadzor nad standardnim merama i tegovima obavqali su odgovaraju}i kuralni zvani~nici (aediles, agronomoi) koji su bili postavqeni u svakom gradu. Zbog u~estalog krivotvorewa novca i drugih malverzacija, dr`ava }e ve} od vremena cara Julijana (363. godine) po~eti da postavqa posebne gradske slu`benike – mera~e (zygostates), pod ~iju nadle`nost su spadali sporovi kupaca i trgovaca.17 POLUGA KANTARA IZ MUZEJA GRADA BEOGRADA18 Poluga kantara (Sl. 2–5; T. I/1) ~uva se u zbirci Sredwovekovnog odeqewa Muzeja grada Beograda zajedno sa jednim mawim tegom (Sl. 6; T. I/2).19 U inventaru ove zbirke, oba predmeta nose isti inventarski broj i zavedena su kao slu~ajni nalaz iz Beograda, bez poznatog mesta i uslova nalaza. Prema navodu iz Kwige inventara Tursko-austrijske zbirke u kojoj su se prvobitno nalazili, poluga kantara i teg otkupqeni su od Milosava M. Stojadinovi}a iz Beograda, {to je, kako }e se pokazati, samo delimi~no ta~no.20 Poluga kantara je izra|ena od mesinga tehnikom livewa, a detaqi su izvedeni urezivawem i punktirawem.21 Dimenzije kantara su: du`ina 61 cm i {irina osovine 2 cm. O~uvana masa kantara danas iznosi 950 g, ali je nekada bila ve}a.22 Kra}i krak kantara je pravougaonog preseka i zavr{ava se bikoni~nim dugmetom. Du`i krak je romboidnog preseka (a = 1,3 cm) i ima dve kalibrisane i naspramno postavqene skale za vagawe lak{ih odnosno te`ih mera. Zavr{en je dugmetom u obliku ~etvorostrane piramide. Sude}i prema kompletno o~uvanim primercima ranovizantijskih kantara, poluzi kantara iz Muzeja grada Beograda nedostaje jedna kuka fulkruma. U posebnom le`i{tu na kra}em kraku poluge o~uvana je samo fragmentovana karika za kuku predvi|enu za ka~ewe kantara pri vagawu u te`em mernom re`imu. Kantaru nedostaje jo{ i lan~ani sistem za ve{awe merenog tereta, ali mu, kako }emo ne{to kasnije pokazati, nedostaje i odgovaraju}i kontrateg. i riba, maslac, sir, uqe, med, povr}e, mahunarke, konopqa, lan, razli~ite posude i gvo`|arska roba: Dagron 2002, 461; TÕ ^parcikÕn bibl/on, 10.1, 5–6; 11.8; 13.1. 16 Hendy 1994, 331–332. 17 Jones 1940, 216, 255, 317–318. 18 Jankovi} 1997, 312, 333, kat. br. 577. 19 Poluga kantara i teg se od 1965. godine ~uvaju u okviru Tursko-austrijske zbirke Muzeja grada Beograda pod istim inventarskim brojem (TAZ 935). 20 Milosav M. Stojadinovi} bio je potpredsednik beogradske op{tine u godinama pred izbijawe Drugog svetskog rata, a privremeno i gradona~elnik okupiranog Beograda tokom 1941. godine. Kao qubiteq starina, kolekcionar i zvani~nik gradske uprave uticao je na razvijawe rada i na status Muzeja grada. U dokumentaciji Muzeja se pomiwe kao donator, odnosno, u posleratnim godinama i kao prodavac antikviteta i umetnina. 21 Fizi~ko-hemijske analize sastava legure izvr{ene su metodom atomske apsorpcione spektrofotometrije na instrumentu AAS Perkin Elmer 1100 u Vojno-tehni~kom zavodu u Beogradu, u laboratoriji Sektora za materijale. Ove analize su pokazale da je poluga kantara izra|ena od bakarne legure koja po sastavu odgovara mesingu (Cu 80%, Zn 20%). 22 Budu}i da jedna kuka fulkruma nedostaje (kao i polovina jedne karike sa kojom je ona bila povezana), dana{wa masa kantara ne odgovara onoj prvobitnoj. U `eqi da rekonstrui{emo originalnu masu, izmerili smo pojedina~no postoje}u kuku kao i kariku za koju je ova bila pri~vr{}ena. Uz pretpostavku da su mase postoje}ih i nedostaju}ih delova kantara bile identi~ne, o~uvanoj masi kantara (950 g) dodali smo vrednosti masa kuke (75 g) i polovine karike (2,5 g). Wihov zbir od 1027,5 g bi najpribli`nije odgovarao prvobitnoj masi kantara bez pokretnog tega i lan~anog sistema za ve{awe tereta. Dobijena te`inska vrednost pribli`no se podudara sa 3 ranovizantijske litre (logarik» l/tra), koja je iznosila dana{wih 322 g. Za vrednost litre videti: Schilbach 1970, 166. 23 Razdaqina izme|u dve oznake za litru na skali iznosi 26 mm. Merna skala Poluga kantara ima dve merne skale koje su razmerene urezanim podeocima i punktiranim nume- 164 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 Sl. 2 i 3. Poluga ranovizantijskog kantara iz Beograda (lak{a i te`a strana). Muzej grada Beograda (foto: N. Bori}) Fig. 2 and 3. The beam of the Early Byzantine steelyard from Belgrade (lighter and heavier weight range side), Belgrade City Museum (Photo: N. Bori}) zontalama. Numeri~ka vrednost iskazana slovima odnosi se na litre (logarik» l/tra), vizantijski pandan anti~koj rimskoj libri, koje su pomenutim punktiranim znacima obele`ene du` merne skale. Vertikalnim podeocima du` skale obele`ene su i ungije (oÙgg/a), kojih je bilo 12 u litri, {to se jasno vidi i na na{em razmerniku. Sa suprotne, odnosno „debqe“ strane (T. II/2), merna skala ne zapo~iwe posledwim brojem na „tawoj“ strani – slovom E, ve} oznakom IE za broj 15, a potom sledi skala obele`ena du`im podeocima za svaku litru i kra}im podeocima za polovinu wene vrednosti.24 U nastavku se du` ove skale ~itaju slede}a slova u inverznom polo`aju: K, E, V, E, zatim slede slova postavqena pravilno: M, E i, na kraju, N. Numeri~ke oznake upu}uju na to da su na ovoj „debqoj“ mernoj skali merene vrednosti od 15 (IE) do 50 (N) litri. Sude}i prema upisanim numeri~kim vrednostima, ranovizantijski kantar ~ija poluga se danas ~uva u Muzeju grada Beograda mogao je da meri do 50 litri tereta, odnosno, u dana{wim merama, ne{to vi{e od 16 kg. 24 Udaqenost izme|u ve}ih podelaka kojima su obele`ene litre iznosi 12 mm. 165 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) Sl. 4 i 5. Ranovizantijski kantar iz Beograda, detaq kra}eg kraka sa natpisom (lak{a i te`a strana). Muzej grada Beograda (foto: N. Bori}) Fig. 4 and 5. Inscription on the shorter arm of the Early Byzantine steelyard from Belgrade (lighter and heavier weight range side). Belgrade City Museum (Photo: N. Bori}) vinom i, na kraju, sasvim jasno slovo S. Ovaj natpis pro~itali smo kao: +KURSFIL(IKO)S+ Natpis Sa obe strane kra}eg dela poluge kantara iz Muzeja grada Beograda, nalazi se punktirani tekst sastavqen mahom od slova gr~ke kapitale. Tekst oba natpisa po~iwe i zavr{ava se prikazom krsta. 1. Na lak{oj ili „tawoj“ strani kra}eg dela poluge kantara nalazi se natpis (Sl. 4; T. III/1) izveden punktirawem slova gr~kog alfabeta bez jasnog odvajawa zasebnih re~i. Sastoji se od petnaest slovnih mesta i {esnaest slova (visina slova: 0,8–1,2 cm), od kojih su posledwa dva prikazana u ligaturi i malo odvojena od ostalog dela teksta. Slova su sa nagla{enim serifima. Sigma je napisana u alfabetnom minuskulnom maniru ili u vidu majuskulnog latini~nog slova S. Natpis smo pro~itali na slede}i na~in: +PAPAKESOIKONOMOU+ 2. Sa „te`e“ strane nalazi se natpis (Sl. 5; T. III/2) od ukupno jedanaest slovnih mesta. Paleografske osobine ovog natpisa i dimenzije slova odgovaraju natpisu sa suprotne strane (visina slova: 0,9–1,3 cm). Sredwi deo teksta je ne{to vi{e o{te}en korozijom i slabije ~itqiv. Prva ~etiri slova su potpuno jasna i radi se o: KYPS. Nakon toga slede malo o{te}eno ali ipak prepoznatqivo slovo F, pa zatim boqe o~uvana slova I, L i I. Od narednog slova su o~uvani vertikalna crta i kosi, serifno zavr{eni krakovi – dowi i gorwi krak, koji izvesno ukazuju na slovo K. U nastavku je punktirano slovo O, sa slabije o~uvanom desnom polo- Tuma~ewe natpisa 1. U po~etnom delu natpisa +PAPAKESOIKONOMOU+ moglo je biti navedeno crkveno, odnosno sve{teni~ko zvawe p£pa(s), koje se u zna~ewu „otac“ generalno odnosi na sve{tenike.25 Ovo zvawe sre}emo na jo{ jednom suvremenom kantaru iz Metropoliten muzeja u Wujorku.26 Mora se, me|utim, imati u vidu da se istim nazivom jo{ od III veka ozna~avaju i episkopi pojedinih isto~nih crkvenih centara poput Aleksandrije.27 Zatim u nastavku sledi KES i na kraju OIKONOMOY, sa posledwa dva slova u uobi~ajenoj ligaturi. KES bi se moglo pro~itati kao tri po~etna slova li~nog imena poput Kšj(arios), ali i kao ve- 25 ODB 3, 1578 (papas, p£pas). Na ovom kantaru se na obe strane kra}eg kraka poluge javqaju imena navedena u genitivu: +PAPALIOU, PATRIKIOU: Annual Report of the Trustees of the Metropolitan Museum of Art 139 (July 1, 2008 – June 30, 2009), 28 (no. 2008.355a-d); http://metmuseum. org/collections/search-the-collections/478383?img=5 (22.01.2014). 27 Ovim zvawem nagla{avana je duhovna veza sve{tenika i pastve. Znatno kasnije, termin na zapadu po~iwe da se koristi i za rimskog episkopa. Naime, tek od XI veka se u hijerarhiji zapadne crkve ova titula iskqu~ivo odnosi na rimskog episkopa i Svetog oca, odnosno Papu: ODB 3, 1578. 26 166 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) znik „i“, odnosno ka/ ili ke, kako se tako|e nekad navodi. Ovaj veznik se uobi~ajeno javqa u suvremenim natpisima u skra}ewu kao K sa malom repatom sigmom u nastavku dowe kose crte – Ks. Mo`e se lako pretpostaviti mogu}nost da je zanatlija koji je iskucavao natpis na kantar prema datom predlo{ku, ne nu`no i pismen, napravio gre{ku kombinuju}i svakodnevni izgovor sa ligaturom. U posledwem delu ovog natpisa (OIKONOMOU) navedena je u genitivu poznata funkcija ikonoma (o„konÒmoj). Ovim terminom ozna~avana je osoba, sve{teno lice, monah ili laik, koja je upravqala svojinom, prihodima i rashodima crkvenog sedi{ta, manastirskog imawa ili zadu`bine.28 Broj ikonoma zavisio je od zna~aja crkvenog ili monasti~kog centra ali i od veli~ine poseda. Za vlade Justinijana I, na primer, Vaseqensku patrijar{iju Konstantinopoqa opslu`ivalo je ~ak 9 ikonoma razli~itih statusa i zadu`ewa.29 Ikonom manastira je bio stariji monah zadu`en za upravqawe manastirskom imovinom, a posebno poqoprivrednim imawima, kao i za odr`avawe manastirskih zgrada. On je obi~no bio visoko rangiran u monasti~koj hijerarhiji, odmah iza igumana (hegoumenos), koga je naj~e{}e i nasle|ivao. Crkveni ikonomi se pomiwu i na dva mesta u Teodosijevom kodeksu. U ranijoj naredbi careva Arkadija i Honorija, upu}enoj prefektu pretorija Evtihijanu 27. jula 398. godine, ure|uje se problem begunaca koji od zakona na|u uto~i{te u crkvama ili manastirima. Ovde se ikonomi ~ak jasno defini{u kao oni koji obi~no upravqaju crkvenim ra~unima.30 Naredba odre|uje da takvi begunci ne samo moraju biti predati vlastima ve} i da ikonomi moraju podmiriti, o ra~unu crkve koja zlo~ince skriva, svaki javni ili privatni dug tih osoba. Potowa naredba, potpisana 15. decembra 434. godine od strane careva Teodosija i Valentinijana,31 upu}ena je pretorijanskom prefektu Taurusu i odnosi se na pravo crkve ili manastira da nasledi imovinu sve{tenika ili monaha koji premine bez naslednika ili testamenta. Ikonomima, drugim re~ima crkvi ili manastiru, to nasle|e se ne sme odricati. Dve odluke Halkedonskog sabora iz 451. godine odre|uju da episkopi ne smeju samostalno da upravqaju poslovima svog sedi{ta (26. kanon), ve} moraju da imenuju ikonoma iz redova wemu podre|enog sve{tenstva. Ovo je svakako bilo stoga da se crkvena imovina ne bi nekontrolisano rasipala, ali i zato da se crkva i weni velikodostojnici (episkopi) ne bi kompromitovali u slu~aju eventualnih pronevera ili sporova. Ista odredba potvr|ena je i u odlukama sedmog Vaseqenskog sabora u Nikeji 787. godine (kanon 11), s tom razlikom {to je obaveza naimenovawa ikonoma preneta i na manastire. Tako|e, kanonom 25 Halkedonskog sabora regulisano je da u slu~aju smrti episkopa ikonom nastavqa da raspola`e crkvenom svojinom i vodi brigu o woj sve dok se episkopija nalazi u „udovi{tvu“, odnosno sve do izbora novog episkopa.32 Na izuzetan zna~aj koji su ikonomi imali ukazuje i odstupawe od tog pravila onda kada su bili neophodni stru~nost i iskustvo. U Konstantinopoqu su tako, tokom perioda V–VI veka, ikonomi kao najvi{i slu`benici u crkvenoj organizaciji birani i iz redova mirjana.33 2. Natpis +KURSFIL(IKO)S+ punktiran na „te`oj“ strani ukazuje na li~no ime Filika (FilikÒj), osobe svakako vi{eg ranga (kÚrioj) koja je overila ta~nost kantara ili kojoj je on pripadao.34 Za li~na imena koja su ~esto upisana na kra}im krakovima poluga kantara smatra se da predstavqaju imena vlasnika ili zvani~nika koji su proveravali i potvr|ivali ta~nost mere.35 Prema drugom mi{qewu, mogu}e je da ti natpisi upu}uju na manastire i crkve u kojima su kantari i tegovi ~uvani kao merni instrumenti i slu`ili za provere te`inskih vrednosti.36 U slu~aju beogradskog nalaza ~ini nam se najizvesnijom mogu}nost da je Filik ime sve{tenika kome je, kao ikonomu (episkopske?) crkve ili manastira, poveren kantar. Natpis bi u ovoj varijanti glasio: +kÚrioj FilikÒj+ +papa(s) kes(sic) oikonmou+ 28 ODB 3, 1517 (oikonomos). Funkcija glavnog ikonoma (enoikion skrinion) prera{}e od IX veka u polo`aj jednog patrijar{ijskog ikonoma. Od X veka, naimenovawe tog glavnog ikonoma bi}e pod carskom kontrolom, a u isto vreme javqa se i zvawe megas oikonomos kao najvi{e zvawe za ~inovnika u slu`bi vaseqenskog patrijarha. Ikonomi su tako|e bili pridru`eni velikim javnim crkvama u Konstantinopoqu, a javqaju se i u kontekstu carskih zadu`bina i manastirskih poseda: ODB 3, 1517 (oikonomos). 30 CTh 9.45. 3. 31 CTh 5.3. 1. 32 Pililh 1985, 54. 33 Pililh 1985, 53. 34 Pape 1911, 746. 35 Ross 1962, 62; Franken 1993, 106. 36 Ridder 1905, no. 335, 651, 652. 29 167 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) ten muzeja u Wujorku, na kantaru iz Britanskog muzeja nalazi se natpis +KURIAKOU.37 U muzeju u Prilepu nalazi se kantar sa lokaliteta Gradi{te, iz okoline Debre{ta, potpisan imenom Anastazija (ANASTAZIS) koje je dato u nominativu.38 Kompletno o~uvan kantar iz kolekcije Damberton Ouks (Dumberton Oaks), na|en navodno u Konstantinopoqu, datovan je u V vek i signiran imenom Pantalaonta (+PANDWLEWNTWS+).39 Drugi kantar iz iste zbirke datovan je u V–VI vek i signiran je imenom Edesija u genitivu (+HDESIOU+).40 KONTRATEG KANTARA IZ MUZEJA GRADA BEOGRADA Mawi teg (Sl. 6; T. I/2), navodno otkupqen zajedno sa navedenom polugom kantara sa kojom deli isti inventarski broj, izra|en je od bakarne legure livewem u dvodelnom kalupu. Loptastog je tela, sa su`enim i zaravwenim dowim delom (visina: 6,8 cm, {irina: 3,4 cm, masa: 120,15 g). Na vrhu tega je kru`na u{ica kroz koju je provu~ena kuka za ve{awe. Kuka je izra|ena od debqe bronzane `ice koja je u gorwem delu raskovana. Budu}i da unutra{wi pre~nik kuke tega (R 1,2 cm) ne odgovara debqini kalibrisanog du`eg kraka poluge kantara (2 cm), on, iako zaveden u inventarske kwige kao deo istog kompleta, nije mogao da se koristi na mernoj skali tog instrumenta.41 Karakteristi~no raskovana kuka za ka~ewe tega odgovara uobi~ajenom izgledu kuka sa tegova kantara i upu}uje nas da je ovaj teg nekada zaista slu`io kao aequipondium nekog drugog, svakako maweg, mernog instrumenta. Sl. 6. Teg kantara iz Beograda, nepoznato nalazi{te (Dubrova~ka ulica?). Muzej grada Beograda (foto: N. Bori}) Fig. 6. Small steelyard counterweight from Belgrade, unknown site (Dubrova~ka Street?). Belgrade City Museum (photo: N. Bori}) U prevodu: (kantar) gospodina Filika, sve{tenika i ikonoma Mogu}e je, me|utim, i da je Filik ime zvani~nika (komes, prefekt, vir) koji je overio ta~nost kantara, a da je na drugoj strani naveden ili potpisan sve{tenik i ikonom koji je, shodno zakonima o ~uvawu mera, mo`da pripadao episkopskoj crkvi: +kÚrioj FilikÒj+; +papa(s) Kšs(arios) oikonmou U ovom slu~aju bi se natpis sa beogradskog kantara mogao prevesti i na slede}i na~in: (overio) gospodin Filik; (kantar) ikonoma sve{tenika (episkopa?) Kesarija Navo|ewe li~nog imena u genitivu ili nominativu sre}e se relativno ~esto na kantarima ranovizantijske epohe. Osim natpisa +PAPALIOU, PATRIKIOU sa pomenutog kantara iz Metropoli- 37 Entwistle 2008a, 743. Kantar (inv. br. 564, du`. 46 cm) poti~e sa lokaliteta Gradi{te, kod sela Debre{te, i bio je namewen za tri mere: Anti~ka bronza u Jugoslaviji, kat. br. 338; Kosti} 1993, 68, T. I/1; Kosti} 2003, 207–208, Sl. 8. 39 Ross 1962, 61, cat. no. 71, T. XLIV, XL. Wegova du`ina iznosi 60 cm, pa je po veli~ini najsli~niji beogradskom. Tre}i kantar iz Dumberton Oaks kolekcije, koji Ros datuje u V–VI vek, ima o~uvane lance sa tasom ali nije signiran: Ross 1962, 63–64, cat. no. 74, T XLIII. 40 Ross 1962, 63, T. XLII/73. 41 Postoji mogu}nost, prema na{em mi{qewu malo verovatna, da je teg mogao biti kori{}en sa ovim kantarom pri merewu tereta ve}e te`ine od one nazna~ene na skalama, odnosno da je, prema potrebi, ka~en na lan~ani sistem sa tasom ili na kuke za ka~ewe tereta na suprotnoj strani kantara. 38 168 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) Kako i kada je do{lo do spajawa ova dva dela razli~itih kantara delimi~no mo`e da rasvetli uvid u muzejsku dokumentaciju Muzeja grada Beograda. U starim kwigama inventara Muzeja grada, koje se vode od 1955. godine, nije bilo mogu}e na}i podatak o tome kada je ovaj nalaz prvi put evidentiran, {to ukazuje da je dospeo u Muzej pre te godine. U inventarske kwige kasnije formirane Turskoaustrijske zbirke Muzeja grada poluga kantara i teg zavedeni su tek deset godina kasnije, nakon zavr{ene konzervacije u decembru 1965. godine. Poreklo i vreme nabavke tega mo`e se naslutiti iz jednog navoda u Kwizi ulaza i izlaza muzejskih objekata Muzeja grada Beograda (AK 125). U woj se, pod rednim brojem 264, navodi da je 16. 12. 1952. godine, pored nekoliko drugih i hronolo{ki veoma raznorodnih predmeta, nare~eni Milosav M. Stojadinovi} ponudio na otkup i jedne terazije. U dopisu od 5. 1. 1953, koji je on uputio upravi Muzeja, prilo`en je i spisak ponu|enih predmeta gde se, pod brojem 4, navodi slede}e: Male kovane terazije na|ene su pri iskopavawu doweg dela 7. jula ulice /ranije Kr. Petra a u pro{lom veku zvane Dubrova~ke/.42 Smatramo, stoga, da je gore opisani mawi teg zapravo deo kantara (pogre{no nazvanog terazije, koje zapravo ozna~avaju jednakokraku vagu) koji je naveden u ovom dopisu, te da je, iz jo{ uvek nepoznatih razloga, pridru`en poluzi kantara kojoj nikako, prema svojim dimenzijama i tipolo{kim karakteristikama, nije mogao pripadati. Na pitawe kada je i gde otkrivena sama poluga kantara odgovori}emo u nastavku ovog rada. Pre toga izlo`i}emo tipolo{ko-hronolo{ko opredeqewe ovog nalaza. STARINAR LXIV/2014 ispustima, i spojenih nitnama (Sl. 1). Takav na~in ka~ewa kuka, izme|u ostalog, omogu}avao je boqu pokretqivost prilikom balansirawa, a samim tim i ve}u preciznost pri merewu. Osim uobi~ajenih zavr{etaka krakova u obliku bikoni~nih ili piramidalnih dugmadi, na ve}im kantarima tipa Konstantinopoq mogu se na ovim mestima javiti i `ivotiwske protome – glave lava ili divqeg vepra, koje su obi~no nasa|ene na polugu i zalemqene. Tako|e, posebnu karakteristiku ovog tipa predstavqaju i kontrategovi koji mogu biti i figuralno modelovani u obliku poprsja Minerve (Atene) ili predstava careva i carica. Dok se vladari mogu javiti prikazani i kako sede na prestolu,45 ranovizantijske carice su obi~no predstavqene u formi poprsja. Na ovom tipu kantara ~esto se javqaju li~na imena koja, kako je ve} napomenuto, ukazuju na vlasnika ili slu`benika koji je overavao ta~nost mere. Kantari tipa Konstantinopoq datuju se uglavnom u razdobqe V–VII veka, a, sude}i prema najve}oj koncentraciji nalaza (Karta 1) i natpisima na gr~kom jeziku, mesto wihove izrade je isto~ni Mediteran.46 Sli~ni primerci kompletnih kantara i wihovih delova otkriveni su u najve}em broju na tlu dana{we Turske47 i Gr~ke48 kao i na tra~kom prostoru (Gradi{te, Qutibrod, Mihailovgrad), ali i u Ma|arskoj49 i Rumuniji50 tako|e. Kantari ovog tipa na|eni su u Srbiji jo{ i na Gamzigradu i Cari~inom gradu.51 Kantar iz Gamzigrada ima polugu sa tri fulkruma i o~uvan lan~ani sistem 42 Pismo je zavr{eno slede}im re~ima: Nadam se da }e Uprava Gradskog muzeja svojom {to skorijom odlukom o otkupu primqenih stvari omogu}iti da se ova kupovina realizuje i s obzirom na pravi~ne cene pojedinih predmeta, o ~emu molim da me izvoli obavestiti, na ~emu unapred blagodarim. 43 Jankovi} 1997, 312, 333, kat. br. 577. 44 Franken 1993, 89–94, Abb. 11. 45 Alföldi, Ross 1959; 179–183; Franken 1994, 101–102, 192–193, nos. CC1–CC9, pls. 115–17; Pitarakis 2012, 417–418, Fig. 16.15; Franken 1994, 101–102, 192–193, nos. CC1–CC9, pls. 115–117. 46 Franken 1993, 93, Abb. 12. 47 Visy 1994; Kad1oglu 2007; Pitarakis 2012, Fig.16.57. 48 Karud£j 1998. 49 Thomas 1988. 50 ªtefan 1950. 51 Srejovi} i dr. 1983, kat. br. 348, sl. 93, 94; @ivi} 2003, kat. br. 332 (Gamzigrad); Bavant 1990, 243–244, cat. no. 296, Pl. XLIV/296–298 (Cari~in grad). *** Iako prvobitno zavedena u inventarske kwige kasnosredwovekovne zbirke, poluga kantara iz Muzeja grada Beograda je nekoliko decenija kasnije ispravno opredeqena u ranovizantijski period i pojavi}e se u okviru postavke i kataloga izlo`be „Anti~ka bronza Singidunuma“ 1997. godine, kada je datovana u sredinu VI veka.43 Prema svojim osnovnim karakteristikama, poluga kantara iz Muzeja grada Beograda mo`e se preciznije svrstati u tip Konstantinopoq.44 Za ovaj tip je karakteristi~no da su dve ili tri kuke za ve{awe kantara (kuke fulkruma) povezane sa kra}im krakom pomo}u alki umetnutih u kanal, prose~en u uglasto ispup~enim 169 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) Karta I. Rasprostrawenost nalaza kantara tipa Konstantinopoq (prema: N. Franken 1993) Map I. Distribution of the Early Byzantine steelyards type Constantinople (by: N. Franken 1993) za ka~ewe. I na wemu se nalazi punktirani natpis: +ROUSTIKIOU+, sa imenom Rustikija (`Roust/kios, Rusticius) navedenim u genitivu na jednoj strani kra}eg kraka poluge.52 Otkriven je prilikom sistematskih istra`ivawa u dvori{tu ranovizantijske zgrade (Zgrada 2) dogra|ene u blizini bazilike (Bazilika II) i kompleksa zgrade sa atrijumom (Zgrada I) i jednobrodnom crkvom sa krstionicom. Sloj u kome je otkriven kantar datovan je, numizmati~kim i drugim nalazima (imitacija folisa Justina I), u period prve polovine VI veka.53 Od kantara iz Cari~inog grada o~uvana je samo poluga sa tri skale i tri le`i{ta fulkruma kojima nedostaju kuke za ve{awe. Tako|e, kantar iz Cari~inog grada je u ve}oj meri o{te}en korozijom, tako da su i merne skale i eventualni natpis sa poluge ne~itki. Delovi ovakvih ili sli~nih kantara zabele`eni su u ranovizantijskim utvr|ewima na \erdapu (Saldum, Boqetin), Rasu, Jelici i Barju kod Vrawa.54 *** Vratimo se sada pomenutoj nedoumici – gde, kada i pod kakvim okolnostima je otkrivena poluga rano- 52 Za natpis sa kantara iz Gamzigrada videti: Petrovi} 1995, no. 137. 53 Srejovi} i dr. 1983, 123, kat. 348. 54 Jeremi} 2009, 189–190, Fig. 89/cat. 626–628 (Saldum); Popovi} 1999, 116. kat. br. 160, sl. 67 (Ras); Milinkovi} 2010, sl. 39–41 T. XIV/5, 60–61; 229–231 (Jelica), Mitrovi} 2010, kat. 158 (Barje). 170 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) Sl. 7. Teg ranovizantijskog kantara iz Beograda. Narodni muzej, Beograd (foto: V. Ili}) a b Fig. 7. Early Byzantine Steelyard counterweight from Belgrade. National Museum, Belgrade (Photo: V. Ili}) d v g vizantijskog kantara iz Muzeja grada Beograda? U nedostatku bilo kakvih konkretnijih podataka o mestu i uslovima ovog nalaza, usmerili smo na{u pa`wu ka hronolo{ki i stilski bliskim nalazima delova ranovizantijskih kantara sa teritorije Singidunuma. Merni instrumenti ranovizantijske epohe su veoma retki na tlu Beograda, {to na{ primerak ~ini jo{ interesantnijim. Posebno su retki delovi kantara, me|u kojima se izdvaja ve} dobro poznati teg u obliku poprsja vizantijske carice koji se ~uva u Narodnom muzeju u Beogradu (Sl. 7 a–d).55 Po{li smo zato od vrlo slobodne pretpostavke da ovaj teg i poluga kantara iz Muzeja grada Beograda, budu}i suvremeni i jedinstveni na tlu Singidunuma, mo`da predstavqaju deo iste celine. TEG U OBLIKU VIZANTIJSKE CARICE, NARODNI MUZEJ U BEOGRADU Teg iz Narodnog muzeja (Sl. 7) izra|en je {upqim livewem od bakarne legure, a wegova unutra{wost je delimi~no zalivena olovom (visina: 13,4 cm, masa: 1551 g). Modelovan je u obliku `enskog poprsja, na pro{irenoj stopi ukra{enoj punktiranom valovnicom sa trostrukim kru`nim urezima u sredini polukru`nih poqa. Na jednom mestu u zidu postoqa izbu{ena je pravilna kru`na 55 171 Tati}-\uri} 1962, T. I, IIa, III v–g. STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) perforacija. U maniru uobi~ajenom za tu vrstu mernih instrumenata, vizantijska carica je prikazana u bogato ukra{enoj odori, sa jednom rukom podignutom u gestu blagosiqawa i svitkom u drugoj. Detaqi kojima je nagla{ena luksuzna tkanina izvedeni su kombinovawem ureza, punktirawa i kru`nica. Na glavi vladarke nalazi se rasko{na kruna, pod kojom se naslu}uje kosa formirana u visoku frizuru sa kapom i dijademom. Predstava je opredeqena kao lik carice Arijadne, `ene careva Zenona i Anastasija I, i datovana je u kraj V veka.56 O uslovima nalaza ovog kontratega nije mnogo re~eno prilikom prvog objavqivawa 1962. godine, osim da je na|en neobi~nim sticajem okolnosti, me|u objektima XVII veka slu~ajno iskopanim tokom radova u Danijelovoj ulici u Beogradu.57 Preciznije, i za na{ rad presudne, podatke o mestu i uslovima nalaza ovog tega potra`ili smo u kwigama inventara Narodnog muzeja u Beogradu. U delu kwige Inventara istoriske zbirke58, pod rednim brojem 1517, zaveden je popis objekata (predmeta)59 otkupqenih 26. avgusta 1932. godine od Sotira Zari}a: – Gvozdeni sud (mangale) za zagrevawe, dubok – gvozdeni predmeti za raspirivawe vatre – 2 kom. – Krug sa sinxirima i kukama za ve{awe – metalno patinirano kandilo – (metalni patinirani) svetwak veliki 1 m. mo`e se izdu`iti – Kantar, patiniran, – ukovanim imenom Papa Keso Ikonom, s druge strane: Kir… – jaje koje visi na kantaru od bronze sa stilizovanom bistom orientalnom Isti teg je zaveden 1950. godine pod rednim brojem 1718 u kasnijoj, prvoj Kwizi inventara Sredwovekovnog odeqewa Narodnog muzeja kao jaje koje visi na kantaru i pri tome ilustrovan crte`om. Iz sukcesivnih napomena zabele`enih pod istim rednim brojem ovog inventara a nastalih tokom naredne decenije mo`e se jasno videti da je teg sa likom vizantijske carice najpre datovan u XVII vek, da bi kasnije, svakako po objavqivawu tega od strane M. Tati}-\uri} 1962. godine, wegovo stilsko-hronolo{ko opredeqewe u inventaru bilo prepravqeno u slede}e: vizantijski, V–VI vek. Tre}i deo kantara – lan~ani sistem sa kukama – uveden je 1955. godine u novu, drugu Kwigu inventara Sredwovekovne zbirke pod rednim brojem 2860, i to kao sinxir i kuka za ve{awe. On je i tom prilikom opredeqen kao kasnosredwovekovan. Jedna napomena dopisana u istoj rubrici ukazuje na jo{ jedan va`an podatak za daqu sudbinu nalaza iz Danijelove ulice, a posebno za sudbinu poluge kantara – on je predvi|en za rashodovawe odnosno predaju Etnografskom muzeju ili Muzeju grada Beograda. LAN^ANI SISTEM ZA KA^EWE TERETA, NARODNI MUZEJ U BEOGRADU Lan~ani sistem od dva lanca sastavqen je od po 15 dvojnih karika u obliku osmice, ~ije su alke tordirane pod uglom od 90 stepeni (Sl. 8; T IV). Lanci su na krajevima zavr{eni masivnim kukama 56 Tati}-\uri} 1962, 126. Tati}-\uri} 1962, 115. Treba, me|utim, napomenuti da se pod vi{ezna~nim terminom objekti, koji je, kako }e se pokazati, bukvalno prepisan iz Kwiga inventara Sredwovekovnog odeqewa Narodnog muzeja, ovde zapravo ne radi o nekakvim arhitektonskim objektima ve} o grupi zajedno otkrivenih predmeta koji su prvobitno opredeqeni u XVII vek. U inventarnim kwigama nema nikakvih opisa eventualnih arhitektonskih ostataka niti podataka o zvani~noj poseti stru~waka mestu nalaza koji bi potvrdili wihovo postojawe. Ova terminolo{ka zabuna je umnogome doprinela da se atribucija pojedinih nalaza iz Danijelove ulice u kasnosredwovekovni period zadr`i kako u muzejskoj dokumentaciji, tako i u publikovanom radu o ranovizantijskom tegu. 58 Inventar istoriske zbirke (kwiga vo|ena od 1928. do 1947. godine), strana 248, br. 1517, delovodni protokol 573/ 1932. U arhivi Narodnog muzeja u Beogradu, na`alost, nedostaju upravo dokumenti pod ovim brojem delovodnog protokola. 59 Vidi napomenu 57. U nastavku, u odgovaraju}im rubrikama dodato je i slede}e: Na|eno prilikom kopawa temeqa za ku}u u Dani(e)lovoj ulici kod Topovskih {upa u Beogradu. Kao komentar hronologije nalaza dopisano je jo{ i slede}e: iz 17. veka. Iz ovog zapisa se jasno mo`e zakqu~iti da je poluga kantara koja se danas ~uva u Muzeju grada Beograda zapravo ona ista poluga kantara koja se navodi u spisku predmeta otkupqenih 1932. godine za Narodni muzej, {to pre svega potvr|uje identi~an natpis ukovan na kantaru. Ovde nalazimo i podatak da su zajedno sa ovom polugom otkriveni i drugi delovi kantara: figuralno oblikovan teg i lan~ani sistem sa kukama za ka~ewe tereta. 57 172 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) ZAKQU^AK sa za{iqenim dugim krajevima. U gorwem delu su spojeni ne{to ve}im karikama sa V nosa~em, ~ije raskovane u{ice su povezane sa plo~asto raskovanim lukom uobi~ajene „omega“ forme. Te`ina lan~anog sistema iznosi 821 g odnosno oko 2,5 litri (du`ina: 27,5 cm). Razdvajawu delova kantara i ostalih nalaza iz Danijelove ulice u Beogradu prethodilo je prvobitno hronolo{ko opredeqewe ostave u znatno pozniji period. Mo`da zbog izvesne sli~nosti sa kantarima koji su se 30-ih godina pro{log veka uveliko sretali u Beogradu, a mo`da i zbog ostalih predmeta od gvo`|a koji pripadaju istom nalazu, kantar na|en u Danijelovoj ulici je najpre datovan u XVII vek. U prvu kwigu Inventara Sredwovekovne zbirke Narodnog muzeja bi}e zaveden 1950. godine samo teg, budu}i figuralno modelovan i svakako „atraktivniji“ od drugih nalaza, dok }e ostali elementi biti vremenom preba~eni u druge zbirke Narodnog muzeja ili ~ak preraspore|eni u druge nadle`ne muzeje u Beogradu. Pet godina kasnije, u drugoj kwizi Inventara Sredwovekovne zbirke Narodnog muzeja posebno je inventarisan i lan~ani sistem ovog kantara opredeqen kao kasnosredwovekovan. Predlo`eno je rashodovawe ovog predmeta i spomenuta su mogu}a nova odredi{ta, me|u kojima i Muzej grada Beograda. Na taj na~in, iako za to jo{ uvek nemamo direktnih potvrda zbog nedostatka odgovaraju}e dokumentacije, svakako je u Muzej grada Beograda dospela i poluga kantara. Sudbina lan~anog sistema je ipak bila druga~ija. Iako rashodovan, on, prema svemu sude}i, nije napustio Narodni muzej i ostao je u depou Sredwovekovne zbirke do sredine marta 2014. godine, kada mu ponovo ulazimo u trag. Ve} ranije opredeqena kao kasnosredwovekovna, poluga kantara je, dospev{i u Muzej grada Beograda, najpre uvedena u inventarske kwige tada{we Tursko-austrijske zbirke. Na to da se radi o istom kantaru jasno upu}uje identi~an natpis koji doslovce odgovara natpisu sa poluge kantara sa ukovanim imenom navedenom u popisu predmeta otkupqenih od Sotira Zari}a za Narodni muzej 1932. godine. Kada je poluga kantara dospela u Muzej nije poznato, budu}i da u ulaznim kwigama, koje se vode od 1955. godine, nije bilo mogu}e na}i podatak kada je ovaj nalaz prvi put evidentiran, {to ukazuje na to da je najverovatnije dospeo u Muzej grada ranije, verovatno izme|u 1950. i 1955. Poluga kantara je zapravo zavedena u inventar Turskoaustrijske zbirke tek nakon zavr{ene konzervacije u decembru 1965. godine, kada je kantaru, iz nepoznatog razloga, ve} bio prikqu~en i mawi teg sa kojim, kao {to smo pokazali, on nema nikakve Sl. 8. Lan~ani sistem ranovizantijskog kantara iz Beograda. Narodni muzej, Beograd (foto: V. Ili}) Fig. 8. Collar with two hooked chains for hanging loads. National Museum, Belgrade (Photo: V. Ili}) 173 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 Karta II. 1) Mesto nalaza ranovizantijskog kantara iz Beograda (Danijelova ulica); 2) Ranovizantijski Singidunum Map II. 1) Find spot of the Early Byzantine steelyard from Belgrade (Danijelova Street); 2) Early Byzantine Singidunum Danijelovoj ulici (oko 1 km), Topovske {upe su o~igledno navedene samo kao tada najbli`a i jedina urbana referentna ta~ka u onovremenom okru`ewu vo`dova~kih vo}waka i wiva, a ne i kao u`e mesto otkri}a. Iz popisa predmeta na|enih u Danijelovoj ulici se mo`e zakqu~iti da delovi kantara nisu bili i jedini predmeti iz ranovizantijske epohe u ovom nalazu. Verovatno se radi o ostavi koja je sa- veze. Podatak zabele`en u kwizi inventara Muzeja grada da su kantar i teg otkupqeni od Milosava M. Stojadinovi}a iz Beograda, zapravo se odnosi samo na teg koji je pripadao nekom mawem, verovatno poznijem kantaru. Dodatne detaqe o ta~nom vremenu i mestu ovog nalaza dobili smo usmenim putem od Viktorije Alimpijevi} (ro|ene Zari}), unuke nalaza~a Sotira Zari}a († 1964). On je 1929. godine, prilikom kopawa temeqa za porodi~nu ku}u u Danijelovoj ulici (dana{wi broj 25) na Vo`dovcu (Karta 2), otkrio pomenuti skupni nalaz koji je prodao Narodnom muzeju 1932. godine.60 Kao mesto ovog nalaza u inventarskim kwigama i u kasnijim radovima koji su se bavili ranovizantijskim tegom61 navode se Topovske {upe, odnosno prostor nekada{weg artiqerijskog poligona Vojske Kraqevine Jugoslavije i pomo}nih zgrada kasarne Kraqevi} Andrej.62 Budu}i prili~no udaqene od mesta nalaza u 60 Ovaj podatak obja{wava i to da stru~waci iz Narodnog muzeja nisu mogli da iza|u na mesto nalaza neposredno po otkri}u predmeta iz Danijelove ulice jer je do otkupa pro{lo nekoliko godina, {to iskqu~uje konstatovawe eventualnih arhitektonskih objekata (iz XVII veka) na ovom lokalitetu. 61 Jankovi} 1997, 333, kat. br 576. 62 Ova lokacija poznata je i kao mesto zloglasnog koncentracionog prolaznog logora za Jevreje i Rome iz 1941. godine: \uri}-Zamolo, Nedi} 1993–1994, 68; @arkovi} 2009. 174 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 skog mitropolita, pod ~iju nadle`nost je spadao i Singidunum.71 Nakon toga, imena episkopa ovoga grada se dugo ne spomiwu, a nema ni pomena episkopije.72 Razlog tome je svakako nezaustavqivi nalet Gota i Huna, koji je u~inio kraj vizantijske vlasti na ovom tlu. Dele}i sudbinu ostalih velikih gradova na ovoj teritoriji, Singidunum biva razoren. O razmerama tog pusto{ewa, a posebno o stradawu hri{}ana, svedo~i opis bla`enog Jeronima iz 396. godine: Episkopi su zarobqeni, sve{tenici i crkveni qudi razli~itih rodova poubijani. Poru{eni su hramovi. Uz Bo`ije oltare stoje kowi kao u kowu{nicama. Premetnute su mo{ti mu~enika.73 Objavquju}i odluke halkedonskog sabora 451. godine, car Lav I upu}uje 458. godine akt svim episkopima u Carstvu, u kome se ne spomiwu episkopi Gorwe Mezije i Priobalne Dakije. Ovo, izme|u ostalog, potvr|uje da su po~etkom druge polovine V veka gradovi u tim oblastima, ako ne potpuno zapusteli, ono svakako ostali van vizantijske kontrole i bez crkvene organizacije. Singidunum }e u ru{evinama ostati najverovatnije do vremena Justinijana dr`ala i druge predmete, mahom crkvenog karaktera. Osim kantara, ~iji natpis jasno upu}uje na to da je bio u posedu crkvenog ili manastirskog ikonoma, navedeni su jo{ kandilo i svetwak, odnosno kandelabar, najverovatnije oba izra|ena od bakarne legure.63 Dok o izgledu i dimenzijama kandila nema, na`alost, nikakvih podataka, pomenuti svetwak je interesantan ne samo zbog svoje veli~ine (visina: 1 m)64 ve} i zbog komentara mo`e se izdu`iti, koji upu}uje na to da se verovatno radi o podesivom kandelabru. Takvi kandelabri sa teleskopskom konstrukcijom osovine, koja je omogu}avala prilago|avawe visine na kojoj se nalazila svetiqka, poznati su u kasnoanti~kom periodu (III–IV vek) kako u luksuznijim varijantama od srebra, tako i od bakarne legure.65 Pojedini primerci takvih kandelabara, me|utim, javqaju se i kasnije, sve do VI veka.66 Budu}i da se radi o slu~ajnom nalazu bez detaqnijih pokazateqa konteksta, preciznije hronolo{ko opredeqewe kantara iz Beograda nije izvesno. Kantari tipa Konstantinopoq datuju se u {irem rasponu od V do VII veka, kada je svakako kori{}en i na{ primerak. Detaqna stilska i ikonografska analiza tega ukazala je, na osnovu analognih primeraka i predstava sa novca, da je na wemu predstavqena bista carice Arijadne, `ene cara Zenona (425–491).67 Na osnovu ornata i elemenata odore, teg je preciznije datovan u period od 474. do 490. godine.68 Podaci o episkopima i episkopskoj crkvi Singidunuma, pod ~iju nadle`nost je, kako je ve} pomenuto, spadalo i ~uvawe zvani~nih mera i mernih instrumenata,69 veoma su {turi. O ta~nom mestu na kome se ova crkva nalazila ne zna se jo{ uvek ni{ta. Od ranih, arijanstvu naklowenih episkopa Singidunuma, koji su bili pod jurisdikcijom episkopa u Viminacijumu, prvi se, u kontekstu sabora u Triru 335, pomiwe Ursacije koji stoluje do oko 370.70 U wegovo vreme, u Singidunumu biva 366. godine odr`an i jedan sabor arijanskih episkopa Panonije, Mezije i Priobalne Dakije. Wegov naslednik, episkop Sekundijan, tako|e arijanac, bio je 381. godine, nakon crkvenog sabora u Akvileji, osu|en za jeres i prinu|en da napusti Singidunum i svoju dijecezu. Kasnije, odlukama papa Siracija (389–399) i Ino}entija I (401–407), koji su bezuspe{no ali istrajno poku{avali da povrate primat Rima na ovoj teritoriji, ovla{}ewa rukopolagawa episkopa u ~itavom Iliriku preneta su na solun- 63 Na ovo upu}uje kori{}ewe termina metalno patinirano, za razliku od najpre navedenog suda za zagrevawe i predmeta za raspirivawe vatre (`ara~, ma{ice?) koji su jasno kvalifikovani kao gvozdeni. 64 Kandelabri na trono`nom postoqu, visoki i po 120–160 cm, poznati su jo{ iz ranocarskog perioda (Bailey 1996, 91–91; Q 3867–3876). 65 Bailey 1996, 102–104; Q 3917–3918; Xantopolou 2010, 33–34, CD 2.001–2.002. 66 Baratte 1998, 78. 67 ODB I, 166–167(Ariadne). 68 Tati}-\uri} 1962, 126. 69 Treba, naravno, pomenuti i mogu}nost da je kantar sa ostalim nalazima liturgijskog karaktera iz Danijelove ulice mogao biti svojina nekog, nama jo{ uvek nepoznatog, crkvenog ili manastirskog imawa koje se nalazilo bli`e mestu nalaza u okolini Singidunuma. O tome da je neposredno okru`ewe mesta nalaza kantara bilo kori{}eno kao poqoprivredno zemqi{te u kasnoanti~kom periodu svedo~e retki ali interesantni arheolo{ki nalazi, koji upu}uju na mogu}e postojawe vila rustika na padinama mokrolu{kog i kumodra{kog potoka: Bojovi} 1977; Pop-Lazi} 2002, 33, G-226, 227, sl. 23/16. Kolega S. Pop-Lazi} ukazao mi je na ove nalaze, na ~emu mu najiskrenije zahvaqujem. 70 Jankovi} 1996, 306; Popovi} 1995, 187–195. 71 Istorija Beograda I, 106. 72 Tek krajem VI veka, o wenom postojawu usputno saznajemo u kontekstu zakletve avarskog kagana Bajana vizantijskom strategu Setu nad svetim kwigama koje {aqe neimenovani episkop Singidunuma: VIINJ I, 93–94. 73 Popovi} 1995, 199–200. 