Problems of the Late Neolithic Absolute Chronology in Eastern Croatia Marcel Burić Abstract Neolithic and the Late Neolithic sequences (roughly, on an axis running from Polgár-Cső sz- Although the first systematic excavation of a halom in northern Hungary, through Sopot, Sar- Neolithic site in Croatia was completed almost vaš and Bapska in Croatia, Gornja Tuzla in north- 120 years ago, we still lack aclear picture of its eastern Bosnia, down to Okoliš te in Central absolute chronology. The Late Neolithic phase of Bosnia). Except for the type site of Sopot, Croatia the whole Eastern Croatia shares the same des- still does not have atell excavation, which would tiny. As an attempt to clarify the issue, several al- provide asignificant number of stratified abso- ready published and some more recent regional lute dates. This is primarily due to the fact that absolute dates are presented and discussed in radiocarbon dating was not immediately ac- the paper. Despite those dates, the Late Neo- cepted as amethod by the majority of the scho- lithic chronology of the given area – embedded lars in Southeast Europe.3 This includes S. Dimi- in recently available archaeological data for the trijević as the main figure in past research on the Balkans – remains fairly vague. Neolithic Period in Croatia.4 Also, there are no precise absolute chronological intrasite ana- lyses, which would shed some light in that direc- Introduction tion as well. Consequently, there are no chrono- logical schemes relying on Bayesian modelling Fig. 1 .Sites mentioned in The most extensively excavated area in Croatian or other models that might fill the void in the text: 1 Sopot; 2 Vinkovci-Ho- prehistory, particularly the area dated to the chronological framework of Late Neolithic tell tel; 3 Dubovo-Koš no; 4 