175 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) I (527–567), koji ga obnovi i, opasav{i ga ~vrstim bedemom, ponovo stvori sjajan i velike hvale dostojan grad.74 Mogu}e je, me|utim, da je upe~atqiviji povratak Vizantije na ove teritorije zapo~eo ubrzo nakon odlaska Gota 488. godine, odnosno kasnije, tokom vladavine cara Anastasija I (491–518). Taj car sklapa sa Teodorihom mirovni sporazum 510. godine, prema kome Singidunum biva ponovo ukqu~en u sastav vizantijske dr`ave.75 Kako je i koliko brzo pokrenut proces obnove dr`avne i crkvene vlasti u oslobo|enim pograni~nim oblastima te{ko je re}i. Pojava kantara sa tegom u liku carice Arijadne, koja nakon Zenonove smrti postaje i `ena Anastasija I, nije bez zna~aja u ovom pogledu. Pored vojnog prisustva, od nasu{ne va`nosti za organizaciju ponovo zaposednutih krajeva u Podunavqu bilo je uspostavqawe dr`avne kontrole nad finansijama i redovno prikupqawe poreza. To, izme|u ostalog, ilustruje i jedan podatak iz ranije, ali istim te{ko}ama bremenite, istorije ovih prostora. Ekonomske prilike u carstvu izlo`enom upadima varvara nisu bile nimalo povoqne ni u posledwoj ~etvrtini IV veka. U naredbi upu}enoj 29. jula 386. godine Eusigniju, pretorijanskom prefektu Ilirika, carevi Gracijan, Valentinijan II i Teodosije I nala`u da nadle`ni dr`avni slu`benici, koji su zbog dejstva neprijateqa zanemarili svoje du`nosti, nastave sa prikupqawem poreza.76 Kontrola mera je u ovom smislu imala poseban zna~aj, {to je podrazumevalo namensko obezbe|ivawe zvani~nih mernih instrumenata poput tegova, merica i vaga. Najraniji zakon koji se bavi zvani~nom distribucijom tegova je iz vremena Teodosija I (389. godina). On predvi|a zakonsku obavezu javnog izlagawa oficijelnih tegova i mera na svakoj putnoj stanici (statio), sa ciqem da se osujete prevare.77 O izuzetnom zna~aju tih mera svedo~i i ubrzo zatim donet zakon sa preciznijim odredbama koje su propisivale proveru vrednosti robe sakupqane za porez, ali i ka`wavawe onih koji su vr{ili zloupotrebe.78 Ovaj zakon bi}e kasnije prepisan u Justinijanov kodeks, s napomenom da u slu~aju spora izme|u obveznika i sakupqa~a poreza o{te}ene plati{e mogu da sa najvi{eg mesta (Comes sacrorum largitionus) potra`uju referentne mere i tegove, koji se imaju ~uvati u najsvetijoj (episkopskoj) crkvi svakog grada.79 Uloga crkve i crkvenih zvani~nika u ekonomskom `ivotu ranovizantijske dr`ave je bila pre- sudna, naro~ito u pograni~nim oblastima udaqenim od glavnih centara mo}i, vlasti i fiskalne kontrole. Crkva se u takvim kriznim okolnostima javqa ne samo kao duhovni sto`er ve} i kao glavni eksponent dr`ave u ekonomskoj organizaciji, pa i kao direktan u~esnik u finansirawu vojske i vojne logistike. Snabdevawe trupa u kriznim vremenima bio je problem sa kojim se vizantijska dr`ava suo~avala od samog po~etka. Kako bi ovo re{io, Justinijan osniva posebnu pretorijansku prefekturu (quaestura exercitus), u ~iju nadle`nost su spadali Kipar, Karija, egejska ostrva ali i podunavske provincije Skitija i Druga Mezija. Jedna od glavnih briga nadle`nog kvestora je bilo snabdevawe trupa stacioniranih na severnim pograni~nim teritorijama, uveliko opustelim i ispo{}enim stalnim ratovima i epidemijama.80 O ovoj ekonomskoj ulozi crkve i u ne{to kasnijim, ali tako|e kriznim, vremenima Iraklijeve vlade svedo~i natpis sa poznatog kantara signiranog imenom sve{tenikakapetana Georgija (+GEORGIOU PRESBUTEROU NAUKLEROU) iz broda potonulog kod Jasi Ade.81 Kantar sa tegom u liku carice Arijadne iz Beograda mogao bi, dakle, da predstavqa trag tih mera vizantijske dr`ave koje su preduzimane u smislu obnove fiskalne organizacije i snabdevawa vojske u pograni~nim oblastima Podunavqa jo{ u vreme Anastasija I, a sasvim izvesno i kasnije, za 74 Procopii opera, De aedificilis IV, 5; VIIW I, 66. Istorija Beograda I, 109. 76 Ch.T. 1.32.5. 77 Ch.T. XII. 6.19. 78 Ova naredba, potpisana od istih careva kao i prethodna (Valentinijan II, Teodosije I i Arkadije), propisuje da merice za suve ili te~ne mere izra|ene od kamena ili bronze (modii, sextarii) kao i tegovi (pondera) moraju biti sme{teni na svakoj stanici (mansio) i u svakom gradu, kako bi poreski obveznici (tributarii) znali koliko ~ega duguju; Ch.T. XI.6.21. 79 Teoretski, ovaj zakon je predvi|ao da se distribucijom i overom mera i tegova za trgova~ku robu bave prefekt pretorija i gradski prefekti, a monetarnim tegovima komes carske riznice (comes sacrorum largitionum), {to se navodi u 15. glavi CXXVII novele datovanoj u 545. godinu (CI X.72.9). Sude}i po sa~uvanim natpisima na tegovima, ova striktna podela nije za`ivela u praksi, pa tegove obe vrste izdaju i prefekti i komesi. Me|u ostalim titulama koje se pojavquju uz imena zvani~nika na overenim tegovima javqaju se jo{ i comes, comes rei privatae, anthypatos kao i prokonzuli, vir laudibilis i vir clarissimus; Entwistle 2008, 40–41. 80 Van Alfen 1996, 211–213. 81 Sams 1982. 75 176 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 vlade Justina I (518–527)82 i potowem zamahu Justinijanove obnove limesa. Iako smo skloni da prihvatimo rezultate detaqne ikonografske analize M. Tati}-\uri}, koja je opredelila predstavu sa tega beogradskog kantara kao lik carice Arijadne, treba izdvojiti i druga~ija tuma~ewa. Lik carice sa veoma sli~nog tega iz kolekcije Severeanu83 opredeqen je kao predstava Elije Pulherije (399–453), }erke cara Arkadija, koja titulu avguste dobija od senata 414. godine, a kasnije postaje i `ena cara Markijana.84 Time bi datovawe beogradskog kantara bilo za vi{e od pola veka ranije, {to bi dalo osnova da se sakrivawe ostave iz Danijelove ulice u Beogradu pove`e sa hunskim ru{ewem gradova u Podunavqu 441. godine, me|u kojima se nalazio i Singidunum.85 Tako|e, pomenu}emo i mi{qewe prema kome su na figuralno modelovanim tegovima prikazani zapravo uop{teni prikazi vladarki, a ne konkretne osobe.86 One bi, tako, predstavqale personifikaciju vrlina carskog autoriteta ali i ikonografske i apotropejske tekovine razli~itih ranije po{tovanih bo`anstava (Fortune, Pravde i Tihe), {to se generalno uklapa u svet trgovine i razmene. Lik carice prikazane na tegu trebalo je, prema ovom mi{qewu, da bdi nad uspe{nom trgovinom i donese sre}u i bogatstvo korisniku. Kao zvani~ni merni instrumenti kantari su, u svakom slu~aju, a to se vidi i iz pomenutih zakonskih spisa, poveravani na ~uvawe ve}im i zna~ajnijim crkvama, odnosno nadle`nim ikonomima, {to jasno pokazuje i natpis sa beogradskog kantara. Koliko je on bio u upotrebi i kada je i u kojim okolnostima sakriven – te{ko je re}i s ve}om sigurno{}u. Budu}i da nam je kontekst nalaza iz Danijelove ulice u potpunosti nepoznat i po{to, za sada, ne znamo ni ta~an izgled ostalih predmeta mahom liturgijskog karaktera, mo`emo samo da pretpostavimo da je to najkasnije vreme od avarske opsade i pqa~kawa Singidunuma 584. godine do kona~nog uru{avawa dunavskog limesa u prvoj deceniji VII stole}a, kada se vizantijska dr`ava i crkva povla~e pred naletom Avara i Slovena. zeja grada Beograda, koje su mi omogu}ile da detaqno prou~im i publikujem polugu ranovizantijskog kantara koji se ~uva u Sredwovekovnoj zbirci Muzeja grada Beograda. Pored wih, zahvaqujem i Jeleni Vasi}, koja mi je omogu}ila uvid u dokumentaciju i arhivsku gra|u Muzeja grada Beograda. Posebno zahvaqujem mr Jeleni Kondi} na korisnim savetima i mogu}nosti da publikujem ranovizantijski teg iz Kasnoanti~ke zbirke Narodnog muzeja, kao i dr Slobodanu Fidanovskom i dr Emini Ze~evi}, koji su mi pomogli ne samo da u muzejskoj dokumentaciji i depou Sredwovekovne zbirke istog muzeja u|em u trag lan~anom sistemu kantara ve} i da pregledom inventarskih kwiga rekonstrui{em sve elemente nalaza iz Danijelove ulice u Beogradu. Zahvalnost dugujem i mr Veri Bogosavqevi}-Petrovi}, mr Dragani Eremi} i Jeleni ]eriman iz Odeqewa za dokumentaciju Narodnog muzeja u Beogradu. @elim i ovom prilikom da istaknem pomo} mr Du{ana Vra~ari}a, Nata{e Kari{ik i Mirjane Dimi}, iz Vojno-tehni~kog zavoda u Beogradu, u obavqawu fizi~ko-hemijskih analiza sastava legure poluge kantara iz Muzeja grada Beograda kao i tuma~ewu rezultata ovih istra`ivawa. Za pomo} pri ~itawu natpisa sa kantara zahvaqujem profesoru dr Marici [uput, sa Odeqewa za istoriju umetnosti Filozofskog fakulteta u Beogradu, i Ilu Akadu, sa Odeqewa za klasi~ne nauke Filozofskog fakulteta u Beogradu. Wihovo iskustvo i korisne sugestije bile su od velikog zna~aja za formulisawe ponu|enog ~itawa, dok odgovornost za eventualne oma{ke u pogledu iznesenog tuma~ewa i opredeqewa snosi autor li~no. Za kvalitetne fotografije i crte`e kojima je ilustrovan ovaj rad zahvaqujem Neboj{i Bori}u, Veqku Ili}u i Awi Suboti}. Zahvalnost Izuzetnu zahvalnost dugujem kolegama iz Muzeja grada Beograda i Narodnog muzeja u Beogradu, bez ~ije nesebi~ne pomo}i ovaj rad ne bi mogao da bude realizovan. Na prvom mestu `elim da zahvalim mr Milici Jankovi} i Niki Strugar, kustosima Mu- 82 Istom periodu, sude}i po uslovima nalaza, pripada i gamzigradski kantar: vidi napomenu 51. 83 Gramatopol 1985, fig. 38. 84 ODB III, 1757–1758 (Pulcheria). 85 VIINJ I, 10–11. 86 James 2003, 51–56. O zna~aju `enskih personifikacija u kasnoanti~kom periodu videti: Herrin 20003, 35. 177 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 BIBLIOGRAFIJA: Alföldi, Ross 1959 – A. Alföldi, M. Ross, Cornuti: A Teutonic Contigent in the Service of Constantine the Great and its decisive role in the Battle at the Milvian Bridge, Dumberton Oaks Papers 13, 169–183. Van Alfen 1996 – P. G. van Alfen, New light on the 7th-c. Yassi Ada shipwreck: capacities and standard sizes of LRA 1 Amphoras, Journal of Roman Archaeology 9, 198–213. Anti~ka bronza na tlu Jugoslavije 1969 – Anti~ka bronza na tlu Jugoslavije (ur. M. Kolari}), Beograd 1969. Baratte 1998 – F. Baratte, Note à propos d’un trésor de vaisselle de bronze d’époque byzantine découvert à Pupput (Tunisie), Cahiers archéologiques 1998, 73–80. Bavant et all 1990 – B. Bavant, Les petits objects, in: Cari~in Grad II: Le Quartier Sud-est de la Ville Haute, (ed. Bavant, Kondi} and Spieser), Belgrade – Rome 1990, 191–257. Bendall 1996 – S. Bendall, ByzantineWeights: An Introduction, London 1996. Bojovi} 1977 – D. Bojovi}, Singidunum, Beograd, Bukovi~ka ul. br. 88 – vila rustika, Arheolo{ki pregled 17, 1977, 156–158. ChT – Codex Theodosianus, The Theodosian Code and Novels abd Sirmodian Constitutions (ed. C. Pharr), Princeton 1952. CI – Codex Iustinianus (ed. Kruger, Paul), Berlin 1959. Covachev and Potârniche – Z. Covachev, T. Potârniche, Accesorii ale unei balanþe din bronz descoperite în sectorul de est al Capidava vei, Pontica 43, 267–275. Dagron 2002 – G. Dagron, The Urban Economy, Seventh-Twelfth Centuries, 393–461, in: The Economic History of Byzantium The Economic History of Byzantium 2 (ed. A. E. Laiou), Washington DC 2002. Daremberg, Saglio 1877 – C. Daremberg, E. Saglio, Dictionnaire des antiquitis grecques et romaines III, 2, Paris 1887. Darrouzèse 1970 – J. Darrouzèse, Recherches sur les Offikia de l’Eglise Byzantin, Paris 1970. Dr~a 2007 – B. Dr~a, Kantar iz kasnoanti~ke zbirke Narodnog muzeja u Beogradu, Arhaika 1, 188–196. \uri}-Zamolo, Nedi} 1993–1994 – D. \uri}Zamolo, S. Nedi}, Stambeni delovi Beograda i wihovi nazivi do 1941. godine, Godi{wak grada Beograda XL–XLI, 65–106. Entwistle 2002 – C. Entwistle, Byzantine Weights in: The Economic History of Byzantium: From the Seventh through the Fifteenth Century (ed. Angeliki E. Laiou), Washington DC 2002. Entwistle 2008 – C. Entwistle, Late Roman and Byzantine Weights and Weighing Equipment, 38–46, in: The Oxford Handbook of Byzantine Studies (ed. E. Jeffreys and J. Haldon), Oxford 2008. Entwistle 2008a – C. Entwistle, A Catalogue of the Late Roman and Byzantine Weights and Weighing Equipment in the British Museum, London 2008. Franken 1993 – N. Franken, Zur Typologie Antiker Schnellwaagen, Bonn. Jahrb. 193, 69–120. Franken 1994 – N. Franken, Aequipondia: Figürliche Laufgewichte römischer und frühbyzantinischer Schnellwaagen, Alfter 1994. Franken 1995 – N. Franken, Katalog der römischen Schnellwaagen im Rheinschen Landesmuseum Bonn, Bonn. Jahrb. 195, 425–438. Garbsch 1988 – J. Garbsch, Wagen oder Waagen?, Bayerische Vorgeschbl. 53, 1988. Gramatopol 1985 – M. Gramatopol, Portretul roman în Romania, Bucureºti 1985. Hendy 1994 – M. F. Hendy, Studies in the Byzantine Monetary Economy c. 300–1450., Cambridge 1994. Herrin 2000 – J. Herrin, The Imperial Feminine in Byzantium, Past and Present 169, 3–35. Hill 1955 – D. K. Hill, When Romans went Shopping, Archaeology V (March 1955), 51–55. Istorija Beograda – (grupa autora) Istorija Beograda (ur. V. ^ubrilovi}), Beograd 1974. Ivani{evi}, [pehar 2005 – V. Ivani{evi}, P. [pehar, Early Byzantine finds from ^e~an and Gornji Streoc (Kosovo), Starinar 55, 133–159. James 2003 – L. James, Who’s That Girl? Personifications of the Byzantine Empress, 51–56, in: Through a Glass Brightly: Studies in Byzantine and Medieval Art and Archaeology Presented to David Buckton (ed. C. Entwistle), Oxford 2003. Jankovi} 1997 – M. Jankovi}, Seoba naroda, u: Anti~ka bronza Singidunuma (ur. S. Kruni}), Beograd 1997, 305–342. Jenemann 1989 – H. R. Jenemann, Zur Geschichte der Waagen mit variablem Armlängenverhältnis im Altertum, Trierer Zeitschrift 52, 319–352. Jenemann 1995 – H. R. Jenemann, Die Geschichte der Waage im Mittelalter. Internationale Zeitschrift für Geschichte und Ethik der Naturwissenschaften, Technik und Medizin: Neue Serie, 3, 145–166. 178 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 Ridder 1905 – A. de Ridder, Collection de Clercq, Catalogue III, Les bronzes, Paris 1905. Ross 1962 – Catalogue of the Byzantine and Early Medieval Antiquities in the Dumbarton Oaks Collection, Washington DC 1962. Sams 1982 – G. K. Sams, The weighing implements, 217–230, in: Yassi Ada vol. I – A seventh-Century Byzantine Shipwreck, (eds. G. F. Bass, F. H. Van Doorninck Jr), College Station 1982. Schilbach 1970 – E. Schilbach, Byzantinische Metrologie, Munich 1970. Simi} 1987 – Z. Simi}, Utvr|ewe na Kulini u Rogatcu, Novopazarski zbornik 11, Novi Pazar 1987, 13–19. Srejovi} i dr. 1983 – D. Srejovi}, \. Jankovi}, A. Lalovi}, Gamzigrad – kasnoanti~ki carski dvorac, Beograd 1983. ªtefan 1950 – Gh. ªtefan, O balanþã roman din secolul VI e.n. descoperitã în Dobrogea, SCIV I/2, 152–162. [pehar 2010 – P. [pehar, Materijalna kultura ranovizantijskih utvr|ewa u \erdapu, Beograd 2010. Thomas 1988 – E. B. Thomas, Bronzebusten der Licinia Eudoxia in Ungarn, 501–516, in: Ritratto ufficiale e ritratto privato: Atti della II Conferenza Internationale sul Ritratto Romano (ed. N. Bonacasa, G. Rizza), Roma. Tati}-\uri} 1962 – M. Tati}-\uri}, Zbornik Narodnog muzeja III, 115–127. TÕ ^parcikÕn bibl/on – Le Livre du Préfet ou l’édit de l’empereur Léon le Sage sur les corporations de Constantinople (trans. J. Nicole), Genavae 1894. @arkovi} 2009 – N. @arkovi}, Prolazni logor Topovske [upe, Nasle|e 10 (2009),103–112. @ivi} 2003 – M. @ivi}, Felix Romuliana – 50 godina odgonetawa, Zaje~ar 2003. Visy 1994 – Z. Visy, Romische und byzantinische Schnellwaagen aus der Turkei, 435–444, in: Akten der 10. Internationalen Tagung über antike Bronzen, Freiburg, 18–22. July 1988, Stuttgart 1994. VIINJ I – Vizantijski izvori za istoriju naroda Jugoslavije I (ur. G. Ostrogorski), Beograd 1955. Wei{ 2001 – P. Wei{, Schnellwaage, Der Neue Pauly 11, 200–204. Jones 1940 – A. H. M. Jones, The Greek City from Alexandar to Justinian, Oxford 1994. Jovanovi} 1984 – A. Jovanovi}, Hajdu~ka Vodenica, kasnoanti~ko i ranovizantijsko utvr|ewe, Starinar 33–34, 319–335. Kad±oglu 2007 – M. Kad±oglu, Menderes Nysas±’ ndan bir Kantar, PATRONVS, Coºkun Özgünel’e 65. Yaº Armagan› / Festschrift für Coºkun Özgünel zum 65. Geburtstag, ed. Betül Avunc, Bas±m 2007, 223–230. Karud£j 1998 – N. Karud£j, C£lkinoj kampanÒj (kant£ri) apÒ palaiocristianik» špaul thj Qessalon/khj, Mouse/o BuzantinoÚ PolitismoÚ 5/1998, 36–41. Kosti} 1993 – Z. Kosti}, Prilozi uz ranovizantijsku metrologiju, Glasnik Srpskog arheolo{kog dru{tva IX, 67–72. Kosti} 2003 – Z. Kosti}, Marginalije uz nekoliko epigrafskih spomenika iz Makedonije, 195–219, u: Spomenica Jovana Kova~evi}a (ur. R. Bunarxi}, @. Miki}), Beograd. Milinkovi} 2010 – Gradina na Jelici, Beograd 2010. ODB – The Oxford Dictionary of Byzantium (ed. A. P. Kahzdan), Oxford 1991. Pape 1911 – W. Pape; G. E. Benseler, Wörterbuch der griechischen Eigennamen, Braunschweig 1911. Petrovi} 1995 – P. Petrovi}, Inscriptions de la Mésie Supérieure, Vol. III/2, Timacum Minus et la Valée du Timok, Beograd 1995. Pililh 1985 – I. Pilil», Titloi, off/kia kai axièmata en th Buzantin» Autokrator/a kai th Cristianik» OrqodÒxw Ekklhs/a, Aq»nai 1985. Pitarakis 2012 – B. Pitarakis, Daily Life at the Marketplace in Late Antiquity and Byzantium, 399–426, in: Trade and Markets in Byzantium (ed. C. Morrisson), Washington DC 2012. Pop-Lazi} 2002 – S. Pop-Lazi}, Nekropole rimskog Singidunuma, Singidunum 3, 7–100. Popovi} 1999 – M. Popovi}, Tvr|ava Ras, Beograd 1999. Popovi} 1995 – R. Popovi}, Rano hri{}anstvo na Balkanu pre doseqavawa Slovena, Beograd 1995. Procopii, De aedificilis – Procopii opera, De aedificilis (ed. J. Haury), Lipsiae 1913. 179 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 Summary: MIROSLAV VUJOVI], Department of Archaeology, Faculty of Philosophy, Belgrade EARLY BYZANTINE STEELYARD FROM BELGRADE Key words. – Belgrade, Singidunum, steelyard, weight, epigraphy, metrology, early Byzantine period, measuring instruments, oikonomos. ments. In the case of the Belgrade find, it is most likely that the name Philikos refers to the priest who, as an oikonomos of the (episcopal?) church or a monastery, was entrusted with the steelyard. That being the case, the inscription would probably read: +kÚrioj FilikÒj + +papa(s) kes(sic) oikonmou+. In translation: (steelyard) of Master Philikos, the priest and oikonomos. According to its basic features, the steelyard beam from BCM can be more precisely aligned with the Constantinople type (Fig. 1) which is dated to the period from the 5th to the 7th century. Upon the examination of the inventory records of NMB, it was determined that the steelyard beam from BCM is a part of the hoard discovered during the construction works in Danijelova Street in Belgrade in 1929 and repurchased for NMB in 1932. The counterweight (Fig. 7), shaped like the bust of the Byzantine Empress Ariadne (height: 13.4 cm, weight: 1,551 g), also belongs to the same group of finds. Mentioned among these finds are one iron vessel, two iron tools for fanning a fire, one adjustable candelabrum, a censer, a chain system with hooks (Fig. 8; Pl. IV) and a steelyard beam with an inscription matching the one from the steelyard beam from BCM. What preceded the separation of the steelyard parts and other finds from Danijelova Street in Belgrade is the original chronological determination of the hoard to a much later period (17th century), whereby some contents of the hoard were distributed to different museum collections. The chain system with the hooks (length: 27.5 cm; weight: 821 g) which is certainly a part of the same steelyard, remained in NMB, whilst we still do not know anything about what happened to the other finds. Since we are dealing here with an accidental find, a more precise chronological determination of the steelyard from Belgrade is not possible at this time. The detailed stylistic and iconographic analysis of the counterweight indicated that the bust of the Empress Ariadne, the wife of the Emperors Zeno and Anastasius I, is represented on it and, based on the adornments and elements of the gown, it was more precisely dated to the period between 474 and 490. How long the steelyard had been in use and when, and in what circumstances, it had been hidden is difficult to say. Since we do not even know the actual appearance of other objects, of a mainly liturgical character, from this hoard, we can only assume that it is, at the latest, the period between the Avar siege and the plundering of Singidunum in 584 and the final downfall of the Danube Limes in the first decade of the 7th century, when the Byzantine state and church retreated from the Avar and Slav assault. The initial aim of this paper was to present scholarly circles with a detailed report about the insufficiently known Early Byzantine steelyard beam (Figs. 2–5; Pl. I/1) treasured in Belgrade City Museum (further in the text BCM) together with a small counterweight (Fig. 6; Pl. I/2) which, judging by its dimensions and weight, does not belong to it. However, after inspecting the documentation of BCM and the National Museum in Belgrade (further in the text NMB), the supposition that the steelyard beam from BCM represents a part of the same hoard with the well known counterweight (Fig. 7) in the form of a bust of a Byzantine empress, and the unpublished chain system with hooks for hanging loads (Fig. 8; Pl. IV), which are kept in NMB, was confirmed. The detailed examination of the museum inventory records shed light on the set of circumstances which led to the separation of the parts of the Belgrade steelyard. The steelyard beam (length: 61 cm, width: 2 cm) originally had two fulcrum hooks, only one of which has been preserved. It has two incised measuring scales with punched numerical weight marks (Figs. 2, 3; Pl. II) in accordance with the duodecimal system of measurement and its division into litrae (logarik» l/tra) and ounces (oÙgg/a). Judging by the marks ranging from A to N, the Early Byzantine steelyard from BCM could have been capable of weighing up to 50 litrae, that is, according to today’s measures, slightly more than 16 kg. On both sides of the shorter part of the beam of the steelyard from BCM, there is a punched inscription that begins and ends with a cross. On the lighter weight range side of the shorter part of the steelyard beam there is an inscription (Fig. 4; Pl. II/1): +PAPAKESOIKONOMOU+; whilst on its heavier weight range side the inscription (Fig. 5; Pl. II/2) reads: +KURSFILIKOS+. In the first part of the inscription +PAPAKESOIKONOMOU+ a possible ecclesiastical rank p£pa(s) is mentioned, referring to priests. What follows is KES, with OIKONOMOU at the end. KES could be read as the first three letters of a personal name such as Kšj (arios), but also as the conjunction “and”, that is ka/ which usually appears in contemporary signs abbreviated to Ks. In the final part of the inscription (OIKONOMOY), the well known function of oikonomos (o„konÒmoj) is mentioned in the genitive case. This term denoted a person, a clergyman, monk or a layman, who managed the property, income and expenses of the church seat, monastic estate or an endowment. One of his duties entailed taking care of the measuring instru- 180 STARINAR LXIV/2014 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) 1 2 Tabla I – Poluga ranovizantijskog kantara (1) i mawi teg (2) iz Beograda. Muzej grada Beograda (crte`: A. Suboti}) Plate I – 1) Early Byzantine steelyard beam from Belgrade; 2) Small steelyard counterweight from Belgrade. Belgrade City Museum (Drawing: A. Suboti}) 181 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 1 2 Tabla II – Merna skala ranovizantijskog kantara iz Beograda: 1) lak{a strana; 2) te`a strana (crte`: A. Suboti}) Plate II – Measuring scales of the Early Byzantine steelyard from Belgrade: 1) lighter weight range side; 2) heavier weight range side (Drawing: A. Suboti}) 1 2 Tabla III – Natpis sa kra}eg kraka poluge ranovizantijskog kantara iz Beograda: 1) lak{a strana; 2) te`a strana (crte`: A. Suboti}) Plate III – Inscription on the shorter lever arm of the Early Byzantine steelyard from Belgrade: 1) lighter weight range side; 2) heavier weight range side (Drawing: A. Suboti}) 182 Miroslav VUJOVI], Ranovizantijski kantar iz Beograda (161–183) STARINAR LXIV/2014 Tabla IV – Lan~ani sistem ranovizantijskog kantara iz Beograda. Narodni muzej, Beograd (crte`: A. Suboti}) Plate IV – Early Byzantine collar with two hooked chains from Belgrade, Danijelova Street. The National Museum, Belgrade (Drawing: A. Suboti}) 183 MILO[ P. SPASI], Muzej grada Beograda, Beograd ADAM N. CRNOBRWA, Narodni muzej u Beogradu, Beograd VIN^ANSKE ZDELE SA PROTOMAMA e-mail: [email protected] UDK: 903.23"634"(497) DOI: 10.2298/STA1464185S Primqeno: 20. februara 2014. Prihva}eno: 28. jula 2014. Prilog Apstrakt. – Zdele sa protomama pojavquju se od samog po~etka vin~anske kulture i u gotovo neizmewenom obliku traju do kraja neolita centralnog Balkana. Mogu}e je izdvojiti dva osnovna tipa ovih posuda: zdele sa ~etiri i zdele sa osam, odnosno ~etiri udvojene protome. Protome na obodima modelovane su u vidu antropomorfnih, zoomorfnih ili krajwe stilizovanih – {ematizovanih predstava. Na svim analiziranim zdelama potpuno izostaju fizi~ki tragovi bilo kakve upotrebe. U unutra{wosti recipijenata nisu vidqiva nikakva fizi~ka o{te}ewa koja bi upu}ivala na to da se u wima ne{to mrvilo ili drobilo, a izostaju i tragovi izlagawa posuda na vatri ili paqewa nekog sadr`aja u wima. Ako se imaju u vidu metri~ki i tehnolo{ki podaci, najverovatnijom se ~ini teza da su zdele sa protomama slu`ile za konzumirawe ili odlagawe/izlagawe neke materije prilikom nekih sasvim specifi~nih aktivnosti. Posude ovog tipa vidimo kao fizi~ku pojavnost sociokulturne prakse vin~anskih zajednica, kao indikatore odnosa u zajednici, ali i kao konceptualizaciju percepcije neolitskog sveta. Kqu~ne re~i. – zdele sa protomama, neolit, vin~anska kultura, Stubline, ikonografski obrasci, opozicione strukture. P ozni neolit centralnog Balkana obele`en je razvojem vin~anske kulture.1 U poku{aju da sumarno predstavimo {ta je to „vin~anska kultura“ parafrazira}emo obja{wewe D. Bejlija i wegovih saradnika koji termin „vin~anska kultura“ obja{wavaju kao prekomernu generalizaciju, koja je u {irokoj upotrebi kada treba pojednostaviti kompleksnost neolitskih zajednica, wihovog pona{awa i materijalne kulture na teritorijama dana{we Srbije, zapadne Bugarske i jugozapadne Rumunije.2 Tako se pod kategorijom ozna~enom kao „vin~anska kultura“ na{ao nepregledan niz razli~itih poznoneolitskih zajednica, koje, do izvesne mere, dele sli~ne oblike materijalne kulture.3 Iako se fizi~ka sli~nost materijalne kulture mo`e objasniti postojawem ili deqewem jedne op{te ideje, kanona, matrice, u stvarnosti su pojavne sli~nosti mogle imati i potpuno druga~i- 185 je sociokulturne i ideolo{ke odlike.4 U nastavku }emo pokazati kako ta sli~nost pojedinih oblika materijalne kulture mo`e predstavqati i obrazac i specifi~nost. Zdele sa protomama predstavqaju jedan od nekoliko desetina tipova vin~anskih posuda od pe~ene gline koje se ve} vi{e od jednog veka publikuju kao nalazi sa iskopavawa na vin~anskim lokalitetima. Kao i kod gotovo svih posebnih oblika vin~anske keramike (`rtvenici, prosopomorfni poklopci, zoomorfne i antropomorfne posude i dr.) dobro 1 2 3 4 Cf. Chapman 1981; Gara{anin 1951; idem 1979. Bailey, Cochrane, Zambelli 2010, 161. Crnobrnja 2012a, 163. Cf. Spasi} 2012a, 295; idem 2012b, 14. SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) STARINAR LXIV/2014 stavqaju retki nalazi ukra{eni kanelovawem, urezivawem i gla~awem. Kanelovawem se izvode linearne kose trake (Jela, T. I/16) ili naizmeni~ne kose trake koje podse}aju na takozvani parketmuster ornament (Jela, T. I/15), koji se ~esto pojavquje i na drugim tipovima zdela pozne Vin~e. Urezivawem se tako|e izvode jednostavni linearni motivi na obodu, ali i na spoqnim povr{inama zidova (Mal~a, T. III/33; Rast, T. III/34). Na obodu zdele sa Stublina se tako|e pojavquju paralelni linearni urezi, koji su postavqeni u naizmeni~nom nizu od po 4 i 6 ureza izme|u protoma. Izuzetak u ovom krajwe svedenom ornamentalnom izrazu predstavqa nalaz zdele sa protomama sa lokaliteta Beqin–Gradu{ina na obali Save, izme|u Obrenovca i [apca, ~ije su i unutra{wa i spoqna povr{ina zidova ukra{ene slo`enom kompozicijom urezanih {ahovskih poqa u kojima se nalaze naizmeni~no postavqeni kru`no punktirani ornamenti i urezane spirale (Beqin, T. II/21). Ova kompozicija se sasvim retko pojavquje u ornamentalnom repertoaru vin~anske kerami~ke produkcije.9 Jedini primeri zdele sa protomama ukra{ene tehnikom gla~awa je nalaz iz objekta 03/03 u Vin~i, ~ija je unutra{wost ukra{ena horizontalnim linijama (T. I/3), i fragment zdele iz Crnokala~ke bare (T. III/32). Finiju tipolo{ku klasifikaciju ovih zdela mogu}e je izvesti na osnovu broja, polo`aja i vrste protoma na obodu. Ote`avaju}u okolnost pri razvijawu tipologije predstavqa ~iwenica da je re~, izuzev nekoliko nalaza, o veoma fragmentovanim predmetima sa o~uvanim malim delom posude sa protomom na obodu. Rekonstrukcija izgleda ostatka posude, broja i vrste protoma se u takvim slu~ajevima ~esto ~ini plauzibilna. Generalno posma- nam je poznata ve}ina wihovih stilsko-tipolo{kih karakteristika, geografska distribucija, pa gotovo i wihovo precizno datovawe. Ipak, nakon toliko godina i daqe izostaju temeqne studije o tehnologiji izrade vin~anskih posuda i wihovom kori{}ewu, kontekstualne analize sa ostalim oblicima materijalne kulture, tuma~ewe wihove simboli~ke uloge.5 Teku}a studija }e pokazati da teze o ritualnoj ili kultnoj ulozi ovih predmeta ne moraju biti pogre{ne, ve} da moraju imati utemeqewe u opse`noj analizi svih parametara bitnih za interpretaciju wihove uloge. Povod za pisawe ovoga rada bila je zdela sa osam protoma koja je tokom istra`ivawa vin~anskog naseqa na lokalitetu Crkvine kod sela Stubline prona|ena na podu ku}e 1/2010 (Sl. 1).6 Jasno ponavqawe ikonografskog obrasca koji je do tada bio prisutan samo na zdeli otkrivenoj nekoliko godina ranije na Vin~i, kao i prisutnost u istoj ku}i jo{ nekih elemenata koji su upu}ivali na ponavqawe obrazaca vezanih za posebne radwe neolitske populacije u {irem arealu, ukazali su na neophodnost temeqnijeg prou~avawa pojavnosti posuda sa protomama u „vin~anskoj kulturi“. STILSKO-TIPOLO[KA I FUNKCIONALNA ANALIZA7 Vin~anske zdele sa protomama predstavqaju veoma ujedna~enu klasu posuda od pe~ene gline. Re~ je o zdelama izrazito koni~ne profilacije, kosih ili blago zaobqenih zidova i ravnog ili blago prstenastog dna. Izuzetak ~ini jedna zdela ~iji je recipijent, u maniru modelovawa takozvanih `rtvenika, postavqen na tri kratke noge (Vin~a, T. I/5). Pojedine zdele imaju blago zadebqan obod (Jela, T. I/15, II/19), a jedini nalaz koji ima i dve male, lateralno postavqene jezi~aste dr{ke je zdela iz objekta 03/03 sa Vin~e (T. I/3). Zdele su izra|ene od gline sa malom koli~inom neorganskih primesa. Spoqne i unutra{we povr{ine zidova su gla~ane, ponekad i polirane do visokog sjaja.8 Ve}ina nalaza pe~ena je u uslovima potpune ili nepotpune oksidacije, {to je rezultiralo dobijawem oker i okermrkih tonova zidova, a nisu retke ni zdele sa protomomama pe~ene u uslovima redukcije, tamnosive i, re|e, crne boje, kakav je slu~aj sa ve}inom nalaza sa lokaliteta Jela–[abac. Zdele su u ve}ini slu~ajeva neukra{ene, a izuzetak pred- 5 Novi milenijum u srpskoj neolitskoj arheologiji je ipak doneo zna~ajan napredak u prou~avawu navedenih aspekata vin~anske keramike – cf. Vukovi} 2013; Nikoli}, Vukovi} 2008; Por~i} 2012; Tasi} 2007. 6 Crnobrnja 2012b, 53, fig. 14. O istra`ivawima lokaliteta Crkvine–Stubline, cf. Todorovi} 1967; Simi}, Crnobrwa 2008; Crnobrnja et al. 2010. 7 Veliku zahvalnost dugujemo kolegama M. Cerovi}u, iz Narodnog muzeja u [apcu, i M. Igwatovi}u, iz Muzeja grada Beograda, na ustupqenom materijalu i dokumentaciji. 8 Tasi} 2007, Pl. II. 9 U tipologiji M. Gara{anina i S. Stankovi}a, slo`ene kompozicije {ahovskih poqa nose oznaku C3a i C4f, cf. Gara{anin, Stankovi} 1985, 29. 186 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) trano, re~ je o izuzetno stilizovanim predstavama qudskih, `ivotiwskih ili hibridnih glava. Protome su uglavnom blago izvu~ene iz oboda posude, malih su dimenzija (0,5–3 cm) i po pravilu su okrenute ka unutra{wosti posude. One mawe i {ematizovanije su modelovane s izvla~ewem maweg dela gline iz ose oboda, dok su ve}e i detaqnije predstave aplicirane na obod. Predstave su u ve}ini slu~ajeva krajwe {ematizovane, tako da gotovo nije mogu}e odrediti da li se radi o antropomorfnoj ili zoomorfnoj predstavi. Izuzetak ~ine oni nalazi gde su jasno definisani anatomski detaqi na glavi, pre svih rogovi ili veoma izdu`ene u{i, nos ili wu{ka, te detaqi na licu, koji predstavqaju ustaqene obrasce u modelovawu antropomorfnih predstava. S tim u vezi, dva nalaza sa Vin~e (T. I/6,7) i jedan sa lokaliteta Jela – Benska bara (T. I/15) jesu jedini nalazi koji se sigurno mogu definisati kao antropomorfne protome, budu}i da je manir modelovawa nosa, o~iju, pa ~ak i maske i neke vrste kape, identi~an pristupu u oblikovawu antropomorfne figuralne plastike. Sigurne zoomorfne predstave ne mogu se odrediti do odre|ene vrste, budu}i da prisustvo rogova na stilizovanoj glavi mo`e upu}ivati na nekoliko razli~itih `ivotiwskih vrsta koje su vin~anske zajednice gajile ili lovile (bik, jarac, ovan, jelen). U sasvim izuzetnim slu~ajevima je mogu}e identifikovati i neke druge vrste osim pobrojanih. Nalaz zdele iz okoline Ni{a ima ne{to detaqnije modelovanu predstavu koja se, mo`da, mo`e identifikovati s nekom vrstom ptice (T. III/33), dok jedan slu~ajan nalaz sa lokaliteta Jablanica u Mladenovcu ima protom u vidu neke vrste gmizavca na obodu. Ako se uzmu u obzir sve stilsko-tipolo{ke karakteristike, mogu}e je izdvojiti dve koherentne kategorije nalaza. Prvu ~ine zdele sa ~etiri naspramno postavqene protome. U ovoj grupi mogu}e je izdvojiti dva podtipa, i to zdele sa sve ~etiri antropomorfne predstave (Rast, T. III/34; Vin~a, T. I/6,7) i zdele sa sve ~etiri zoomorfne predstave (Vin~a, T. I/5, 8, 9; Mal~a, T. III/33). Kao kod svih zdela sa protomama, i u ovoj grupi veliki deo predstava odlazi na krajwe {ematizovane prikaze glava koje nije mogu}e identifikovati ni kao antropomorfne ni kao zoomorfne (Jela, T. I/16, 17; Rast, T. III/35; Grivac, T. II/28, 29; Gomolava, T. I/14). Jedan takav nalaz zdele sa lokaliteta Jela – Benska bara ima protomu koja odgovara hibridnim predstavama u antropomorfnoj plastici (T. I/16). Vrlo je verovatno da }e budu}i nalazi omogu}iti i izdvajawe tre}eg podtipa ove kategorije zdela sa protomama, i to zdela sa po dve zoomorfne i dve antropomorfne protome. Dok je kod prve kategorije nalaza sa ~etiri protome na obodu donekle bilo mogu}e tipovati i one predmete koji su fragmentovani, u drugoj kategoriji, koju ~ine zdele sa osam protoma na obodu, to je gotovo nemogu}e izvesti sa sigurno{}u. Zdele sa osam protoma imaju zapravo ~etiri naspramno postavqene uparene predstave. Bilo je mogu}e izdvojiti dva podtipa i to na osnovu predmeta koji su o~uvani u celosti. Prvom tipu pripada zdela iz takozvanog ritualnog seta iz Vin~e, sa po dve uparene naspramno postavqene zoomorfne i po dve uparene antropomorfne protome (T. I/3).10 Drugom podtipu pripada zdela iz ku}e 1/2010 sa Stublina, sa po ~etiri istovetne, uparene, krajwe stilizovane i {ematizovane predstave, ~ija je bli`a identifikacija nemogu}a (T. I/1).11 Ovom podtipu verovatno pripada i fragmentovana zdela sa uparenim {ematizovanim predstavama sa lokaliteta Jela – Benska bara (T. II/19).12 Fragment zdele sa dve uparene antropomorfne predstave sa istog lokaliteta (T. I/15) ostaje za sad van izvedene tipologije, budu}i da je nemogu}e pretpostaviti da li je zdela imala 2 x 2 antropomorfne i 2 x 2 zoomorfne protome, ili su sve predstave bile antropomorfne.13 Zdela sa uparenim protomama sa iskopavawa na Gorwem gradu Beogradske tvr|ave predstavqa jedinstven nalaz u ovoj kategoriji, budu}i da su dve krajwe stilizovane i {ematizovane predstave razli~itih dimenzija (T. II/25). Zna~aj tog nalaza vidimo i u ~iwenici da su nam do sada ostali potpuno nepoznati primerci zdela gde jedan par protoma ~ine dve razli~ite predstave. Kao prilog razmatrawu funkcije zdela ovog tipa, u obzir uzimamo metri~ke podatke, tehnologiju izrade i tragove upotrebe. Gotovo sve analizirane zdele sa protomama su posude malih dimenzija, ~ija visina, gotovo po pravilu, ne prelazi 10 cm, dok pre~nik oboda varira izme|u 7 cm i 12 cm. Sasvim izuzetno se pojavquju i zdele ve}ih dimenzija, 10 11 12 13 187 Tasi} 2007. Crnobrnja 2012b, 57, fig. 14. Stoji}, Cerovi} 2011, 321, T. CL/65. Stoji}, Cerovi} 2011, 315, T. CXLIV/25. STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) i najkompletniji stilsko-tipolo{ki presek posuda ovog tipa. Zdele sa protomama su pronala`ene u razli~itim kontekstima. Najvi{e nalaza poti~e iz naseobinskih konteksta – iz ru{evina nadzemnih ku}a, ispuna otpadnih jama i sli~nog (Vin~a, Jakovo, Stubline). Zdele iz otpadnih jama i nalazi koji su u drugim kontekstima deponovani kao otpad su po pravilu veoma fragmentovani, dok su primerci iz ku}a ponekad i potpuno o~uvani (Vin~a, Stubline). Ne postoji striktan obrazac u mestu deponovawa/~uvawa zdela sa protomama u ku}ama. Zdela sa osam protoma iz Vin~e prona|ena je u veoma malom nadzemnom objektu ~ija funkcija nije bila rezidencijalnog karaktera;21 nalaz sa Stublina poti~e iz nadzemnog objekta u kom su prona|eni brojne instalacije i predmeti vezani za proizvodwu, obradu i ~uvawe hrane;22 ku}u u Anzi u kojoj je prona|ena zdela sa protomama, autorka tako|e opredequje kao objekat posebne namene ili ~ak kao hram.23 S tim u vezi, oba nalaza sa novijih sistematskih iskopavawa ukazuju na to da se zdele sa protomama pojavquju u objektima ~ija osnovna funkcija nije stanovawe. Tako|e, oba pomenuta nalaza izgleda da ~ine integralni deo grupe predmeta – seta kerami~kih posuda.24 Zdela sa Gomolave predstavqa izuzetak i veoma va`an prilog ovoj studiji pre svega zato {to je u celosti o~uvana zdela, prona|ena u grobu. Re~ je o mu{kom skeletnom grobu u kome su prona|ene jo{ jedna kerami~ka posuda, bakarna narukvica i sekira od gla~anog kamena.25 Razmatrawe raspolo`ivog, jo{ uvek nedovoqnog, uzorka ide u prilog tezi da ne postoji veza izme|u odre|enog tipa zdela sa protomama i odre- kao {to je slu~aj sa nalazima sa Gorweg grada na Kalemegdanu i sa lokaliteta Jela – Benska bara kod [apca, ~iji pre~nici iznose i skoro 30 cm (T. II/20, T. II/25). U celosti o~uvane zdele sa Vin~e, Stublina i Gomolave pokazuju da odnos visine i pre~nika zdela iznosi pribli`no 1:2. U ve}ini slu~ajeva, zapremina zdela kod kojih je bila mogu}a metri~ka analiza ne prelazi 0,3 litra, dok je zapremina zdela sa Stublina i sa iskopavawa na Vin~i 2006. i znatno mawa.14 Na svim analiziranim zdelama potpuno izostaju fizi~ki tragovi bilo kakve upotrebe. U unutra{wosti recipijenata nisu vidqiva nikakva fizi~ka o{te}ewa koja bi upu}ivala na to da se u wima ne{to mrvilo ili drobilo. Tako|e, za razliku od srodnih posuda tipa `rtvenika, na spoqa{wim i unutra{wim povr{inama zdela izostaju tragovi izlagawa posuda na vatri ili paqewa nekog sadr`aja u wima.15 Ako se imaju u vidu metri~ki i tehnolo{ki podaci, najverovatnijom se ~ini teza da su zdele sa protomama slu`ile za konzumirawe ili odlagawe/izlagawe nekog sadr`aja prilikom nekih sasvim specifi~nih aktivnosti.16 U prilog tome da su okolnosti upotrebe ovakvih posuda nesvakida{we (ritualne?) aktivnosti ide i ~iwenica da sa svih pomenutih lokaliteta postoji ~itav niz kerami~kih oblika, sli~nih zapremina i dimenzija, namewenih konzumirawu hrane i te~nosti. GEOGRAFSKA DISTRIBUCIJA I HRONOLO[KI POLO@AJ ZDELA SA PROTOMAMA Zdele sa protomama pojavquju se u gotovo svim oblastima centralnog Balkana gde su `ivele zajednice vin~anske kulture (Sl. 2). Najseverniji nalaz vezujemo za naseqe na lokalitetu Roszke-Ludvar u Ma|arskoj, koje se datuje u veoma rani period vin~anske kulture.17 Kod Rasta u Rumuniji, lokaliteta koji ujedno predstavqa i jednu od najisto~nijih pozicija na kojima prepoznajemo vin~ansku materijalnu kulturu, tako|e su prona|ene zdele sa protomama.18 Najju`niji nalaz poti~e sa iskopavawa u ~uvenoj Anzi,19 dok su najzapadnije zdele sa protomama prona|ene u vin~anskim naseqima u zapadnoj Srbiji i Podriwu.20 Najve}i broj zdela sa protomama poti~e sa lokaliteta u Podunavqu i Posavini (Vin~a, Jakovo, Gomolava, Stubline, Jela). Nalazi sa tih lokaliteta ujedno predstavqaju 14 Tasi} 2007, 204. Marangou 2008. 16 Iz pojedinih zdela sa protomama se nije mogla konzumirati te~nost. Takav je slu~aj sa zdelom iz Stublina, na kojoj su uparene protome toliko gusto postavqene da iz wih nije mogu}e direktno konzumirawe te~nosti. 17 Horvath 2006, 117, Pl. II/6–10. 18 Dumitrescu 1980, Pl LXXV/96, 97, LXXVI/98. 19 Gimbutas 1974, 59, fig. 30. 20 Trbuhovi}, Vasiqevi} 1983, T. XXI–XXII. 21 Tasi} 2007, 207. 22 Crnobrnja 2012b. 23 Gimbutas 1974, 61. 24 Crnobrnja 2012b, 57; Tasi} 2007, 204, 205. 25 Petrovi} 1984, 23; Bori} 2009, fig. 36. 15 188 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) STARINAR LXIV/2014 Sl. 1. Zdela sa protomama iz ku}e 1/2010 na lokalitetu Crkvine–Stubline Fig. 1. Bowl with protoma from house 1/2010 at the site of Crkvine, Stubline |ene prakse i mesta kori{}ewa/deponovawa. Hronolo{ki posmatrano, zdele sa protomama se pojavquju od samog po~etka vin~anske kulture do wenih najpoznijih perioda. Nalazi iz Roszke-Ludvar-a, Anze i iz jame M sa Belog Brda u Vin~i pripadaju hronolo{ki najstarijim primercima.26 Nalazi iz vin~anske zemunice M sa kote 8.9–8.7 sigurno se datuju u najraniju fazu vin~anske kulture. Radiokarbonski datumi sa Vin~e pokazuju da se ovi nalazi mogu okvirno datovati u period 5310–5200. pre n. e.27 S druge strane, najpozniji nalazi koji poti~u iz objekta 03/03 sa Vin~e, objekta 1/2010 iz Stublina, sa Gorweg grada na Kalemegdanu, te iz groba 12 na Gomolavi imaju jasne asocijacije sa poznovin~anskom keramikom, ali i radiokarbonsko datovawe ukazuje na sam kraj vin~anske kulture.28 Zdele sa Jakova i Jele su tako|e veoma pozne, ali se verovatno mogu sinhronizovati sa ne{to starijom Vin~a D fazom, pre svega zahvaquju}i nalazima posuda ukra{enih urezanim trakama ispuwenim kru`nim 26 Horvath 2006, 117, Pl. II/6–10; Gimbutas 1974, 59, fig. 30; Vasi} 1936, 42. 27 Bori} 2009, 203. 28 Cf. Bori} 2009, 199. 189 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) 1) Crkvine, Stubline; 2) Belo Brdo, Vin~a; 3) Gomolava; 4) Jela, [abac; 5) Gradu{tina, Beljin; 6) Kormadin, Jakovo; 7) Kalemegdan, Belgrade; 8) Kara{, Sremski Karlovci; 9) Supska; 10) Grivac; 11) Gradac, Zloku}ani; 12) Vito{evac, Ra`anj; 13) Crnokala~ka Bara; 14) Mala~a, Ni{; 15) Rast; 16) Anza Sl. 2. Prostorna distribucija zdela sa protomama Fig. 2. Spatial distribution of bowls with protoma punktirawem.29 Trenutna evidencija pokazuje da ipak postoje hronolo{ke razlike u pojavi izdvojenih tipova zdela sa protomama. Naime, najraniji nalazi su oni sa ~etiri naspramno postavqene {ematizovane, antropomorfne ili zoomorfne protome, i taj obrazac }e se zadr`ati do kraja vin~anske kulture, dok zdele sa osam protoma za sada nisu poznate iz ranijih horizonata vin~anske kulture, pa se mo`e pretpostaviti da je obi~aj uparivawa protoma verovatno poznijeg datuma. me|u ta dva tipa posuda. Tipologija oblika vin~anskih `rtvenika je mnogo raznovrsnija nego {to je to slu~aj sa gotovo uniformisanim oblikom zdela sa protomama.31 Recipijent `rtvenika je plitak, pravougaonog, trougaonog i, re|e, kru`nog oblika . Mala zapremina recipijenta je jo{ jedna sli~nost sa pojedinim zdelama sa protomama. Koncept modelovawa antropomorfnih, zoomorfnih ili {ematizovanih protoma na obodu recipijenta je mo`da najva`nija stilska odlika obe kategorije nalaza. Kod `rtvenika su protome obi~no modelovane na DISKUSIJA Zdele sa protomama se obi~no tretiraju zajedno s drugim posebnim oblicima vin~anske keramike, ali i u vezi sa antropomorfnom i zoomorfnom plastikom. Najsrodnija tipolo{ka grupa bili bi takozvani kultni stolovi–`rtvenici–oltari.30 U morfolo{kom smislu, ne postoje direktne veze iz- 29 Bulatovi} i dr. 2010, T. I, II. Cf. Stankovi} 1986. 31 Ornamentalne kompozicije na `rtvenicima su tako|e mnogo raznovrsnije nego kod zdela sa protomama. Izvesna morfolo{ka veza vidi se samo kod jedne zdele iz Vin~e, koja je, u maniru `rtvenika, polo`ena na tri vaqkaste noge – cf. Vasi} 1936, T. LXXVIII, sl. 334; Stankovi} 1986, T. IX/12. 30 190 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) Sl. 3. Hronolo{ka distribucija zdela sa protomama na lokalitetu Belo brdo, Vin~a Fig. 3. Chronological distribution of bowls with protoma at the site of Belo Brdo, Vin~a Dugotrajna uniformnost oblika zdela sa protomama tokom vi{e od 700 godina i nepostojawe znatne regionalne diversifikacije tipova ~ini ove nalaze jedinstvenim. Ostali pobrojani oblici materijalne kulture su znatno vi{e hronolo{ki i regionalno osetqivi. Tako su brojni tipovi `rtvenika ili figurina karakteristi~ni za odre|ene periode i regije vin~anske kulture. Nave{}emo poznati tip grada~kih `rtvenika koji se izvan oblasti severoisto~ne Srbije i Pomoravqa retko kad pojavquje.34 Sli~no je i sa antropomorfnim figurinama sa nagla{enim ovnujskim rogovima, karakteristi~nim za ju`no Pomoravqe,35 ili figurinama tipa kentaur koje se retko pojavquju van uglovima recipijenta i, u zavisnosti od oblika posude, pojavquju se dve, tri ili ~etiri protome. Na `rtvenicima se tako|e pojavquju i antropomorfne i zoomorfne i {ematizovane predstave, a nisu retke ni kombinacije razli~itih vrsta predstava. Osnovnu razliku izme|u zdela sa protomama i `rtvenika predstavqa orijentacija protoma u odnosu na recipijent. Dok je kod zdela sa protomama lice iskqu~ivo orijentisano ka unutra{wosti posude, lica protoma na `rtvenicima su okrenuta ka spoqa. Jedan od retkih izuzetaka od ovog na~ina orijentisawa predstavqa nalaz `rtvenika sa lokaliteta Mali Po`arevac, ~ija je jedna, neznatno ve}a (antropomorfna?), protoma orijentisana ka spoqa, dok je druga (zoomorfna?) okrenuta ka unutra{wosti recipijenta.32 Drugi oblik vin~anske materijalne kulture u vezi s kojim bi trebalo razmatrati zdele sa protomama je antropomorfna i zoomorfna plastika. Na~in modelovawa glava na antropomorfnim i zoomorfnim figurinama ima dobre paralele u na~inu predstavqawa protoma na zdelama.33 32 Petrovi} i dr. 2009, 159, sl. 218. Gotovo identi~ne predstave glava figurina i protoma na zdelama prona|ene su i na istim lokalitetima – cf. Stoji}, Cerovi} 2011, 348/313. 34 Jovanovi} 1982; Peri} 2006, T. V/35, 36. 35 Gara{anin 1979, T. XXXI/5. 33 191 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) 0 STARINAR LXIV/2014 5m Sl. 4. Osnova ku}e 1/2010, Crkvine–Stubline: 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) ju`ni lep; 16) zdela sa protomama Fig. 4. Plan of house 1/2010, Crkvine–Stubline, left – plan of house, right – orthogonal projection: 1) wall; 2) post holes; 3) floor; 4) oven 1; 5) oven 2; 6) storage container; 7) small clay table; 8) clay millstone structure; 9) stone work-surfaces (slabs of millstones); 10) groups of pottery on the house floor; 11) group of loom weights; 12a) eastern head; 12b) western head; 12c) southern head; 13) pit 1; 14) pit 2; 15) southern daub; 16) bowl with protoma oblasti Kosova i Metohije i ju`nog Pomoravqa.36 Hronolo{ki i teritorijalno se razlikuju ~ak i dimenzije `rtvenika i figurina.37 Za figuralnu plastiku i `rtvenike se tako|e ~esto navodi da postoji jasna tendencija u opadawu kvaliteta izrade, {ematizaciji i hibridizaciji predstava u poznoj vin~anskoj kulturi. U isto vreme, zdele sa protomama se gotovo ne mewaju ni hronolo{ki ni regionalno. [ematizovane, antropomorfne i zoomorfne predstave na protomama se pojavquju i u ranoj i u poznoj Vin~i, a oblik zdela i oskudna ornamentika su tako|e veoma postojani. Jedinu specifi~nost 192 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) to da su za gradwu zidova bile kori{}ene {iroke tesane talpe, a ne uobi~ajeni pleter; iznad severnog dela ku}e uo~eni su tragovi koji ukazuju da je ona imala i gorwi eta` odnosno potkrovnu konstrukciju; supstrukcija masivne podnice gra|ena je od delova zidova ranije izgorelih ku}a; u ku}i su se nalazile dve velike pe}i kao i obiqe pokretnog materijala. Izme|u pe}i 1 i zapadnog zida ku}e, unutar mno{tva ulomaka velikih skladi{nih posuda (pitosa) koji su ~inili jednu grupu keramike (GK–17), prona|ena je i zdela sa osam protoma (Sl. 4/16). Ona je bila okrenuta otvorom nadole i nalazila se ispod masivnog sloja jako zape~enih ostataka ku}e, sa tragovima izgorelih greda i elemenata potkrovne konstrukcije. Izme|u zdele i podnice nalazio se grumen amorfnog lepa, dok su se neposredno iznad i oko we nalazili veliki delovi trbuha pitosa (Sl. 5). Sama zdelica, kao i fragmenti pitosa uz koje se nalazila, imali su intenzivne tragove sekundarnog gorewa. U neposrednoj blizini posude sa osam protoma nalazilo se jo{ nekoliko predmeta ~ija bi se pretpostavqena namena mogla, u naj{irem smislu, opredeliti kao kultna. Na oko 0,8 m severno od posude sa osam protoma, tako|e zatrpana ulomcima iste grupe pitosa (GK–17), nalazila se loptasta posuda na tri kratke noge, a na oko 1 m isto~no prona|ena je bikoni~na zdelica sli~ne zapremine kao i zdela sa osam protoma. Pri tome svakako treba imati u vidu i nalaze dva bukraniona koji su se nalazili na stubovima ispred pe}i 1.44 O eventualnim me|usobnim vezama ovih predmeta bi}e vi{e re~i u nastavku, a za sada treba podvu}i da su nalazi zdela sa osam protoma sa Vin~e i Stublina bili deo nekog ve}eg seta posuda ili drugih oblika materijalne kulture, kao {to je to slu~aj i sa opisanim kontekstima nalaza Hajd vaze i antropomorfne posude iz Vin~e. Dodatnu bi mogla predstavqati hronolo{ka osetqivost dva izdvojena tipa zdela sa protomama. Zdele sa osam, odnosno ~etiri para protoma nisu poznate u ranijim horizontima Vin~e i, kako se ~ini na osnovu sada{we evidencije, one predstavqaju odliku poznije vin~anske kulture. Istovremeno, zdele sa ~etiri protome su prisutne tokom celog trajawa vin~anske kulture. U nastavku }e biti pokazano da pojava zdela sa ~etiri uparene protome ne mora biti samo naznaka promena u stilu i na~inu modelovawa, ve} i signal neke markantnije promene sociokulturnog ili ideolo{kog diskursa. Najve}a razlika izme|u ostalih posebnih oblika vin~anske keramike i zdela sa protomama bi ipak mogla da bude ona upotrebna. Tako se u recipijentima `rtvenika ~esto nalaze tragovi paqewa ne~ega, dok zdele sa protomama nemaju nikakve tragove upotrebe. S tim u vezi, `rtvenici su naj~e{}e tuma~eni kao recipijenti u kojima je paqena neka materija koja je osloba|ala svetlost ili neki miris, {to su fizi~ko-hemijske analize i potvrdile.38 Kontekstualne asocijacije sa ostalim klasama predmeta i praksom deponovawa su sli~ne i ukazuju na neku vrstu ritualne aktivnosti. Analiza grupe posuda sa tzv. Hajd vazom i grupe sa antropomorfnom posudom sa Vin~e donela je zna~ajne rezultate koji govore u prilog ritualnog karaktera obe grupe posuda.39 Hajd vaza je prona|ena sa deset, u celosti o~uvanih, posuda u plitkoj jami, dok je antropomorfna posuda, zajedno sa jo{ 11 posuda, dva prosopomorfna poklopca, 16 minijaturnih posuda i amforom sa dve predstave lica, prona|ena u jednom nadzemnom objektu.40 Kona~na interpretacija vrste i karaktera rituala u kojima su kori{}ene antropomorfna posuda i Hajd vaza je nemogu}a, ali je i sama ~iwenica da su posude bile namewene svakodnevnoj upotrebi i da su kori{}ene i/ili deponovane zajedno sa posebnim oblicima veoma va`an pokazateq. Va`nim se ~ini i podatak da u objektu u kojem je prona|ena grupa sa antropomorfnom posudom nije postojala pe}.41 Na istom lokalitetu, jedan vek kasnije, u objektu u kojem tako|e nije bilo pe}i prona|en je ritualni set sa zdelom sa ~etiri para protoma.42 Uslovi nalaza zdele sa osam protoma sa Stublina su naizgled potpuno druga~iji.43 Zdela je prona|ena u ku}i 1/2010, dimenzija 13,10 m h 5,10 m, koja se nalazi na jugozapadnom delu centralne zone naseqa. Ku}a je bila veoma solidno gra|ena. Malobrojni o~uvani delovi zidnih platana upu}uju na 36 Tasi} 1957. Galovi} 1966. 38 Marangou, Stern 2008. 39 Nikoli}, Vukovi} 2008. 40 Za Hajd vazu vidi: Nikoli}, Vukovi} 2008, 51–58, Fig. 5; za grupu sa antropomorfnom posudom vidi: Nikoli}, Vukovi} 2008, 58–64, Fig. 8. 41 Nikoli}, Vukovi} 2008, 58, 60. 42 Tasi} 2007, 206–207. 43 Crnobrnja 2012b, 57. 44 Crnobrnja 2012b, 58–59, Fig. 16, 17, 20; Spasi} 2012a, 300, 301, Fig. 11/1, 2. 37 193 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) Sl. 5. Zdela sa protomama in situ (ku}a 1/2010, Crkvine–Stubline) Fig. 5. Bowl with protoma in situ (house 1/2010, Crkvine–Stubline) vih namena samo su slu~ajno zaostali materijalni ostaci koji su slu`ili u procesu artikulacije kompleksnih verovawa, radwi i mi{qewa. Ako razmi{qamo na taj na~in, asocijacije bi se mogle tra`iti i u nagove{tajima obrazaca pona{awa i mi{qewa qudi toga vremena. Interesantan iskorak u tom smeru na~inili su P. Racki i A. Anders, sagledavaju}i na~ine upotrebe prostora i delatnosti koje su se odvijale u dva dela poznoneolitskog naseqa u Polgar ^o{halomu: rondelu okru`enom rovovima i velikom ravnom nasequ unutar koga se rondel nalazio.49 Detaqnom analizom svih dostupnih podataka (artefakata, polo`aja i vrste analogiju u obi~aju deponovawa tih posuda vidimo i u praksi postavqawa sudova obodom ka dole, {to je slu~aj sa zdelom sa osam protoma sa Stublina ali i s antropomorfnom posudom sa Vin~e.45 Jedina zdela sa protomama bez analogije u praksi deponovawa je nalaz iz groba 12 sa Gomolave.46 Simboli~ki aspekti zdela sa protomama i{~itavaju se iz wihove pojavnosti, vrste predstava na obodu kao i iz wihove asocijacije sa ostalim oblicima materijalne kulture sa markantnim simboli~kim asocijacijama i zna~ewem. Udvojene figuralne predstave na protomama imaju analogije u konceptualizaciji parnih predstava i u drugim klasama materijala. Takav je slu~aj sa dvoglavim antropomorfnim figurinama47 i posudom sa dva plasti~no predstavqena lica.48 Me|utim, ako kao verovatnu uzmemo pretpostavku da su posude sa protomama, kao i druge klase predmeta posebne namene (`rtvenici, prosopomorfni poklopci), svoju osnovnu namenu imale u okviru kultne/obredne prakse, analogije ne bi trebalo tra`iti samo u oblicima artefakata. Predmeti tak- 45 Nikoli}, Vukovi} 2008, 60. Premala evidencija o nekropolama vin~anske kulture ograni~ava raspravu o praksi polagawa bilo kog oblika materijalne kulture u grobovima. 47 Petrovi} i dr. 2009, 81. 48 Nikoli}, Vukovi} 2008, Fig. 8/14. 49 Raczky, Anders 2010. 46 194 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) vin~anske kulture.54 Imaju}i u vidu ve} navedena razmatrawa I. Hodera, P. Rackog i A. Anders, postojawe principa binarnih parova i u vin~anskoj kulturi mo`emo smatrati o~ekivanim, a kada ih najzad i prepoznajemo, mo`emo se zapitati i zbog ~ega nismo u mogu}nosti da taj princip prepoznamo i u vi{e sfera tada{weg `ivota. Mo`da bi i suprotna orijentacija protoma na zdelama koje razmatramo (ka unutra) i istovremenim `rtvenicima (ka spoqa) tako|e mogla da bude jedna od manifestacija toga principa, i to ne samo u fizi~kom smislu ve} i u sferi obi~aja, verovawa, mi{qewa? Razmi{qawe o nameni zdela sa protomama je te`ak zadatak sa neizvesnim i dalekim ishodom, ali bi se pritom neizostavno moralo istovremeno tragati i za namenom i smislom `rtvenika kao wihovog opozita, kroz wihove fizi~ke i stilske karakteristike kao i kontekste u kojima su nala`eni. Takav na~in razmi{qawa nosi veliku prednost, jer se slika o nameni predmeta dobija iz dva izvora: iz samoga razmatranoga predmeta i iz wegovog opozita, kroz ~iju drugost mo`emo sagledati i {ta razmatrani predmet nije – da bismo dobili kompleksnije saznawe o tome {ta je sve mogao biti. Pri tome, sve vreme treba imati na umu da se zdele sa protomama pojavquju u veoma dugom vremenskom rasponu i na veoma {irokoj teritoriji, kao i da je wihova pojavnost i u vremenskom i u prostornom smislu jasno ograni~ena. One se mogu vezati, barem na osnovu do sada poznatih podataka, iskqu~ivo za pojavnost koju nazivamo vin~anskom kulturom. Prisutne su ta~no onoliko koliko i sama vin~anska kultura, pojavquju se kad i ona i zajedno sa wom i nestaju. Prve posude sa protomama okrenutim ka unutra{wosti recipijenta javqaju se gotovo sinhrono na samom po~etku vin~anske kulture, to jest poznog neolita Balkana, u sredi{tu wenog areala (Vin~a i Grivac) ali i u krajwim, rubnim zonama (Roszke-Ludvar i Anza). Ve} na po~etku pojave ovih posuda jasno je definisan obrazac wihovog fizi~kog izgleda: koni~an profil i objekata, `ivotiwskih kostiju), autori su do{li do zakqu~ka o postojawu vi{estrukih opozita izme|u praksi koje su se odvijale u dva dela toga naseqa – u rondelu i ravnom nasequ: divqe/pitomo (nalazi kostiju `ivotiwa), mu{ko/`ensko (prisustvo i na~in sahrawivawa), radijalno/linearno (orijentacije ku}a), vertikalno/horizontalno (pravci obnavqawa ku}a), gore/dole (vertikalni rast tela naspram postojawa jama i bunara), fragmentarnost/celovitost (kerami~ki nalazi), luksuzno/svakodnevno (kori{}ene sirovine, keramika). Na taj na~in je op{teprepoznat princip domus/ agrios u evropskoj praistoriji50 dobio svoju punu potvrdu upravo u nasequ Polgar ^o{halom. Sli~an princip parova binarnih opozita mo`emo naslutiti i u vin~anskoj kulturi, a putokaz u tom smeru bio nam je i raspored posebnih predmeta unutar ku}e 1/2010 u Stublinama, u kojoj je prona|ena zdelica sa osam protoma. U neposrednoj blizini nalaza zdele sa protomama nalazila se velika pe}, ispred koje su prona|ena dva velika glinena modela bukraniona. Oba su le`ala na podu ku}e licem okrenuta nadole i nalazila su se pokraj predwih uglova pe}i. Gledano iz ugla posmatra~a, levi bukranion je sasvim sumarno modelovan i kraj wega su prona|eni ostaci ro`ine gove~eta, dok je desni ne{to pa`qivije modelovan, sa jasno nazna~enim o~ima (u najosnovnijim crtama podse}a na likove vin~anskih figurina), i kraj wega su otkriveni ostaci jelewih rogova. Ikonografski obrasci sa dve razli~ite antropomorfne predstave ili kombinacije antropomorfne i zoomorfne predstave nisu nepoznanica u neolitu centralnog Balkana i okoline.51 Sli~ne predstave prisutne su na rekonstruisanom „`rtveniku“ sa vin~anskog lokaliteta u Parci.52 Ponavqawe dva ikonografska obrasca udvajawa razli~itih vrsta predstava danas je vi{e nego evidentno. S jedne strane imamo nalaz dve za sada jedine cele, zdele sa ~etiri para protoma okrenutih ka unutra{wosti posude (Vin~a i Stubline), dok, s druge strane, imamo kompoziciju od dva bukraniona (glave) koja simbolizuju dva opozitna principa (mu{ko/`ensko, divqe/ pitomo) i flankiraju kultni objekat („`rtvenik“ u Parci i pe} u Stublinama53). Prepoznavawe opozita divqe/pitomo ili mu{ko/`ensko u razli~itim parovima protoma na posudi iz Vin~e kao i u kompozicijama bukraniona iz Parce i Stublina, jedno je od prvih mesta na kojima taj princip mo`emo prepoznati i u arealu 50 Hodder 1990, 20–92. Cf. Gimbutas 1989, 121/86, 122/90, 229, 76, 77. 52 Lazarovici, Lazarovici 2006, 314–317. 53 O mogu}im kultnim aspektima koji su mogli biti vezani za pe}i u neolitskim ku}ama pisano je vi{e puta – cf. Petrovi} 2000/2001; Naumov 2010, 230, 232. 54 Cf. Spasi} 2012a, 305, 306. 51 195 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) ravan obod na kome se nalaze ~etiri naspramno raspore|ene protome okrenute ka sredi{tu recipijenta. Taj obrazac osta}e jedinstven, uz kasniju pojavu ~etiri para protoma, kroz ~itavo vreme i na ~itavom prostoru na kome su bile u upotrebi. Na Vin~i, eponimnom lokalitetu, mo`e se najkompletnije sagledati vremenski raspon u kome se javqa ova vrsta posuda (Sl. 3). Hronolo{ki najmla|e pojave ovih posuda prona|ene su u Vin~i, Crkvinama u Stublinama i na Gomolavi. Ovako dug period, od gotovo 700 godina, u kome bez znatnijih izmena (izuzimaju}i stilske karakteristike samih protoma) opstaje jedan kompleksan igonografski obrazac, ukazuje pre svega na dugove~an i nepromewen obrazac obi~aj/verovawe/obred koji se o~igledno mo`e vezati iskqu~ivo za populacije unutar vin~anske kulture. Obi~aji, verovawa i obredi koji ih prate predstavqaju osnovu identiteta i kohezioni faktor zajednica u svim epohama. Predmeti vezani za wih najsporije se mewaju i uporno zadr`avaju svoju prvobitnu i osnovnu formu, ~esto je ne mewaju}i ~ak i kada dolazi do izvesnih promena u praksama. Ovde je va`no ista}i da zdele sa protomama pokazuju naj- ve}i stepen nepromenqivosti „kanona“, ve}i ~ak i od `rtvenika i prosopomorfnih poklopaca.55 Mo`emo li se usuditi da pretpostavimo kako pred sobom imamo jasno kanonizovan materijalni obrazac kroz koji se mo`e naslutiti i jasno definisan na~in mi{qewa/verovawa jedne odre|ene populacije? Pojava ovih zdela je nejasna, prve posude s takvim ikonografskim obrascem javqaju se u samoj Vin~i kao i na rubnim zonama vin~anske kulture u okviru materijala drugih „kulturnih grupa“,56 ali wihov nagli nestanak, zajedno s nestankom vin~anske kulture, odnosno poznoneolitskog na~ina `ivota, ne treba da iznena|uje. Promena koja dolazi sa napu{tawem posledwih vin~anskih naseqa nije samo promena koja se prime}uje u materijalnoj kulturi ve}, i iznad svega, u promenama dru{tvenog sistema i organizovawa zajednice.57 Korenite promene dru{tvenog ure|ewa sredinom V milenijuma pre n. e. su o~igledno dovele i do uru{avawa i nestanka mnogih temeqnih vrednosti tada{wih zajednica, pa i samog sistema verovawa i predawa, {to je rezultiralo i nestankom iz upotrebe zdela sa protomama, jedne od najpostojanijih materijalnih manifestacija vin~anske kulture. 55 56 57 196 Cf. Stankovi} 1986, 15, 28. Horvat 2006, 117; Gimbutas 1974. Crnobrnja, in press. SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) STARINAR LXIV/2014 * 1. 2. 3. 4. Lokalitet / Site Crkvine, Stubline Crkvine, Stubline Belo Brdo, Vin~a Belo Brdo, Vin~a Kontekst / Context Pod ku}e / House floor Nepoznat / Unknown Pod ku}e / House floor Kota 4.1 m / Elevation 4.1 m Period / Period Vin~a D Vin~a D Vin~a D/D2 Vin~a D 5. Belo Brdo, Vin~a Kota 6.3 m / Elevation 6.3 m Vin~a B2 6. 7. 8. Belo Brdo, Vin~a Belo Brdo, Vin~a Belo Brdo, Vin~a Kota 6.5 m / Elevation 6.5 m Kota 6.5 m / Elevation 6.5 m Kota 7.1 m / Elevation 7.1 m Vin~a B2 Vin~a B2 Vin~a B1/B2 9. Belo Brdo, Vin~a Kota 8.0 m / Elevation 8.0 m Vin~a A/B1 10. Belo Brdo, Vin~a Vin~a A Stankovi} 1988, 43, T. X/6 11. Belo Brdo, Vin~a Kota 8.45 m /Elevation 8.45 m Zemunica M, kota 8.7–8.9 m / Dugout M, elevation 8.7-8.9 m Vin~a A Vasi} 1936, T. XX, 42 12. Belo Brdo, Vin~a Nepoznat / Unknown Vin~a C/D 13. Belo Brdo, Vin~a Nepoznat / Unknown Vin~a C/D 14. Gomolava Grob 12 / Grave 12 Vin~a D2 15. Jela, Benska Bara kv. G/13, 6. o.s. / sq. G/13, 6. e.l. Vin~a C/D 16. Jela, Benska Bara kv. K/14, 6. o.s. / sq. K/14, 6. e.l. Vin~a C/D 17. Jela, Benska Bara 18. Jela, Benska Bara nepublikovano, Muzej grada Beograda, inv. br. AV 271 nepublikovano, Muzej grada Beograda, inv. br. AV 493 Petrovi} 1984, 23, sl.15; Bori} 2009, fig. 36 Trbuhovi}, Vasiqevi} 1983, T. XXI; Stoji}, Cerovi} 2011, 315, T. CXLIV/25 Trbuhovi}, Vasiqevi} 1983, 28, I/8; Stoji}, Cerovi} 2011, 336, CLXV/191 Stoji}, Cerovi} 2011, 345, T. CLXXIV/277 Stoji}, Cerovi} 2011, 320, T. CXLIX/59 19. Jela, Benska Bara 20. Jela, Benska Bara Nepoznat / Unknown Vin~a C/D 21. Gradu{ina, Beljin Slu~ajni nalaz / Accidental find Vin~a D kv. L/16, 5. o.s. / sq. L/16, 5. e.l. kv. H/17, 2. o.s/ sq. H/17, 2. e.l. kv. H/18, 3. o.s/ sq. H/18, 3. e.l. Vin~a C/D Vin~a C/D Vin~a C/D Referenca / Reference Crnobrnja 2012, 53, fig. 14 Tasi} 1973, 70, T. XXXIV/126 Tasi} 2007 Stankovi} 1986, 44, T. X/10. Vasi} 1936, LXXXVIII, 334; Stankovi} 1988, 43, T. IX/12 Vasi} 1936, 162, sl. 339 Vasi} 1936, 162, sl. 341 Stankovi} 1988, 44, T. X/8 Vasi} 1936, T. XX, 43; Stankovi} 1988, 44, T. X/7 Stoji}, Cerovi} 2011, 321, T. CL/65 Trbuhovi}, Vasiqevi} 1983, T. XXI; Stoji}, Cerovi} 2011, 435, sl. 325 Trbuhovi}, Vasiqevi} 1983, T. XXII; Stoji}, Cerovi} 2011, 248, T. XIX/10 Bulatovi}, Kapuran i Strugar 2010, 21, T. I/14 Tasi} 1973, 67, T. XXXII/118 Tasi} 1973, 67, T. XXXII/119 nepublikovano, Muzej grada Beograda 22. Kormadin, Jakovo Horizont I / Horizon I Vin~a C/D 23. 24. Kormadin, Jakovo Kormadin, Jakovo Kalemegdan, Beograd Kara{, Sremski Karlovci Nepoznat / Unknown Nepoznat / Unknown Vin~a C/D Vin~a C/D Nepoznat / Unknown Vin~a D Nepoznat / Unknown Vin~a C Ga~i} 1990, 74–75. 27. Stublina, Supska Sloj 8 / Stratum 8 Vin~a A/B 28. Grivac 29. 30. 31. 32. 33. 34. 35. 36. 37. Grivac Gradac, Zloku}ani Vito{evac, Ra`anj Crnokala~ka Bara Radaj~e, Mal~a Rast (Rumunija) Rast (Rumunija) Rast (Rumunija) Anza (Makedonija) Sonda A, jama 2 / Sondage A, pit 2 Nepoznat / Unknown Nepoznat / Unknown Nepoznat / Unknown Nepoznat / Unknown Nepoznat / Unknown Nepoznat / Unknown Nepoznat / Unknown Nepoznat / Unknown Pod ku}e / House floor Vin~a B2/C Vin~a Vin~a C/D Vin~a Vin~a Vin~a B2/C Vin~a B2/C Vin~a B2/C Anza IVb/ Vin~a A Gara{anin, Gara{anin 1979, 33, T. XXXI/3 Bogdanovi} 1996, 30, T. VI/6; Bogdanovi} 2006, 195, Pl. VII/4 Nikoli} 2004, 246, T. 927/e Stalio 1972, 19, T. IV/58 Krsti} 1964, T. II/5 Tasi}, Tomi} 1969, 44, sl. 23 Stoji}, Joci} 2006, 135, T. LII/135 Dumitrescu 1980, Pl LXXV, 96 Dumitrescu 1980, Pl LXXV, 97 Dumitrescu 1980, Pl LXXVI, 98 Gimbutas 1974, 59, fig. 30 25. 26. Vin~a A Tabela 1. *Redni brojevi u tabeli odgovaraju rednim brojevima slika na prilo`enim tablama Table 1. *Ordinal numbers in the table correspond to those of pictures presented in plates 197 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) BIBLIOGRAFIJA: Gimbutas 1989 – M. Gimbutas, The Goddesses and Gods of Old Europe, Berkeley – Los Angeles 1989. Dumitrescu 1980 – V. Dumitrescu, The Neolithic Settlement at Rast, BAR International Series 72, Oxford 1980. Jovanovi} 1982 – B. Jovanovi}, Rudna Glava: najstarije rudarstvo bakra na centralnom Balkanu, Bor – Beograd 1982. Krsti} 1964 – D. Krsti}, Neolitsko naseqe u Vito{evcu kod Ra`wa, Zbornik Narodnog muzeja IV, 1964, 51–63. Lazarovici, Lazarovici 2006 – C. M. Lazarovici, G. Lazarovici, Arhitectura neoliticului si epocii cuprului din Romania I (Neoliticul), Iasi 2006. Marangou, Stern 2008 – C. Marangou, B. Stern, Neolithic zoomorphic vessels from Eastern Macedonia, Greece: Issues of function, Archaeometry 51, 2008, 397–412. Naumov 2010 – G. Naumov, Neolithic anthropocentrism: the principles of imagery and symbolic manifestation of corporeality in the Balkans, Documenta Praehistorica XXXVII, 2010, 227–238. Nikoli} 2004 – D. Nikoli}, Kerami~ko posu|e, u: Grivac: naselja protostar~eva~ke i vin~anske culture, (ur.) M. Bogdanovi}, Kragujevac 2004, 203–315. Nikoli}, Vukovi} 2008 – D. Nikoli}, J. Vukovi}, Vin~a ritual vessels: Archaeological context and possible meaning, Starinar LVIII, 2008, 51–69. Petrovi} 1984 – J. Petrovi}, Gomolava: arheolo{ko nalazi{te, Ruma – Novi Sad 1984. Petrovi} 2001/2002 – B. Petrovi}, Model pe}i iz Progara, Godi{wak muzeja grada Beograda 47/48, 2001/2002, 11–23. Petrovi} i dr. 2009 – B. Petrovi}, V. Kati}, M. Spasi}, @ivot u glini: neolitska umetnost na tlu Beograda – figuralna plastika iz zbirki Muzeja grada Beograda, Beograd 2009. Peri} 2006 – S. Peri}, The Gradac Period in the Neolithic Settlements in the Middle Morava Valey, in: Homage to Milutin Gara{anin, (eds.) N. Tasi}, C. Grozdanov, Belgrade 2006, 235–251. Por~i} 2012 – M. Por~i}, De facto refuse or structured deposition, Starinar LXII, 2012, 19–43. Raczky, Anders 2010 – P. Raczky, A. Anders, Activity loci and data for spatial division at a Late Neolithic sitecomplex (Polgár-Csoszhalom: a case study), in: Leben auf dem Tell als soziale Praxis, (ed.) S. Hansen, Kolloquien zur Vor- und Frühgeschichte 14, Bonn 2010, 143–163. Bailey, Cochrane, Zambelli 2010 – D. Bailey, A. Cochrane, J. Zambeli, Unearthed: a comparative study of J÷mon Dogø and Neolithic figurines, Norwich 2010. Bogdanovi} 1996 – M. Bogdanovi}, Neolitska naseqa u Grivcu, Glasnik Srpskog arheolo{kog dru{tva 11, 1996, 27–40. Bogdanovi} 2006 – M. Bogdanovi}, Early Vin~a in Central Serbia, in: Vorgi} B., Brukner B. (eds.), Current problems of the Transition Period: From Star~evo to Vin~a Culture, Zrenjanin 2006, 179–196. Bori} 2009 – D. Bori}, Absolute Dating of Metallurgical Innovations in the Vin~a Culture of the Balkans, in: T. K. Kienlin, B. W. Roberts (eds.): Metals and Societies: Studies in honour of Barbara S. Ottaway, Bonn 2009, 191–245. Bulatovi}, Jovi} 2010 – A. Bulatovi}, S. Jovi}, Leskovac: kulturna stratigrafija praistorijskih lokaliteta u leskova~koj regiji, Beograd – Leskovac 2010. Bulatovi} i dr 2010 – A. Bulatovi}, A. Kapuran, N. Strugar, Neolitski stratum na lokalitetu Kormadin u Jakovu – iskopavawa 2008. godine, Godi{wak grada Beograda LVII, 2010, 11–42. Vukovi} 2013 – J. Vukovi}, Fragmenti grn~arije kao alatke u kasnoneolitskoj Vin~i, Starinar LXIII, 2013, 191–207. Galovi} 1966 – R. Galovi}, The Monumental Prehistoric Clay Figures of the Middle Balkans, American Journal of Archaeology 70/4, 1966, 370–371. Gara{anin 1951 – M. Gara{anin, Hronologija vin~anske grupe, Ljubljana, 1951. Gara{anin 1979 – M. Gara{anin, Centralnobalkanska zona, u: Praistorija jugoslavenskih zemalja II, (ur.) A. Benac, Sarajevo 1979, 79–212. Gara{anin, Gara{anin 1979 – D. Gara{anin, M. Gara{anin, Supska „Stublina“ – praistorijsko naseqe vin~anske grupe, Beograd 1979. Gara{anin, Stankovi} 1985 – M. Gara{anin, S. Stankovi}, Razrada tipologije vin~anske grupe: prilog jedinstvenoj arheolo{koj dokumentaciji, Glasnik Srpskog arheolo{kog dru{tva 2, 1985, 10–30. Ga~i} 1990 – \. Ga~i}, Novi rezultati istra`ivawa lokaliteta Laka Staza kod Sremskih Karlovaca, Glasnik Srpskog arheolo{kog dru{tva 6, 2010, 73–76. Gimbutas 1974 – M. Gimbutas, Anza, ca. 6500–5000 B.C.: A Cultural Yardstick for the Study of Neolithic Southeast Europe, Journal of Field Archaeology Vol. 1, No 1/2, 1974, 26–66. 198 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) STARINAR LXIV/2014 Todorovi} 1967 – J. Todorovi}, Crkvine, Stubline, Obrenovac – naselje vin~anske grupe, Arheolo{ki pregled 9, 1967, 17–18. Trbuhovi}, Vasiqevi} 1983 – V. Trbuhovi}, M. Vasiqevi}, Najstarije zemqoradni~ke kulture u Podriwu, [abac 1983. Hodder 1990 – I. Hodder, The domestication of Europe, Oxford, Cambridge 1990. Crnobrnja 2012a – A. Crnobrnja, Group identities in the Central Balkan Late Neolithic, Documenta Praehistorica XXXIX, 2012, 155–165. Crnobrnja 2012b – A. Crnobrnja, Investigations of Late Vin~a House 1/2010 at Crkvine in Stubline, Starinar n.s. LXII, 2012, 45–64. Crnobrnja, in press – A. Crnobrnja, The (E)neolithic settlement Crkvine at Stubline, Serbia, in: The Transition from the Neolithic to the Eneolithic in Central and South-Eastern Europe in the Light of Recent Research, Prähistorische Archäologie Südosteuropas, (eds.) W. Schier, F. Drasovean. Crnobrnja et al. 2010 – A. Crnobrnja, Z. Simi}, M. Jankovi}, Late Vin~a Culture Settlement at Crkvine in Stubline: Household organization and urbanization in the Late Vin~a culture period, Starinar n.s. LIX, 2010, 9–25. Chapman 1981 – J. Chapman, The Vin~a Culture of South East Europe, Oxford, 1981. Simi}, Crnobrwa 2008 – Z. Simi}, A. Crnobrwa, Sonda`no iskopavawe lokaliteta Crkvine u selu Stubline, Arheolo{ki pregled n.s. 4, 2008, 44–46. Spasi} 2012a – M. Spasi}, Cattle to settle – bull to rule: on bovine iconography among Late Neolithic Vin~a culture communities, Documenta Praehistorica XXXIX, 2012, 285–309. Spasi} 2012b – M. Spasi}, Metahousing: Neolithic and Modern Dwelling in Belgrade, Belgrade 2012. Stalio 1972 – B. Stalio, Gradac: praistorijsko naseqe, Beograd 1972. Stankovi} 1986 – S. Stankovi}, @rtvenici i prosopomorfni poklopci iz Vin~e, Beograd 1986. Stoji}, Joci} 2006 – M. Stoji}, M. Joci}, Ni{ – kulturna stratigrafija praistorijskih lokaliteta u ni{koj regiji, Beograd – Ni{ 2006. Stoji}, Cerovi} 2011 – M. Stoji}, M. Cerovi}, [abac – kulturna stratigrafija praistorijskih lokaliteta u Podriwu, Beograd – [abac 2011. Tasi} 1957 – N. Tasi}, Zavr{na istra`ivanja na praistorijskom naselju kod Vala~a, Glasnik muzeja Kosova i Metohije IV–V, 1957, 11–77. Tasi} 1973 – N. Tasi}, Neolitska plastika, Beograd 1973. Tasi} 2007 – N. N. Tasi}, Ritual Pottery Set from Vin~a, Glasnik Srpskog arheolo{kog dru{tva 23, 2007, 203–210. Tasi}, Tomi} 1969 – N. Tasi}, E. Tomi}, Crnokala~ka bara: naselje star~eva~ke i vin~anske kulture, Kru{evac – Beograd 1969. 199 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) Summary: MILO[ P. SPASI], Belgrade City Museum ADAM N. CRNOBRNJA, The National Museum in Belgrade VIN^A BOWLS WITH PROTOMA Key words. – Bowls with protoma, Neolithic, Vin~a culture, Stubline, iconographic patterns, oppositional structures. and their usage lasts for exactly the same time as the Vin~a culture itself, appearing at the same time and together disappearing. The first vessels with their protoma facing the inside, appear almost synchronously at the very beginning of the Vin~a culture, that is the Late Neolithic of the Balkans, in the centre of its area (Vin~a and Grivac), but also in its furthermost peripheral areas (RöszkeLúdvár and Anza). Such a long period of almost 700 hundred years, in which one complex iconographic pattern survives without any significant changes (with the exception of the stylistic characteristics of the protoma themselves), primarily points to a long-standing and unchanged custom/belief/ritual that can evidently be associated solely with the communities within the Vin~a culture. It is important to highlight the fact that bowls with protoma show “canonic” consistency to the utmost degree, even more so than the concurrent sacrificial alters and prosopomorphic lids. Dare we assume, on the basis of all that has been said, that in front of us we have a clearly canonised material pattern through which a clearly defined way of thinking/beliefs of one distinct community can be sensed? Their abrupt disappearance, together with the disappearance of the Vin~a culture, that is to say the Late Neolithic way of living, should not be surprising. The change that comes along with the abandonment of the last Vin~a settlements is not only perceptible in the material culture, but also, and above all, in the social system and the organisation of the community. Fundamental changes in the social structure in the middle of the 5th millennium BC, evidently led to the downfall and disappearance of many deep-rooted values of the communities of that time, as well as the very system of beliefs and sagas. This resulted in the disappearance of the vessels with protoma’s utilisation, one of the most steadfast material manifestations of the Vin~a culture. Bowls with protoma represent one of several tens of types of Vin~a vessels made of baked clay which have been published as finds excavated at Vin~a sites for more than a century. We are dealing here with bowls of a markedly conical profile, with angled and slightly rounded walls and with a flat or slightly annular bottom. In most cases these bowls are not decorated, with the exception of the rare finds decorated by fluting, engraving and polishing. Their most significant peculiarity is the presence of four or eight protoma on the rim, facing the inside of the vessel. Taking into consideration all stylistic-typological characteristics, it is possible to divide them into two coherent categories of finds. The first consists of bowls with four oppositely placed protoma. Within this group it is possible to single out two subtypes, bowls with all four as anthropomorphic (Pl. I/6, 7; Pl. III/34) or all four as zoomorphic representations (Pl. I/5, 8, 9; Pl. III/33, 36). Bowls with eight protoma have actually got four pairs of oppositely placed representations. It was possible to discern two subtypes based on the objects that were completely preserved. The first subtype includes a bowl from the so-called Vin~a ritual set, with two pairs of zoomorphic and two pairs of anthropomorphic protoma, both oppositely placed (Pl. I/3). The second subtype includes a bowl from house 1/2010 from Stubline with four identical, paired, totally stylised and schematised representations, whose identification is impossible (Fig. 1; Pl. I/1). Taking into consideration the metric and technological data, the thesis that seems most possible is that the bowls with protoma served for the consumption of or the storing/displaying of the content during some quite specific activities. Bowls with protoma appear in almost all regions of the central Balkans populated by Vin~a culture communities and we do not note them in the areas of the neighbouring Late Neolithic communities. These bowls appear in a very long and clearly defined time span 200 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) 1 4 2 3 5 8 6 9 12 7 10 13 15 11 14 16 17 Tabla I – 1–2) Crkvine, Stubline; 3–13) Belo Brdo, Vin~a; 14) Gomolava; 15–17) Jela, Benska bara Plate I – 1–2) Crkvine, Stubline; 3–13) Belo Brdo, Vin~a; 14) Gomolava; 15–17) Jela, Benska Bara 201 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) 18 19 20 21 23 22 24 27 25 28 26 29 Tabla II – 18–20) Jela, Benska bara; 21) Gradu{tina, Beqin; 22–24) Kormadin, Jakovo; 25) Kalemegdan; 26) Kara{, Sremski Karlovci; 27) Stublina, Supska; 28, 29) Grivac Plate II – 18–20) Jela, Benska Bara; 21) Gradu{tina, Beljin; 22–24) Kormadin, Jakovo; 25) Kalemegdan; 26) Kara{, Sremski Karlovci; 27) Stublina, Supska; 28, 29) Grivac 202 STARINAR LXIV/2014 SPASI], CRNOBRWA, Vin~anske zdele sa protomama (185–203) 30 31 32 33 34 35 36 37 Tabla III – 30) Gradac; 31) Vito{evac; 32) Crnokala~ka bara; 33) Radaj~e; 34–36) Rast; 37) Anza Plate III – 30) Gradac; 31) Vito{evac; 32) Crnokala~ka Bara; 33) Radaj~e; 34–36) Rast; 37) Anza 203 RASTKO VASI] Archaeologisches Institut, Belgrad EIN NACHTRAG ZU DEN PBF-BÄNDEN; DIE DEN ZENTRALBALKAN BETREFFEN e-mail: [email protected] UDK: 903-034.3"63"(497) DOI: 10.2298/STA1464205V Abgegeben: Februar 20, 2014 Akzeptiert: Juli 12, 2014 Beitrag Abstract. – Der Autor betrachtet neue Bronzefunde, Sicheln, Fibeln, Nadeln und Halsringe aus dem Zentralbalkan, die seit der Publikation von vier PBF- Bänden zu diesen Themen bekannt geworden sind. Er stellt fest, dass sie gro{teils mit den vorigen Zuordnungen nach Typen und Varianten korrespondieren, anders gesagt, dass die Reihe Prähistorische Bronzefunde (PBF) ein sehr nützliches Instrument für das Studium der Metallzeiten ist. Key words. – Prahistorische Bronzefunde, Zentralbalkan, Sicheln, Fibeln, Nadeln, Halsringe. N ach etwa 30 Jahren einer intensiven und erfolgreichen Zusammenarbeit mit den entgegenkommenden und liebenswürdigen Kollegen in Frankfurt und Münster bin ich als Betroffener nicht im Stand, ein objektives Bild über das PBF-Unternehmen zu zeichnen, dennoch darf ich wenigstens sagen, dass diese Reihe der Prähistorischen Bronzefunde ein wichtiges archäologisches Monument ist. Ohne die fast 200 Bände würde eine ziemlich gro{e Lücke in unseren Kentnissen zur europäischen Bronzezeit bestehen. Es gibt aber ein Problem. Im Moment des Erscheinens eines Bandes mit dem typologisch, chronologisch und geographisch eingeordneten Fundstoff gibt es bereits nach den Phasen einer mühsamen und jahrelangen Arbeit, nach der Abgabe des Manuskripts und der Recherchen eines Redakteurs oder einer Redakteurin, nach den Manuskriptkorrekturen und dem Druck neues Material, das ausgegraben oder in Museumsdepots und Privatsammlungen entdeckt wor- den ist. Grossenteils sind diese Stücke schon bekannte Formen, die den vorher festgestellten Zuordnungen zu Typen, Varianten und einzigartigen Stücken entsprechen und das Mosaikspiel komplettieren. Manchmal kommen aber auch unerwartete Formen zum Vorschein, die Probleme bei der engültigen Lösung machen könnten. Was ist dann zu tun? Einen Anhang zu schaffen, ist sicher nicht sinnvoll, weil in der Zwischenzeit bis zu dessen Druck wiederum neues Material erscheinen wird. Und so weiter… Auch im Zentralbalkan wurden in den letzten Jahren neue und interessante Funde gemacht, die hier dem bisher publizierten Material in den vorhandenen zentralbalkanischen PBF-Bänden zugeordnet werden sollen. Es gilt, den Stand der Forschung in diesem Feld zu ergänzen. Ich präsentiere diesen Beitrag, der kein Katalog sondern mehr ein Kommentar zum Thema ist. Es ist mit bewusst, dass das keine Lösung des Problems sondern nur der Versuch ist, wenigstens etwas zu tun. * Die Arbeit resultiert aus dem Projekt: Archäologie Serbiens: Kulturidentität, Integrationsfaktoren, technologische Prozesse und die Rolle des Mittelbalkans in der Entwicklung europäischer Vorgeschichte (Nr. OI 177020), des Ministeriums für Bildung, Wissenschaft und Technologische Entwicklung der Republik Serbien. 205 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) STARINAR LXIV/2014 1 2 3 4 Abb. 1. Sicheln: 1) Veliki Vetren; 2) Kon~ulj; 3) Novi Itebej; 4) Banatska Topola Sl. 1. Srpovi: 1) Veliki Vetren; 2) Kon~uq; 3) Novi Itebej; 4) Banatska Topola SICHELN Uioara 2, 2 – Uioara 8, und 26 Fragmenten) und aus [etonje in Ostserbien5 mit 18 Sicheln (3 – Uioara 2, 2 – Uioara 4 und 13 Fragmente). Es gibt mehrere neue Einzelfunde. Besonders wichtig sind Knopfsicheln mit mittlerer Rippe, die in Zentralserbien in Veliki vetren bei Para}in (Abb. 1/1)6, in Südserbien, in Kon~ulj bei Bujanovac (Abb. 1/2)7 Im Band „Die Sicheln im Zentralbalkan“ (PBF XVIII,5) haben wir vermutet dass Sicheln im Arbeitsgebiet südlich der Save- und Donautäler relativ selten vorkommen, aber dass man neue Funde trotzdem erwarten kann. Die Sichel von Kalandria (Kilindir)1, die wahrscheinlich mit dem Norden verbunden sein dürfte, gab einen Hinweis in diese Richtung. Neue Funde bestätigen generell diese Annahme: Vier Depots mit Sicheln sind inzwischen publiziert und zwar aus Kuzmin II in Syrmien2, mit wenigen Gegenständen, darunter ein Sichel des Typs Uioara 1, aus Jarak–Ruma3, mit vier Sicheln (2 Stücke – Uioara 1, 1 Stuck – Uioara 2, 1 Stuck – Uioara 3), aus Ugrinovci in Südostsyrmien4 mit 38 Sicheln (5 – Uioara 1, 5 – 1 2 3 4 5 6 7 206 Vasi} 1994, 5, Taf. 31A. Vasi} 1996, 71, Sl. 3. Popovi} 1996. Igwatovi} 2008/9, 13–17, Br. 1–44. Jacanovi}, Radoj~i} 2003, 8–10, Sl. 7–43. Stoji} 2003, 118. Bulatovi} 2007, 192, T. LII, 65. Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) STARINAR LXIV/2014 Lofkënd), die ins 9. Jahrhundert datieren23 und ohne weiteres in irgendeiner Verbindung mit unserem Stück stehen. Aufgrund der halbrunden Fussplatte und des Bogenquerschnittes ist sie auch mit den zweischleifigen Fibeln mit rundem Fuss verbunden, die in Südserbien und Makedonien etwas später verbreitet waren24. Neulich gab es eine zufällige Entdeckung einer solchen Fibel an der westlichen Morava bei dem Dorfes [anac, bei Kru{evac25. Sie indiziert, dass dieser Typ bis Zentralserbien vorgedrungen war. Eine zweischleifige Fibel mit beschädigtem blechförmigen Fu{ von Gorno Pole bei [tip ist möglicherweise, wie schon vermutet, als eine lokale Variante in diesem Fibelkreis entstanden.26 Ein eisernes Stück mit dreieckiger Fu{platte und glattem Bügel wurde zufällig auf Crni vrh neben Jagodina (Abb. 2/4)27 und ein anderes aus Bronze in Mojsinje ausser Gräber endeckt28. Zwei eiserne Fibeln mit engen dreieckigen Fussplatten wurden im Gräberfeld in ^itluk bei Sokobanja gefunden, zusammen mit Kalaka~a-Keramik, und datieren auch spätenstens ins 8 Jahrhundert v. Chr.29 Nennenswert ist ein Lesefund aus Ostserbien mit etwa 300 eisernen Gegenständen, darunter einer Anzahl eiserner zweischleifiger Bogenfibeln mit glattem Bügel und dreieckigem Fu{ und doppelkonischen Knoten auf dem Bügel und solche sowie in Novi Itebej bei Zrenjanin im Banat (Abb. 1/3) entdeckt worden sind.8 Sie repräsentieren einigermassen die lokale Fazies mit mittlerer Rippe, die bisher aus Klenje und Nova Bingula9 bekannt war. Das Stuck von Kon~ulj stärkt die vermutete Route nach Makedonien bis zum Stück aus das Kilindir (Kalandiani). Hinzu kommen Zungensicheln aus der Umgebung von Zemun (Uioara 2)10, Aradac (Uioara 2)11 und Bordjo{, Novi Be~ej (Uioara 3)12. Zusammen mit den Zungensicheln aus den oben erwähnten Depots zeigen sie, dass die Typen Uioara 1 und 2 die zahlreichsten sind. Das stimmt mit den vorigen Ergebnissen überein. Am Ende eine Korrektur: im Depot von Banatska Topola wurden auch zwei Sichelschneiden gefunden (Abb. 1/4). Ich habe im Buch irrtümlich das Depot als sichellos notiert13. FIBELN Nach der Publikation des Bandes „Die Fibeln im Zentralbalkan“ (PBF XIV,12) wurden viele neue Funde mit Fibeln, einem der wichtigsten Typen für das Studium der Chronologie des Arbeitsgebietes, entdeckt. Meistens handelt es sich um schon bekannte Formen, doch erscheinen auch neue Varianten. Auch neue Kenntnisse über die Verbreitung einiger Typen lieferten wichtige Indizien. Eine Anzahl der spätbronzezeitlichen Typen kommt zum Vorschein, so ein Unterradl-Stück aus der Umgebung von Zemun (Abb. 2/1)14 und zwei oder drei Posamenteriefibeln von Se~anj, vielleicht von der gleichen Fundstelle „Radovanova ciglana“ (Abb. 2/2)15 wie die schon entdeckten Exemplare16. Möglicherweise handelt es sich hier um Funde aus einem zerstörten Gräberfeld und nicht aus einem Depot. Brillenfibeln sind auch zu erwähnen, so eine Fibelhälfte mit Achterschleife aus [etonje in Ostserbien17 und eine komplette Fibel aus Südmakedonien18. Ohne Achterschleife, meistens kleineren Formats gibt es sie aus Mojsinje bei ^a~ak19 und Lisi~in Dol, Marvinci20. Beide Fundorte gehören etwa dem 6. Jahrhundert v. Chr. an. Viele neue zweischleifige Bogenfibeln wurden inzwischen in ganz Serbien entdeckt21. Die grosse zweischleifige eiserne Fibel mit halbrundem Fu{ aus Mojsinje (Abb. 2/3)22 ist in einem Grab mit Kalaka~aKeramik gefunden worden, gehört also spätestens dem Anfang des 8 Jahrhunderts v. Chr. an. Sie hat Parallelen in Albanien und im Südwesten Griechenlands (Typ 8 Unpubliziert, ich danke Frau Sne`ana Marinkovi} für die Möglichkeit die Sichel hier zu veröffentlichen. 9 Vasi} 1994, Nr. 45–47. 10 Najhold 2010, 41, Sl. 04. 11 Slg. J. Bakalov, Zrenjanin. Unpubliziert. 12 Ibid. 13 Vasi} 1994, 62, Taf. 36B. 14 Najhold 2010, 50, Sl. 42. Ich danke Herrn Branko Najhold für alle Fotos aus seiner Sammlung. 15 Stoji} 2006, 24, Sl. 17, 18. 16 Vasi} 1999, Nr. 92, 93. 17 Stoji}, Jacanovi} 2008, 281, Sl. 151; T. CXX, 11. 18 Videski 2003, Br. 191. 19 Nikitovi}, Stoji}, Vasi} 2002, T. XII, 92. 20 Videski 2003, Br. 291, 292. 21 Vgl. Vasi} 1999, 48 ff. 22 Nikitovi}, Stoji}, Vasi} 2002 T. III, 1; XVI, 141. 23 Papadopulos 2010, 233–252. I danke Frau Daniela Heilmann fur den Nachweis. 24 Vasi} 1999, 54–55. 25 Unpubliziert. Feundliche Mitteilung Gordana ^a|enovi}. 26 Sanev 1999, 64, Sl. 1; Videski 2003, Br. 175. 27 Unpubliziert. Freundliche Mitteilung Milorad Stoji}. 28 Nikitovi}, Stoji}, Vasi} 2002, T. XII, 91. 29 Stoji}, Vasi} 2005, 178, Sl. 2. 207 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) mit sanduhrförmigem Fuss30. Der letzte Typ ist auch in Davidovac bei Bujanovac aus Bronze entdeckt worden31, während in Rusce in der gleichen Gegend eine zweischleifige Fibel mit gerippten Bügel aus Bronze und eine zweischleifige Fibel mit drei kugeligen Knoten und eiserner Nadel gefunden worden war32. Eine Fibel mit vierkugeligem Knoten aus Bronze und Eisen kam als Einzelfund aus Gradi{te, Praskov~e bei Kru{evac33 zu Tage. Eine einschleifige Fibel mit drei Bügelknoten kam aus Leskovac bei Zaje~ar (Abb. 2/5)34. Der sehr verbreitete zweischleifige Fibeltyp mit dem Fu{ in der Form eines Böotischen Schildes kam neulich aus Zemun (Abb. 2/6) in Syrmien35, aus Beranje bei Po`arevac (Fibelbügel)36 in Ostserbien, aus Veliki Trnovac bei Bujanovac37 (Fu{platte) in Südserbien und aus Marvinci38 in Südmakedonien zum Vorschein. Zu bemerken ist, dass es in Sudwestserbien, im Kosovo und in Metohija diese Fibeln noch nicht entdeckt worden sind. Sie sind eine sehr populäre Form und können deshalb zur Antwort auf gewisse kulturelle und auch ethnische Fragen herangezogen werden. Zweischleifige Fibeln mit viereckiger durchlochten Fu{platte, eine ganz typische Glasinacform, die sich aus den Fibeln mit Schildfu{ entwickelte, wurden im Arbeitsgebiet bisher nur aus Ra`ana bekannt39. Jetzt sind sie auch in Mojsinje40 in der Gegend von ^a~ak entdeckt worden. Auch neue einschleifige Glasinac-Fibeln mit verschiedenen Bogenformen und dreieckigen oder trapezoiden Fü{en sind im Arbeitsgebiet gefunden worden41, sie sprechen für einen Einfluss der Glasinac-Kultur auf Serbien im 6. und 5. Jahrhundert v.Chr. Das Stück von Cernica bei Gnjilane (Abb. 2/7) im Kosovo ist eine nahe Variante dieser Fibeln mit kahnförmigem Bügel42, sie wurde wahrscheinlich in einer lokalen Werkstatt gefertigt. Es ist wichtig zu erwähnen, dass diese Fibelart, und zwar jene mit Verdickungen auf dem Bügel, neulich auch in Mojsinje im ^a~ak-Gebiet43 entdeckt wurde, wo sie früher nicht bekannt waren. Der Glasinac-Einfluss in dieser Gegend ist auch durch den Fund eines typischen Glasinacschwertes in Gornja Kravarica bei Gu~a (Abb. 2/8)44, datiert in das 6. Jahrhundert v. Chr., bestätigt worden. Einschleifige Fibeln mit rechtwinkligem Fu{ vom Typ Marvinci–Gogo{u, kamen in Serbien in Poljna bei Trstenik45, Beo~i} bei Rekovac (Abb. 3/1)46 und Leskovac bei Zaje~ar (Abb. 3/2)47 vor. Zwei solche beschädigte Fibeln wurden im Fürstengrab von Pe}ka Banja im Kosovo gefunden (Abb. 3/3)48, die ich irrtümlich als einschleifige Glasinac-Fibeln mit dreieckigem Fu{ STARINAR LXIV/2014 identifiziert hatte49. In Makedonien ist die Nekropole von Lisi~in Dol bei Marvinci besonders wichtig, wo viele einschleifige Fibeln vom Typ Marvinci gefunden wurden50, darunter auch eine einschleifige Fibel mit asymetrischem Fu{51. Einige Exemplare stammen auch aus unbekannten Lokalitäten in Südmakedonien52. Auch neue Stücke vom Typ Novi Pazar sind aus Zemun (Abb. 3/4)53, aus „Serbien“54, aus Nikinci55 und aus Velika Krsna56 aufgetreten. Beschädigte Fibeln aus der Lokalität Mihajlov Ponor bei dem Dorf Miro~ in Ostserbien gehören auch zu diesem Typ57. Ein neues Exemplar des Typs mit rechtwinkligem Novi Pazar-Fu{ und Scharnierkopf wurde in Hisar bei Leskovac (Abb. 3/5) entdeckt58. Die Form ist schon aus Gu~a, Lok. 30 Vgl. Vasi} 2010, 40, Nr. 196. Bulatovi} 2007, 168, T. XLII, 2. 32 Ibid., 239, T. LXXII, 11, 12. 33 ^a|enovi} 2007, 7, T. 7, 3. 34 Unpubliziert, PrivatSlg. Freundliche Mitteilung Milorad Stoji}. Vgl. Vasi} 1999, 46–47. 35 Najhold 2010, 51, Sl. 44. 36 Stoji}, Jacanovi} 2008, 73, T. IV, 4. 37 Bulatovi} 2007, 171, T. XLIV, 5. 38 Videski 1999, T. II, 2; ders., 2003, Br. 271. 39 Vasi} 1999, 71. 40 Nikitovi}, Stoji}, Vasi} 2002, 31, T. V, 2. 41 Vasi} 1999, 89–98. 42 Vasi} 2004, 24, Fig. 5. 43 Nikitovi}, Stoji}, Vasi} 2002, T. V, 3; XIV, 173. 44 Blago Narodnog muzeja. Istra`ivanja 2002–2012, ^a~ak 2012, 18, Br. 25. 45 ^a|enovi} 2011, 37 T. XVI, 5. 46 Unpubliziert, PrivatSlg. Freundliche Mittelung Milorad Stoji}. 47 Unpubliziert, PrivatSlg. Freundliche Mitteilung Milorad Stoji}. 48 Das Foto aus der Dokumentation des Museum in Pri{tina, Belgrad. 49 Vasi} 1999, Nr. 809, 810. 50 Videski 1999, T. II, 1.3.4.6; III, 3–6; ders., 2003, Br. 184–187, 270, 272, 273. 51 Videski 2003, Br. 183. 52 Ibid., Br. 188, 189. 53 Najhold 2010, Br. 45. 54 Unpubliziert, Narodni muzej, Beograd. Freundliche Mitteilung Aca \or|evi}. 55 Unpubliziert, Slg. Milutin Mi{i}, Beograd. Vgl. Vasi} 2005, 72 Anm. 14. 56 Unpubliziert, gefunden in einem reichen Kriegersgrab. Freundliche Mitteilung Bisenija Petrovi}. 57 Jevti}, Pekovi} 2007, T. I, 7, 14. 58 Stoji} 2007, Fig. 30, 31. 31 208 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) STARINAR LXIV/2014 1 2 3 4 5 6 7 8 Abb. 2. Fibeln: 1, 6) Zemun, Umgebung; 2) Se~anj; 3) Mojsinje; 4) Crni Vrh; 5) Leskovac bei Zaje~ar; 7) Cernica bei Gnjilane; 8) Gornja Kravarica bei Gu~a Sl. 2. Fibule: 1, 6) Zemun, okolina; 2) Se~aw; 3) Mojsiwe; 4) Crni Vrh; 5) Leskovac kod Zaje~ara; 7) Cernica kod Gwilana; 8) Gorwa Kravarica kod Gu~e 209 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) STARINAR LXIV/2014 1 2 3 4 5 6 7 8 9 Abb. 3. Fibeln: 1) Beo~i}; 2) Leskovac bei Zaje~ar; 3) Pe}ka banja; 4, 7–9) Zemun, Umgebung; 5) Hisar, Leskovac; 6) Gu~a, Grotnica Sl. 3. Fibule: 1) Beo~i}; 2) Leskovac kod Zaje~ara; 3) Pe}ka bawa; 4, 7–9) Zemun, okolina; 5) Hisar, Leskovac; 6) Gu~a, Grotnica 210 STARINAR LXIV/2014 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) 1 2 3 Abb. 4. Fibeln: 1–3) Nikinci Sl. 4. Fibule: 1–3) Nikinci Ko~ani69) und mit zwei Gliedern sind bekannt geworden (Zemun – Abb. 3/970, Kr{evica71). Diese letztgenannte Variante, die meistens im Westmakedonien, in der Umgebung von Prilep gefunden worden war, ist jetzt nördlich von der Save und Donau wie auch in Südserbien vertreten. Es ist zu vermuten, dass diese Grotnica bekannt (Abb. 3/6)59, aber nach der Ausführung bestimmter Einzelheiten stammen die zwei Stücke nicht aus der gleichen Werkstatt. Wahrschenlich war die Form nicht sehr beliebt, sie wurde einzeln in verschiedenen Werkstätten unabhängig von einander hergestellt. Ein neuer Fibeltyp wurde aus Marvinci, Fundstelle Lisi~in dol, bekannt die Fibel mit einem achterschleifigen Bügel, die nach den Beifunden im Grab ins 6. Jahrhundert. v.Chr. datiert werden kann60. Einige Certosa- und Armbrustfibeln sind auch entdeckt in Zemun (Abb. 3/7) worden61, ein Grab mit zwei solchen Fibeln aus Belnja~a bei [id62 soll man besonders erwähnen. Die Certosafibeln aus der Nekropole Stubarlija bei Mo{orin63 sind inzwischen publiziert worden64. Scharnierfibeln kommen auch in grosser Zahl vor und zwar alle Varianten65. Jene mit fünf Ziergliedern am Bügel sind die zahlreichsten (Ni{ 66, Kr{evica67, Oraovica aus Silber 68, Zemun – Abb. 3/8), aber auch Fibeln mit drei Gliedern (Pilovo, Dorf Bur~ilovo bei 59 Vasi} 1999, Nr. 656. Videski 1999 T. II, 5; 2003 Sl. 277. 61 Najhold 2010, Sl. 50. 62 Koledin 2012, 108, Sl. 1, a.b. 63 Vgl. Vasi} 1999, Nr. 827–834. Zwei Fibeln von Katlanovo sind inzwischen publiziert, vgl. Ristov 1999, T. II, 9, 10. 64 Medovi} 2007, 10–19; T. III, 4–7; V, 2–5. 65 Vgl. Vasi} 1999, 102–117. 66 Vasi} 2006, 121, Fig. 1. 67 Popovi} 2012, 94, Br. 114, 68 Bulatovi} 2007, 268, T. XXXIII, 47. 69 Atanasova, Karpuzova 2006, 117, Sl. 2. 70 Najhold 2010, Sl. 49. 71 Popovi} 2012, Sl. 94, Br. 113. 60 211 STARINAR LXIV/2014 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) 4 1 2 3 5 6 Abb. 5. Nadeln: 1) Kaludjerske livade; 2) Crljenac, Glavica; 3) Aradac, Orbara; 4–6) Zemun, Umgebung Sl. 5. Igle: 1) Kalu|erske livade; 2) Crqenac, Glavica; 3) Aradac, Orbara; 4–6) Zemun, okolina Stücke aus Westmakedonien im Norden importiert wurden. Schlie{lich ist eine neue Variante der Scharnierfibeln in Syrmien entdeckt worden, die Variante Nikinci, nach demselben Fundort benannt (Abb. 4/1–3)72, die gro{ und reich verziert und aus Gold und Silber gefertigt wurde.Am wahrscheinlichsten ist sie eine Fazies lokaler Werkstätten. Sie datiert in die Zeit vor dem Einfall der Kelten. Zusammen mit der Variante ^urug stellt sie verschiedene Produkte lokaler Werkstätten für Scharnierfibeln im 4. Jahrhundert v. Chr. dar. kamen zum Vorschein, z.B. eine Petschaftkopfnadel mit geripptem Hals aus Kalu|erske livade bei Sur~in (Abb. 5/1)73, sie stammt aus einem Grab, das ich nicht vorher zu publizieren bekommen habe. Es gehört in die späte Phase der Mittelbronzezeit (BzC). Etwa gleichzeitig ist eine typische Petschatkopfnadel aus Crljenac, Fundstelle Glavica, bei Po`arevac (Abb. 5/2)74, die das Auftreten dieses Typs auch in Ostserbien bestätigt. Ein wenig später ist ein Einzelfund aus Orbara bei Aradac im Banat (Abb. 5/3)75. Die Nadel hat einen NADELN 72 Vasi} 2005, 67, Sl. 1–5; Ru{evljan, Jevti} 2006, 291–293, Fig. 4. Neue Nadelfunde, als Ergänzung zum Band „Die Nadeln im Zentralbalkan“ (PBF XIII,13), sind nicht so zahlreich wie die Fibeln. Einige bronzezeitliche Stücke 73 74 75 212 Petrovi} 2006, 100, T. XXXV, 4; 57. Stoji}, Jacanovi} 2008, 77, T. VII, 1. Slg. J. Bakalov, Zrenjanin. Unpubliziert. Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) 1 STARINAR LXIV/2014 2 Abb. 6. Halsringe: 1) Velesnica; 2) Korbovo, Obala Sl. 6. Torkvesi: 1) Velesnica; 2) Korbovo, Obala typischen bikonischen Kopf, doch die Verdickung um den Hals ist mit vertikalen Einritzungen, nicht horizontalen, verziert, was im Arbeitsgebiet selten geschieht. Aus der Spätbronzezeit sind eine Nadel mit bikonischem Kopf und eine Keulenkopfnadel, die in einem zerstörten Hügel bei Maovi, Fundstelle Bus u gaju, bei [abac76 gefunden worden, und vier Nadeln mit bikonischem Kopf sowie zwei Nadeln mit turbanartigem Kopf aus Stari Kostolac bei Po`arevac (Fundstelle unbekannt).77 Die Nadel mit konischem Kopf aus Lisi~in dol bei Marvinci78 gehört einer frühen Eisenzeitphase (8. bis 7. Jahrhundert v. Chr.) an und hat gute Parallelen in Visoi bei Bitola79. Ein kugelförmiger Nadelkopf mit geripptem Schaftansatz aus der Umgebung von Zemun (Abb. 5/4)80 hat die Parallele in dem Fürstengrab von Arareva gromila auf der Glasinac-Ebene81, er datiert in die zweiten Hälfte des 6. Jahrhunderts v. Chr. Sie spricht wieder einmal für ein „Drang nach Osten“ der Glasinac-Kultur. Doppelnadeln82 erschienen in grosser Zahl im Arbeitsgebiet, vorwiegend jene mit „M“ Kopf (Typ IV). Eine Menge stammt aus Ostserbien, wo sie bisher fast unbekannt waren (Zlotska pe}ina83, Hisar bei Leskovac84, Mihajlov ponor on Miro~85, [etonje86, Signal in Svrljig87, Kr{evica88, Fundstelle Sinja~ko polje in Sinjac bei Dimitrovgrad89). Ihr Auftreten sollte man wahrscheinlich mit verstärkten Kontakten zwischen dem Zentralbalkan und Makedonien in Verbindung bringen und deshalb meistens nach der Mitte des 5 Jahrhunderts v. Chr. datieren, obwohl es auch möglich ist, dass einige Stücke etwas älter sind. Andere Typen sind nicht so zahlreich. Eine Doppelnadel mit akzentuierten Schultern (Typ IIa) aus Zlotska Pe}ina90 kann noch dem 6. Jahrhundert angehören. Gleichzeitig ist eine Doppelnadel mit zweischleifigem Kopf und Knoten (Typ IIIb) aus Südmakedonien91, wobei zwei Doppelnadeln mit zweischleifigem Kopf (Abb. 5/5,6) aus der Umgebung von Zemun (Typ IIIa und IIId)92 wahrscheinlich ins 5. Jahrhundert zu datieren sind. Eine gehört der späteren Variante mit verdicktem Kopf (IIId) an, die auch für das Donautal charakteristisch ist. In Südmakedonien wurden auch neuerdings bronzene griechische Scheibenkopfnadeln93 entdeckt. 76 Stoji}, Cerovi} 2011, 112, Sl. 79, 80. Stoji}, Jacanovi} 2008, 260, T. LVIII, 9–14. 78 Videski 2003, Br. 199. 79 Vasi} 2003, Nr. 347, 348. 80 Najhold 2010, Sl. 90. 81 Benac, ^ovi} 1957, 79, T. XLI, 6. 82 Vgl. Vasi} 2003, 109–129. 83 Vasi} 2008, 265, Sl.1. 84 Stoji} 2007, Fifs. 25, 32. 85 Freundliche Mitteilung Milo{ Jevti}. 86 Stoji}, Jacanovi} 2008, 283, T. CXX, 14. 87 Filipovi}, Bulatovi} 2010, 76, Figs. 2, 6. 88 Popovi} 2012, 93, Br. 108, 109. Auch Webgewichte mit „M“ Nr. 98. 101. 89 Freundliche Miteilung Aleksandar Kapuran. 90 Vasi} 2008, 263, Sl. 1. 91 Videski 2003, Br. 212. 92 Najhold 2010, Sl. 37, 38. 93 Videski 2003, Br. 200–203. 77 213 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) HALSRINGE STARINAR LXIV/2014 Statuetten sind später und korrespondieren mit den Stücken von Ostrovul Mare und Orsoja99. Die Halsringgrö{e bei diesen letzten Stücken nähert sich der natürlichen, was indizieren kann, dass sie etwa früher als jene von Orsoja und Ostrovul Mare sind. In verschiedenen Werkstätten im Donautal scheint man versucht zu haben, diese Dekoration auf den Statuetten darzustellen. Neue Funde von Sicheln, Fibeln, Nadeln und Halsringen ergänzen in erfreulicher Weise unsere PBF-Ergebnisse und verbreitern Möglichkeiten für ein tieferes Studium der kulturellen aber auch ethnischen Verhältnisse in den Metallzeiten des Arbeitsgebiets. Nicht nur auf Vermutungen sondern auf Tatsachen gestützt kann man jetzt das vorher gefundene und verarbeitete Material um das neu entdeckte erweitert offerieren.100 Der Band „Die Halsringe im Zentralbalkan“ (PBF XI,7) ist vor relativ kurzer Zeit erschienen. Dennoch kommen im Arbeitsgebiet neue Stücke vor, wie zum Beispiel ein tordierter Halsring in der Umgebung von Zemun94. Auch neue Halsringdarstellungen auf anthropomorphen Figurinen sind bekannt geworden, und zwar Einzelfunde aus der Siedlung Velesnica (Abb. 6/1)95, dann aus Korbovo, Fundstelle Obala (Abb. 6/2)96, und aus der Nekropole Stubarlija bei Mo{orin97. Die Figurine von Stubarlija ist in der Struktur ähnlich jener von Gardinovci98 in der Südba~ka, sie kann etwas früher in Rahmen der späten Phase der Mittelbronzezeit datiert werden. Der dargestellte Halsring scheint nach den Details eine Tordierung gehabt haben. Die zwei andere 94 Najhold 2010, Sl. 63. Bulatovi}, Kapuran, Jawi} 2013,72, T. X, 29. 96 Krsti} 1980, Sl. 81, 2. 97 Medovi} 2007, 51, Sl. 49. 98 Vasi} 2010, 49, Abb. 5. 99 Ibid., Abb. 7. und 8. 100 Herrn Professor Bernhard Hänsel danke ich herzlich für die Hilfe bei der Übersetzung des Textes. 95 214 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) STARINAR LXIV/2014 LITERATURVERZEICHNIS: Popovi} 1996 – D. Popovi}, Prilozi prou~avawu ostava poznog bronzanog doba u Sremu, Zbornik Nar. muz. XVI/1, arheologija, 1996, 105–109. Popovi} 2012 – Popovi} et al., Centralni Balkan izme|u gr~kog i keltskog sveta. Kale–Kr{evica 2001–2011. Narodni muzej, Beograd 2012. Ristov 1999 – K. Ristov, Dva groba od Katlanovska banja, Mac. Acta Arch. 15, 1999, 113–134. Ru{evljan, Jevti} 2006 – V. Dautova Ru{evljan, M. Jevti}, Silver Jewelry of Hellenistic and Celtic type from Hrtkovci in Srem, Starinar LVI, 2006, 291–307. Sanev 1999 – G. Sanev, Novi naodi od Gorno Pole, Balabanica, [tipsko, Mac. Acta Arch. 15, 1999, 63–68. Stoji} 2003 – M. Stoji}, Veliki Vetren, Beograd 2003. Stoji} 2006 – M. Stoji}, Nekoliko novih arheolo{kih nalaza iz Srema i Banata, Rad Muzeja Vojvodine 47/48, 2006, 21–28. Stoji} 2007 – M. Stoji}, Hisar in Leskovac at the end of the Early Iron Age, Starinar LVII, 2007, 175–189. Stoji}, Cerovi} 2011 – M. Stoji}, M. Cerovi}, [abac. Arh. gra|a Srbije VII, [abac 2011. Stoji}, Jacanovi} 2008 – M. Stoji}, D. Jacanovi}, Po`arevac. Arh. gra|a Srbije IV, Beograd– Po`arevac 2008. Stoji}, Vasi} 2005 – M. Stoji}, R. Vasi}, ^itluk kod Soko Bawe. Nekropola sa inhumiranim pokojnicima iz gvozdenog doba, Kru{eva~ki zbornik 11, 2005, 177–182. Vasi} 1994 – R. Vasi}, Die Sicheln im Zentralbalkan, PBF XVIII,5, 1994. Vasi} 1996 – R. Vasi}, Bronzani srpovi u Srbiji i Vojvodini, Glasnik SAD 11, 1996, 65–75. Vasi} 1999 – R. Vasi}, Die Fibeln im Zentralbalkan, PBF XIV, 12, 1999. Vasi} 2003 – R. Vasi}, Die Nadeln im Zentralbalkan, PBF XIII,11, 2003. Vasi} 2004 – R. Vasi}, Gli Autariati, la tribù illirica più grande e più forte, nella storia e nell’archaeologia, u: Gli Illiri e L’Italia, Convegno Internazionale di Studi, Treviso, 2004. Vasi} 2005 – R. Vasi}, Srebrni nalaz iz Nikinaca, Starinar LV, 2005, 67–73. Vasi} 2006 – R. Vasi}, Connections between Serbia and Macedobnia in the 4th century BC, Folia archaeologica Balkanica 1, Skopje 2006. Vasi} 2008 – R. Vasi}, Dvojne igle opet, Zbornik MS za klasi~ne studije 10, Novi Sad 2008. Atanasova, Karpuzova 2006 – I. Atanasova, S. Karpuzova, Ranoanti~ka keramika od Pilovo – s. Buril~evo, Ko~ani, Mac. Acta Arch. 17, 2006, 115–139. Benac, ^ovi} 1957 – A. Benac, B. ^ovi}, Glasinac II, Sarajevo 1957. Bulatovi} 2007 – A. Bulatovi}, Vrawe. Arh. gra|a Srbije III, Beograd–Vrawe 2007. Bulatovi}, Kapuran, Jawi} 2013 – A. Bulatovi}, A. Kapuran, G. Jawi}, Negotin, Arh. Gra|a Srbije VIII, Negotin 2013. ^a|enovi} 2011 – G. ^a|enovi}, Gvozdeno doba u zoni stava Ju`ne i Zapadne Morave, Kru{eva~ki zbornik XV, 2011, 23–95. ^a|enovi} 2007 – G. ^a|enovi}, Praistorijske gradine u zoni stava Zapadne i Ju`ne Morave, Kru{eva~ki zbornik XII, 2007, 7–32. Filipovi}, Bulatovi} 2010 – V. Filipovi}, A. Bulatovi}, An Early Iron Age Necropolis in Eastern Serbia, Glasnik SAD 26, 2010, 73–83. Igwatovi} 2008/9 – M. Igwatovi}, Ostava kasnog bronzanog doba iz Ugrinovaca kod Beograda, Godi{wak Muzeja grada Beograda 55/56, 2008/2009, 11–53. Jacanovi}, Radoj~i} 2003 – D. Jacanovi}, N. Radoj~i}, Praistorijska ostava metalnih predmeta iz sela [etowa kod Petrovca na Mlavi, Viminacium 13/14, 2003, 7–43. Jevti}, Pekovi} 2007 – M. Jefti}, M. Pekovi}, Mihajlov Ponor on Miro~ – Tribal cult places, Starinar LVII, 2007, 191–219. Koledin 2012 – J. Koledin, O jednom novijem nalazu iz gvozdenog doba u Sremu, Starinar LXII, 2012, 107–112. Krsti} 1980 – D. Krsti}, Obala–Korbovo, Djerdapske sveske II, 1980, 101–107. Medovi} 2007 – P. Medovi}, Stubarlija. Nekropola naselja Feudvar, Novi Sad 2007. Najhold 2010 – B. Najhold, Taurunum Skordiska, Zemun 2010. Nikitovi}, Stoji}, Vasi} 2002 – L. Nikitovi}, M. Stoji}, R. Vasi}, Mojsiwe. Nekropola pod humkama iz bronzanog i gvozdenog doba, ^a~ak 2002. Papadopoulos 2010 – J. K. Papadopulos, A new type of Early Iron Age Fibula from Albania and Northwest Greece, Hesperia 79, 2010, 233–252. Petrovi} 2006 – B. Petrovi}, Kalu|erske livade. Nekropola bronzanog doba, Beograd 2006. 215 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) Vasi} 2010 – R. Vasi}, Die Halsringe im Zentralbalkan, PBF XI,7, 2010. Videski 1999 – Z. Videski, Lisi~in Dol–Marvinci, nekropola ot `eleznoto vreme, Maced. Acta Archaeologica 15, Skopje 1999. Rezime: STARINAR LXIV/2014 Videski 2003 – Z. Videski, Makedonski bronzi, Skopje 2003. RASTKO VASI], Arheolo{ki institut, Beograd PRILOG PBF SVESKAMA KOJE SE ODNOSE NA CENTRALNI BALKAN Kqu~ne re~i. – Praistorijski bronzani nalazi, Centralni Balkan, srpovi, fibule, igle, torkvesi. pada tipovima Ujoara 1 i 2, {to potvr|uje ranije rezultate istra`ivawa. Tako|e, treba napomenuti da je u monografiji pogre{no navedeno da ostava iz Ba~ke Topole ne sadr`i srpove. U woj su na|ena dva se~iva od srpova. Me|unarodna serija Praistorijski bronzani nalazi – Prähistorische Bronzefunde (PBF) – obra|uje hronolo{ki i teritorijalno razli~ite tipove praistorijskih metalnih objekata ve} preko ~etrdeset godina (do sada je objavqeno gotovo dvesta svezaka), tako da predstavqa zna~ajan izvor i istinski priru~nik za prou~avawe metalnih predmeta i metalurgije u Evropi a i van we. Sa na{e teritorije, odnosno sa Centralnog Balkana, publikovano je nekoliko kwiga, izme|u ostalog srpovi, fibule, igle i torkvesi. Priprema jedne PBF sveske za {tampu traje dugo, dok se respektivni materijal ne sakupi i prou~i, a onda po predaji rukopisa pro|e katkad jo{ dosta vremena dok se na|e redaktor, izvr{i redakcija i kwiga od{tampa. U me|uvremenu, otkrivaju se novi nalazi sa iskopavawa ili se javqaju oni zaboravqeni iz zbirki i muzeja, koji ne mogu vi{e da se ukqu~e u publikaciju. Da se prave aneksi u okviru serije, bilo bi slo`eno i nezahvalno, jer bi se opet pojavili novi nalazi dok se ti aneksi ne od{tampaju. Da bi se ne{to u~inilo u tom pravcu, ovaj prilog daje nove srpove, fibule, igle i torkvese koji su otkriveni posle objavqivawa na{ih monografija, odnosno one komade koji su nama poznati, u `eqi da se dopune znawa o tim oblicima materijalne kulture i istovremeno poka`e da PBF monografije predstavqaju ~vrstu osnovu za daqa prou~avawa, jer se najve}i broj novih komada uklapa u tipolo{ki sistem u objavqenim kwigama. Fibule Od izlaska sveske o fibulama (1999) otkriveni su brojni nalazi fibula na Centralnom Balkanu, sistematskim i divqim iskopavawima ili kopawem po privatnim zbirkama i muzejima. Jedan broj fibula pripada kasnom bronzanom dobu, ali znatno vi{e starijem gvozdenom dobu, gde se javqaju svi poznati tipovi. Najzna~ajniji nalaz je svakako velika gvozdena dvopetqasta fibula sa polukru`nom nogom, na|ena u Mojsiwu kod ^a~ka, u grobu sa keramikom tipa kalaka~a, pa je stoga mo`da najstarija dvopetqasta fibula kod nas. Poznato je vi{e dvopetqastih fibula sa razli~ito ukra{enim lukom i razli~ito oblikovanom nogom od bronze i gvo`|a, od kojih je najva`niji skupni nalaz iz Isto~ne Srbije, koji me|u tri stotine predmeta obuhvata i nekoliko desetina fibula. Uz nove dvopetqaste fibule sa nogom u obliku beotskog {tita, treba pomenuti i dvopetqaste fibule sa pravougaonom nogom i sa dva otvora, nastale iz prethodnog tipa, karakteristi~ne za Glasinac a retke u Srbiji, sada na|ene i u Mojsiwu kod ^a~ka. U nekropoli u Mojsiwu na|ene su i glasina~ke jednopetqaste fibule sa trougaonom i trapezoidnom nogom, koje nisu bile poznate do sada u ^a~anskoj regiji, dok je jedna varijanta takve fibule na|ena u Cernici na Kosovu. Nove jednopetqaste fibule sa pravougaonom nogom i dugmetom na kraju noge (tip Marvinci-Gogo{u) javqaju se u ve}em broju u Srbiji i Makedoniji i pokazuju da je ovaj tip u 6. veku bio dobro poznat u na{im krajevima. Ovde navodimo i dve takve fibule iz kne`evskih grobova u Pe}koj bawi, koje su u kwizi pogre{no pripisane glasina~kim fibula- Srpovi Sveska o srpovima, objavqena 1994. g., nije obuhvatila dve mawe ostave bronzanih predmeta, ranije otkrivene – Kuzmin 2 i Jarak-Ruma, i dve nove – Ugrinovci i [etowe. Tako|e se javqa ve}i broj pojedina~nih nalaza, od kojih su najzna~ajniji srpovi sa dugmetom i sredwim rebrom iz Novog Itebeja, Velikog Vetrena i Kon~uqa, koji ukazuju na postojawe jedne lokalne varijante srpova sa dugmetom na Centralnom Balkanu. Najve}i broj srpova sa jezi~kom pri- 216 Rastko VASI], Ein Nachtrag zu den PBF-Bänden; die den Zentralbalkan betreffen (205–217) STARINAR LXIV/2014 Gorwi deo jedne vi{eglave igle iz Zemuna ima bliske paralele na Glasincu u Ararevoj gomili i ukazuje na glasina~ke uticaje u Srbiji, iskazane ve} pojedinim tipovima fibula, u 6. i 5. veku pre n. e. Najvi{e novih nalaza odnosi se na dvojne igle, gde je konstatovano vi{e dvojnih igala sa „M“ glavom u Isto~noj Srbiji, u regiji gde do sada gotovo nisu bile poznate i koje treba, najverovatnije ve}inom, datovati posle sredine 5. veka i pripisati poja~anim uticajima sa juga na ovo podru~je. ma sa trougaonom nogom. Tip Novi Pazar, koji ih sledi, tako|e je reprezentovan novim primercima. [arnirske fibule su u me|uvremenu tako|e na|ene u ve}em broju, kako one sa pet ukrasa na luku, tako i, mawe popularne, sa dva i tri ukrasa na luku. U Sremu, oko Nikinaca, konstatovana je nova varijanta {arnirskih fibula od zlata i srebra ve}ih dimenzija, koja je najverovatnije proizvod doma}ih radionica. Igle Kwiga o iglama objavqena je 2003. godine, a u me|uvremenu je otkriveno vi{e novih primeraka koji pripadaju raznim tipovima i periodima. Me|u nekoliko igala iz sredweg bronzanog doba isti~u se dve igle sa zadebqalim vratom iz Srema i Banata, i jedna igla sa pe~atastom glavom iz isto~ne Srbije, koja pokazuje da je ovaj tip bio ra{iren i na istoku na{e oblasti. Torkvesi Kwiga o torkvesima je iza{la 2010. Novih torkvesa ima, naro~ito onih tordiranih, ali su tako|e zanimqivi novi nalazi `utobrdskih statueta sa jasno nazna~enim torkvesima iz Mo{orina, Korbova i Velesnice. 217 VUJADIN IVANI[EVI], Institute of Archaeology, Belgrade SONJA STAMENKOVI], Institute of Archaeology, Belgrade LATE ROMAN FORTIFICATIONS IN THE LESKOVAC BASIN IN RELATION TO URBAN CENTRES UDK: 904:623.1"652"(497.11) ; 904:711.42"653"(497.11) DOI: 10.2298/STA1464219I e-mail: [email protected] Received: February 17, 2014 Accepted: July 23, 2014 Short communication Abstract. – Due to a general insecurity and the need to protect the population and communications, towards the end of the 4th century a large number of Late Roman fortifications were built in the region of the Leskovac basin, mainly towards the edges. Their distribution was determined by the level of the region’s population density, its resources and by the need to control the roads. These were predominantly smaller fortifications whose primary role was the protection of the local population, who lived off the land and bred cattle. However, the largest number of these is in the western part of the basin, in the mountainous regions of Goljak, Majdan, Radan and Pasja~a, whilst the highest density of fortifications is in the Banjska Reka valley, around the village of Sijarina. The whole region was known for its mining activity in previous centuries. A particular group comprises the fortifications around Cari~in Grad – Justiniana Prima, whose main role was the defence of the access to the city. Key words. – Leskovac basin, Late Antiquity, fortifications, economy, Cari~in Grad, Justiniana Prima. T he central part of the Leskovac basin comprises the South Morava river valley and the valleys of its tributaries, the Pusta Reka, Jablanica, Veternica and the Vlasina, and is bordered by mountains which surround the entire area from the western, southern, eastern and, to a degree, the northern side. The mountain ranges of Radan and Pasja~a stretch to the west, Goljak and Kukavica to the south, Bukovik, Kru{evica, Babi~ka Gora and Seli~evica to the east and Dobri~ to the north, the peaks of which do not exceed a height of 1,500 m. The basin covers an area of 2,250 km². The region of the Leskovac basin was only Romanised in the 1st century AD. Its settlement followed two basic routes – the occupation of the river terraces in the valleys of the South Morava, Veternica, Jablanica and the Pusta Reka and the settlement of the mountainous regions surrounding the mining areas. The contact zones between these regions were less densely populated, judging by the distribution of the sites in the area of the basin. The highest intensity of settlement was noted around the river valleys, with the highest concentration around the South Morava.1 In this area, at the entrance to the Grdeli~ka Klisura (Grdelica Gorge) in Mala Kopa{nica, was the largest Roman settlement in this area, covering approximately 20 ha. The only vicus with an unknown name was noted on a votive relief dedicated to Liber from the village of Pusto [ilovo.2 1 2 Stamenkovi} 2013, 36. Stamenkovi} 2013, 54–55, sl. 40. * The article results from the project: Urbanisation Processes and Development of Mediaeval Society (no. 177021), funded by the Ministry of Education, Science and Technological Development of the Republic of Serbia. 219 IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 1. Gornji Gajtan, Sokolov Vis; 2. Miljkovica; 3. Pesti{–Bukoloram; 4. Mali Kamen; 5. Gornji Statovac, Mlinarov Kr{; 6. Bu~ince; 7. Arbana{ka; 8. Rgaje, Grad; 9. Velika Braina, Brainski vis; 10. Gubavce, Gradina; 11. Mrkonje, Mrkonjski Vis; 12. Ma}edonce, Crkvi{te; 13. Vrapce, on the bank of the Tularska Reke; 14. Sijarinska Banja; 15. Sijarina, Gradina I; 16. Sijarina, Gradina II; 17. Svirce; 18. Zbe`i{te, Kopriva; 19. Zbe`i{te, Skobalji} Grad; 20. Pade`; 21. Gradi{te, Gradac; 22. Samarnica, Visoki Mori~; 23. Dedina Bara; 24. Novo Selo, Leskova Padina; 25. Grdelica, Kale; 26. Svo|e, confluence of the Lu`nica and the Vlasina; 27. Zlati}evo, Prokop–Rimski Grad; 28. Gornji Prisjan; 29. Gornja Loko{nica, Gradi{te; 30. Golema Njiva, Gradi{te; 31. Golema Njiva, Le{je; 32. Stupnica, ^ukar; 33. Crna Bara, Gradac or Rsa; 34. Grada{nica, Gradac or Gradi{te; 35. Bregovina, Kale; 36. Marino Kale; 37. Braj{orski Vis; 38. Lece, Kuline; 39. [tulac, St. Elias; 40. Prekop~elica, Jezero; 41. Sekicol, Gradi{te; 42. Svinjarica, Gornje Gradi{te; 43. Rujkovac, Kaljaja; 44. confluence of the Zabr|ski Potok and the Lapa{tica; 45. Popovac, Gradi{te; 46. Lapotince, south–east of the village; 47. Gornje Brijanje, Kale; 48. Me|a, Gradi{te; 49. Me|a, Kuli{te; 50. Kutle{, [iljegarnik; 51. Zlata, Kale; 52. Lipovica, Kuli{te; 53. Priboj, Gradi{te; 54. Leskovac, Hisar; 55. Konopnica, Gradac; 56. Konopnica, Latinsko (Cigansko) Groblje or Ciganski ^ukar. Map 1. The Late Roman fortifications in the Leskovac basin Karta 1. Kasnoanti~ka utvr|ewa u Leskova~koj kotlini The whole area gravitated towards the distant Naissus, the closest key city, with most of this area officially belonging to the territory of Naissus.3 Towards the end of the 4th century, as a result of the barbarian invasions and their subsequent settling of the region of northern Illyricum, significant socio-economic turmoil started to occur, which was particularly reflected in the changing roles of the settlements. Roman lowland settlements gave way to new fortified settlements on the dominant and well-guarded higher grounds above the river valleys or in the mountainous regions. Due to a general insecurity and the need to protect the population and communications in the area of the Leskovac basin, a large number of Late Roman fortifi- cations were built, mainly on the edges of the basin. Fortifications were erected on Radan along the river basins of the Lepa{tica, Lecka and Gazdarska Reka, on Goljak in the area around the rivers Tularska and Banjska Reka, around Cari~in Grad, in the area around Rujkovac and Radinovac, and in the valleys of the [umanska Reka, Pusta Reka, Jablanica and Veternica. The slopes of Kukavica, the South Morava river valley, the complex around the Kozara~ka Reka, that is the Rupska Reka, the lower course of the Vlasina, the 3 220 Petrovi} 1976, 89. IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 Map 2. The analysis of the spatial distribution of the Late Roman fortifications in the Leskovac basin Karta 2. Analiza prostornog raspored kasnoanti~kih utvr|ewa u Leskova~koj kotlini slopes of Babi~ka Gora and Kru{evica, and the confluence of the Lu`nica and the Vlasina, were also fortified.4 Most of these fortifications were built on the higher grounds that dominate the wider surroundings (Map 1). Chronological determination of the fortifications, based on the archaeological and numismatic material, was possible to a certain degree. Coins from the 4th century were discovered within a large number of fortifications such as those in Lece, Rujkovac, on Hisar in Leskovac, Stupnica, Grdelica and Skobalji} Grad near Vu~je. The circulation of coins in the 5th century was testified to in Rujkovac, which represents important proof of its survival during these turbulent times.5 The dating of the fortifications into the 6th century is not only supported by the coin finds but also by other archaeological material, primarily ceramic vessels.6 Into this period were also classified most of the fortifications located in Sijarina, Mrkonje, on Mali Kamen, in Rujkovac, Gornje Brijanje, Gornja Loko{nica, Grada{nica, Konopnica, Gradi{te, on Hisar in Leskovac and Grdelica. In a small number of fortifications, mainly those where archaeological excavations have been carried out, such as in Lece, Sijarinska Banja, Rujkovac, on Hisar in Leskovac, Konopnica and Grdelica, horizons from the 4th and 6th centuries have been confirmed. The anthropogeographic features of the area, the economic resources in particular, played a significant role in the settlement of the Leskovac basin. Most of all, the river valleys stand out, the South Morava in particular, as well as the Veternica, Jablanica, Pusta Reka and Toplica, as being suitable for agricultural production and wheat cultivation. It is worth remembering that annona represented the foundation of the economy for urban and rural populations.7 On the other hand, the foot hills on the margins of the Leskovac basin were suited to cattle breeding. Viticulture was also present as one of the most important cultures of that time, judging by the finds of stone winepresses from Vrbovac,8 in the vicinity of Cari~in Grad and grape seeds at Cari~in 4 5 6 7 8 221 Stamenkovi} 2013, 23, map 6. Ivani{evi}, Stamenkovi} 2010, 59–84. Stamenkovi} 2013, 119–122. Morrisson, Sodini, 2002, 196. Stamenkovi} 2013, 58–59, sl. 42. IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) Grad. It is quite certain that in this area, as indicated by the numerous remains of different kinds of grain and fruit at Cari~in Grad, other agricultural produce was also grown. In Roman times this territory was distinctive for its mining, the traces of which were confirmed in the western and north-western parts of the basin, on Goljak, Radan, Majdan, Pasja~a, in the wider area of Lece, in the region around the Banjska Reka, in Ravna Banja and Marovac, as well as in the eastern and north-eastern parts of the basin, in the region of Ruplje and on the slopes of Babi~ka Gora and Kru{evica.9 As evidence of mining, mine shafts, smelting complexes and slag heaps are singled out. Roman mine shafts were recorded on Rasova~a, which is well-known for deposits of gold, sulphides of lead, copper and silver, and semi-precious stones (amethyst, opal and agate).10 (Fig. 1) They are also testified to in Ravna Banja and Marovac, where remnants of a smelting complex and a slag heap have been confirmed. In Roman times, mines in the far east of the basin were also in use, in the area of Ruplje, where silver, lead and gold were mined, as well as in the region of Crna Trava, well known for iron ore. Not far from here, important material proof of ore processing was found,11 in the form of an iron ingot from the Visoki Mori~ site in Samarnica.12 In addition to the obvious traces of Roman mining, a significant contribution to the evidence of this activity are the epigraphic monuments found in the area of the Leskovac basin, given that the most frequently occurring deities on them are Hercules and Liber who have been attributed with, among other things, the role of protecting mines and mining activity.13 The distribution of the Late Roman fortifications, as is the case with the Roman settlements, was determined by the level of the region’s population density, its resources and the need to control the roadways. Analysis of the position of the fortifications indicates a concentration of forts on the higher grounds which surround the river valleys of the South Morava and its tributaries. Their highest density is found on the slopes of Dobra Gora, situated between the valleys of the Pusta Reka, Jablanica and the South Morava. These are mainly smaller fortifications whose primary role was to protect the local population, which was most certainly engaged in agriculture. However, the largest number of these forts is located in the western part of the basin, in the mountainous regions of Goljak, Majdan, Radan and Pasja~a. The highest density of the fortifications is in the valley of the Banjska Reka, around Sijarina.14 (Map 2) STARINAR LXIV/2014 Fig. 1. Rasova~a – Roman mine shaft Sl. 1. Rasova~a – Rimsko okno The entire area was well known for mining in previous centuries. The re-establishment of mining activity during the 6th century should not be excluded as a possibility. A separate group is comprised of a smaller cluster of fortifications located around Cari~in Grad – Justiniana Prima, whose predominant role was to guard the access to the city.15 Justiniana Prima represents the only city – pÒlij erected in around 535 in the centre of this region, as the seat of the archbishop ('Illurièn £rciereÚj) with 9 Stamenkovi} 2013, 86–96. Pe{ut 1976, 33. 11 Stamenkovi} 2013, 66–67, 86–89. 12 Stamenkovi} 2013, 88, sl. 75. 13 Du{ani} 1999, 130–135; Stamenkovi} 2013, 95–98. 14 Stamenkovi} 2013, map 5. 15 Kondi}, Popovi} 1977; Bavant, Ivani{evi} 2003. 10 222 IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 of enemy incursions, assumed the role of defenders – soldiers. From a previous soldier – peasant category, from the time of Constantine, a new peasant – soldier category appeared in the 6th century.24 Procopius himself states that peasants from around the forts, at the first sign of approaching enemies, switched to their role of soldiers as required. Due to their inexperience, as Procopius says, they were easy prey for their enemies.25 The economic activities of some castella were connected to agriculture, which particularly applies to those fortifications situated along the plains and valleys. This is indicated by the names of castella in some areas which, according to M. Mirkovi}, bear the names of former landowners: Tim/ana – Timiana, OÙrbr/ana – Urbriana, Kassia – Cassia.26 Likewise, the names of some fortifications from Procopius’ listing point to the places connected with mining activities and ore processing ’Erar/a – Eraria in the vicinity of Naissus, Frerrar/a – Frerraria and D£lmataj – Dalmatae in the area of Remesiana, and ’Argšntarej – Argentares in the area of Ad Aquas.27 The castella certainly had a military character, particularly those situated on roadways, as in the fortification in Rujkovac. This fortification, built as far back as the 4th century BC, also had a significant role in Roman times judging by the numerous finds of Roman coins from the 2nd all the way to the 6th century. The fortification itself was occupied again in the 9th century, indicated by the find of a follis of Leo VI.28 The second fortification at Zlata, based on its size and imposing dam, represented an important centre.29 It is possible that it was a town – polihnion. This fortification, like the one at Bregovina, could have been jurisdiction over provinces of both the Dacias, Moesia I, Dardania, Praevalitana, Macedonia II and the region of Bassiana in Pannonia II. In addition, it was anticipated that the city was to become the seat of the Illyrian prefecture.16 This most likely did not happen since, in a later novella, only the ecclesiastical jurisdiction over the diocese of Dacia was confirmed.17 Procopius of Caesarea, who gave us the most complete data about our region, also mentions Justiniana Prima as a city – pÒlij. The lower category of a settlement, according to this Byzantine chronicler, is policn/on – town. Aquae – ’Akušj pol/cnion had this status. Next in rank is frour/on – castellum (fortress). Most of the fortifications in our region belong exactly to this category – a castellum guarded by ramparts and other defensive mechanisms. By this term Procopius implies two types of castella, those with a military garrison (stratiwtèn froura/), and those functioning as refugia – shelters for the population (šrÚmata).18 In them, the bishop could have resided, as is the case with Meridio which had the status of phrourion.19 At the bottom of the scale was cwr/on – village. In describing the construction of Justiniana Prima, Procopius gives an example of the transformation of the village of Taurisium, where Emperor Justinian was born, into a castellum: “Having hastily encircled this village with a wall of a quadrangular shape and placing a tower at each corner, he made a castellum with four towers (Tetrapurg/an), and thus he named it.”20 The distinction between the meanings of these terms is important for understanding the relationship between a fortification and a city. Procopius, in De aedificiis, outlines the administrative and territorial organisation. This chronicler grouped most of the restored and newly built castella according to provinces. On the other hand, he grouped the castella of two Illyrian provinces, both Dacias, according to urban centres or regions. Novella XI, in which it is stated: Aquensis autem episcopus habeat praefatam civitatem et omnia eius castella et territoria et ecclesias,21 also indicates that the city is at the centre of the territorial organisation. Castella, village territories and churches, that is church estates, were governed by the city. This provision, according to Gilbert Dagron, portrays a society grounded in the military, peasantry and the church, with a city at its core.22 The fortifications in the Leskovac basin belong to the category of castella – phrouria, whose main role was to protect the local population.23 The population in its most part consisted of peasants who, in the event 16 Nov. 11 (14th April 535). Nov. 131, 3 (545). 18 Proc. De aedif. IV, 1, 6; 1, 33; 2, 13–14; 2, 28. 19 Proc. De aedif. IV, 4. 20 Proc. De aedif. IV,1.17–27. 21 Nov. 11 (14th April 535). 22 Dagron 1984, 9–10; Ducellier 1985, 123–126. 23 Decker 2006, 508. 24 Mac Mullen 1963, 14–22. 25 Proc. De aedif. IV,2.11–16. 26 Mirkovi} 1996, 63. 27 Mirkovi} 1996, 63, 68–71. 28 Ivani{evi}, Stamenkovi} 2010, 59–84; Stamenkovi} 2013, 90–92, sl. 78. 29 Evans 1883, 157–160; Kanic 1985, 327–330; Milinkovi} 2007, 193–199. 17 223 IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 Map 3. Late Roman towns: 1) Theoretical delineation according to Thiessen’s polygons; 2) Relation to the Late Roman fortifications based on a 15 km radius of movement Karta 3. Kasnoanti~ki gradovi: 1) Teoretsko razgrani~ewe prema Tisenovom modelu; 2) Odnos prema kasnoanti~kim utvr|ewima na osnovu radijusa kretawa od 15 km linked to mining, since evidence of mining is located on the slopes of Rasova~a. Large amounts of slag at Glasovik stand out in particular. The castellum at Bregovina stands out with its basilica with rich architectural plastic which, up until now, has not been seen in other fortifications.30 (Fig. 2) Undoubtedly, it is the work of a wealthy benefactor, who could have acquired his wealth from ore extraction and processing. There are numerous examples of castella built by private individuals. Let us mention the case of the fortification at Androna in Syria, built between 558 and 559, under the patronage of Thomas, a rich individual of note.31 The aforementioned fortifications were situated along the local roads or in their vicinity. These together with the other forts, arranged along the river valleys cut deep into the wide mountain ranges, protected the main local communications. These valleys are those of the Jablanica and Tularska Reka, through which the road connecting the provinces of Inner Dacia and Dardania passed. The valley of the Pusta Reka represented an important communication link, not only connecting Naissus and Justiniana Prima, but also the central part of the Leskovac basin with Hammeum, and further to the north. A section of the local network also passed through the valley of Vlasina further to the east, towards Remesiana. The main roadway passed through the South Morava river valley. (Map 1) In the region of the Leskovac basin there were also villages in the river valleys, evident by the rare finds of Early Byzantine coins in Re~ica, Turjane, Rafuna (Crkvena Livada site), Lipovica, Ora{ac (Padina site) and Rujkovac (Vaskina Porta). These are coin finds which can be dated to the period of the reigns of Justin I and Justinian I. What is particularly indicative are the finds of a solidus of Justin I in Re~ica and a tremissis of Justinian I in Turjane.32 The cessation of coin circu- 30 31 32 224 Jeremi}, Milinkovi} 1995, 209–225; Jeremi} 2004, 111–137. Decker 2006, 511. Stamenkovi} 2013, kat. br. 24, 29, 45, 107, 131, 207. IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) lation in the second half of the 6th century in lowland parts of the basin points to the fact that the population, by and large, abandoned the valleys and retreated to hilltop fortifications. This corresponds with the intensified building of structures within the empty areas of the city and porticos at Cari~in Grad. The incessant incursions of the Kutrigurs, Slavs and Avars most certainly contributed to the reduction of the population and degradation of the economy in the region of central Illyricum. Undoubtedly, along with the barbarian raids, some other factors influenced large changes in the Late Roman society, such as a large scale plague epidemic33 and climate changes, prolonged spells of cold in particular. Marcellinus Comes noted the plague epidemic in 543, which spread across Italy and Illyricum.34 It was a large scale epidemic which originated in Egypt in 541 and spread throughout the Mediterranean until 544. Evidence of this epidemic was not only recorded by chroniclers, but also by numerous epigraphic monuments, especially in Rome.35 The gravestone of Petrus, son of Thomas the vicar, buried in Naissus, provides direct testimony to the plague epidemic in the region of central Illyricum. In the epitaph it is stated that the sisters and two sons of Thomas the vicar died within a short time of each other – in uno Mensa simul vita(m) finirunt.36 It is important to note that in this inscription, along with the Christian names of Petrus and Thomas, an Illyrian name of Gentio (variant of Gentius) is also mentioned.37 The plague most definitely left its mark on the population of this region. Contributing factors to its cyclical occurrence in the second half of the century were the prolonged spells of cold weather and subsequent hunger, which additionally reduced the population.38 The territory of Naissus, which Procopius singles out as a separate area – ÚpÕ pÒlin b_ [Na=su~]39– undoubtedly included the northern parts of the Leskovac basin, and the possibility exists that the jurisdiction of this city also extended further to the south, especially along the South Morava valley and to the west, towards the mining areas. Likewise, the region of Remesiana – šn cèrv 'Remesianis/v40 – could have included the eastern parts of the Leskovac basin, particularly those linked to mining production. A more precise determination of the territory of the cities is not possible, since a large number of fortifications have not been identified. An idea of the theoretical delineation between cities is provided by the spatial analysis according to Thiessen’s polygons. (Map 3.1) STARINAR LXIV/2014 Which parts of this region were under the jurisdiction of this newly founded metropolis of Justiniana Prima is not known. The city could have had an important administrative role judging by the large number of lead seals found in recent years. The new reorganisation was undoubtedly important for the functioning of the economy and, as will be seen later, could have been determined by the great distances between the castella and villages situated in the central part of the Leskovac basin, and the old urban centres. On the other hand, this new city, the endowment of the Emperor, required resources. One of the significant elements of the economic life of an ancient society were market days/village fairs (nundinae), which had a significant role in supplying both the urban and rural population.41 In urban areas, periodical market days were held where the townspeople could buy produce. The residents of nearby villages sold their goods there and thus obtained much needed money for rent or taxes, as well as for buying required goods or services.42 The Theodosian Codex points to the importance of selling produce by stating that peasants were freed from paying lustral tax if they sold produce from their own farms.43 Products bought for farming were exempt from the same taxes.44 These regulations clearly demonstrate the importance of sustaining agricultural production on both, large and small estates. The decrees of Emperor Justinian I also stated this view. In Novella XXXII, addressed to Dominicius, the praetorian prefect of Illyricum, a series of decrees were passed to guard against the greed of creditors who took pawned land from peasants who were unable to repay, due to poor harvests, the loan of crop seeds.45 In certain areas itinerant traders also had a significant role.46 33 34 35 36 37 38 39 40 41 42 43 44 45 46 225 Grmek 1998, 787–794. Marcellinus Comes, Chronicle, 107. Stathakopoulos 2006, 102. Vuli} 1934, 47–48, br. 38; PLRE 3, 1320 – Thomas 22. IMS IV, 92, no. 51. Stathakopoulos 2006, 102–103. Proc. De aedif. IV.4. Proc. De aedif. IV.4. Shaw 1981, 43. Choi 2005, 7–25; Papaconstantinou 2012, 412. Cod. theod. 13.1.3, 10, 12. Cod. theod. 4.13.2–3. Nov. 32–34. Choi 2005, 18. IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 Fig. 2. Bregovina – Basilica Sl. 2. Bregovina – Bazilika could reach the city within a day.50 These parameters were also valid in the case of the castella in the area of the Leskovac basin. The largest number of forts, based on these parameters, gravitated towards Cari~in Grad – Justiniana Prima, except for those in the far eastern part of the basin, which were orientated towards Remesiana. The most favourably positioned were the castella in the Morava river valley, which gravitated towards Cari~in Grad – Justiniana Prima, as well as Naissus and Remesiana. The connection with Remesiana was certainly impeded by the wide mountain ranges between these regions. It should also be mentioned that the castella situated in the south-western part of the basin were closest to Justiniana Secunda. At Cari~in Grad, a large number of artisans, such as potters, blacksmiths, goldsmiths, glaziers and others were testified to, clearly indicating that it was a regional centre which supplied both the local area and those further afield. The very concept of a newly built centre with wide streets and porticos points to the fact that the city was planned as an administrative and trade centre.47 (Fig.3) An indication of the lively local and regional trade is the presence of camels and mules at Cari~in Grad,48 which were particularly used in the 6th century for transporting goods.49 Castella, in the wide area of the Leskovac basin, as well as in other parts of the central Illyricum, were mostly built in non-urban areas. Spatial analysis indicates that most of the castella were situated outside a 10 mile – 15 km radius from the city, which represents a journey that could have been taken to the city and back within a day. (Map 3.2) In the urbanised regions of the Empire, in central Italy, the distance between cities was from 11 to 13 km, which enabled strong economic ties. In Lower Galilee, on the other hand, peasants 47 48 49 50 226 Saliou 2005, 207–224. Markovi} 2013. Morrisson, Sodini 2002, 200. Choi 2005, 11–12: with quoted literature. IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 Fig. 3. Cari~in Grad – Porticos of the Upper Town’s eastern street Sl. 3. Cari~in grad – Portik isto~ne ulice Gorweg grada on the polis, which represented the pillar of the administrative system of Byzantine power in the area of central Illyricum. The former importance of this region is testified to by the prefecture having been situated in Thessalonica which, as late as 604, consisted of Dacian and Macedonian offices. The prefect of Illyricum was soon to disappear, ceding his place to the eparch of Thessalonica, by which the end of the administration over Illyricum was confirmed.51 The study of the Late Roman fortifications in relation to the urban centres is important when considering their role and significance. In spite of the fact that new cities were built in this region, the degree of urbanisation remained relatively low. Barbarian incursions, epidemics and climate change resulted in the weakening of the villages and, subsequently, the castella, which together led to the decline of the city’s importance and the disappearance of the administrative system based 51 227 Lemerle 1979; 126 § 106; Lemerle 1981, 69–70 and 176. IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 SOURCES: Codex Theodosianus, (ed.) Th. Mommsen, P. Meyer, Berlin 1905. The Chronicle of Marcellinus, (ed.) B. Croke. Sydney 1995. Corpus Iuris Civilis III. Novellae, (eds.) R. Schoell, G. Kroll, Berlin 1912. Procopii Caesarensis De Aedificiis, (ed.) J. Haury. Leipzig 1913. Cod. theod. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marcellinus Comes, Chronicle . . . . . . . . . . . . . . . . . . . Nov. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proc. De aedif. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BIBLIOGRAPHY: Ivani{evi}, Stamenkovi} 2010 – V. Ivani{evi}, S. Stamenkovi}, Nalazi novca 5. veka iz Rujkovca, Leskova~ki zbornik 50, 2010, 59–84. Jeremi} 2004 – M. Jeremi}, Arhitektonska dekorativna plastika crkve u Bregovini, Starinar 53–54, 2004, 111–137. Jeremi}, Milinkovi} 1997 – M. Jeremi}, M. Milinkovi}, Die Byzantinische Festung von Bregovina (Südserbien), Antiquité tardive III, Paris 1997, 209–225. Kanic 1985 – F. Kanic, Srbija, zemlja i stanovni{tvo, 2, Beograd 1985. Kondi}, Popovi} 1977 – V. Kondi}, V. Popovi}, Cari~in Grad, utvr|eni grad u vizantijskom Iliriku, Beograd 1977. Lemerle 1979 – P. Lemerle, Les plus anciens recueils des miracles de saint Démétrius, I (texte et analyse), Paris, 1979. Lemerle 1981 – P. Lemerle, Les plus anciens recueils des miracles de saint Démétrius, II (commentaire), Paris, 1981. Mac Mullen 1963 – R. Mac Mullen, Soldier and Civilian in the Later Roman Empire, Cambridge Mass., 1963. Markovi} 2013 – N. Markovi}, Promene u eksploataciji `ivotinja na ranovizantijskom nalazi{tu Cari~in grad: ostaci faune iz kompleksa jugoisto~ne ugaone kule Donjeg grada. Unpublished MA thesis, Faculty of Philosophy, Belgrade 2013. McCormik et al. 2012 – M. McCormick, U. Büntgen, M. A. Cane, E. R. Cook, K. Harper, P. Huybers, T. Litt, S. W. Manning, P. A. Mayewski, A. F. M. More, K. Nicolussi, W. Tegel, Climate Change during and after the Roman Empire: Reconstructing the Past from Bavant, Ivani{evi} 2003 – B. Bavant, V. Ivani{evi}, Ivstiniana Prima – Cari~in Grad, Beograd 2003. Choi 2005 – A. Choi, The Travelling Peasant and Urban-Rural Relations in Roman Galilee, in: Travel and Religion in Antiquity. Canadian Society of Biblical Studies Seminar, 2005. Available at http://www. philipharland.com/travelandreligion.htm (10th December 2013). Dagron 1984 – G. Dagron, Les villes dans l’Illyricum protobyzantin, in: Villes et peuplement dans l’Illyricum protobyzantin. Actes du colloque de Rome (12–14 mai 1982), Rome, 1984, 1–20. Decker 2006 – M. Decker, Towns, Refuges and Fortified Farms in the Late Roman East, Liber Annus 56, 2006, 499–520. Ducellier 1985 – A. Ducellier, Le problème des autonomismes urbains dans les Balkans: origines, continuités et ruptures (VIe–XIIIe siècles), In: Actes des congrès de la Société des historiens médiévistes de l’enseignement supérieur public. 16e congrès, Rouen, 1985, 121–141. Du{ani} 1999 – S. Du{ani}, The Miners’ Cults in Illyricum, in: Mélanges C. Domergue, Pallas 50, Toulouse, 1999, 129–139. Evans 1883 – A. J. Evans, Antiquarian researches in Illyricum, parts I and II, Westminster 1883. Grmek 1998 – M. Grmek, Les conséquences de la peste de Justinien dans l’Illyricum, in: Radovi XIII me|unarodnog kongresa za starokr{}ansku arheologiju, 2, (eds.) N. Cambi, E. Marin, Split 1998. 2, 787–794. IMS IV – Inscription de la Mésie Supérieure IV. Naissus – Remesiana – Horreum Margi, (ed.) P. Petrovi}, Belgrade 1979. 228 IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) STARINAR LXIV/2014 Petrovi} 1976 – P. Petrovi}, Ni{ u anti~ko doba, Ni{ 1976. PLRE 3 – The Prosopography of the Later Roman Empire, vol. 3, (ed.) J. R. Martindale, Cambridge 1992. Saliou 2005 – C. Saliou, Identité culturelle et paysage urbain : remarques sur les processus de transformation des rues à portiques dans l’Antiquité tardive, Syria 82, 2005, 207–224. Shaw 1981 – B. Shaw, Rural markets in North Africa and the political economy of the Roman Empire, Antiquités africaines 17,1981, 37–83. Stamenkovi} 2013 – S. Stamenkovi}, Rimsko nasle|e u Leskova~koj kotlini, Beograd 2013. Stathakopoulos 2006 – D. Stathakopoulos, Crime and Punishment: The Plague in the Byzantine Empire, 541–749, in: Plague and the End of Antiquity, The Pandemic of 541–750, (ed.) L. Little, Cambridge 2006, 99–118. Vuli} 1934 – N. Vuli}, Anti~ki spomenici na{e zemlje, Spomenik, Srpske kraljevske akademije LXXVII, 1934, 3–54. Scientific and Historical Evidence, Journal of Interdisciplinary History, 43:2 (Autumn, 2012), 169–220. Milinkovi} 2007 – M. Milinkovi}, O potrebi nau~nog prou~avanja lokaliteta Zlata–Kale, Ni{ i Vizantija V, 2007, 191–203. Morrisson, Sodini 2002 – C. Morrisson, J.-P. Sodini, The Sixth-Century Economy, in: The Economic History of Byzantium: From the Seventh through the Fifteenth Century, (ed.) A. Laiou, Washington D.C. 2002, 171–220. Papaconstantinou 2012 – A. Papaconstantinou, Les propriétaires ruraux en Palestine du sud et en Egypte entre la conquête perse et l’arrivée des Abbassides, Mélanges de l’École francaise de Rome – Moyen Âge 124/2, 2012, 405–416. Pe{ut 1976 – D. Pe{ut, Geolo{ki sastav, tektonska struktura i metalogenija leckog andezitskog masiva, Rasprave Zavoda za geolo{ka i geofizi~ka istra`ivanja XIV, Beograd 1976. 229 IVANI[EVI], STAMENKOVI], Late Roman fortifications in the Leskovac basin… (219–230) Rezime: STARINAR LXIV/2014 VUJADIN IVANI[EVI], Arheolo{ki institut, Beograd SOWA STAMENKOVI], Arheolo{ki institut, Beograd KASNOANTI^KA UTVR\EWA U LESKOVA^KOJ KOTLINI I WIHOV ODNOS PREMA URBANIM CENTRIMA Kqu~ne re~i. – Leskova~ka kotlina, kasna antika, fortifikacije, privreda, Cari~in grad, Justiniana Prima. nih pravaca. Analiza polo`aja utvr|ewa ukazuje na koncentraciju utvrda po visovima koji uokviruju re~ne doline Ju`ne Morave i wenih pritoka. Najve}u gustinu tih utvr|ewa bele`imo upravo na pobr|u Dobre gore, sme{tenom izme|u dolina Puste reke, Jablanice i Ju`ne Morave. Radi se prevashodno o mawim fortifikacijama ~ija je primarna uloga bila za{tita lokalnog stanovni{tva, koje se, sasvim izvesno, bavilo poqoprivredom. Ipak, najve}i broj tih utvrda nalazi se u zapadnom delu kotline, u planinskim oblastima Goqaka, Majdana, Radana i Pasja~e. Najve}a gustina utvr|ewa je u dolini Bawske reke, oko Sijarine. Cela oblast je bila poznata po rudarewu u prethodnim stole}ima. Ne bi trebalo iskqu~iti mogu}nost da je tokom 6. veka rudarska aktivnost bila obnovqena. Posebnu grupaciju ~ini mawa grupa utvr|ewa sme{tena oko Cari~inog grada – Justinijane Prime, ~ija je prevashodna uloga bila odbrana prilaza gradu. Justinijana Prima predstavqa jedini grad – polis – podignut oko 535. godine u sredi{tu ove oblasti kao sedi{te arhiepiskopa. Utvr|ewa na podru~ju Leskova~ke kotline pripadaju kategoriji kastela – fruria, ~iji je osnovni ciq bio za{tita lokalnog stanovni{tva. Najve}i deo utvrda gravitirao je prema Cari~inom gradu – Justinijani Primi, izuzev onih na krajwem isto~nom delu kotline koji su bili okrenuti ka Remesijani. Najboqe pozicionirani su bili kasteli u dolini Morave, koji su gravitirali ka Cari~inom gradu – Justinijani Primi, ali su to bili i Naisu i Remesijani. Veza sa Remesijanom je bila svakako ote`ana, budu}i da se izme|u tih oblasti nalaze {iroki planinski venci. Dodajmo da su kasteli sme{teni u jugozapadnom delu kotline bili najbli`i Justinijani Sekundi. Smatramo da je posmatrawe kasnoanti~kih utvr|ewa u odnosu na gradska sredi{ta va`no za razmatrawe wihove uloge i zna~aja. I pored podizawa novih gradova, u ovoj oblasti je stepen urbanizacije ostao slabo razvijen. Upadi Varvara, epidemije i klimatske promene odrazili su se na slabqewe sela, a potom i kastela, {to je sve zajedno uslovilo opadawe zna~aja grada i nestanak upravnog sistema zasnovanog na polisu kao stubu upravnog sistema vizantijske vlasti na prostoru centralnog Ilirika. Krajem 4. veka, usled upada i naseqavawa Varvara na podru~je severnog Ilirika, dolazi do velikih dru{tveno-ekonomskih potresa, koji }e se posebno odraziti na promenu uloge naseqa. Rimska ravni~arska naseqa ustupaju mesto novim utvr|enim stani{tima podignutim na dominantnim i dobro brawenim visovima iznad re~nih dolina ili unutar planinskih oblasti. Usled op{te nesigurnosti i potrebe za{tite stanovni{tva i komunikacija na podru~ju Leskova~ke kotline, podi`e se veliki broj kasnoanti~kih fortifikacija, mahom na rubovima kotline. Fortifikacije su podignute na Radanu, uz slivove Lepa{tice, Lecke i Gazdarske reke, Goqaku, sa podru~jem oko Tularske i Bawske reke, oko Cari~inog grada, oblasti oko Rujkovca i Radinovca, zatim u dolini [umanske reke, Puste reke, Jablanice i Veternice. Fortificirani su i pobr|a Kukavice, dolina Ju`ne Morave, kompleks oko Kozara~ke, odnosno Rupske reke, dowi tok Vlasine, pobr|a Babi~ke gore i Kru{evice i u{}e Lu`nice u Vlasinu. Najve}i deo ovih utvrda podignut je na visovima koji dominiraju nad {irom okolinom. Hronolo{ko opredeqewe utvr|ewa na osnovu arheolo{kog materijala i numizmati~ke gra|e omogu}eno je u izvesnim slu~ajevima. Na velikom broju utvr|ewa na|en je novac iz 4. veka, i to unutar fortifikacija u Lecu, Rujkovcu, na Hisaru u Leskovcu, Stupnici, Grdelici i Skobaqi} gradu kod Vu~ja. Cirkulacija novca u 5. veku posvedo~ena je u Rujkovcu, {to predstavqa va`an dokaz wenog opstajawa u nemirnim vremenima. Datovawe utvr|ewa u 6. vek, pored novca, upotpuwuje i drugi arheolo{ki materijal, a pre svega kerami~ke posude. U to vreme opredeqen je i najve}i broj fortifikacija, koje su ubicirane u Sijarini, Mrkowu, na Malom kamenu, u Rujkovcu, Gorwem Brijawu, Gorwoj Loko{nici, Grada{nici, Konopnici, Gradi{tu, na Hisaru u Leskovcu i Grdelici. Na mawem broju utvr|ewa, uglavnom na onima na kojima su sprovedena arheolo{ka iskopavawa, potvr|eni su horizonti iz 4. i 6. veka. Re~ je o fortifikacijama u Lecu, Sijarinskoj Bawi, Rujkovcu, Hisaru u Leskovcu, Konopnici i Grdelici. Raspored kasnoanti~kih fortifikacija, kao {to je slu~aj sa rimskim naseobinama, bio je uslovqen stepenom naseqenosti oblasti, resursima i potrebom kontrole put- 230 MILOJE VASI], Arheolo{ki institut, Beograd GORDANA MILO[EVI], Univerzitet u Beogradu, Arhitektonski fakultet, Beograd NADE@DA GAVRILOVI], Arheolo{ki institut, Beograd ISKOPAVAWA MEDIJANE U 2010. I 2011. GODINI e-mail: [email protected] UDK: 902.2:902.01(497.11)"2010/2011" ; 904:728"652"(497.11) DOI: 10.2298/STA1464231V Primqeno: 17. februara 2014. Prihva}eno: 24. jula 2014. Pregledni rad Apstrakt. – Iskopavawa u 2010. i 2011. godini, koja su trajala ukupno devet meseci, u potpunosti su zaokru`ila analizu stratigrafije i arhitekture vile s peristilom, bez obzira {to bi trebalo ispitati i isto~ni deo trema i vestibul ispred ju`nog ulaza. Stratigrafija, analiza arhitekture i prate}i materijal, posebno novac, omogu}uju da se postavi jasna hronologija i vile s peristilom i Medijane. Konstatovana su tri horizonta: horizont 1 (prve decenije 4. veka); horizont 2 (330–378); horizont 3 (380 – 441–443. godina). Utvr|eno je, suprotno ranijim mi{qewima, da je vila u svom osnovnom planu nastala u horizontu 1, a da su u horizontu 2 naknadno izvr{ene prepravke (stibadijumi A i B, ju`ni trem), i da je u to vreme bila ukra{ena mozaicima, freskama, mermernom oplatom i skulpturama. U horizontu 3, posle velikih razarawa, oko vile je formirana arhitektura koja indicira ruralizaciju Medijane. Kqu~ne re~i. – Medijana, vila s peristilom, stratigrafija, arhitektura, pokretni nalazi, kasna antika. U 2014. godini navr{ava se 150. godina istra`iva~kih radova u Medijani, koja su zapo~eta 1864. godine skromnim iskopavawima istra`iva~a Feliksa Kanica.1 Interesovawe za ovaj lokalitet je nastavqeno i u periodu izme|u dva svetska rata. Ispitivawa arheologa amatera okupqenih oko Muzejskog dru{tva za delokrug Moravske banovine i Istorijsko-etnografskog muzeja u Ni{u stvorili su uslove za daqe radove. Istra`ena je prostorija centralnog rasporeda sa mozaikom izuzetnog kvaliteta, nad kojom je podignuta muzejska zgrada.2 Ve}i obim radova tokom 1961. godine u organizaciji Arheolo{kog instituta utemeqio je savremene metodolo{ke principe prou~avawa i postavio dobru osnovu za budu}a sistematska istra`ivawa. Utvr|ena je granica prostirawa lokaliteta i evidentiran veliki broj objekata, od kojih je 20 u celosti ili delimi~no definisano.3 Centralno mesto na lokalitetu zauzima kompleks vile sa peristilom. Tokom vi{e arheolo{kih kampawa 1969, 1972, 1976, 1978–1979. i 1980. godine u organizaciji Arheolo{kog instituta iz Beograda sprovedena su iskopavawa na ovom objektu.4 Istra`eni su peristil, sve~ana dvorana/trikli- 1 Kanitz 1892, 77–78. Or{i}-Slaveti} 1933/34, 303–310; Bratani} 1938, 199–204. 3 Milo{evi} 1996, 59–68. 4 Rukovodilac radova bila je dr Q. Zotovi}, Dokumentacija Arheolo{kog instituta u Beogradu. 2 * ^lanak predstavqa rezultat rada na projektu: Romanizacija, urbanizacija i transformacija urbanih centara civilnog i vojnog karaktera u rimskim provincijama na tlu Srbije (br. 177007), Ministarstva prosvete, nauke i tehnolo{kog razvoja Republike Srbije. 231 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 1. Plan vile sa obele`enim sondama istra`enim u 2010. i 2011. godini Fig. 1. Plan of the villa with marked sondages explored in 2010 and 2011 232 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 1. Sonde severozapadno od severne fasade vile (sl. 2, 3, 4) Ovaj sektor obuhvata sonde koje nose oznake od zapada ka istoku: 72 i 73; 88 i 90; 86 i 87; 42/2004, 89 i 91 (videti plan). Iako je na prvi pogled stratigrafija u ovim sondama jednostavna, treba ipak ukazati na neke osnovne karakteristike. Iznad sondi 72, 73, 88 i 90 prolazio je put koji je nekada vodio u selo Dowu Vre`inu, a potom je slu`io kao prilaz zgradi Dokumentacionog centra. Time je dobijen konglomerat recentnih slojeva nastalih povremenim nasipawem puta. Osim toga, nad sondama 42, 89 i delom 90, mo`da i 86 i 87, bila je deponovana zemqa sa iskopavawa sve~ane dvorane/triklinijuma i dela severnih prostorija vile. U tim sondama se humusni sloj me{a sa slojem {uta (sloj A), koji je najve}im delom nastao va|ewem zidova vile. Ispod jakog sloja {uta (sloj B) u sondama 72/3, 88–90 i 42–89 konstatovan je sloj tegula i imbreksa od krovova zgrada podignutih u tre}em horizontu Medijane, posle ¯ 380. godine (sl. 5). Ovaj sloj je u dnevnicima prvobitno obele`en kao nivo a i nivo b – zbog preciznijeg razdvajawa, a naro~ito stoga {to je on najvi{i u sondama 88 i 90 (blok A) i pada prema zapadu, prema zapadnoj ivici sondi 72/3. U kona~noj konkordanci slojeva ove nivoe obele`avamo kao sloj C. nijum i prostorije kancelarija na isto~nom, zapadnom i jugoisto~nom delu objekta. No, i pored ovako opse`nih radova, nije bilo mogu}e utvrditi u potpunosti gabarit osnove i hronologiju gra|ewa gra|evine. Radovima maweg obima u periodu 2003–2006. izvr{ena su reviziona ispitivawa i zapo~eta nova na jugoisto~nom i severozapadnom delu.5 Rezultati tih istra`ivawa pokrenuli su niz novih pitawa o izgledu i funkciji gra|evine, ali je obim radova bio skroman i prekinut je 2007. godine usled promene koncepta prezentacije.6 U februaru 2010. godine, odlukom Ministarstva kulture Republike Srbije, imenovan je Projektni tim ~iji je zadatak bio organizacija i realizacija pripreme, izvo|ewa i prezentacije projekta „Konstantinova vila na Medijani“ u povodu obele`avawa 1700 godina Milanskog edikta.7 Osnovni koncept projektnog zadatka bilo je prikupqawe relevantnih podataka neophodnih za izradu idejnog re{ewa i glavnog projekta prezentacije vile sa peristilom, {to je podrazumevalo arheolo{ka i arhitektonska istra`ivawa, kao i ispitivawa ukupnog stawa arhitekture vile. Tokom 2010. i 2011. godine, u trajawu od devet meseci, sprovedena su sistematska arheolo{ka istra`ivawa na severoisto~nom, severozapadnom i jugoisto~nom delu vile sa peristilom, kao i na prostoru ispred muzejske zgrade. Predvi|ena povr{ina za istra`ivawa iznosila je oko 1400 m², a u potpunosti je ispitano ukupno 1230 m² (sl. 1). Ciq iskopavawa je bio definisawe gabarita vile na severozapadnom i jugozapadnom delu, izgleda ulaza u vilu, kao i ispitivawe temeqa zgrade Muzeja.8 5 Rukovodilac radova bio je dr M. Vasi}. Bilo je planirano da se nad vilom sa peristilom u termama podigne balon konstrukcija (rukovodilac projekta i radova M. Kora}). 7 Projektni tim su ~inili: dr Gordana Milo{evi} (rukovodilac), Arhitektonski fakultet BU, dr M. Vasi}, arheolog i M. Veqkovi}, arhitekta (~lanovi tima); Re{ewe Ministarstva kulture Republike Srbije broj 110-01-17/2010-02. 8 Iskopavawa su izvo|ena od 9. avgusta do 15. decembra 2010. i od 27. maja do 26. jula, potom od 5. septembra do 1. oktobra 2011. U istra`ivawima su u~estvovali: dr Gordana Milo{evi} (Arhitektonski fakultet, BU), rukovodilac projekta; dr Miloje Vasi}, rukovodilac arheolo{kih iskopavawa; dr Nade`da Gavrilovi}, rukovodilac terenskih istra`ivawa (Arheolo{ki institut u Beogradu). ^lanovi ekipe: Marica Maksimovi} i Vesna Crnoglavac (Narodni muzej, Ni{); mr Aleksandar Aleksi} (Zavod za za{titu spomenika kulture, Ni{); Vladimir Mihailovi} (Filozofski fakultet u Novom Sadu); Kosana Oqa~i}, dia; studenti arheologije i arhitekture. Konsultanti: dr Ivana Popovi} (Arheolo{ki institut u Beogradu) i prof. dr Ulrike Wulf-Reidt (Arhitekturreferat, DAI Berlin). 9 Kao ilustracija, za svaki sektor je prilo`en najkarakteristi~niji profil, kako bi se dobila op{ta slika o slojevima vile s peristilom. 6 STRATIGRAFIJA Tokom iskopavawa su uo~ene odre|ene razlike u stratigrafiji vile s peristilom i dela koji je iskopavan s wene spoqwe strane. Zbog toga smo ova iskopavawa podelili na pet sektora: 1. Sonde severozapadno od severne fasade vile 2. Sonde severoisto~no od severne fasade vile 3. Severozapadne prostorije 4. Severoisto~ne prostorije 5. Jugozapadne prostorije Ova podela omogu}ava da se lak{e prati stratigrafska situacija na ~itavom iskopavanom prostoru.9 233 Fig. 2. North profile of sondages 73, 90, 87 (sketches 239–250/11) 234 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) Sl. 2. Severni profil sondi 73, 90, 87 (skice 239–250/11) STARINAR LXIV/2014 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) Sl. 3. Zapadni profil sonde 86 (skica 109/11) Fig. 3. West profile of sondage 86 (sketch 109/11) Sl. 4. Isto~ni profil sonde 87 (skica 81/10) Fig. 4. East profile of sondage 87 (sketch 81/10) 235 STARINAR LXIV/2014 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 5. Sonda 72, severni deo, nivo ru{ewa u sloju V–S, snimak sa istoka Fig. 5. Sondage 72, northern part, level of destruction in stratum B–C, photographed from the east Radi lak{eg snala`ewa, izradili smo tabelu sredwih vrednosti slojeva (tabela 1) i tabele novca (tabela 1a) i mermerne oplate (tabela 1b) iz svih analiziranih sondi severozapadnog sektora. Na osnovi tako dobijenih rezultata mo`emo da damo slede}i opis slojeva, polaze}i od najni`ih (najstarijih) do najmla|ih. 1.1. Temeqi severne fasade vile i velike dvorane/triklinija bili su ukopani u sloj E – protohumus iznad zdravice. Zemqa je tamnije mrke boje i konstatovana je u prostoru prefurnijuma u sondi 42 i malim delom u sondi 88. U sondi 42 su u ovaj sloj bila ukopana ~etiri jaka direka, ~etvrtastog preseka, koja su nosila krovnu konstrukciju koja je {titila prostor ispred lo`i{ta hipokausta sve~ane dvorane i prostorije 24a. Delovi imbreksa i tegula na|eni su na nabijenoj zemqi u ovom prostoru. Sokl severne fasade i sokl sve~ane dvorane javqaju se na koti 198.6, {to je i kota poda prostora prefurnijuma. Blizu severozapadnog ugla vile nalazio se drugi prefurnijum koji je trebalo da zagreva prostorije 22 i 23. U probnoj sondi u okviru sonde 72 konstatovane su krovne opeke u prostoru koji je bio ukopan u sloj E na toj strani, ali, zbog stabilnosti lo`i{ta hipokausta, ovaj prefurnijum nije u potpunosti istra`en. Sru{eni prefurnijum bio je iznivelisan zemqom sloja E. 1.2. Izme|u kota 198.8 i 198.6, u sondama 72, 86, 90 i 89/42, iznad sloja E (osim prefurnijuma u sondi 42) konstatovana je po ~itavoj povr{ini malterna podnica. Razlika od dvadesetak centimetara na tako velikoj povr{ini nije zna~ajna. Ova podnica, prakti~ki u nivou sokla vile, pripada wenoj prvoj gra|evinskoj etapi. 1.3. Iznad malterne podnice nasut je sloj D, svetlija mrka zemqa sa sitnim {utom.10 Ovaj sloj je prekrio prefurnijum sa spoqne strane prosto- 236 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Tabela 1a / Table 1a Novac Sloj B 199.47 199.43 199.4 199.1 199.03 355–361. 364–375. 4. vek 404–406. 4. vek Sloj D 198.98 4. vek 198.93 5. veka 198.91 364–375. 198.86 4. vek 198.84 364–367. 198.83 364–367. 198.79 355–361. 198.76 330–335. 198.69 341–346. 198.67 330–335. prefurnijum stibadijuma B 198.51 364–375. Sonda 45 – prostorija 24a 198.96 364–367. Sonda 46 – ulaz u stibadijum B 198.85 382–389. 355–361. 335–337. 364–375. 4. vek 364–375. 364–375. 364–367. ispod tegula 4. vek 198.89 364–375. 4. vek 364–375. Tabela 1 / Table 1 Tabela 1b / Table 1b Sredwa vrednost stratigrafije slojeva A i B 200.63–200.45 200.44–200.2 put 200.07–199.98 Humus+A ±199.5 kraj A ±199.51 po~etak B ±199.25 kraj B kraj B iznad poda 199.01 stibadija B Sredwa vrednost stratigrafije slojeva C i D ±199.24 po~etak C kraj C, ±199 malterna supstrukcija 199.15 gorwa ivica „kasete“ ±198.9 po~etak D ±198.8 do kraj D u s. 72, 86, 90, 89/42 ± 198.67 izme|u 198.82 malt. pod. u s. 72, 86, 90, i 198.6 89/42 ±198.6 po~etak sloja E Mermerna oplata Sloj B 199.8 Sloj C 199.1 Sloj D 198.9 425–450. dno kasete 199.2 39 komada 199 27 komada 198.4 16 komada rije w-22, a isto tako i deo prefurnijuma u sondi 42. Sloj se od sonde 89 spu{tao od severa prema jugu u sondu 42, do otvora hipokausta sve~ane dvorane. U sloj D ukopani su temeqi stibadijuma B i prefurnijum s wegove severne strane, kao i temeqi prostorije w-26. Ova stratigrafija jasno pokazuje da je stibadijum B, a samim tim i stibadijum A, iz druge gra|evinske etape kada je vila do`ivela 10 Zemqa za ovaj nasip verovatno je doneta s mesta na kojima su jo{ bili ostaci sloja medijanske faze I, a koja je bila namerno poru{ena. 237 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Takva situacija pokazuje da i oko stibadijuma B nije bilo znatnijeg sloja {uta od prvobitnog ru{ewa vile u 378. godini, ve} je direktno na gaznoj povr{ini starijeg horizonta 2 nastao novi horizont 3. U sondama 89 i 42, gorwi nivo sloja D je zape~en i ozna~en je kao nivo a. Ovaj nivo se spu{ta u sondi 42 do otvora hipokausta sve~ane dvorane/triklinija. Te{ko je odgovoriti na pitawe da li je nivo bio pod jakim uticajem vatre u prvobitnom ru{ewu vile 378. godine ili tek tokom stradawa ku}a horizonta 3. 1.4. Horizont sloja D, „skrivnica“ i zidovi 1, 2 i 3 bili su pokriveni slojem tegula i imbreksa – sloj C. Kako je napomenuto, tokom iskopavawa je ovaj sloj prvobitno obele`en kao nivoi a i b. Sloj je bio najvi{i oko zapadnog aneksa stibadija B (kota ¯199.63). Ova situacija jasno pokazuje da su zidovi stibadijuma bili znatno vi{e o~uvani i verovatno kori{}eni u horizontu 3. Sloj C se od stibadijuma spu{tao prema zapadu i preko zida 3, a u sondama 72 i 73 dostizao je do kote ¯ 198.7. Ovde je on nalegao na najstariju malternu podnicu, a direktno ispod wega, izme|u kota 198.87 i 198.74, na|eni su fragmenti skulptura, od kojih neki pripadaju pojedinim statuetama iz ostave u prostoriji w-4, otkrivenoj 1972. godine. O ovom nalazu }e biti re~i u daqem tekstu. Za definitivnu tabelu stratigrafije uzeta je vrednost 199.0 kao kraj sloja C, posmatranog u odnosu na ivicu „skrivnice“ i deo ovog sloja u sondama 89 i 42. 1.5. Sloj C je pokriven jakim slojem ru{ewa zidova stibadijuma, severnih prostorija vile i velike sve~ane dvorane – sloj B. On bi predstavqao sekundarno ru{ewe, kroz vekove, zidova preostalih posle hunske invazije. To je jasno kada se pogleda unutra{wost stibadijuma B, gde ovaj sloj dopire do tankog sloja gline pome{ane s delovima nagorelog mozaika od staklenih tesera iznad mozai~nog poda stibadijuma. Mo`da bi se moglo pretpostaviti da je stibadijum kori{}en i u horizontu 3, ali sloj gline i mozai~nih tesera nije sklawan. O tome na izvestan na~in govori i prokopan rov po Sl. 6. Severozapadni deo vile, van severne fasade vile, skrivnica izme|u zidova 33a i 58, u toku pra`wewa, sa zapada (sonda 88, sloj S) Fig. 6. North-western part of the villa, outside the villa’s northern facade, hiding place between walls 33a and 58, whilst being emptied, from the west (sondage 88, stratum C) znatnije prepravke.11 U sloj D su ukopani i temeqi zida 1 pravca severoistok–jugozapad, severno od stibadijuma B. Gorwi nivo sloja D poslu`io je i kao stambeni nivo u zgradama koje su podignute oko stibadijuma B (zidovi 2 i 3 u sondi 88 i zid u sondi 90). Temeqi ovih zidova, gra|eni od oblutaka vezanih blatom, bili su ukopani u sloj D. Kota wihovih kruna je na koti ¯ 199.13, a sokl na ¯ 198.65. Nije iskqu~eno da su nadzemni zidovi bili izgra|eni i od drveta ili bondruka i da je krov pokriven tegulama i imbreksima. U uglu severnog zida prostorije w-26 i zapadnog zida stibadijuma B bila je, s gorweg nivoa sloja D, ukopana „skrivnica“ (gorwa kota ivice 199.15), sa~iwena od nasati~no postavqenih opeka na pod od opeka (sl. 6). Skrivnica je bila pokrivena tegulama koje su bile naknadno uba~ene u wu posle wenog pra`wewa. U tankom sloju zemqe na dnu kasete, na koti 198.89 na|en je komad bronzanog novca iz perioda 425–450. godine. Nalaz ovog komada jasno pokazuje da je horizont sloja D kori{}en i u periodu posle 380. godine i da su tada izgra|ene zgrade oko stibadijuma B, a da je kraj Medijane bio u hunskom upadu 441. godine. U gorwem nivou sloja D na|ena su i dva komada novca iz perioda 382–389. godine i jo{ dva komada iz intervala 404–450. godine. 11 U za{titnim iskopavawima, koja je 2013. godine vodio mr Aleksandar Aleksi}, stru~ni saradnik Zavoda za za{titu spomenika u Ni{u, nesumwivo je dokazano da je i stibadijum A bio prizidan uz severnu fasadu vile, kao i da stoji upravno na wu. Konstatovano je i da je stibadijum A imao hipokaust (otkriven je prefurnijum). 238 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 7. Isto~ni profil sonde 103 (skica 241/11) Fig. 7. East profile of sondage 103 (sketch 241/11) sredini stibadijuma, kada je uni{tena fontana u wegovom centru, ~iji su ostaci na|eni tokom iskopavawa. Verovatno je to ura|eno da bi se izvukle olovne cevi od fontane.12 Karakteristi~no je da su u zapadnom i isto~nom aneksu, koji su na vi{oj koti od mozaika u centralnom delu, najve}im delom uni{teni mozaici, a na malternoj podlozi se vide rupe, najverovatnije od koqa koje je dr`alo krov (mo`da sredwovekovnih lakih objekata). Mozaik je uni{ten i u ulaznom delu u stibadijum. Sloj B je rastresitija siva zemqa pome{ana s krupnim {utom od kamewa, maltera i znatnim koli~inama fragmentovane mermerne oplate i prozorskog stakla. U ovom sloju mogu se opaziti mnogi proslojci, koji su posledica kori{}ewa tog terena bilo kao majdana kamena, obrade zemqe ili ukopavawa tokom sredweg veka. Sredwa visina sloja je izme|u kota ¯ 199.51 i ¯ 199.25, s varirawima u pojedinim sondama, {to ne remeti op{ti uvid u stratigrafiju. Posledica raznih ukopavawa u ranije slojeve su nalazi novca (7 komada) u sloju B izme|u kota 199.47 i 199.03 u rasponu od 335–337. do 404–406. godine, {to je u saglasnosti s op{tom cirkulacijom novca u Medijani u horizontima 2 i 3. 1.6. Posledwi, najmla|i sloj je sloj A nastao kao posledica intenzivnog va|ewa zidova na ~itavoj povr{ini vile i na profilima se jasno vide rovovi od povr{ine do sa~uvane krune zidova ili ~ak temeqa. Ovi rovovi su ispuweni tamnijom sivom zemqom koja je rasplanirana i iznad sloja B. Sloj A se u pojedinim sondama me{a s humusom. Verovatno je deo sloja A nastao od zemqe s iskopavawa tokom 1933–1936. i sedamdesetih godina dvadesetog veka. 1.7. Kako je ranije pomenuto, u krajwem zapadnom delu istra`ivane povr{ine nalazio se nasip recentnog puta, a izme|u puta i sve~ane dvorane/ triklinija otkriveni su delovi stubova od stare ograde Muzeja i delovi staze koja je vodila ka zgradi Muzeja, kao i temeqi biletarnice. 12 Pomenuta iskopavawa stibadijuma A pokazala su da je i deo mozai~nog poda u zapadnom pravougaonom aneksu bio prokopan. Izvesno je da je u jednom periodu u Medijani postojala potreba za olovom, te su prona|ene i iskopane vodovodne cevi i u stibadiju A i u stibadiju B. Prilikom iskopavawa sonde 30 (jugozapadni ugao isto~ne stra`are – ekskubitorije uz kapiju) na|ena je ve}a koli~ina amorfnog olova (mo`da ono delom poti~e od vodovodnih cevi). 239 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) 2. Sonde severoisto~nog sektora (sl. 7) Ovaj sektor je obuhvatio prostor isto~no od isto~ne fasade zgrade Muzeja i severnog dela fasade vile. Otvorene su sonde 71, 97, 98, 103 i 104. I na ovoj povr{ini treba ra~unati s nasutim slojem od ranijih iskopavawa, a ovde su po~iwali ba{ta i vo}wak porodice Mili}, koja je stanovala na lokalitetu i brinula se o Medijani. Stratigrafija je u ovom delu jednostavnija nego u severozapadnom sektoru. U tabeli dajemo sredwe vrednosti debqine slojeva. sta do kote 198.36. Prefurnijum je u jednom trenutku izgubio funkciju i wegovi zidovi su sru{eni i iznivelisani do kote ¯198.6. Ostaci prefurnijuma bili su zatrpani i iznivelisani zemqom sloja D. U konkordanci, sloj D bi odgovarao sloju E u severozapadnom sektoru. 2.2. Iznad sloja D i prefurnijuma, posle wegove nivelacije nasut je sloj C – svetlomrka kompaktna zemqa pome{ana sa sitnim {utom od maltera i opeka. Ovaj sloj odgovara sloju D u severozapadnom sektoru. U sloj C ukopan je i temeq prostorije prislowene na severnu fasadu vile. Od sredine severnog zida ove prostorije pru`a se prema severu zid (zid 7d) koji prolazi kroz sonde 97, 98 i 104. Zid je gra|en od oblutaka vezanih blatom. Kruna ovog zida je na koti ¯198.8. Prostorija i zid 7d odgovaraju zidovima 2 i 3 u sondama 72/3, 88 i 90, kao i zidu u sondi 91 u severozapadnom sektoru. Sa gorweg horizonta sloja C u sondi 103 ukopana je ve}a jama za otpatke. Konstatovana je na koti 198.76, a weno dno je na koti 197.38. Jama je prokopana kroz slojeve C i D u zdravicu. U jami su na|eni fragmentovani lonci koji pripadaju punom sredwem veku. Ovo bi ukazivalo da je taj deo dugo bio otvoren i da nije bio zatrpan {utom od prvobitnog ru{ewa vile, tako da su se na istom nivou sustigli kasnoanti~ki horizonti 2 i 3 i horizont sredweg veka. 2.3. Sloj C, zid od oblutaka i jamu u sondi 103 prekriva sloj B, tamnosiva rastresita zemqa pome{ana s krupnijim {utom. Te{ko je re}i kada je izvr{ena nivelacija sredwovekovnog horizonta, jer u sloju B nije na|en hronolo{ki osetqiv materijal. Sloj B se me{a s recentnim materijalom, a u konkordanci bi odgovarao sloju B u severozapadnom sektoru. 2.4. Iznad sloja B nalazi se sloj humusa koji se u dowem delu me{a sa svetlosivom zemqom (sloj A), i on je deo pomenutog vo}waka i ba{te. Tabela 2 / Table 2 Severoisto~ni sektor ±199.8 Humus +A ±199.3 kraj humusa + A 199.01 do 198.78 zid 7d ±199.3 po~etak B ±198.83 kraj B ±198.83 po~etak C ±198.75 kraj C ±198.75 po~etak D Prefurnijum ±198.6 zidovi I i II ±198.4 dno 198.58 kraj iskopa Tabela 2a / Table 2a Novac Godine 355–361. 364–375. 4. vek 355–361. 330–335 294–307. severoisto~ni sektor Sonda Sloj 97 C 97 C 98 C 98 C 103 C 104 C STARINAR LXIV/2014 Kota 198.86 198.91 198.84 198.76 198.85 198.81 Posmatraju}i slojeve od najstarijeg ka najmla|em, dobijamo slede}u sliku: 2.1. Sloj D – tamnomrka kompaktna zemqa, koja u gorwem sloju ~ini protohumus koji prelazi u zdravicu. U ovaj sloj bio je ukopan temeq vile i sokl severne fasade koji je na koti 198.02. Temeq je ukopan do kote 197.94. U ovaj sloj je ukopan i temeq prefurnijuma kojim su grejane severoisto~ne prostorije 22 i 23. Dno prefurnijuma je ~inila malterna podloga na severnom delu na koti 198.54 i spu{ta se prema jugu, prema otvorima hipokau- 3. Severozapadne prostorije vile s peristilom (sl. 8) Tokom kampawa 2010. i 2011. godine istra`ene su prostorije u okviru ranije postavqenih sondi 45 i 46. To su prostorije w-1, w-22, w-24, w-24a, kao i deo prostorije w-25, koji je bio ispod kontrolnog profila. Iako se stratigrafija su{tinski ne razlikuje od one u severozapadnom i severoisto~nom sektoru, postoje izvesne druga~ije 240 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 8. Isto~ni profil sonde prostorije w-24 (sonda 46) (skica 33/10) Fig. 8. East profile of sondage in room w-24 (sondage 46) (sketch 33/10) prema mla|im. Osim slojeva A i B, ostali slojevi ne mogu da se stave u konkordancu sa slojevima dva prethodno diskutovana sektora. karakteristike. Pre svega, sve ove prostorije imaju podove od opeka ili maltera. Na te podove padala je tokom ru{ewa konstrukcija tavanice, a preko we krovna konstrukcija. No, u prostoriji w-24a, delom w-24 i w-22, ovaj prvobitni ru{evinski sloj bio je ra{~i{}en do podova. U drugim delovima je ostavqen vidqiv, te je bio gazna povr{ina u horizontu 3 (sloj C). Tako su na ovoj povr{ini, u prostoriji w-1 u sekundarnoj upotrebi na|eni baza stuba iz peristila i glava bogiwe Venere. Hipokausti su uz unutra{we lice severnog zida vile bili probijeni i kori{}eni, naro~ito u prostoriji w-22, a u ovom velikom ukopu na|ena je sredwovekovna keramika. Tako se u tim o~i{}enim i probijenim prostorima, kao i na deo podova, naslojava sekundarni ru{evinski sloj (sloj B). I u prostorijama se mogu uo~iti rovovi od najkasnijeg va|ewa zidova (turski period ?). U tabeli su date sredwe vrednosti debqine slojeva, a u daqem tekstu je dat op{irniji komentar. I ovde su slojevi posmatrani od najstarijih Tabela 3 / Table 3 Sonda 46, p. 22; 24 ±200.6 ±200.17 ±200.3 ±199.8 ±199.2 put kraj puta humus + A kraj humusa +A po~etak B kraj B ±199.2 po~etak C ±198.8 ±198.8 ±198.7 ±198.7 kraj C po~etak D kraj D po~etak E ±199.8 241 Sonda 46 ±200.64 ±199.99 ±199.99 ±199.79 ±199.99– ±199.79 ±199.09 humus + A kraj humusa +A po~etak B kraj B po~etak C kraj C VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) 3.1. Prostorija w-1: 3.1.1. Malterna podnica, konstatovana na koti 198,70 m. 3.1.2. Sloj D – tamnomrka zemqa, sa mestimi~nom pojavom maltera i velikim oblucima. Bez nalaza. Sloj/proslojak kompaktne glinovite „zape~ene“ zemqe, relativne debqine 0,05 m, koji prekriva pod prostorije. Predstavqa ostatak me|uspratne konstrukcije stradale u po`aru. U ovom sloju se konstatuju ve}i komadi fresaka i malterne podnice. 3.1.3. Sloj C – na koti ¯199.18 rastresit sloj sa~iwen od kre~nog malternog {uta. Pored vi{e komada lomqenih opeka i kamena, ~esta je pojava gare`i i komada malternog poda; dominiraju i fragmenti krovnih opeka. Sporadi~no se nailazi na fragmente fresaka. Ovaj sloj predstavqa ostatke gorewa i ru{ewa. Uza zid 41 (uz wegovo lice u prostoriji 1), a na 0,70 m severno od zida 29, prona|en je fragment baze stuba na koti 199.33 (AR – 113). Ru{evinski nivo u sloju C se u ju`nom delu nalazi na koti 198.99, dok je u severnom delu na koti 199.08. U severnom delu prostorije w-1 pored zida 32, na koti 199.14, na 0,60 m ju`no od zida 32 i 1,40 m isto~no od zida 41 prona|en je veoma korodiran gvozdeni vrh kopqa (C-2). Na 0,5 m isto~no od zida z-41 i 3,50 m severno od zida 29 prona|ena je, kako je ve} napomenuto, glava mermerne skulpture Venere (C-4). Skulptura je bila polo`ena na desni obraz, licem prema severu, na koti 199.12, na nivou koji predstavqa gorwi nivo sloja C. Ru{evinski nivo u sloju C se u ju`nom delu nalazi na koti 198.99 m, dok je u severnom delu na koti 199.08. 3.1.4. Sloj B3 – javqa se samo u severnom delu profila i predstavqa sloj kre~nog maltera koji je nabijen, kompaktan. Relativne je debqine od 0,4 m. Gorwa linija se jasno grani~i sa slojevima B2 i B1, dok se dowa grani~i sa slojem C. U jugoisto~nom delu konstatovani su delovi zida koji se pru`a pravcem istok–zapad na koti 199.71. Visinske kote o~uvanog isto~nog zida su ¯199,66. 3.1.5. Sloj B2 – na koti ¯199.74 sloj intenzivno rastresitog kre~nog malternog {uta. Nailazi se na fragmente keramike i fresaka. 3.1.6. Sloj B1 – na koti ¯199.74 sloj intenzivno rastresitog kre~nog malternog {uta. Nailazi se na fragmente keramike i fresaka i ve}e i mawe komade malternog poda na koti 199.30. Na koti ¯199.33, na licu zidova 29, 32 i 41 konstatuju se ostaci fresko-maltera in situ. STARINAR LXIV/2014 3.1.7. Humus + A – sloj vegetacije i mrke rastresite zemqe. Bez nalaza, osim recentnih. U sredi{woj koncentraciji kre~nog, malternog {uta, na x = 3,10; y = 6,70 i koti 199.83, ukopana je do kote 199.58 definisana JAMA 1. Jama 1 je recentna, ukopana za betonski stub. Na koti 199.7 konstatovana je JAMA 2 – ukop za betonski stub. U SI uglu sonde konstatovana je i JAMA 3 na koti 199.77. 3.2. Prostorija w-22: 3.2.1. malterna podnica se konstatuje na koti 198.70 i u padu je ka zapadu i severu. 3.2.2. Sloj D – glinoviti sloj, proslojak „zape~ene“ glinovite zemqe, sadr`i i intenzivno gare` i le`i na podu, na koti 198.81. Konstatovano je i vi{e ve}ih komada fresaka in situ. Ovi komadi fresaka su okrenuti licem nagore, ali i licem ka podu prostorije. 3.2.3. Sloj C – rastresit sloj sa~iwen od kre~nog malternog {uta izme{anog sa rastresitom zemqom {qunkovite strukture. U ovom sloju se konstatuje ve}a koli~ina gare`i, lomqene opeke i kamena, kao i krovne opeke i ve}i i mawi komadi malternog poda. 3.2.4. Sloj B2 – sivomrka rastresita zemqa u kojoj se konstatuju komadi lomqenog kamena i lomqene opeke. Na koti 199.79, koncentracija kamena i opeke u sloju intenzivno rastresite zemqe je pepeqastosive boje i gotovo peskovite strukture. Sredina je na koti 199.65. Na koti 199.56 konstatovana je kruna zida 33 (severni zid prostorije w-22). 3.2.5. Sloj B1 – sloj kre~nog malternog {uta. Jugozapadni ugao je na koti 199.52, a jugoisto~ni ugao na koti 199.41. U ju`nom delu, na koti 199.66, konstatovana je kruna zida 32. Predstavqa sloj ru{ewa objekta. 3.2.6. Humus + A – sloj naboja za put. 3.3. Prostorije w-24 i w-24a (sonde 46 i 45) 3.3.1. Pod prostorije 24 – sredi{wi deo ka istoku je na koti 198.79; sredi{wi deo ka zapadu je na koti 198.78. Sredina poda prostorije 24a nalazi se na koti 198.81. 3.3.2. Sloj E – sloj svetlomrke glinovite zemqe u kojoj dominira grumewe gare`i i koji je na vi{e mesta zape~en, odnosno bio je izlo`en gorewu. Sloj le`i na podu prostorije 24. Karakteri- 13 242 AP = inventar arhitektonske plastike VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) 1a. Pravougaonog je preseka, paralelnih vertikalnih ivica. Dno ovog ukopa konstatovano je na nivou malterne supstrukcije, na koti 198.77. Prilikom iskopavawa se ovi ukopi nisu mogli me|usobno razdvojiti, pa su tretirani kao jedan ukop. 3.3.8. Humus + A – sloj sivomrke rastresite zemqe, predstavqa ostatak modernih aktivnosti. Povr{inske kote su: u severnom delu 200.39 i ju`nom delu 200.31. Dowe kote su: u severnom delu 199.80 i ju`nom delu 199.92. Tabela 4 / Table 4 Sonda 45, p. 24 i 24a ±200.46 ±200.13–199.5 ±200.13–199.5 ±199.63–199.06 ±199.63–199.06 ±199.43–198.86 ±199.43–198.86 ±198.84 STARINAR LXIV/2014 humus + A kraj humus + A po~etak B kraj B po~etak C kraj C zape~ena zemqa kraj zape~ene zemqe sti~ni su nalazi gvozdenih klinova, koji su u kasnijim slojevima pronala`eni znatno re|e. 3.3.3. Sloj D – sloj sivomrke rastresite zemqe u kojoj se sporadi~no nailazi na grumewe gare`i. U dowem nivou ovog sloja, u sredi{wem delu prostorije, konstatovan je nivo intenzivnog ru{ewa, sa~iwen od obru{enih krovnih opeka, komada malterne podnice i fragmenata mermerne oplate (zona 1). Karakteristi~ne su velike koli~ine fragmenata mermerne oplate i mozai~ke kockice. Sloj D se javqa izme|u slede}ih niveleta: gorwe kote – u severnom delu 199.14 i ju`nom delu 199.07; dowe kote – u severnom delu 198.89 i ju`nom delu 198.88. 3.3.4. Sloj C – sloj svetlobraon boje, rastresitog kre~nog malternog {uta s mawe lomqenih opeka i kamena, ali s velikom koli~inom mermerne oplate i mozai~kih kockica. Sloj C se javqa izme|u slede}ih niveleta: gorwe kote – u severnom delu 199.41 i ju`nom delu 199.29; dowe kote – u severnom delu 199.14 i ju`nom delu 199.07. 3.3.5. Sloj B – sloj bele boje, intenzivno rastresitog kre~nog malternog {uta u kome je konstatovana velika koli~ina kamena i lomqene opeke. U severnom delu sonde je prese~en ukopima 1a i b, dok sa dowe strane ovaj sloj zatvara sloj C, koji je stariji. Javqa se izme|u slede}ih niveleta: gorwe kote – u severnom delu 199.80 i ju`nom delu 199.92; dowe kote – u severnom delu 199.41 i ju`nom delu 199.29. U ovom sloju su konstatovani anti~ki i sredwovekovni nalazi. 3.3.6. Ukop 1a – relativne je {irine 1,40 m, levkastog je preseka i ukopan je iz recentnog sloja do nivoa ispod malterne supstrukcije poda, u sloj mrke glinovite zemqe (sloj F). Dno ukopa je konstatovano na koti 198,70 m, a nivo ukopa u sloju humusa + A. 3.3.7. Ukop 1b – relativne je {irine 0,80 m, ukopan je iz recentnog sloja, ali je mla|i od ukopa 4. Severoisto~ne prostorije (sl. 9, 10) Ove prostorije bile su iskopavane u ranijim istra`ivawima vile, tako da su u iskopavawima 2010/11. godine izvr{ena samo reviziona iskopavawa. Tako su dobijeni podaci o hipokaustnom postrojewu, koje je bilo u jednom trenutku negirano, a na starijem podu postavqena je nova podnica. 4.1. Prostorija e-22, kontrolna sonda 1 Dimenzija je 1 m u pravcu zapad–istok i 4,90 m sever–jug od profila ispred zgrade muzeja i ulaza u prostoriju e-22. Malterni pod je na ∇198.94; ispod malternog poda nalazi se supstrukcija od kamena i opeka, koja le`i na starijem podu od opeka. Stariji pod u ovom delu le`i na dva paralelna zida koji su bili omalterisani, pravca S–J, izme|u kojih su stubi}i. Na mestu ulaza u prostoriju e-22, na jugu, utvr|eno je postojawe stubi}a hipokausta. U severnom profilu kontrolne sonde 1 vidi se proslojak gare`i, koji se nije mogao izdvojiti prilikom kopawa slojeva, i to samo u tom, severnom delu sonde. U sredi{wem delu kontrolne sonde 1 utvr|eno je postojawe proslojka podnice (?), koja se prostire od 1,84 m od severnog profila kontrolne sonde do 3,10 m, na pre~niku oko 2 m i koti ∇ 198.60 m. Ispod hipokausta je utvr|eno postojawe podnice: ∇ 198.29. 4.2. Prostorija e-22, kontrolna sonda 2 Postavqena je kontrolna sonda 2 u zapadnom delu prostorije e-22, izme|u prolaza zida u du`ini od 3, 2 m u pravcu sever–jug i 0,7 m zapad–istok. Ispod tankog sloja ru{ewa nai{lo se na dovratnik i deo podnice. Dovratnik je prona|en u ju`nom delu kontrolne sonde, uz ju`ni deo zida 3, i sa~uvan je kao deo poplo~awa od malterske supstrukcije. Iznad wega je prona|ena podnica, u severoisto~nom delu kontrolne sonde 2, a zapadno od we nekoliko pore|anih oblutaka kamena. Kotirani su poplo~awe u ju`nom delu (∇ 198.77), kamewe i osnova 243 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 9. Severoisto~ne prostorije: prostorija e-23, kontrolna sonda 1, osnova iskopa, snimak sa zapada Sl. 10. Severoisto~ne prostorije: sonda 71, konstrukcija prefurnijuma, sa severa Fig. 9. North-eastern rooms: room e-23, control sondage 1, base of excavation, photographed from the west Fig. 10. North-eastern rooms: sondage 71, praefurnium construction, from the north u zapadnom delu (∇ 198.77, 198.70, 198.78) i podnica u isto~nom delu (∇ 198.75). 4.3. Iskopavawe severoisto~nog ugla prostorije 23 Na povr{ini ovog prostora, ispod recentnog sloja tamnosive rastresite zemqe, otkriven je originalni deo zida Z7 (∇ 198.67 m), supstrukcija poda prostorije 23 (sa~iwena od kre~nog maltera pome{anog sa kamenom i lomqenom opekom: najvi{i nivo na ∇ 198.63 m, najni`i na ∇ 198.52 ), kao i jedna opeka in situ (∇ 198.70), koja predstavqa ostatak originalnog poda prostorije 23. 4.4. Prostorija e-23, kontrolna sonda 1 Kontrolna sonda je postavqena u severozapadnom delu prostorije 23. Iskopano je oko pola metra sloja ru{ewa, gde se uo~ava recentni sloj ru{ewa (nastao prilikom konzervacije zidova prostorije), koji je pome{an sa anti~kim slojem (velika koli~ina maltera pome{ana sa opekom, kamewem i fresko-malterom). Opeke sa~uvane ispod sloja ru{ewa su na ∇ ¯198.51. Ispod tog poda nai{lo se na deo hipokausta, koji je o~uvan na razli~itim nivoima u sondi. Kotirane su opeke u zapadnom delu pored zida (∇ 198.52), zatim velika opeka u ju`nom profilu kontrolne sonde (∇ 198.74) i opeka u isto~nom delu sonde (∇ 198.33). Izme|u stubi}a hipokausta nabijena je tamnocrvena zemqa pome{ana sa malterom. Sloj je tanak (od nekoliko centimetara). 4.5. Prostorija e-23, kontrolna sonda 2 Kontrolna sonda 2 je postavqena u pravcu sever–jug za 2 m (ukupno 3,5 m) i 1,6 m u pravcu zapad–istok uz zapadno lice zida prostorije 22 i ju`no od zida u hipokaustu. U isto~nom profilu je ustanovqeno postojawe dve podnice – gorwi nivo malterne podnice = ∇ 198.95, dowi nivo (podnica od opeka) = ∇ 198.82, izme|u kojih se nalazi {ut (supstrukcija malternog poda). Ispod doweg, starijeg poda nalazi se sloj tvrde, nabijene zemqe (zdravica). 4.6. Prostorija e-24 – sonda 70 Veli~ina sonde iznosi: AB = 3.20 m, CD = 3.07 m, AC = 7.70 m, BD = 7.53 m. Vrednosti kota uglova sonde 70 pre po~etka iskopavawa iznose: A = ∇ 199.5, B = ∇ 199.5, C = ∇ 199.81, D = ∇ 199.8. [ut sa krupnim fragmentima kamewa, opeke, maltera i nalazima recentnih predmeta konstatovan je kroz celu sondu, i to su ostaci gra|ewa muzejske zgrade 1936. godine. Konstatovana je podnica od opeka, ispod kojih je prime}ena podnica `ute boje: ∇ ¯198.9. 5. Jugozapadni sektor (sl. 11) 5.1. Jugozapadne prostorije Pre komentara stratigrafije da}emo neke uvodne napomene. Tokom kampawe 2010/11. istra`ivane su prostorije w-21 (sonda 78), w-9 (sonda 75), w10 (sonda 76), w-11 (sonda 77) i zapadni deo w-12 (sonda 79). Treba odmah re}i da su prostorije w-21 244 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 11. Isto~ni profil sonde 81 (skica 201/11) Fig. 11. East profile of sondage 81 (sketch 201/11) i w-9, delom i w-10, bile intaktne, dok su u ostalim ra|ena reviziona iskopavawa jer nisu bile dokopane u radovima 1972. godine. Od jugoisto~nih prostorija reviziono su istra`ene: sredwi deo prostorije 12 (sonda 107) i isto~ni deo prostorije e-12 (sonda 80). Ispred tih prostorija iskopavan je portik, a wegova jugozapadna polovina bila je podeqena na slede}e sonde: 81; 83; 84 i 86, i ulazni deo u vilu (sonda 101). Sonde 81, 83 i delom 84 bile su netaknute, a u sondama 86 i 107 izvr{eno je reviziono iskopavawe, kao i u sondama po~etka jugoisto~ne polovine portika: sonde 82 i 105. U reviziji je iskopavan mali deo zaostalog sloja C (ru{evinski sloj) i gorwi nivo sloja D. Stratigrafija se i u ovom sektoru unekoliko razlikuje od severozapadnih prostorija. Pre svega, ovde nema podova od opeka ili maltera, ve} je nabijena zemqa (gorwi nivo sloja D, u kome je na|eno dosta novca). Iznad je jak ru{evinski sloj krovnih opeka (sloj C) i on potpuno zatvara sloj D. Ovaj podatak je dragocen, jer je na|eni novac iz zatvorene celine, a ve}ina pripada periodu 364–375. godine, neposredno pred prvo ru{ewe vile, koje upravo zbog ovih nalaza mo`emo da pripi{emo provali Gota posle hadrijanopoqske bitke 378. godine. Iznad sloja C je drugi ru{evinski sloj ozna~en kao nivo a. I sloj i nivo su nastali od ru{ewa krova, zidova i stubaca trema. U okviru ta dva sloja konstatovani su ~itavi delovi zidova i stubaca, i to onako kako su pali prilikom ru{ewa. Iznad nivoa a nalazi se sloj sekundarnog ru{ewa (sloj B). U jugoisto~nom delu, me|utim, u sondi 80 je konstatovano malo postrojewe (za kupawe? kada?) koje je sagra|eno na gare`i koja le`i na sloju D. Ova konstrukcija bi pripadala medijanskom horizontu 3, {to je jasno, jer je „kada“ prese~ena bunarom u kome je na|ena kasnosredwovekovna keramika. Sli~an bunar otkriven je i u sondi 105. I ovde dajemo tabelu sredwih vrednosti debqine slojeva: 245 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) 5.2. Jugoisto~ne prostorije 5.2.1. Prostorija e-12 (isto~ni deo – sonda 80) Ova sonda je iskopavana u 1978. godini, ali nije bila zavr{ena. Kota osnove sonde 80, od koje je po~elo iskopavawe 2010. godine, iznosi: ¯198.6. U sredi{wem delu sonde otkriven je ostatak zida debqine 0,9 m, koji se pru`a u pravcu sever–jug, a oko wega, tako|e u sredi{wem delu sonde, zaravwene povr{ine prekrivene kre~nim malterom. U isto~nom delu sonde je otkrivena formacija od opeka koje le`e na tankom sloju sitnog, tro{nog kre~nog maltera (sloj je debqine oko 0,5 cm), a ispod tog sloja nalazi se ne{to kompaktniji sloj maltera (podnica?). Ve}i deo ove podnice uni{ten je iskopavawem iz prethodnih kampawa. Konstrukcija je bila gotovo pravilnog kvadratnog oblika, stranica du`ine 1,6 m. Uo~ava se da platforma od opeka le`i direktno na malternoj podlozi, koja se kontinuirano nastavqa na krunu zida 59 i preko wega vezuje, ne{to ju`nije od platforme, i za kvadratni bazen~i}. Zapadno od zida, u nivou povr{ine prekrivene kre~nim malterom, otkrivena je olovna vodovodna cev, koja se u blagom luku, u du`ini od pribli`no 5 metara, pru`a u pravcu severozapad–jugoistok, a zalazi u zapadni i ju`ni profil sonde. Olovna vodovodna cev u ju`nom delu sonde prolazi kroz osnovu zida 59 koji se pru`a u pravcu sever–jug, u sredi{wem delu sonde, te je najverovatnije da je bila postavqena u toku izgradwe samog zida. Osnova na koti je 198,40. „Kada“ – izgra|ena je izme|u zidova 17, 19 i 59. Dimenzije su joj 1,6 m h 1,5 m, a wena oplata od kre~nog maltera povezana je sa sva tri zida, dok je jedino za zid 59 utvr|eno da sloj maltera prelazi i preko krune ovog zida (verovatno tada ve} nije postojao wegov nadzemni deo). Sredi{wi i severozapadni deo „kadice“ uni{ten je ukopavawem bunara. Bunar – Kote wegove gorwe ivice su na 198.2, a dna na 196.05; ukupna dubina bunara iznosi pribli`no 2,15 m, gledano od nivoa dna bazena do wegovog dna. Wegov gorwi pre~nik iznosi 1,35 m. Dno bunara ukopano je u `utu zemqu – zdravicu. Na dnu, ~etiri nagorele daske formiraju kvadratnu konstrukciju. U gorwim slojevima bunara otkriveno je nekoliko fragmenata kasnosredwovekovne zeleno gle|osane keramike. 5.3. Ju`ni trem ispred jugozapadnih prostorija 5.3.1. Sloj E – sloj u sondi 81, u delu izme|u zida i konstrukcije 1, mrke kompaktne zemqe koja je gotovo sterilna. Od kote 197.75 postepeno prelazi Tabela 5 / Table 5 Prostorije ±199.6 ±199.37 ±199.43 ±198.55 ±198.55 ±198.39 ±198.39 ±197.45 STARINAR LXIV/2014 put humus + A po~etak B kraj B po~etak C kraj C po~. D kraj iskopa Tabela 6 / Table 6 Novac jugozapadni sektor 2010. godina Sonda Prostorija Sloj Kota 364–375. 78 21/22 sokl 198.59 B 335–336. 75 9 198.75 C 271–275. 75 9 198.49 B 355–361. 76 10 198.52 B 355–361. 76 10 198.53 B 355–361. 76 10 198.54 B 355–361. 76 10 198.54 2011. godina Prostorija Sloj Kota D 4. vek w-21 198.48 D 364–375. w-21 198.46 D 364–375. w-21 198.46 D 364–375. w-21 198.46 D 364–375. w-21 198.45 D 360–361. w-21 198.52 5.1.1. Sloj D – sloj tamnomrke kompaktne glinovite zemqe, usitweno grumewe maltera i sasvim sporadi~no grumewe gare`i. U prostoriji 12 (sredi{wi deo – sonda 107), uz severoisto~ni ugao, konstatovana je na koti 198.49 olovna vodovodna cev, koja se pru`a i kroz sondu 80. 5.1.2. Sloj C – intenzivno rastresit sloj kre~nog malternog {uta, ali je boja sloja od svetlomrke do crvenkaste. Konstatovana je velika koli~ina lomqene opeke i kamena, kao i komada malternog poda. Ru{evinski nivo od opeka i kamena konstatovan u navedenom sloju odgovara ru{evinskim nivoima koji su kasnije otkriveni u sondama 78, 81, 83 i 84, gde su ozna~eni kao nivo a. 5.1.3. Sloj B – ru{evinski sloj od opeke, kamewa i maltera. Pri dnu ovog sloja, u ju`nom delu sonde, uza zid 49, konstatuje se koncentracija gare`i. 5.1.4. Humus + A – sloj recentnog naboja za nekada{wi put. 246 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) sru{enog zida in situ. Ovi delovi zidova le`e na gorwem nivou sloja C, odnosno dowem nivou sloja B, na ¯198.8. U sondi 81 niveleta ru{ewa zidova nalazi se preko nivelete ru{ewa krovne konstrukcije, tako da tu zapravo imamo dva nivoa ru{ewa. Sloj C predstavqa nivo ru{ewa krova i krovne konstrukcije koja je stradala u po`aru i obru{ila se na sloj D. Tragovi gorewa spojnice su na koti 198.7. Nivo ru{ewa (¯198.77–198.6) le`i na svetlomrkoj kompaktnoj zape~enoj zemqi u kojoj ima dosta grumewa gare`i. 5.3.4. Sloj B2 – sloj od sive do tamnosive rastresite zemqe u sredi{wem i severozapadnom delu sonde. 5.3.5. Sloj B1 – sloj sivomrke kompaktne zemqe koja je „pro{arana“ grumewem maltera. Osnova iskopa je na koti 198.05. 5.3.6. Sloj B – sloj intenzivno rastresitog kre~nog malternog {uta, krupnih komada kamena i lomqene opeke. U severoisto~nom delu se u ovom sloju sporadi~no nailazi na koncentraciju pepela i usitwene gare`i. Na koti 198.92 je kruni{te zida 50. Kameni blokovi su na ¯198.82. 5.3.7. Sloj A1 – proslojak svetlosivo-mrke kompaktne zemqe u kojoj dominira kre~ni malter – konstatuje se ispod humusnog sloja, na ¯199.2. 5.3.8. Humus + A – sloj naboja za nekada{wi put. Sivomrka rastresita zemqa uglavnom sadr`i ostatke aktivnosti iz mla|eg ili recentnog perioda, a relativne je debqine 0,50 m. Sadr`i ostatke naboja za put, ali i komad asfalta u severnom delu. Ovom sloju pripadaju i dva ukopa iz mla|ih ili recentnih perioda. 5.4. Ulazni deo trema (sonda 101)14 5.4.1. Sloj C – svetlomrka kompaktna zemqa sa grumewem maltera i usitwenom opekom, uz zapadnu bazu stuba ulaza u Vilu sadr`i ostatke paqevine – gare`. 5.4.2. Sloj D – kote osnove iskopa iznose: ¯198.34. U ovom sloju konstatuje se kruna zida z50a na koti 198.49. Tabela 7 / Table 7 Jugozapadni sektor Trem Sonda 81 ±199.5 put ±199.6 ±199.37–198.34 kraj puta ±199.37 ±199.37–199.07 po~. B 199.2 ±198.77 kraj B ±199.22 po~. C ±198.77 ±198.8 ±198.27 kraj C 198.05 ±198.27 po~. D ±197.64 kraj iskopa ±198.8 ±198.5 ±198.5 ±198.24 Sonda 83 ±199.82–199.51 humus + A ±198.24 ±199.57 kraj h. + A ±197.75 ±199.57–198.69 po~. B ±199.27 kraj B ±199.27 po~. C ±198.62 kraj C ±198.62 po~. D ±198.4 kraj iskopa STARINAR LXIV/2014 Dnevnik put h. + A A2 B kraj B B1 B2 C kraj C D kraj D E kraj iskopa Tabela 8 / Table 8 Novac jugozapadni sektor 2010. godina Sonda Sloj Kota 364–375. 81 trem B 198.59 367–375. 81 trem B 198.59 2011. godina A 271. zid 49 0 D 198.41 355–361. trem u sloj svetlomrke kompaktne glinovite zemqe – zdravice. 5.3.2. Sloj D – sloj mrke kompaktne zemqe. Sporadi~no se nailazi na grumewe gare`i i tragove gorewa, a uza zid z-49 gare` se konstatuje i u osnovi iskopa. Na koti 198.49 na|ena je kasnolatenska bronzana fibula s posuvra}enom stopom (S-270). Konstatovan je i jedan fragment praistorijske keramike. 5.3.3. Sloj C i nivo a – intenzivno rastresit sloj kre~nog malternog {uta koji sadr`i ogromnu koli~inu opeke iz zida, komade malternih spojnica i sporadi~no komade kamena. Sporadi~no se nailazi na grumewe gare`i i pepela. Konstatovani su i komadi zidova/stubova u ru{ewu, a vidi se i deo 14 U kv. Q/11/72 i 12/72 teren je bio nasut. Taj nasip je nastao 1933. godine kada je Aleksandar Nenadovi} iskopao statuu Dea Dardanica. Ova prostorija bila je otvorena prema ulaznom prostoru peristila. U SZ delu prostorije, na nivou poda, na|en je novac: …Val… (Valens), a na reversu GLORIA ROMANORUM (Jankovi}-Mihald`i} 2008, 51, br. 111 – (t. i. br. 448): 366–367. godina). 247 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 12. Severozapadne prostorije 24 i 24a, snimak sa severa Sl. 13. Sonda 45, prostorija w-24a, deo sloja ru{ewa sa fragmentima fresaka, snimak sa istoka Fig. 12. North-western rooms 24 and 24a, photographed from the north Fig. 13. Sondage 45, room w-24a, part of the layer of destruction with fresco fragments, photographed from the east temeqnog pro{irewa. Na osnovu sa~uvanih otisaka opeka mo`e se pretpostaviti da su bili prose~ne debqine od 0,68 m, zidani kamenom i opekom u tehnici opus mixtum. Spoqwi severni i zapadni zidovi zapadnih prostorija imaju debqinu 0,80 m i izgra|eni su tako|e u tehnici opus mixtum. Severozapadni ugao prostorije w-23 izgra|en je u celosti od opeka u du`ini od oko 1,40 m u pravcu istoka i juga. Temeqi su bili ukopani u temeqni rov {irine do 0,80 m i dubine od 0,80 do 1,00 m. Ra|eni su od lomqenog kamena i opeke i zaliveni kre~nim malterom sa dodatkom sitno lomqene opeke i re~nog {qunka. Podovi zapadnih prostorija izvedeni su razli~ito. U prostorijama w-22 i w-24a podovi su bili od dobro i kvalitetno zagla~anog maltera sa izra`enim malternim lajsnama uza zidove, dok su podovi u prostorijama w-24, w-23a i w-13 slo`eni od opeka razli~itog formata i ukra{enih reqefima u obliku dijagonala i valovnica. Pod u prostoriji w-1 je od zemqe (zdravice? – sloj E). Pod u prostoriji w-24 bio je prete`no od opeka, ali su pojedine povr{ine bile od maltera (sl. 12). 5.5. Isto~ni deo ju`nog portika (sonda 105) 5.5.1. Sloj D – mrka kompaktna zemqa; kota osnove iskopa iznosi 198.49. Bunar (jama 5) – pre~nika je oko 2,00 m. U woj su, osim {uta, na|eni i fragmenti kasnosredwovekovne keramike. Na koti 196.12 otkrivene su ~etiri drvene daske, {irine oko 10–15 cm i debqine oko 5 cm, pore|ane tako da obrazuju kvadrat. Dno bunara (pesak i {qunak) nalazi se na koti 195.96 m. ANALIZA OTKRIVENE ARHITEKTURE 1. Severozapadne prostorije i prostor severozapadno od severne fasade vile (sektori 3 i 1) Tokom 2010. i 2011. godine su u potpunosti istra`ene prostorije w-22, w-24–w-24a i w-1. Na ovaj na~in su u zapadnim prostorijama objediwena ukupna istra`ivawa zapo~eta u ju`nom delu prostorija w-1 i w-25 u 1969. i 1972. godini (Q. Zotovi}),15 i u sondama 45–47 iz 2003, 2005. i 2006. godine (M. Vasi}). Utvr|eno je sedam prostorija, raspore|enih u dva reda, izme|u sve~ane dvorane/ triklinijuma (isto~no) i hodnika prema termama (zapadno). O~uvanost zidova je nejednaka. U visini od oko 0,80 m sa~uvani su pregradni zidovi izme|u prostorija w-23 i w-22, i w-13 i w-1, dok su ostali zidovi sa~uvani neznatno (do 0,20 m) ili na nivou 15 Obele`avawe prostorija preuzeto je iz Dnevnika istra`ivawa Q. Zotovi}. Prostorije na isto~nom i zapadnom krilu imaju iste arapske brojeve, uz koje su dodate skra}enice w i e. 248 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 14. Deo hipokausta i otvora zatrpanog prefurnijuma u prostoriji w-22, snimak s juga Fig. 14. Part of hypocaust and the opening of covered praefurnium in room w-22, photographed from the south Na sa~uvanim zidovima otkrivene su freske in situ, na kojima su prikazani motivi mermernih oplata. Znatan broj fragmenata fresaka i mermerne oplate otkriven je u ru{evnim slojevima iznad podova. Velika koncentracija gorelog fresko oslikanog lepa, s otiscima trske, od tavanice na podovima pojedinih prostorija ukazuje na drvenu slikanu tavanicu (sl. 13). Na severnom, spoqwem zidu u prostorijama w-22 i w-24a na|eni su ostaci lu~nih konstrukcija od prefurnijuma. U o{te}ewima poda, uz severnu fasadu sa~uvani su delovi sistema za podno i zidno grejawe, koji su ~inili hipokausti i tubulusi (sl. 14). Na osnovu rasporeda hipokausta moglo bi se zakqu~iti da su sistemom grejawa bili obuhva}eni obimni i centralni delovi podova. Zidno grejawe bilo je primeweno samo na severnim zidovima tih prostorija. Severno od prostorije w-24, u wenoj osi, otkrivena je prostorija kru`ne osnove – stibadijum B, sa dva aneksa na istoku i zapadu (sl. 15). Zidovi sti- badijuma sa~uvani su u visini od 0,20 m do 0,40 m. Pod je bio ukra{en dobro o~uvanim mozaikom visokog kvaliteta, a zidovi su imali oplatu od mermernih plo~a na|enih in situ (sl. 16). Na podu se prepoznaje motiv od kru`nih poqa, sa rozetama, ograni~enih prepletenim trakama. Na mozaiku su sporadi~no le`ali, licem okrenuti nadole, delovi mozaika od staklenih tesera koji poti~u sa svoda i sa zidova stibadija B. Na pojedinim od wih bili su sa~uvani tragovi zlatnih listi}a. Sredi{wi deo stibadijuma o{te}en je rovom. Ima indicija da se u sredi{tu kru`ne prostorije nalazila fontana. Na severnoj strani, otkriven je razru{en prefurnijum.16 Sa zapadne strane ulaza, u uglu, naknadno je prizidana prostorija patosana opekama 16 Rov nije ~i{}en zbog mogu}nosti o{te}ewa mozaika i nepoznavawa stawa hipokausta. 249 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 15. Severozapadni deo vile, van severne fasade vile, stibadijum B, pogled sa severa Fig. 15. North-western part of the villa, outside the villa’s northern facade, stibadium B, viewed from the north Sl. 16. Detaq ornamenta mozaika iz isto~nog dela stibadijuma B Sl. 17. Severozapadni deo vile, van severne fasade vile, sonda 90, blok A, zid z-2 Fig. 16. Detail of a mosaic ornament from the eastern part of stibadium B Fig. 17. North-western part of the villa, outside the villa’s northern facade, sondage 90, block A, wall z-2 250 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 18. Severoisto~ni deo, van severne fasade, ostaci prostorije, posle 380. godine; sonda 97, snimak sa severozapada (konstrukcija u sloju V) Sl. 19. Jugozapadni deo vile, sonda 78 – prostorija w-21 i 81, nivo ru{ewa (u pozadini sonde 75 prostorija w-9, 76 – prostorija w-10, sonde 81, 85), snimak sa zapada Fig. 18. North-eastern part, outside the northern facade, remnants of a room, after 380, sondage 97, from the north-west (the construction in stratum B) Fig. 19. South-western part of the villa, sondage 78 – rooms w-21 and 81, level of destruction (in the background, sondage 75 – room w-9, sondage 76 – room w-10, sondage 81 and 85), from the west i s freskama na zidovima (w-26). U neposrednoj blizini stibadijuma B, sa wegove spoqwe severne i zapadne strane (sonde 72, 74, 90, 87, 89) otkriveni su plitko fundirani zidovi, sa~uvani u jednom redu slo`enih oblutaka u suvozidu (sl. 17). Severno od wih je na|en deo zida, pravca zapad–istok, oja~an pilastrima. Zid je gra|en u alternaciji kamena i opeke u kre~nom malteru. 2. Severoisto~ne prostorije i prostor severoisto~no od vile (sektori 4 i 2) Na ovom delu gra|evine, u prostorijama e-22 – e-24 i e-24a, istra`ivawa su bila usmerena ka prikupqawu podataka koji su se ve}im delom odnosili na razumevawe funkcije i konstrukcije, a predstavqaju reviziju ranijih istra`ivawa iz 1934, 1969, 1972. i 1978–1979. godine. I na tom delu vile potvr|en je istovetan raspored prostorija kao i u zapadnom delu, koji se sastoji od sedam prostorija raspore|enih u dva reda – ~etiri na severnom delu i tri na ju`nom. Zbog stabilnosti muzejske zgrade, severni profili sondi bili su udaqeni od trema zgrade za 2,00 m i nije bilo mogu}e utvrditi polo`aj i stawe severnog spoqweg zida, kao i vezu pregradnih zidova sa wim. Izgled i konstrukciju gradwe u severoisto~nim prostorijama bilo je te{ko sagledati i usled konzervatorskih radova izvede- nih 1980. godine, kao i nerazumevawa pojedinih elemenata arhitekture tokom tih radova. Kako su konzervirani zidovi bili u stawu ru{ewa, pristupilo se wihovom uklawawu. Tom prilikom je utvr|eno da je severoisto~ni ugao vile bio u punoj visini izveden opekom na du`ini od 1,40 m u pravcu zapada i juga, kao {to je to potvr|eno i na severozapadnom uglu. Pregradni zid izme|u prostorija e-24 i e-24a pova|en je do temeqa. Polo`aj prolaza na ovom zidu mogao se samo pretpostaviti prema sa~uvanom dovratniku pregradnog zida izme|u prostorija e-22 i e-24. Na osnovu sa~uvanih tragova nadzemnog dela utvr|ena je debqina pregradnih zidova (0,68 m) i debqina spoqwih zidova (0,80 m). Zidovi su gra|eni u alternaciji kamena i opeke – tehnikom opus mixtum. Temeqi su bili istog kvaliteta, debqine i izgleda kao i u prostorijama na severozapadnom delu. Smewivawe podova od izgla~anog maltera i od opeke primetno je i u isto~nim prostorijama. Prostorije e-24a, e-22–e-23 imale su malterni pod, a prostorija e-24 pod od opeka. U prostoriji e-13 je pod od zemqe. U ju`nim prostorijama e-16a i e-1 isto~nog krila nalazi se pod poplo~an opekama, konzerviran u vreme radova 1969. godine. Otvorene su i dve probne sonde u prostorijama e-22 i e-23, 251 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 20. Sonda 81, jugozapadni ugao vile, nivo ru{ewa (u pozadini sonda 78 – prostorija w-21), snimak sa jugozapada Sl. 21. Sonda 81, situacija na jugozapadnom uglu vile, snimak sa jugoistoka Fig. 20. Sondage 81, south-western corner of the villa, level of destruction (in the background, sondage 78 – room w-21), from the south-west Fig. 21. Sondage 81, south-western corner of the villa, photographed from the south-east photographed from the south Sl. 22. Sonda 82, u toku iskopavawa, snimak sa juga Fig. 22. Sondage 82, whilst being excavated, photographed from the south 252 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 Sl. 23. Sonda 101, osnova sonde nakon iskopavawa (u drugom planu: zapadni deo ju`nog trema vile), snimak sa severoistoka Sl. 24. Sonda 101, baza zapadnog stuba ulaza u vilu, otisak u zidu z-50, snimak sa juga Fig. 23. Sondage 101, base of the sondage after excavation (in the background the western part of the villa’s southern porch), photographed from the north-east Fig. 24. Sondage 101, base of west column at the villa’s entrance, impression in wall z-50, photographed from the south storijama w-21 i w-9–10 (sonde 78, 70, 76) (sl. 19), kao i elemente za analizu konstrukcije i izgleda trema (sonde 81, 83–85, 101) (sl. 20, 21, 22). Izvr{ena su i reviziona istra`ivawa u prostorijama w-10–12 (sonda 77). Odnos sloja {uta od ru{ewa krovne konstrukcije ukazao je na na~in gradwe tih prostorija. Utvr|eni su i veli~ina, konstrukcija i izgled trema na jugu. Posebno su bili dragoceni podaci od ru{ewa stubaca od opeke, ~iji su se postamenti (kamene baze) sa~uvali du` ju`nog zida trema. Iskopavawa su pokazala da je prvobitni ju`ni trem bio pro{iren prema jugu (sl. 23, 24). Od starijeg trema sa~uvan je temeq. radi kontrole temeqa i sistema za grejawe. Potvr|en je hipokaustni sistem podnog grejawa i zidnog grejawa tubulusima, sli~no nalazima u severozapadnim prostorijama vile (sl. 9). U sondama severoisto~no od vile (sektor 2) ispitan je temeq severnog zida vile i utvr|eni su wegova dubina i na~in fundirawa. Tom prilikom je utvr|eno postojawe prefurnijuma, koji nije zabele`en u istra`ivawima 1978–1979. godine, a konzervatorskim radovima je i potpuno negiran (sl. 10). Ostali delovi arhitekture, van gabarita vile, istovetni su sa zidovima od oblutaka u suvozidu otkrivenom i na prostoru oko stibadijuma B. Konstantovana je i pravougaona konstrukcija prislowena uz severnu fasadu vile, kao i zid pravca sever– jug na du`ini od 17,00 m (sl. 18). Wihova konstrukcija, izgled i orijentacija su razli~iti od zidova vile. O~i{}ena je i sonda B iz 2008. godine (M. Kora}) i ponovo analizovana veza spoqwih zidova vile i prislowenog zida od opeka u slogu „ribqe kosti“, koji verovatno predstavqa deo konstrukcije trema postavqenog na isto~nom krilu vile. 3. Jugozapadne prostorije i ulazni deo vile (sektor 5) Arheolo{ka i arhitektonska istra`ivawa na jugozapadnom i ulaznom delu vile pru`ila su podatke za potpuno sagledavawe arhitektura u pro- POKRETNI ARHEOLO[KI MATERIJAL Pokretni materijal na|en u ove dve kampawe je, osim novca, malobrojan. U ovom izve{taju je dat samo sumaran spisak na|enih predmeta, kako bi se stekla osnovna slika. Materijal nije konzerviran te se ne mogu dati neki precizniji opisi, a i bilo bi suvi{e op{irno za ovu namenu. Za neki drugi rad ostaje da se dâ kompletna katalo{ka obrada. Otkriveni predmeti, iako dobro stratificirani, nisu hronolo{ki osetqivi i ne pru`aju neke jo{ 253 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 (dim. 8,7 cm). Gvozdene i bronzane pincete su u anti~ko doba predstavqale vrlo rasprostrawene instrumente, pri ~emu se smatra da su kori{}ene u medicinske i kozmeti~ke svrhe (kao toaletni pribor i kao ukosnice).23 Na prvi pogled, materijal na|en u sloju V pripadao bi periodu posle 380. godine, dakle vi{e odgovara materijalu sloja C. Postoje dve mogu}nosti. Ili je materijal u sloj V dospeo kasnijim prekopavawem sloja C, ili }e detaqnija analiza pokazati da se delom radi o sredwovekovnom materijalu koji bi mogao da se na|e u sloju V. Sloj S – u sondi 72, u kampawi 2011. godine, na kotama izme|u 198.87 i 198.74, prona|eno je na mawoj povr{ini vi{e fragmenata skulptura od porfira i beloga mermera: S-253 – fragmentovana `ivotiwska {apa (lav, tigar?) od belog mermera; S-282 – fragment glave, sa delimi~no o~uvanom frizurom, jednim okom i korenom nosa od belog mermera (sl. 25); S-294 – fragment porfirne skulpture, o~uvana leva {aka u kojoj se nalazi glob/jaje, sa naborima draperije preba~ene preko ruke24 (sl. 26); S-295 – fragment porfirne skulpture koji predstavqa zmiju prislowenu uz skut Higijine haqine (deo zmije koju Higija hrani);25 S-296 – fragment desnog stopala sa o~uvanim delom sandale i draperije;26 S-297 – fragment skulpture od crvenog porfira, vaqkastog oblika; S-298 – fragmentovana mermerna zmija obmotana oko drveta/{tapa; sigurnije potvrde zakqu~aka koji su ve} izneti u vezi sa stratigrafijom. 1. Sonde severozapadnog sektora (severozapadno od severne fasade vile sa peristilom, sonde 72, 73, 88, 90, 86, 87, 42/2004, 89 i 91) Sloj A – dve fragmentovane gvozdene strelice S-327/10 i S-332/1017 (sonda 88), pqosnatog trougaonog oblika, sa koso zase~enim dr{kama, ~iji je tip poznat i kao tip pqosnatih trougaonih strelica, a kao analogije se mogu navesti primerci sa lokaliteta Pontes i U{}e Pore~ke reke.18 Sloj B – prona|eni su slede}i nalazi: fragmentovani bronzani struga~ S-286/10 (sonda 86), ~ije se trakasto telo lepezasto {iri na jednom kraju;19 fragment bronzanog predmeta S-287/10 (sonda 86), trakastog oblika, pravougaonog preseka tela koje je blago savijeno i su`ava se prema jednom kraju; ko{tani rog, sa tragovima crvene boje S-328/10 (sonda 86), koji ima primetne tragove se~ewa i gla~awa i pretpostavqa se da je pripreman za izradu dr{ke. Sli~ni primeri ko{tanih rogova se tako|e datuju u IV–V vek n. e.;20 gvozdena alka S-322/10 prona|ena u kontrolnom profilu izme|u sondi 86 i 87. Sloju V tako|e pripada i bronzana kop~a S-278/10 (sonda 87), koja je, sude}i po wenim malim dimenzijama (2,2 cm h 1,4 cm h 0,1 cm), najverovatnije kori{}ena kao kop~a za obu}u. Kop~a je ovalnog oblika, s trnom pravougaonog oblika i preseka, i okovom sa perforacijama po obodu. U sloju V je prona|en i fragment crne staklene narukvice S-264/11 (sonda 87), koji pripada najjednostavnijem tipu staklenih neukra{enih, jednobojnih narukvica. Staklene narukvice ovoga tipa se datuju u razdobqe od III do prve polovine V veka n. e.21 Jo{ jedan fragment narukvice od crnog stakla S-85/11 (sonda 90) tipa staklenih jednobojnih narukvica ukra{enih ispup~ewima razli~itih oblika prona|en je u istom sloju. Usled razli~itog ukra{avawa, postoji vi{e podtipova, a na{ nalaz S-85/11 pripada tipu narukvica ukra{enih blago zaobqenim rebrima. Ovaj podtip staklenih narukvica se datuje u razdobqe od III do kasnog IV veka n. e.22 U sloju V sonde 90 prona|eni su jo{ i staklena dvo~lankovita perlica S-91/11, tamnoplave boje, sa perforacijom za nizawe, i fragment `ute staklene narukvice S-92/11, koji pripada tipu jednobojnih neukra{enih staklenih narukvica. Na preseku slojeva V i S prona|ena je pinceta S-98/10 (sonda 73) koja ima vrh nepravilnog kru`nog oblika, od kojeg se ra~vaju dva ravna kraka 17 Brojevi 10 i 11, koji }e slediti nakon broja S nalaza, ozna~avaju 2010, odnosno 2011. godinu arheolo{kih istra`ivawa vile sa peristilom na Medijani. 18 [pehar 2010, 129. 19 Ibid, 103. 20 Petkovi} 1995, 56, 99. 21 Jelin~i} 2009/2010, 111–112. 22 Spaer 1988, 55–56; Jelin~i} 2009/2010, 112. 23 Pincete se po obliku krakova mogu podeliti na one sa ravnim i i one s malo zavijenim krakovima, dok im dimenzije variraju od 5 cm do 10,5 cm: Deringer 1954, 152; Grupa autora 1997, 252, br. 415 i 416; Gregl 1982, 181–182. 24 Ustanovqeno je da porfirni fragment skulpture C-294, kao i C-296, pripada skulpturi boga Asklepija prona|enoj 1972. godine u prostoriji w-4 vile sa peristilom (u anti~koj zbirci skulptura boga Asklepija je pod inv. br. 985). 25 Identi~na predstava kao kod Higijine porfirne statuete s natpisom na|ene 1972. godine u prostoriji w-4 vile s peristilom (u anti~koj zbirci skulptura Higije je pod inv. br. 986). 26 Ibid. 27 Grupa autora 1997, 69, br. 47; Grupa autora 1969, 87–88. 254 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) S-299 – fragmentovano mermerno postoqe skulpture sa fragmentovanim levim stopalom i urezanim tekstom na gr~kom jeziku; S-300 – fragment skulpture od belog mermera koji predstavqa deo leve {ake sa okruglim atributom (glob/jaje?) (sl. 27); S-301 – fragment dela prsta od belog mermera; nalazi S-302, S-303, S-331, S-332 i S-340 koji predstavqaju delove sklupturalnog ukrasa (delovi grana od podupira~a skulpture ?); fragment S-305 koji verovatno predstavqa deo kopita, i nalaz S-306 kojim je obuhva}eno vi{e sitnih delova skulptura od belog mermera, koje, usled wihovog fragmentovanog stawa, nije mogu}e pravilno identifikovati. Nalaz ovih fragmenata na jednom mestu izvan vile pokazuje da je tu izvr{eno saka}ewe jednog dela medijanskih skulptura (odsecawe ruku i glave). Deo tako osaka}enih statueta, prvenstveno Asklepija i Higije, odnet je naknadno u prostoriju w-4 u kojoj su bile pohrawene i druge, isto tako saka}ene skulpture. Neke skulpture, od kojih su na|eni STARINAR LXIV/2014 fragmenti u sondi 72, jo{ nisu otkrivene i mo`da su zauvek nestale. Sloju S tako|e pripada bronzana dr{ka u obliku poprsja `enskog bo`anstva i s dowim delom u koji je uglavqena gvozdena spatula S-140/11 (sonda 90) (dim. 10, 9 cm), koja je najverovatnije kori{}ena u kozmeti~ke svrhe (sl. 28). Iako su glava i torzo bo`anstva prili~no sumarno ura|eni, na osnovu {lema sa perjanicom i karakteristi~nog oklopa – egide, prikazanog na grudima bogiwe, mo`emo sa prili~nom sigurno{}u konstatovati da je u pitawu bogiwa Minerva. Ako se ima u vidu da je Minerva bila i ijatri~ko bo`anstvo i za{titnica lekara, potpuno je logi~no weno predstavqawe na predmetu verovatno upotrebqavanom u kozmeti~ke a mo`da i medicinske svrhe. Dosta sli~nosti sa primerkom sa Medijane u pogledu oblikovawa glave i atributa bogiwe, pronalazimo u figuralnom poprsju i figurama Minerve iz Obrenovca, Zagreba, sa nepoznatih lokaliteta u Podunavqu i iz Gra~anice.27 C–282 C–300 Sl. 25. Fragment mermerne glave (S–282/11) Sl. 26. Fragment porfirne leve {ake sa okruglim atributom (glob/jaje?) (S–294/11) Sl. 27. Mermerni deo leve {ake sa okruglim atributom (glob/jaje?) (S–300/11) Fig. 25. Fragment of marble head (C–282/11) Fig. 26. Fragment of porphyry left hand with round attribute (globe/egg?) (C–294/11) Fig. 27. Part of marble left hand with round attribute (a globe/egg?) (C–300/11) C–294 255 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 C–268 Sl. 28. Bronzana dr{ka, u obliku poprsja `enskog bo`anstva, u koju je uglavqena gvozdena spatula (S–140/11) Sl. 29. Ko{tani predmet (S–268/11) Fig. 28. Bronze handle, shaped like a bust of a female deity, with iron spatula embedded in it (C–140/11) Fig. 29. Bone artefact (C–268/11) C–140 U sloju S sonde 90 severozapadnog sektora je prona|en i fragment crne staklene narukvice S239/11, koji pripada tipu jednobojnih neukra{enih staklenih narukvica, a konstantovana je i bronzana alatka S-290/11, pravougaona, s poluobli~astim predwim delom tela. 2. Sonde severoisto~nog sektora (isto~no od isto~ne fasade zgrade Muzeja i severnog dela fasade vile sa peristilom) Sloj V – fragment gvozdenog predmeta (no`?) S-124/11 (sonda 98), pravougaonog oblika, deo trakaste gvozdene oplate S-133/11 (sonda 98), po sredini iskrivqene. Sloj S – konstantovani su: u sondi 103 gvozdeni stilus S-241/11, piramidalnog oblika, kvadratnog preseka, za{iqen pri jednom kraju, dok se ka drugom kraju postepeno trapezasto {iri (du`.13,6 cm); teg od opeke S-242/11, kru`nog oblika, jednostrano zagla~an, sa perforacijom na sredini; fragment gvozdenog predmeta S-276/11, mo`da jednoseklog no`a, i dva bronzana fragmenta S-316/11, vaqkastog oblika tela i blago povijenih krajeva jednog ka drugom, tako da predmet ima lu~an oblik (razvu~ena narukvica ?). Krajevi oba fragmenta su u obliku glave divqe ma~ke (leopard, gepard), sa perforacijama u vidu o~iju, u{iju i {ara. Iz sonde 104 poti~e ko{tana cilindri~na perla S-268/11 sa perforacijom po du`ini i sa urezanim ornamentima linija i okastih ureza (sl. 29). Ovaj tip perli je ~est tokom ~itavog carskog perioda, odnosno od I do kraja IV veka n. e. i ima analogije u mnogobrojnim primercima sa vi{e lokaliteta.28 3. Severozapadne prostorije vile sa peristilom (sonde 45 i 46) Sloj A – u ukopu 1 prona|eno je bronzano kalotasto dugme S-12/10 (sonda 45), elipsoidnog oblika, sa dve perforacije na vrhu. Prostorija w-1 Gorwa povr{ina sloja S – u prostoriji 1 (kontrolni profil izme|u sondi 46 i 47) prona|eni su glava Venere29 od belog mermera S-4/11 (sl. 30) i 28 Petkovi} 1995, 40. Detaqnije pogledati: Vasi}, Gavrilovi} 2012, 137–149. Prema usmenom saop{tewu Vesne Crnoglavac, vi{eg kustosa Narodnog muzeja u Ni{u, izvr{ena je fizi~ko-hemijska analiza glave i baze sa stopalima i glavom delfina i utvr|eno je da pripadaju istoj statui, pa je time i glava sigurno opredeqena kao Venerina. Rezultati ovog ispitivawa publikovani su u ~lanku: Crnoglavac, I{qamovi}, Todorovi} 2014, 73–78. Snimawe FTIR spektara ra|eno je na Tehnolo{kom fakultetu u Leskovcu, u laboratoriji za spektroskopiju. Jo{ jedanput srda~no zahvaqujemo na korisnim informacijama koleginici Crnoglavac. 30 Petkovi} 1995, 40–41, kat. br. 432–436. 29 256 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) vrh gvozdenog kopqa S-2/11 listolikog oblika, s o~uvanim tulcem kru`nog preseka. Prostorija w-24 Sloj B – fragment ko{tanog predmeta S-72/10 (sonda 46), izdu`enog oblika, blago savijen po sredini sa `lebom, ~iji je vrh na kraju koso zase~en i ima kru`nu perforaciju. Pretpostavqamo da je u pitawu privezak/amulet i to najjednostavniji tip.30 Sloj C – fragment staklene narukvice crne boje S-107/10 (sonda 46), okruglog preseka, koji pripada tipu staklenih jednobojnih neukra{enih narukvica. Sloj D – gvozdeni no` S-165/10 (sonda 45) sa usadnikom i jednoseklim se~ivom, ~ija je gorwa ivica blago povijena, u jako korodiranom stawu. Prostorija w-24a Sloj B – fragmentovano se~ivo no`a S-77/10 (sonda 45), zakrivqeno pri vrhu i u jako korodiranom stawu. Sloj C – fragmentovani trakasti gvozdeni lim S-105/10 (sonda 45) sa sedam perforacija u nizu s jedne strane, veoma korodiran; perla od pe~ene zemqe S-133/10 (sonda 45), tamnozelene boje i perforirana po sredini. Sloj D – bronzano dleto/struga~ S-166/10 (sonda 45) sa o~uvanom o{tricom, lepezastog oblika, na jednom kraju i s trakastim telom. Ni iskopavawe prostorija nije dalo ve}u koli~inu materijala, a pogotovo ne onog hronolo{ki osetqivog. Jedino mo`e da se konstatuje da sav pokretni materijal poti~e iz vremena posle prvog ru{ewa vile s peristilom, o ~emu najboqe svedo~e mermerna glava i kopqe iz prostorije w-1. 4. Jugozapadni sektor: prostorije w-21, 9, 10, 11, 12 (sonde 78, 75, 76, 77 i 79) Prostorija w-11 Sloj B – fragment kremenog se~iva S-182/10 (sonda 77), bele boje. Prostorija w-21 Sloj C – gvozdena alka S-202/10 (sonda 78), kvadratnog preseka. Sloj D – fragment bronzanog okova S-229/11 (sonda 78) s o~uvanom alkom za pri~vr{}ivawe. 4. Jugozapadna polovina portika (81, 83, 84, 85) Sloj C – prona|en je fragment pravougaonog bakarnog lima S-336/11 (sonda 83). Sloj D – prona|en je kasnolatenski tip fibule s posuvra}enom nogom S-270/11 (sonda 83) u sloju u kome se mestimi~no nalaze i fragmenti praistorijske keramike (sl. 31). STARINAR LXIV/2014 ZAKQU^AK Rezultati istra`ivawa vile s peristilom tokom 2010. i 2011. godine pru`ili su nove podatke koji su omogu}ili korekciju i redefinisawe graditeqskih faza. Sama vila je, osim prilaznog vestibula, istra`ena, ali u nekim budu}im radovima treba obratiti pa`wu na prostore izme|u ogradnog zida i vile, koji su samo na~eti sonda`nim is- C–4 C–270 Sl. 30. Mermerna glava Venere (S–4/11) Sl. 31. Kasnolatenska fibula s posuvra}enom nogom (S–270/11) Fig. 30. Marble head of Venus (C–4/11) Fig. 31. Late La Tène bent foot fibula (C–270/11) 257 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) tra`ivawima. Rezimiraju}i stratigrafske podatke, treba odmah da uka`emo na to da su, za razliku od drugih delova Medijane, u vili i oko we sa~uvani slojevi ve}e debqine. Preciznim iskopavawima bilo je mogu}e da se oni dobro prate, {to je omogu}ilo i ozbiqnu analizu ovih slojeva i sagledavawe wihove istorije. Stratigrafija je pokazala da su najstariji temeqi vile s peristilom bili ukopani u protohumus (sloj E u sektoru 1 i slojevi D u sektorima 2 i 5). Vila je, prema tome, izgra|ena ex nihilo i samo je spoqwi hodnik, izgra|en uz zapadnu fasadu, koji je spajao vilu s termama, jednim delom prelazio preko ostataka arhitekture iz starijeg horizonta.31 U drugoj gra|evinskoj fazi, teren u sektorima 1 i 2 je nasut (sloj D u sektoru 1 i sloj C u sektoru 2) i u wega su ukopani temeqi stibadijuma A i B. Mogu jasno da se prate slojevi prvobitnog ru{ewa vile 378. godine, potom weno delimi~no naseqavawe posle 380. godine i ru{ewe te faze. Posle te faze dolazi do ru{ewa zidova koji su jo{ bili vidqivi, a potom i do wihove razgradwe i prekopavawa tog {uta zbog gra|evinskog materijala (sloj B u svim sektorima). Preostali zidovi su pova|eni do temeqa, {to se jasno vidi po otiscima rovova na profilima (sloj A u svim sektorima). U tekstu je znatno insistirano na preciznom odre|ivawu slojeva, {to na prvi pogled deluje zamorno, ali je to bilo neophodno da bi se ova dobro sa~uvana stratigrafija uporedila s ostalim delovima Medijane gde su slojevi mnogo tawi i nekada te{ko uo~qivi. Tehnika gradwe otkrivenih zidova, dubina fundirawa i pojedini konstrukcioni elementi ukazuju na vi{e graditeqskih faza i na mogu}e izdvajawe osnova razli~ite organizacije, izgleda i namene. Tako su otkrivena arhitektura i wene faze u potpunoj saglasnosti s rezultatima dobijenim analizom stratigrafije, novca i pokretnog arheolo{kog materijala koji je hronolo{ki osetqiv. Izdvojene su tri osnovne graditeqske faze, koje se na osnovu arheolo{ke stratigrafije i stratigrafije ru{ewa objekta mogu opredeliti u razdobqe od po~etka 4. veka do polovine 5. veka, i to kao: horizont 1 (prve decenije 4. veka), horizont 2 (oko 330 – do gotskih upada 378. godine) i horizont 3 (posle 383 – do provale Huna 441–443. godine). Odre|ene intervencije i nalazi ukazuju i na kori{}ewe prostora tokom sredweg veka, ali u znatno mawem obimu. Osnovni gabarit objekta u prvoj i drugoj graditeqskoj fazi nije se bitno izmenio, posebno u delu STARINAR LXIV/2014 s administrativnim prostorijama – kancelarijama na isto~nom i zapadnom krilu (prostorije w-2–w-21 i e-2–e-21). Najve}e promene bile su upravo na severozapadnom i severoisto~nom delu gra|evine, kao i na ju`nom prostoru trema i ulaznog dela. Prvobitni objekat (graditeqska faza 1) bio je jednostavne, izdu`ene forme, sa velikim dvori{tem u sredini, sve~anom dvoranom/triklinijem na severu i ulazom na jugu. Brojne prostorije, razli~ite namene i veli~ine, bile su raspore|ene oko dvori{ta i povezane tremom. Na bo~nim krilima i na jugu bile su raspore|ene administrativne prostorije (officium), iste povr{ine i organizacije. Prostorije severnog krila, isto~no i zapadno od sve~ane dvorane/triklinijuma, bile su grejane i namewene du`em boravku. Ne mo`e se sa sigurno{}u utvrditi broj prostorija, ali se mo`e pretpostaviti da su one bile nejednake veli~ine, i to po dve ve}e, na severu i tri mawe, na jugu (?). Utvr|eni su prefurnijumi (~etiri) i sistemi za grejawe podova i zidova u severnim prostorijama. U prvoj graditeqskoj fazi su prostorije e-22–23; e-24–24a, w-24–24a i w-22–23 predstavqale jedinstven prostor, koji je zagrevan preko lo`i{ta na severnom zidu. U ju`nom krilu gra|evine nalazio se niz od devet prostorija, jednakih po veli~ini, a ispred wih je bio trem. U graditeqskoj fazi 2 (oko 330 – do gotskih upada 378. godine) izvr{ena je velika rekonstrukcija na severnom i ju`nom krilu kao i u dvori{tu objekta, koja je u popunosti izmenila izgled i funkciju objekta. Isto~ne i zapadne prostorije severnog krila su pregra|ene i smawene. U pojedinim prostorijama su postavqeni podovi od opeka. U prostorijama (e-23 i w-22) otkrivena su dva sloja malternog poda, koja ukazuju na obnovu podova iz prve graditeqske faze. Zidovi i tavanice su oslikani. Na severoisto~nom i severozapadnom delu prizidana su dva reprezentativna triklinijuma – stibadijumi A i B. Zidovi stibadijuma ukra{eni su oblo`nim mermernim plo~ama i mozaikom, a podovi su dekorisani mozaikom. U sredi{tu svakog od wih nalazila se fontana. Do fontane u severoisto~nom stibadijumu A dopremana je voda olovnim cevima (istra`ivawa 1972. godine), a zatim je daqe bila sprovedena do fontane u severozapadnom sti- 31 258 Sonda 41 kopana 2003. i sonda 53 kopana 2005. godine. VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) badijumu B (istra`ivawa 2011. godine). Promewen je i sistem grejawa. Prefurnijum u prostoriji w-24a nije vi{e bio u funkciji – zbog nemogu}nosti prilaza u taj deo objekta prilikom izgradwe stibadijuma B. I prefurnijum u isto~nom delu, bli`e severoisto~nom uglu, nije bio vi{e u funkciji zbog postavqawa pregradnog zida i deobe velike ugaone severoisto~ne prostorije na dve (e-22 i e-23). Mogu}e je da su preostala lo`i{ta kori{}ena i u drugoj graditeqskoj fazi. Postavqeni su novi prolazi, {irine oko 1,00 m, sa dovratnicima od opeka i nadvratnikom u vidu ravnog luka. Na ju`nom delu objekta izvr{ene su velike intervencije na ulaznom delu i potpuna promena izgleda i konstrukcije prilaznog ju`nog trema. Trem je pro{iren, i na wegovom ju`nom zidu izgra|eno je 14 stubaca, zidanih opekom, sa monumentalnim kamenim bazama. Preoblikovan i pove}an je i trem ka dvori{tu. U rekon- STARINAR LXIV/2014 strukciji su kori{}eni materijali visokog kvaliteta za izradu stubova, podova i fresko-oslikavawe. Primeweni su i novi konstrukcioni sistemi tipa svoda, kalota i polukalota. Po zavr{enoj rekonstrukciji je u potpunosti promewen izgled, ali i zna~ewe gra|evine u kojoj je tada izra`en javni, administrativni i sve~ani ceremonijalni karakter. Ovaj koncept ure|ewa zadr`ao se do gotskih upada 378. godine. Tre}oj graditeqskoj fazi (posle 380. godine) pripadaju pregradwe i zidovi od oblutaka konstatovani van gabarita stibadijuma A i B, na severozapadnom i severoisto~nom delu kompleksa. Koliko u ovoj fazi istra`ivawa mo`e da se utvrdi, radi se o velikim pravougaonim zgradama, te izgleda da naseqe posle 380. godine nije bilo nimalo bezna~ajno. U ovom trenutku, ovaj problem ostavi}emo po strani. 259 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 BIBLIOGRAFIJA: Kanitz 1892 – F. Kanitz, Römische Studien in Serbien. Der Donau-Grenzwall, das Strassennetz, die Städte, Castelle, Denkmale, Thermen und Bergwerke zur Römerzeit im Königreiche Serbien. Denkschr. Kaiserl. Akad. Wiss., phil.-hist. Classe 41,2, Wien 1892. Milo{evi} 1996 – G. Milo{evi}, Archaeological park „Mediana“ by Ni{ (Naissus), in: XIII U.I.S.P.P. Congress International Union of Prehistoric and Protohistoric Siences, Colloquium XXXVI Archaelogical Parks (ed. G. B. Montanati, S. Narayan), Forlì, 1996, 59–68. Or{i}-Slaveti} 1933/34 – A. Or{i}-Slaveti}, Arheolo{ka istra`ivawa u Ni{u i okolini, Starinar (3. s.), 8/9 (1933/34), 303–310. Petkovi} 1995 – S. Petkovi}, Rimski predmeti od kosti i roga sa teritorije Gornje Mezije, Beograd 1995. Spaer 1988 – M. Spaer, The Pre-Islamic glass bracelets of Palestine, Journal of Glass Studies 30, 1988, 51–61. [pehar 2010 – P. [pehar, Materijalna kultura iz ranovizantijskih utvr|enja u \erdapu, Beograd 2010. Vasi}, Gavrilovi} 2012 – M. Vasi}, N. Gavrilovi}, Venus or Diana from Mediana, Starinar, LXII, Beograd 2012, 137–149. Bratani} 1938 – R. O. Bratani}, Arheolo{ka istra`ivawa u Brzom Brodu, Starinar (3. s.), 13 (1938), 199–204. Crnoglavac, I{qamovi}, Todorovi} 2014 – V. Crnoglavac, D. I{qamovi}, B. Todorovi}, Primena Fourier-ove transformacione infracrvene spektroskopije u identifikaciji fragmenata skulpture sa Medijane, Zbornik radova Narodnog muzeja iz Ni{a 22, 2014, 73–78. Deringer 1954 – H. Deringer, Die medizinschen Instrumente des Ennser Museums, Forschungen Lauriacum, 2, Linz 1954, 144–155. Gregl 1982 – Z. Gregl, Rimski medicinski instrumenti iz Hrvatske I, Vjesnik Arheolo{kog muzeja u Zagrebu, 3, XV, Zagreb 1982, 175–198. Grupa autora 1969 – Grupa autora, Anti~ka bronza u Jugoslaviji, Beograd 1969. Grupa autora 1997 – Grupa autora, Anti~ka bronza Singidunuma, Beograd 1997. Jankovi}-Mihald`i} 2008 – D. Jankovi}-Mihald`i}, Pojedina~ni nalazi rimskog novca na Medijani (rez. Single finds of Roman Coins from Mediana) Naissus I, Ni{ 2008, 25–73. Jelin~i} 2009/2010 – K. Jelin~i}, Rimske staklene narukvice kao prilog poznavanju anti~koga Ore{ca, Opuscula Archaeologica, 33, Zagreb 2009/2010, 105–124. 260 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) Summary: STARINAR LXIV/2014 MILOJE VASI], Institute of Archaeology, Belgrade GORDANA MILO[EVI], University of Belgrade, Faculty of Architecture, Belgrade NADE@DA GAVRILOVI], Institute of Archaeology, Belgrade EXCAVATIONS OF MEDIANA IN 2010 AND 2011 Key words. – Mediana, villa with peristyle, stratigraphy, architecture, portable finds, late antiquity. 1.2. Between the elevations of 198.8 and 198.6 in sondages 72, 86, 90 and 89/42, above stratum F (with the exception of the praefurnium in sondage 42), a mortar base was noted over the entire surface. 1.3. Above the mortar base, a light brown soil with fine rubble of stratum D was deposited. This stratum covered the praefurnium on the outer side of room w-22, as well as part of the praefurnium in sondage 42. The foundations of stibadium B and the praefurnium on its northern side were dug into stratum D, as were the foundations of room w-26. This stratigraphy clearly indicates that stibadium B, and consequently stibadium A, belong to a different construction phase when the villa underwent significant alterations. The foundations of wall 1, of a north-east to south-west orientation, north of stibadium B, were also dug into stratum D. The upper level of stratum D was also used as a residential level of the buildings erected around stibadium B (walls 2 and 3 in sondage 88 and the wall in sondage 90). In the corner between the northern wall of room w-26 and the western wall of stibadium B, a “hiding place” (upper edge at an elevation of 199.15), made of bricks laid edgewise on the brick floor, was dug into the upper level of stratum D. The hiding place was covered and subsequently filled with tegulae at a later date, after it had been emptied. In a thin layer of soil at the bottom of the box, at an elevation of 198.89, a bronze coin dating from 425–450 was found. The discovery of this piece clearly shows that the horizon of stratum D was also used in the period after 380, that the buildings around stibadium B were built at that time and that Mediana came to its end with the invasion of the Huns in 441. 1.4. The horizon of stratum D, the “hiding place” and walls 1, 2 and 3, were covered with a layer of tegulae and imbrices – stratum C. The stratum was at its highest around the western annex of stibadium B (at an elevation of ¯ 199.63). This situation clearly shows that the walls of the stibadium were much better preserved and probably used in horizon 3. From the stibadium, stratum C descended towards the west and over wall 3 where, in sondages 72 and 73, it reached an elevation of ¯ 198.7. There it rested against the oldest mortar base and, directly below it, between an elevation of 198.87 and 198.74, fragments of sculptures were discovered, some of which belong to some statuettes from the hoard in room w-4, discovered in 1972. 1.5. Stratum C is covered with a heavy layer of debris caused by the destruction of the walls of the stibadium, the villa’s northern rooms and the large audience hall – stratum B. It represents a secondary, centuries-long, destruction of the walls that remained after the invasion of the Huns. 1.6. The final, most recent stratum, is stratum A, created as a result of the intensive extraction of the walls throughout the Throughout 2010 and 2011, a systematic nine month archaeological exploration was conducted on the north-eastern, northwestern and south-eastern parts of the villa with peristyle, as well as the area in front of the museum building (see note 8). The envisaged area of exploration was around 1,400 m². In total, 1,230 m² of this area was thoroughly examined (Fig. 1). The excavations were performed with the aim of defining the dimensions of the north-western and south-western parts of the villa, the appearance of the villa’s entrance, as well as to explore the foundations of the museum building. In the course of the excavations, certain differences were perceived in the stratigraphy of the villa with peristyle and the part excavated outside the villa. This is why these excavations were divided into five sectors, as follows: 1. Sondages north-west of the villa’s northern facade 2. Sondages north-east of the villa’s northern facade 3. North-western rooms 4. North-eastern rooms 5. South-western rooms This division simplifies the monitoring of the stratigraphic situation in the entire area of excavation. The results of the 2010 and 2011 explorations of the villa with peristyle provided fresh data, which allowed for the redefinition of the construction phases. With the exception of the vestibule, the whole villa was explored. However, in the future, attention will have to be paid to the area between the enclosing wall and the villa, where sondage explorations have only just begun. Summing up the stratigraphic data, straightaway it should be pointed out that, unlike in other parts of Mediana, in and around the villa thicker strata were preserved. The precise excavations allowed for their careful inspection, which enabled a thorough analysis of the strata and gave an insight into their history. In the text, considerable importance was ascribed to defining the strata accurately, which at first seems tiresome, but was essential for comparing this well preserved stratigraphy with the other parts of Mediana, with much thinner and, sometimes, hardly discernible strata. 1. Sondages to the north-west of the villa’s northern facade (Fig. 2, 3, 4, 5, 6, 13, 15, 16, 17) 1.1. The foundations of the villa’s northern facade and the audience hall/triclinium were dug into stratum E – proto-humus above the subsoil. In sondage 42, four sturdy posts of square section, supporting the roof frame which shielded the area in front of the hypocaust furnace of the aula and room 24a, were dug into this stratum. Imbrex and tegula fragments were found on the compacted soil in this place. Near the north-western corner of the villa there was another praefurnium, intended to heat rooms 22 and 23. 261 VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 4. North-eastern rooms (Fig. 9, 10) These rooms had already been excavated in the previous explorations of the villa, consequently only revisional excavations were carried out in 2010/11. Thus, the data about the hypocaust installation was obtained, which was, at one point in time, questioned, and on the older floor, a new floor was set. Room e-22: a mortar floor at an elevation of 198.94, below which there is a stone and brick substructure resting on the older brick floor. The older floor in this part lies on two parallel walls, of a north – south orientation, which are plastered, and have pillars between them. At the entrance to room e-22, in the south, hypocaust pillars were noted. Below the hypocaust, at an elevation of 198.29, a floor was noted. In the western part of room e-22, below a thin layer of destruction, a door jamb and a part of the floor were discovered. The door jamb was found in the southern part of the sondage, by the southern section of wall 3, and was preserved as a part of the tiling of the mortar substructure. Above it, the floor was discovered. The elevation of the tiling in the southern part was measured (at an elevation of 198.77), as well as that of the stones and base in the western part (at an elevation of 198.77, 198.70, and 198.78) and the floor in the eastern part (at an elevation of 198.75). Room e-23: on the surface of this space, below the more recent layer of loose dark grey soil, the original part of wall 37 was discovered (at an elevation of 198.67) along with a floor. The substructure of the floor of room e-23 is composed of lime mortar mixed with stone and broken brick, with its highest level at an elevation of 198.63 and lowest at an elevation of 198.52, and one brick in situ at an elevation of 198.70, which represents the remains of the original floor of room e-23. Below this floor a part of a hypocaust was discovered, which is preserved at different levels in the sondage. In the eastern profile, the existence of two floors was ascertained, the upper level of a mortar floor at an elevation of 198.95, the lower level (a base of bricks) at an elevation of 198.82, with rubble (the substructure of the mortar floor) between them. Below the lower, older floor, there is a layer of hard, compacted soil (subsoil). Room e-24 – sondage 70: Rubble with larger stone, brick and mortar fragments and finds of more recent artefacts, noted throughout the entire sondage, represent the remains of the construction of the museum building in 1936. A brick floor was noted, below which a yellow coloured floor was observed at an elevation of ¯198.9. whole area of the villa and, on the profiles, the trenches, going from the surface to the preserved tops of the walls, or even their foundations, are clearly visible. These trenches are filled with a darker grey soil which is loose, and above stratum B. 2. Sondages of the north-eastern sector (Fig. 7, 16) 2.1. Stratum D – dark brown, compact soil which is, in its upper layer, proto-humus going into the subsoil. This stratum contained the foundations of the villa, and the plinth of the northern facade is at an elevation of 198.02. In concordance, stratum D would correspond to stratum F in the north-western sector. 2.2. Above stratum D and the praefurnium, after its levelling, stratum C was deposited, a light brown, compact soil mixed with mortar and fine brick rubble. This stratum corresponds to stratum D in the north-western sector. The foundations of both stibadium A1 and the room attached to the villa’s northern facade were dug into stratum C. From the middle of this room’s northern wall, a wall stretches to the north (wall 7d) and passes through sondages 97, 98 and 104. This wall was built of stones fixed with mud. The top of this wall is at an elevation of ¯ 198.8. The room and wall 7d correspond to both walls 2 and 3 in sondages 72/3, 88 and 90, and to the wall in sondage 91 in the north-western sector. From the upper horizon of stratum C, in sondage 103, a debris pit was dug. The pit cut through strata C and D into the subsoil. Fragmented pots from the height of the Middle Ages were found in the pit. This would indicate that this part was open for a long time and was not covered with rubble from the villa’s original destruction, consequently enabling the Late Roman horizons 2 and 3 to meet up with the horizon of the Middle Ages. 2.3. Stratum C, the stone wall and the pit in sondage 103 are covered by stratum B, a loose dark grey soil mixed with larger pieces of rubble. It is difficult to say when the levelling of the mediaeval horizon took place, since in stratum B chronologically sensitive material has not been found. Stratum B mixes with recent material and in concordance it would correspond to stratum B in the north-western sector. 3. North-western rooms of the villa with peristyle (Fig. 8, 12, 14) Although the stratigraphy does not essentially differ from that of the north-western and north-eastern sectors, certain different characteristics exist. Primarily, all these rooms have brick or mortar floors. In the process of its destruction, the ceiling fell onto these floors, followed by the roof. However, in room w-24a and, to a degree, in rooms w-24 and w-22, this original layer of ruins was cleared to the floors. In other areas, this layer (stratum C) remained visible and was a floor in horizon 3. In such a way, on this surface in room w-1, in their secondary use, the base of the column from the peristyle and the head of the goddess Venus were found.2 On the inner face of the villa’s northern wall, hypocausts were broken and used, particularly in room w-22, and in this large excavated area mediaeval ceramics were discovered. That is how, in these cleared and perforated areas, and on the parts of the floor, a secondary layer of ruins piled up (stratum B). The trenches, created by the latest extraction of walls (during the Ottoman period?), can also be perceived in the rooms. With the exception of strata A and B, the other strata cannot be placed in concordance with the strata of the two previously discussed sectors. 5. South-western sector (Fig. 11, 19, 20, 21, 22, 23, 24) 5.1. The south-western rooms The stratigraphy in this sector also differs somewhat from the north-western rooms. First of all, there aren’t any brick or mortar floors, but compacted soil (the upper level of stratum D, where numerous coins were discovered). Above it, there is a substantial layer of debris from the roof bricks (stratum C) and it completely covers stratum D. This information is invaluable, since the coins were discovered as an undisturbed group, and most of them belong to the period 364–375, immediately prior 1 2 262 According to the excavations of 2013. Vasi}, Gavrilovi} 2012, 137–149. VASI], MILO[EVI], GAVRILOVI], Iskopavawa Medijane u 2010. i 2011. godini (231–263) STARINAR LXIV/2014 In construction phase 2 (from around 330 to the Gothic incursions in 378) a large reconstruction was carried out on the north and south wings and the courtyard of the building, which totally altered the appearance and function of the building. The eastern and western rooms of the north wing were reduced in size by partitioning. In some rooms, brick floors were installed. In rooms e-23 and w-21 two layers of a mortar floor were discovered, which indicate the restoration of the floors from the first construction phase. The walls and the ceilings were covered with paintings. On the north-eastern and north-western part, two magnificent triclinia were added – stibadium A and B. The walls of the stibadia were lined with decorative marble panels and mosaic, and the floors were decorated with mosaic. In the centre of each, there was a fountain. Water was supplied to the fountain in the north-eastern stibadium A via lead pipes (research from 1972), and then piped to the fountain in the northwestern stibadium B (2011 research). The heating system was also altered. The praefurnium in room w-21 was no longer functional, as a result of the access to this part of the building having been blocked during the construction of stibadium B. Also, the praefurnium in the eastern part, closer to the north-eastern corner, was no longer in use due to the erection of the partition wall and the division of the spacious north-eastern corner room into two (e-22 and e-23). It is quite possible that the remaining furnaces were also used in the later construction phase. New passageways, about 1.00m wide, with brick jambs and flat arch lintels were built. In the southern part of the building, much reconstruction took place on the entranceway and a complete alteration of the appearance and construction of the southern entrance porch was carried out. The porch was extended and, on its southern wall, 14 brick columns with monumental stone bases were built. Furthermore, the porch facing the courtyard was remodelled and enlarged. In this reconstruction, high quality materials were used for the building of the columns, floors and the fresco painting. In addition, new construction methods such as arches, calottes and semi-calottes were applied. After the reconstruction had been completed, the appearance of the building was totally changed, along with its meaning, which now conveyed a public, administrative and solemn, ceremonial character. This organisational concept was preserved until the Gothic incursions in 378. Room partitions and stone walls, noted outside stibadium A and B, in the north-western and north-eastern parts of the complex, belong to the third construction phase (after 380). As far as we can determine at this stage of research, these were large rectangular buildings and it seems that, after 380, the settlement was not at all insignificant. At this moment in time, however, we will set this problem to one side. to the first villa’s destruction, which, based on these finds, we can attribute to the Gothic invasion after the Battle of Hadrianopolis in 378. Above stratum C, there is a second layer of ruins, marked as level a. Both the stratum and the level were formed by the destruction of the roof, walls and columns of the porch. Within these two layers, whole parts of walls and columns were noted, in the manner in which they fell during the destruction. Above level a, there is a layer of secondary destruction (stratum B). The building technique of the discovered walls, the depth of the foundations and certain construction elements, indicate more construction phases and the possible separation of the bases into different positions, appearances and purposes. Thus, the discovered architecture and its phases are in full accordance with the results attained by analysing stratigraphy, coins and the chronologically sensitive, movable archaeological material. Three main construction phases were singled out and, based on the archaeological stratigraphy and the stratigraphy of the building destruction, they can be dated to the period from the beginning of the 4th to the middle of the 5th century as horizon 1 (the first decades of the 4th century), horizon 2 (from around 330 to the Gothic incursions in 378) and horizon 3 (from after 383 – to the invasion of the Huns in 441–443). Certain interventions and finds indicate that the space was used in the Middle Ages, although to a significantly lesser degree. The basic dimensions of the building in the first and second construction phases did not significantly change, particularly in the area with the administrative rooms (officium) in the east and west wings (rooms w2–w21 and e2–e21). The most significant changes were on the north-western and north-eastern parts of the building, as well as on the southern porch and the entrance area. The original structure (construction phase 1) was of a simple, elongated form, with a spacious courtyard in the middle, an audience hall/triclinium in the north and the entrance in the south. Numerous rooms of different purposes and sizes were arranged around the courtyard and joined by the porch. On the side wings and in the south, the administrative rooms of the same surface area and arrangement were positioned. The rooms of the northern wing, east and west of the audience hall/triclinium, were heated and intended for longer stays. The number of rooms cannot be determined with any certainty, although it can be assumed that they were of different sizes, the two larger in the north and three smaller in the south (?). Four praefurnia, along with the floor and wall heating systems, were ascertained in the northern rooms. In the first construction phase, rooms e22–23, e24–24a, w24–24a and w22–23 represented one complete area that was heated by a furnace in the northern wall. In the south wing of the building there was a sequence of nine rooms, all of the same size, with a porch in front of them. 263 STARINAR LXIV/2014 KRITIKE I PRIKAZI – COMPTES RENDUS (265–270) KRITIKE I PRIKAZI – COMPTES RENDUS Rachel S. Opitz, David C. Cowley (eds.), INTERPRETING ARCHAEOLOGICAL TOPOGRAPHY. 3D DATA, VISUELISATION AND OBSERVATION, Occasional Publication of the Aerial Archaeology Research Group No. 5, Oxbow Books, Oxford – Oakville 2013, 268 pp. znawa, koja su ubrzo oplo|ena rezultatima drugih metoda prospekcije i arheolo{kih rekognoscirawa i iskopavawa (Ivani{evi}, Bugarski 2012; 2013; Rummel et al. 2012; Ivani{evi} 2012). U narednom ~lanku Rej~el Opic daje obiman pregled istorijata i tehnika laserskog skenirawa iz vazduha i sa zemqe za potrebe arheologije (pp. 13–31). Osnovnih podataka o lidarskim snimawima iz vazduha ima i u na{oj literaturi (Ivani{evi}, Bugarski 2012, 246–248). Opic pi{e i o terestrijalnom lidaru, tj. razli~itim tipovima laserskih skenera koji se koriste za snimawe arhitekture ili izradu terenske dokumentacije. Autorka op{irno ukazuje na mogu}e gre{ke, bilo u radu sa opremom, bilo tokom prili~no detaqno opisanih procesa obrade i vizuelizacije podataka. Iza svakog poglavqa ovog preglednog rada daje se relevantna bibliografija. Mihael Doneus i Tomas Kitrajber (pp. 32–50) skre}u pa`wu na neophodnost stalne saradwe stru~waka za lidar tehnologiju i arheologa, od osmi{qavawa projekata do izrade finalnih vizuelizacija, kao i na potrebu ukr{tawa lidar podataka s rezultatima drugih metoda prospekcije i rada na istorijskim dokumentima. Kao primer, autori nude prou~avawe kompleksa karmelitskog manastira Sv. Ana u divqini (St Anna in der Wüste) u Austriji. Manastir se nalazi u brdovitom kraju, pod {umom, a tragovi kompleksa su samo delom vidqivi. Bez upu{tawa u tehni~ke detaqe snimawa i obrade podataka, potcrtao bih metodolo{ki ispravan postupak interpretacije digitalnog modela kroz pore|ewe sa dostupnom pisanom i likovnom gra|om. Neobi~na, skoro srcolika trasa bedema kompleksa dovodila se u vezu s pojmom Svetog srca. Me|utim, na osnovu niza vizuelizacija digitalnih podataka utvr|eno je da se ona zapravo poklapa s granicom reqefne „nevidqivosti“ kompleksa sa okolnih pozicija, koja mo`e da se tuma~i potrebom karmelitskog reda da se izoluje od spoqnog sveta. Ole Risbol je predstavio lidarska snimawa u Norve{koj (pp. 51–62). Budu}i da je 40% teritorije te zemqe pod {umom, zbog ~ega postoje velike praznine u arheolo{koj evidenciji, izbor tehnologije ne ~udi. Prilog daje i genezu arheolo{ke literature o lidaru, pri ~emu se naslu}uje da }e se sve ve}i broj radova baviti analizama {irih prostornih zona. S tim u vezi, autor nudi i teorijsku raspravu o pojmu „landscape“. Termin se na srpski prevodi kao pejza` ili krajolik, pa je „landscape archaeology“ pejza`na arheologija (Ivani{evi}, Bugarski 2013, 79) ili arheolo- Odli~no od{tampan i bogato ilustrovan zbornik, u skra}enom A4 formatu, ure|en je kao peti tom serije Povremenih izdawa Istra`iva~ke grupe za arheologiju iz vazduha (Aerial Archaeology Research Group – AARG), a prvi koji je izdat u saradwi sa projektom ArchaeoLandscapes Europe. Kada se ima na umu zajedni~ko poqe delovawa i isprepletanost istra`iva~a, takva saradwa je potpuno prirodna. Dok projekat umre`ava ve} stotinak institucija nauke i kulture, me|u kojima je i na{ Arheolo{ki institut, AARG okupqa pojedince koji se bave razli~itim vidovima arheolo{ke prospekcije iz vazduha. Specijalisti~ka interesovawa stru~waka, ali i visok nivo wihovih nau~nih dostignu}a, jasno se pokazuju u 20 ~lanaka kwige koja je predmet ovog prikaza. Uz ve{te upute na pojedine radove u zborniku, u uvodnom ~lanku Rej~el Opic i Dejvida Kaulija daje se kriti~ki prikaz rastu}e primene lidar (LiDAR, lidar, ALS) tehnologije u arheologiji (pp. 1–12). Ne spore}i veliku korist koju metod pru`a, autori s pravom upozoravaju na probleme obrade dobijenih podataka i pouzdanost wihove interpretacije i nagla{avaju nu`nost kombinovawa lidar podataka s drugim metodama prospekcije i dokumentacije i, prirodno, sa terenskim arheolo{kim radom. Potreba bele`ewa prostora u tri dimenzije vrlo jasno se o~itava jo{ u ostvarewima starih kartografa, a nametala se i u arheolo{koj praksi. U posledwih desetak godina, ako i ostavimo lidar po strani, 3D dokumentovawe je vidno olak{ano s pojavom totalnih stanica, povr{inskih laserskih skenera (TLS) i sa razvojem fotogrametrije. Kako je duhovito sa`eto, razli~iti pristupi „idu od terenca blatwavih ~izama zadubqenog u topografiju pobr|a do kompjuterskog moqca koji sedi u zamra~enoj sobi i, uz zujawe opreme, pi{e kompleksan program kako bi stvorio virtuelnu okolinu!“ Prevedena re~enica dosta govori o {irini dana{wih arheolo{kih zahvata, a sa autorima vaqa podeliti bojazan da puki razvoj metoda prospekcije, koji ponekad predstavqa pre ciq nego sredstvo, kao ni wihovo nekriti~ko prihvatawe („let’s do lidar“) ne}e dovesti do `eqenih rezultata. Saradwa stru~waka iz oblasti arheolo{ke detekcije i tradicionalnih istra`ivawa ipak je dovela do toga da u novije vreme, pa i u zborniku koji su uredili Opic i Kauli, preovla|uju radovi koji oslikavaju integralno metodolo{ko usmerewe. Takva su, dodajmo, i na{a iskustva. Primena lidar tehnologije u zonama Marguma/Morave i Cari~inog grada donela je brojna nova sa- 265 STARINAR LXIV/2014 KRITIKE I PRIKAZI – COMPTES RENDUS (265–270) Radom @ige Kokaqa, Klemena Zak{eka i Kri{tofa O{tira otpo~iwe niz stru~nih priloga koji su uglavnom li{eni teorijskih i introspektivnih osvrta. Kokaq i ostali bave se vizuelizacijom lidarskih modela reqefa (pp. 100–114). Doti~u se raznih na~ina osvetqavawa terena, uz zakqu~ak da pre vizuelizacije lidar podataka treba utvrditi prioritet – prikazivawe topografije terena ili detekciju arheolo{kih pojava. Ukoliko se te`i prikazu topografije, dovoqne su standardne metode osvetqavawa, koje za podrobnu analizu mikroreqefnih struktura treba kombinovati s nekom drugom tehnikom vizuelizacije. U svakom slu~aju, osvetqavawe je posebno va`no u vizuelizaciji lidar podataka, gde je mogu}e birati ugao pod kojim „sunce“ osvetqava teren (hillshading), ne bi li se naglasile reqefne razlike (Doneus, Briese 2006, 99–104). Kao nama posebno va`an, izdvojio bih ~lanak Fabija Remondina o kori{}ewu fotogrametrije u modelovawu trodimenzionalnih DEM (pp. 115–122). Iako je re~ o veoma starom metodu, u posledwe vreme je do{lo do velikog napretka u wegovoj primeni. U osnovi, fotogrametrijskim metodom se vr{i povezivawe slika tako da daju geometriju scene. Trodimenzionalni model sa fotografskom teksturom dobija se kombinacijom triangulacije i tzv. SIFT algoritma. Opisano modelovawe dobija sve {iru primenu u dokumentovawu arheolo{kih celina (kod nas u istra`ivawima Viminacijuma), a vr{i se u nizu ra~unarskih programa. Zahvaquju}i sve ve}oj upotrebi bespilotnih letelica, raste i primena fotogrametrije u modelovawu krajolika. To je danas znatno jeftinija, ali u odre|enim uslovima i boqa alternativa kori{}ewu lidar tehnologije. Fotogrametrijski DEM je ~ak znatno precizniji od lidarskog, ali metod ne omogu}ava pravqewe DTM-a bez vegetacije, pa mo`e da se koristi pre svega za modelovawe zona koje nisu pod {umom. O primeni fotogrametrije u pejza`noj arheologiji bilo je govora na nedavno odr`anom skupu i radionici Recovering Lost Landscapes u Beogradu, pod okriqem projekta ArchaeoLandscapes Europe (Bugarski, Ivani{evi}, u {tampi). Edrijen Evans, Majri Maksvel i Gema Kruik{ans pokazuju kako tehnologija sli~na lidarskoj mo`e da se koristi i u analizi povr{ina arheolo{kih artefakata. Laserski skeniraju}i konfokalni mikroskop (LSCM), budu}i da snima nesrazmerno mawe povr{ine nego lidar, daje neuporedivo ve}u rezoluciju. Ako kvalitetni lidarski snimci iz vazduha daju 20–60 ta~aka po metru kvadratnom, LSCM daje osam miliona ta~aka po istoj mernoj jedinici, {to je rezolucija od 120 nanometara! Tako detaqni snimci predmeta, razume se, mnogo govore o wihovoj upotrebi, ukra{avawu i sl., {to je pokazano na nekoliko primera (pp. 123–135). Sajmon Kra~li, iz zavoda English Heritage, jedan je od najiskusnijih arheologa u poslovima sa lidar snimcima iz vazduha. Prilo`io je vrlo koristan ~lanak o dobrim i lo{ijim stranama upotrebe ove tehnologije u Engleskoj (pp. 136–145). Lidarom niske rezolucije dosad je pokriveno oko 70% teritorije Ujediwenog Kraqevstva. Ti snimci su lako dostupni, a Kra~li je bio u prilici da radi i sa kvalitetnijim, namenski pribavqenim skenovima. ^lanak je namewen i po~etnicima, pa je pa`wa posve}ena i osnovnim stvarima kao {to je razlika izme|u upotrebe slike, jedne mogu}e vizuelizacije podataka snimawa, i modela, ko- gija pejza`a, {to je naziv kursa na doktorskim studijama na Odeqewu za arheologiju Filozofskog fakulteta u Beogradu, odnosno, kako je predlo`eno u Hrvatskoj, arheologija krajolika (Novakovi} 2008, 15, nap. 1). Pritom, treba imati na umu da arheolo{ko poimawe re~i „landscape“ podrazumeva prostor, tj. reqef, vegetaciju, geolo{ku i hidrolo{ku osnovu sa tragovima svih qudskih aktivnosti. Kao takav, on nije stati~an, ve} je izlo`en neprestanim promenama. Norve{ko iskustvo ekstenzivnog lidarskog snimawa pokazuje da {umoviti predeli, koji se obi~no do`ivqavaju kao priroda, zapravo ~esto obiluju tragovima qudske delatnosti (cf. Ivani{evi} 2012). Zakqu~ak da su nam neophodna empirijska saznawa o krajolicima, bez obzira da li se wihovom razumevawu pristupa iz ugla procesualnog tra`ewa „objektivnosti“ ili postprocesualnog zagovarawa kognitivnih aspekata (cf. Novakovi} 2008, 29–30, 39–44), svakako je od {ireg zna~aja. Stretford Halidej prime}uje, na osnovu konkretnih primera, da su arheolo{ke ocene spomenika i krajolika obele`ene evolucijom znawa i iskustva, {to nije ograni~eno samo na terenske spoznaje, ve} se odnosi i na interpretaciju lidar podataka. Wen posledwi korak je, opet, terenska evaluacija (pp. 63–75). Na znawe i iskustvo u ~itawu slika iz vazduha poziva se, sasvim prikladno, i veteran arheolo{ke aerofotografije Rox Palmer (pp. 76–87). Vredi uzgred spomenuti da je Palmer jedan od urednika nove, izuzetno informativne i obimne el. kwige o arheologiji iz vazduha, koja je dostupna na internetu (Musson et al. 2013). Iskustvo u obradi snimaka iz vazduha sti~e se, naravno, kroz praksu, a ekspertiza u tuma~ewu kroz akumulaciju raznorodnih znawa sakupqenih tokom arheolo{kih rekognoscirawa i uo~avawa savremenih promena krajolika. Nakon {to je dao nekoliko primera, Palmer se bavi subjektivno{}u u analizi snimaka iz vazduha, na koju ne gleda negativno. Ona je nu`na, a kvalitet interpretacije zavisi upravo od iskustva i znawa tuma~a. Uvek ima nekoliko mogu}ih interpretacija, a nikad samo jedan „ta~an“ odgovor, pa nijedno ~itawe, bilo arhivske aerofotografije ili lidarskog snimka, ne mo`e da se uzme kao kona~no. U ~lanku Dimitrija Mleku`a daje se teorijska rasprava o lidaru i krajolicima, tj. o tome kako upotreba te tehnologije mewa percepciju pejza`a (pp. 88–99). Na lidar snimku vidi se mnogo vi{e nego na tradicionalnim topografskim snimcima, koji pokazuju samo najizra`ajnije reqefne pojave. Da bi pristupio raspravi o zna~aju tehnologije za razumevawe prostora, Mleku` najpre daje postoje}e definicije pojma „landscape“, od kojih je ovom prikaziva~u najbli`a jedna ve} dosta stara, koja tu zvu~nu, a danas i pomodnu re~ obja{wava kao interakciju prirode i kulture, u kojoj one oblikuju jedna drugu. U svakom slu~aju, ne postoji op{teprihva}eno tuma~ewe (cf. Novakovi} 2008, 15–17). Potom se ukratko razmatraju metodi rada i rezultati pejza`ne arheologije, s neveselim zakqu~kom da oni mewaju i pojednostavquju „pravu“ sliku krajolika. Sli~no, po Mleku`u, va`i i za kartirawe. Lidar tehnologija, pak, nije selektivna, i bele`i dugotrajne uticaje ~oveka na prirodu. ^esto su primetni tragovi raznih epoha, pa ima slu~ajeva da oni – sli~no principu vertikalne stratigrafije na arheolo{kim iskopavawima – negiraju jedni druge. Tako se na krajolik katkad gleda kao na palimpsest. 266 STARINAR LXIV/2014 KRITIKE I PRIKAZI – COMPTES RENDUS (265–270) (npr. stare granice wiva), pa bi pre mogle da se opi{u kao „elementi kulturnog krajolika“ nego {to predstavqaju arheolo{ka nalazi{ta. Ipak je to neuporedivo vi{e arheolo{kih podataka nego {to ih je bilo pre skenirawa. Od oko 15.400 dotada{wih unosa, wih 6.060 moglo je da bude potvr|eno lidar snimcima, jer ostali poti~u iz urbanih sredina, ili je re~ o pojedina~nim nalazima, ili pak o grobovima bez prate}ih podataka o poziciji… Upravo navedene brojke pokazuju mo} lidar tehnologije u arheologiji, posebno ako se ima na umu da je glavnina podataka izvedena iz snimaka male gustine koji nisu prilago|eni potrebama na{e struke. Pore|ewa radi, novim, veoma detaqno izvedenim i primerno obra|enim i publikovanim arheolo{kim rekognoscirawima podru~ja Op{tine Novi Kne`evac, ~ija je povr{ina 305 km², broj od 130 ranije poznatih nalazi{ta narastao je na 314 (Trifunovi} 2012). Mo`e se zasnovano pretpostaviti da bi se u digitalnom modelu reqefa Op{tine Novi Kne`evac ukazalo jo{ antropogenih pojava. S druge strane, podaci pribavqeni na terenu su jasno datovani, dok }emo, kako Hese prime}uje, uz upotrebu lidara u pejza`noj arheologiji morati da se naviknemo na izvesnu nesigurnost podataka, jer wihov ogroman broj ~esto nije mogu}e oceniti u celosti. Poku{aj bli`eg odre|ewa antropogenih pojava na terenu dat je u narednom radu. Nikolas Purije, Rej~el Opic, Lori Nuninger i Kri{tof O{tir analizirali su lidarske snimke ravnog terena oblasti Ma`ijo na samom jugu Francuske, na obali lagune, su~eqavaju}i ih s drugim izvorima (pp. 184–196). Taj plavqen i mo~varan teren poznaje duga~ku istoriju regulacije voda i intenzivne zemqoradwe. U pro{losti se puno radilo na detekciji, snimawima iz satelita i aviona, kao i na temeqnim arheolo{kim rekognoscirawima ove zone. Uz to, raspola`e se i prili~no preciznim katastrom iz Napoleonovog doba. Sve predradwe vodile su ka zakqu~ku da je katastar iz 19. veka sledio rimsku podelu zemqi{ta. Nakon izvr{enih analiza i vizuelizacije lidar snimaka taj utisak mo`e da ostane, ali autori nude i drugu mogu}nost – da su pejza`i u razna vremena organizovani po istim principima usled sli~nih prirodnih i topografskih uslova. Kao primer, istina iz sasvim druga~ijeg konteksta, naveo bih da je preklapawe savremenog katastarskog plana, DTM-a i ortofotografije {ire zone Cari~inog grada pokazalo upravo „pam}ewe“ parcelizacije, koja je sasvim druga~ije organizovana u odnosu na to da li je unutar ili izvan nekada{weg perimetra grada. Slede}i ~lanak, autora Rebeke Benet, Kejt Velam, Rosa Hila i Endrjua Forda, nagove{tava prednosti multisenzorskog pristupa arheolo{koj daqinskoj prospekciji (pp. 197–205). Nagla{ava se da mogu}nosti tzv. multispektralnih (hiperspektralnih) snimaka nisu ni izbliza sagledane. Multispektralno snimawe se vr{i u specifi~nim frekvencijama elektromagnetskog spektra, kao {to je infracrvena, koja je van vidqivog RGB (Red-Green-Blue) odse~ka. Iste scene se snimaju u razli~itim opsezima, pa se wihovim preklapawem dobija multispektralni snimak. Autori opisuju po~etne britanske poku{aje da kombinacijom lidarske i spektralne detekcije zone pa{waka stvore slojevitu osnovu za daqi rad. Kompleksna dokumentacija je po`eqna zato {to, iz mnogih razloga, nijedan senzor zasebno nije u stawu da zabele`i sve pojave, ba{ kao ni arheolog na ji sadr`i sve topografske informacije i apsolutne koordinate ta~aka. Autor predo~ava potencijalnim korisnicima, bilo arheolozima ili amaterima, granicu do koje je rad sa razli~itim lidarskim podacima mogu} bez u~ewa komplikovanih programa, a upu}uje i na preispitivawe ciqeva za pribavqawe skupih, detaqnijih modela. Kra~lijev zakqu~ak je da je u arheologiji i lidar niske rezolucije, od jedne ta~ke po kvadratnom metru, dovoqan da se zabele`i najve}i broj arheolo{kih tragova. To se ne odnosi na {umovite krajeve, za koje je potrebno gu{}e snimawe. Bogata iskustva lidarskih snimawa spomenika svetske ba{tine u Irskoj dele sa ~itaocima Entoni Korns i Robet [ou (pp. 146–160). Za razliku od engleskih kolega, Irci su za ove namene nabavqali guste snimke, od kojih su snimci brda Tare, sa 60 ta~aka po kvadratnom metru, najgu{}i dosad ura|eni za arheolo{ke potrebe. Spektakularna preciznost, me|utim, donela je i neke probleme, jer je bilo te{ko obraditi tako detaqne podatke. U dobre strane se, sem vrhunskog kvaliteta dobijenih modela i stru~ne i nau~ne primene, ubraja i prijem rezultata kod {ire publike. U novije vreme se i u Irskoj razmi{qa o jeftinijoj alternativi, pre svega fotogrametriji. Kit ^elis i Endi Hauard ispituju do sada prili~no zanemarenu upotrebu intenziteta odjeka laserskog zraka (pp. 161–170). Ti podaci mogu da se uzmu iz tzv. oblaka ta~aka, tj. sirovog lidar snimka, iz kojeg se klasifikacijom ta~aka izvode modeli reqefa. Vizuelizacija tih podataka je donekle jo{ uvek u eksperimentalnoj fazi. Wome se ne dobijaju iste slike kao na digitalnim modelima, ali je jasno da intenzitet odjeka sadr`i podatke o snimqenom terenu koji ne mogu da se dobiju u modelu reqefa. Intenzitet odjeka svedo~i o karakteru zemqi{ta, sedimenata i vegetacije u specifi~nim topografskim uslovima, pa uspostavqawe takve korelacije jo{ vi{e doprinosi saznawima o odre|enom krajoliku. Ralf Hese je stru~wak nesvakida{weg iskustva koji ispred slu`be za{tite radi na lidarskim snimcima nema~ke pokrajine Baden-Virtemberg, ~ija je ukupna povr{ina 35.751 km² (pp. 171–183). Dr`avna topografska slu`ba skenirala je ~itav taj prostor sa gustinom merewa od jedne ta~ke po metru kvadratnom, dok su zone pojedinih lokaliteta skenirane sa po ~etiri ta~ke. Ustupqeni topografski podaci, pritom, nisu bili posebno obra|eni za arheolo{ke potrebe, pa se uo~ava jasna razlika u detaqnosti modela. Ogromnom gra|om nije bilo lako rukovati, a probleme su predstavqali i obrada i tuma~ewe podataka. Rade}i na ovoj gra|i, Hese je razvio i odre|ene na~ine vizuelizacije. Kako se objektivno navodi, podvrgavawe ovolikih prostranstava lidar tehnologiji daje mogu}nost zna~ajnog pove}awa broja detektovanih arheolo{kih tragova i za{ti}enih lokaliteta, ali i umawuje izglede za uverqive interpretacije. Ipak, nesporno je da je dobijena vrlo vredna dokumentacija, na osnovu koje se koriguju raniji podaci o lokacijama pojedinih nalazi{ta, kartiraju i unose u baze podataka saznawa o novim potencijalnim nalazi{tima i planiraju obilasci terena. U vreme pisawa ~lanka, analizirani su rezultati skenirawa tre}ine povr{ine pokrajine, i zabele`eno je oko 215.000 prethodno nepoznatih potencijalnih arheolo{kih pojava. One su svakako antropogene, ali je jasno da su mnoge od wih sasvim male spomeni~ke vrednosti 267 STARINAR LXIV/2014 KRITIKE I PRIKAZI – COMPTES RENDUS (265–270) s na{im iskustvom skenirawa Cari~inog grada. Nakon ravno 100 godina istra`ivawa tog nalazi{ta, u DTM-u se jasno ukazao jo{ jedan prsten bedema – kako se kasnije ispostavilo, zidan u tehnici opus mixtum – koji je zatvarao prostrana i tek uzgredno naslu}ivana podgra|a (Ivani{evi}, Bugarski 2013: 82–83). Kit ^elis i Mark Kinsi istra`uju kako u vizuelizaciji podataka sa rekognoscirawa i terestrijalnog i vazdu{nog laserskog skenirawa mogu da pomognu softveri ra~unarskih igara (pp. 238–251). Prikazi kulturnog nasle|a u kompjuterskim igrama su „endemski“, ali kori{}ewe softvera i paradigme kompjuterske igre za „ozbiqan“ prikaz nasle|a predstavqa novost. Potreba za takvim pristupom javila se kada su stru~waci po`eleli da kao rezultat rekognoscirawa dobiju ne{to vi{e od stati~ne vizuelizacije podataka u GIS-u. ^elis i Kinsi, naime, smatraju da koli~ina i kvalitet podataka koji se danas prikupqaju umnogome prevazilaze tradicionalne vidove prezentacije rezultata, bilo na papiru ili u digitalnom obliku. Re{ewe su potra`ili u kori{}ewu game engine sistema, osmi{qenih za pravqewe i razvoj video igara, koji o~igledno imaju dovoqno snage da vizuelizuju obiqe preciznih podataka. Ponu|ene ilustracije, makar, vrlo su ubedqive. Zakqu~uju}i zbornik, Entoni Bek pi{e o saradwi kao na~inu istra`iva~kog rada koji se name}e u eri interneta, i razmatra kako bi saradwa stru~waka razli~itih specijalizacija, ali i amatera, mogla da izgleda u budu}nosti. Razmi{qawa autora se prelamaju kroz primer kori{}ewa lidar tehnologije, uz opravdanu ocenu da su delovi „lidarske zajednice“ primer kolegijalnog deqewa podataka, a wegov esej bi mogao da se ~ita i nezavisno od osnovne teme zbornika (pp. 252–265). Kwiga se zavr{ava spiskom ukratko definisanih kqu~nih tehni~kih termina. terenu. Nagla{avaju}i problem kalibracije senzora za snimawe arheolo{kih pojava na razli~itim topografskim i pedolo{kim osnovama, autori navode da je sli~na fuzija tehnologija ranije vr{ena na Kipru i prilikom istra`ivawa Akvileje. Engleska iskustva su takva da je kombinovanim metodom, uz kori{}ewe arhivskih aerofotografija i geofizi~kih podataka, broj nalazi{ta i arheolo{kih pojava porastao sa 68 na 170. Odlu~uju}u ulogu je ipak imala lidar tehnologija, i u detekciji i u georeferencirawu nalazi{ta, pa je sna`na preporuka da se i budu}a spektralna snimawa izvode u kombinaciji s lidarskim. Sli~na iskustva prenose Stjuart Ejnsvort, Al Osvald i Dejv Vent, koji su tako|e ispitivali nepo{umqen krajolik, {iroku zonu rudarewa olova (pp. 206–222). Autori opisuju slo`en proces izrade osnova za rekognoscirawe, u kojoj su se spektralni snimci jako dobro pokazali u bele`ewu {qaki{ta. Osnovni plan, me|utim, bio je lidarski, a podaci su preuzimani i sa aerofotografija. U ovom slu~aju nije se pristupalo detaqnim vizuelizacijama, ve} je kori{}ena osnovna, ortogonalna projekcija uobi~ajeno osvetqenog DTM-a, pa je ~itav taj pristup nazvan „ortolidar“. U GPS ure|aje arheologa na terenu uno{eni su jo{ i DSM u istoj projekciji, aerofotografije i dostupne istorijske i moderne karte, a arheolozi su svoja zapa`awa unosili na DTM osnovu. Omiqeniji, br`i metod bele`ewa podataka izvo|en je ru~nim ucrtavawem simbola na od{tampane DSM listove, aerofotografije i karte. DSM je izabran za osnovu jer, za razliku od DTM-a, sadr`i jasne orijentire, zgrade, vegetaciju i sl. Lidarske osnove su bile od velike koristi za prepoznavawe arheolo{kih pojava i tuma~ewe praistorijskih elemenata krajolika, pre nego za razumevawe spomenika koji su ranije ve} bili temeqno obra|eni. U nekoliko slu~ajeva, me|utim, na lidarskom snimku bile su pretpostavqene arheolo{ke pojave koje nisu potvr|ene tokom obilaska terena, {to nas opet opomiwe na opasnosti procene rezultata daqinske detekcije bez terenske evaluacije podataka. Naredni ~lanak nas vra}a u Irsku. Stiven Dejvis, Konor Brejdi, Vilijam Migeri i Kevin Barton opisuju iskustva primene lidar tehnologije na lokalitetu Bru na Bojn sa liste svetske ba{tine (pp. 223–237). O tom vi{eslojnom nalazi{tu, najpoznatijem po desetinama neolitskih grobnih humki, od kojih tri najve}e imaju i po 80 metara u pre~niku, bilo je re~i u ~lanku Entonija Kornsa i Roberta [oa. Zona neposredne za{tite ima povr{inu od oko 780 hektara, na koju se naslawa zona sekundarne za{tite, povr{ine 2.500 hektara. Prema prethodnom stawu, na tom prostoru bila su zabele`ena 93 spomenika. Rezultate novih snimawa trebalo je uklopiti u postoje}u dokumentaciju, koja se sastojala od izve{taja sa brojnih rekognoscirawa i iskopavawa, geofizi~kih snimaka i aerofotografija. Sem o~ekivano dobrih rezultata lidara u dokumentaciji poznatih i detekciji nepoznatih spomenika, pri ~emu su se koristili razli~iti na~ini vizuelizacije, ponu|eni su i odgovori na pitawa o vidqivosti tri glavne grobnice, kao i pogledu iz wih, i izgledu pejza`a u doba neolita. Sve tri dominantne grobnice bile su vidqive sa jednog mesta „u srcu ritualnog krajolika“. Iako Boj na Brun predstavqa arheolo{ki najboqe istra`en predeo u Irskoj, snimawe lidar tehnologijom je, izme|u ostalog, dovelo do otkri}a ~etiri velika ogradna zida, koje je opisano kao „otkrovewe“. Name}e se paralela *** Zbornik Interpreting Archaeological Topography je pou~na i zanimqiva zbirka radova visokog kvaliteta, koja na dobar na~in oslikava arheolo{ku upotrebu lidar tehnologije u 21. veku. Nakon desetak godina kori{}ewa, lidar u na{oj struci vi{e nije nov, nego mlad metod. Dostupna je ve} prili~no obimna literatura, mahom sa engleskog govornog podru~ja, ali ima i kwiga na „mawim“ jezicima koje pokazuju postepeno {irewe tehnologije na istok (Gojda, John 2013). Godine de~a~ke fascinacije novom igra~kom su iza nas. Danas se, pored neporecivih vrednosti tehnologije, koje se ne ogledaju samo u prospekciji, ve} i u razumevawu odnosa ~oveka i wegovog okru`ewa, skre}e pa`wa i na neke mawe primamqive aspekte wene upotrebe, bilo finansijske ili tehni~ke prirode. Ispostavilo se da lidar nije „lek za sve“ i da metod treba kombinovati s razli~itim drugim izvorima i vidovima daqinske prospekcije, uz terensku evaluaciju podataka. U nekim slu~ajevima, pak, treba tra`iti jeftinije alternative. Ipak je izvesno da primena lidara u arheologiji nije ni blizu svog zalaska (a verovatno ni zenita), samim tim {to druge tehnologije ne mogu da ogole reqef, {to vizuelizacije ve} dobijenih modela ne mogu da budu iscrpqene, i {to se ~itavi setovi prate}ih podataka uglavnom jo{ uvek ne analiziraju. 268 STARINAR LXIV/2014 KRITIKE I PRIKAZI – COMPTES RENDUS (265–270) kao i ~lanak o primeni fotogrametrije. O narastaju}oj upotrebi satelitskih snimaka u arheologiji, od kojih su mnogi postali komercijalno dostupni, ~itaoci mogu da se obaveste u drugim publikacijama (e. g. Fowler 2010). Uz sve pohvale autorima i urednicima, jedini prigovor bi se ticao rasporeda ~lanaka, jer se sti~e utisak da su mogli da budu svrstani u koherentnije tematske celine. Pro{lo je, ipak, ve} dovoqno vremena da se pi{u i pregledni radovi, kakvih ima u prikazanoj kwizi, i da doprinos lidar tehnologije razumevawu krajolika postane predmet arheolo{ke teorije. Stru~wacima u na{oj zemqi, u kojoj je, kao i u mnogim drugim sredinama, upotreba lidara u arheologiji jo{ uvek u povoju, verovatno }e od ve}e koristi biti radovi koji prikazuju mogu}nosti tehnologije, BIBLIOGRAFIJA: pretation and mapping for archaeology, Occasional Publication No. 4 of the Aerial Archaeology Research Group in partnership with the ArchaeoLandscapes Europe (ArcLand) Project of the European Union, 2013, http:.//www.archaeolandscapes.eu/. Novakovi} 2008 – P. Novakovi}, Arheologija prostora i arheologija krajolika, Povijest u kr{u. Zbornik radova projekta „Naselja i komunikacije u kontekstu veza Jadranskog priobalja i unutra{njosti“, Alpium Illyricarum Studia I, Zagreb 2008, 15–54. Rummel et al. 2012 – C. Rummel, D. Peters, G. Schafferer, Report on the Geomagnetic Survey at Margum in October 2011, Starinar LXII/2012, 229–238. Trifunovi} 2012 – S. Trifunovi} (ur.), Arheolo{ka topografija Op{tine Novi Kne`evac, Arheolo{ka topografija Banata 1, Novi Sad 2012. Fowler 2010 – M. J. F. Fowler, Satelite imagery and archaeology, Landscapes through the Lens. Aerial photographs and Historic Environment, Occasional Publication of the Aerial Archaeology Research Group No. 2, Oxford – Oakville 2010, 99–110. Bugarski, Ivani{evi}, u {tampi – O`ivqavawe izgubqenih krajolika, Glasnik Dru{tva konzervatora Srbije 38, u {tampi. Doneus, Briese 2006 – M. Doneus, C. Briese, Full-waveform airborne laser scanning as a tool for archaeological reconnaissance, From Space to Place. Proceedings of the 2nd International Conference on Remote Sensing in Archaeology, BAR International Series 1568, Oxford 2006, 99–106. Gojda, John 2013 – M. Gojda, J. John (eds.), Archeologie a letecké laserové skenování krajiny, Plze² 2013. Ivani{evi} 2012 – V. Ivani{evi}, Akvedukt Cari~inog grada – Justinijane Prime, Saop{tewa XLIV, 13–31. Ivani{evi}, Bugarski 2012 – V. Ivani{evi}, I. Bugarski, Primena LiDAR tehnologije u analizi topografije Marguma/Morave i Kuli~a, Starinar LXII/2012, 239–255. Ivani{evi}, Bugarski 2013 – V. Ivani{evi}, I. Bugarski, Prva doma}a iskustva u dokumentovawu {irih zona arheolo{kih nalazi{ta putem LiDAR tehnologije, Glasnik Dru{tva konzervatora Srbije 37, 79–84. Musson et al. 2013 – Ch. Musson, R. Palmer, S. Campana (eds.), Flights into the Past. Aerial photography, photo inter- Ivan BUGARSKI Ifigenija Radulovi}, Gordan Mari~i}, ANDOKID, BESEDNIK KOJI JE JEDNOM POGRE[IO, Beograd – Novi Sad 2013 (izdava~i: NNK internacional i Filozofski fakultet Univerziteta u Novom Sadu), 256 strana teksta Andokid, kome je posve}ena ova monografija, ne nalazi se na spisku retora koji ~ine „Kanon“ (kao autori, ja stavqam ovde znakove navode, svestan koliko je izbor utemeqeno li~an, a ne sasvim kanonski) samo zato {to mu je, razumqivo, posve}ena ve}a pa`wa. Tako je u slede}em poglavqu opisan wegov `ivot (str. 46–50), a potom je data jedna zrela i uspela analiza Andokidovog besedni~kog delovawa, usko vezanog za neke od najva`nijih politi~kih doga|aja u Atini krajem 5. veka pre Hrista (str. 50–64). Naro~ito je u ovom poglavqu primetna pedago{ka qubaznost dvoje autora prema ~itaocu: uprkos tome {to pi{u o komplikovanim doga|ajima, oni ne ostavqaju na cedilu ~itaoca koji unapred nije upu}en u atinsku politiku poznog 5. veka. Ali, ta qubaznost, a ~ini mi se da je to dragocen kvalitet ove kwige, nije odvela tekst u publicisti~ke vode – svaki ambiciozniji student (upotrebqavam ovde ovu Ifigenija Radulovi} i Gordan Mari~i}, profesori novosadskog i beogradskog Filozofskog fakulteta, autori su jedne nove monografije o besedniku Andokidu, monografije koja sadr`i i prevod wegovih govora sa komentarom. U pitawu je ambiciozno delo, rezultat nemalog truda i odnegovane u~enosti. Uvod (str. 11–45) predstavqa kratku i efektnu studiju o anti~koj retorici, studiju koja, pored sistematskog dela (Retorika. Nastanak, razvoj i zna~aj, str. 11–28), sadr`i tako|e istorijski odeqak (Retorski „Kanon“. Antifont, Lisija, Isokrat, Isej, Demosten, Likurg, Hiperid, Dinarh, Eshin, str. 28–45). ^itaoci su tako dobili priliku da se temeqno upoznaju sa retorskim pojmovima, a zatim i da se, uz jedan kratak ali vrlo efektan pregled politi~kih biografija glavnih retora 5. i 4. veka, upoznaju sa glavnim besedama gr~kog polisnog sveta. 269 STARINAR LXIV/2014 KRITIKE I PRIKAZI – COMPTES RENDUS (265–270) Zatim, prevodi sve ~etiri besede (str. 132–183). Prvo o srpskom jezi~kom dostignu}u: jezik je jasan, re~enica te~na i bistra. Sadr`aj samih beseda, pa i wihova forma, nisu sasvim saglasni sa modernim senzibilitetom, ali ~ini mi se da su prevodila~ki talenat i ulo`eni napor doneli odli~an rezultat. ^itati danas Andokida na srpskom ne mo`e se porediti sa onim naporom koji je potrebno ulo`iti da se pro~ita neki drugi besednik (prevodi su, na`alost, i po stotinu godina stari). Pritom, spretno i ta~no je prenesen besedni~ki izraz, kako u formi tako i u su{tini, {to je zahtevalo poseban napor, jer je sve~ani, politi~ki i javni format izra`avawa danas gotovo nestao iz srpskog jezika i dru{tva. Posledwa celina u kwizi jesu bele{ke uz tekst beseda (str. 184–206). Uz potrebna specifi~na obja{wewa, autori su imali prilike da tu pobli`e objasne pojedine probleme, koje nije bilo mogu}e razjasniti u uvodima uz pojedine besede. Mislim da su ove bele{ke to the point i da su korisne, pa ~ak i neophodne. Zavr{avam nagla{avaju}i visoko dostignu}e koje predstavqa ova kwiga za na{e klasi~ne studije: dobili smo jedan odli~an prevod, uz opravdano op{irnu, korisnu i ta~nu uvodnu studiju i ve{to sro~ene bele{ke uz tekst Andokidovih beseda. re~ na na~in kako to Anglo-Saksonci ~ine, blisko latinskom smislu) ima}e puno koristi od ovog poglavqa, u~enog i dinami~nog. Slede}e poglavqe je centralno u ovom delu kwige. Ono je posve}eno obja{wewima koja idu uz besede Andokidovog korpusa (O svom povratku, str. 65–76; O misterijama, str. 76–82; O miru sa Lakedemowanima, str. 105–118; Protiv Alkibijada, str. 118–129), uz korisne, gotovo neophodne ekskurse o hermokopidama (str. 82–95) i aferi Eleusinskih misterija (str. 95–105), dvema aferama koje imaju centralno mesto u Andokidovom `ivotu i besedama. ^italac je ve} dobro pripremqen u prethodnim poglavqima, tako da su uvodi u ove besede dobro odmereni i jasno napisani. Strogost nau~nog teksta je odr`ana tokom ~itavog sadr`aja – taj utisak bi bio ja~i da su autori odustali od nekih upu}ivawa na moderne romane koji se bave ovim doga|ajima. (U~inilo mi se da je ponegde, u spisku literature naro~ito, pla}en obol ~ar{ijskim realnostima; ako sam u pravu, za tim nije postojala potreba, ta vi{a potreba koja je vodila autore da sastave primereno izdawe jednog anti~kog besednika.) Jo{ jedna lektura bi dobrodo{la prilikom pre{tampavawa ili novog izdawa: tipografskih gre{aka u napomenama ima vi{e nego {to bi se o~ekivalo; me|utim, te gre{ke jedva da uti~u na utisak o pedantnom i u~enom tekstu. @arko PETKOVI] 270 CIP – Katalogizacija u publikaciji Narodna biblioteka Srbije, Beograd 902/904 STARINAR / Arheolo{ki institut Beograd; urednik Slavi{a Peri}. – God. 1, br. 1 (1884) – god. 12, kw. 1/4 (1895) ; novi red, god. 1, br. 1 (1906) – god. 4, br. 2 (1909) ; novi red, god. 5, br. 1/2 (1910) – god. 6, br. 1/2 (1911) ; tre}a serija, kw. 1 (1922) – kw. 15 (1940) ; nova serija, kw. 1, god. 1 (1950) – . – Beograd : Arheolo{ki institut Beograd, 1884–1940; 1950 – (Beograd : Alta Nova). – 30 cm. Godi{we ISSN 0350–0241 = Starinar COBISS.SR–ID 8111874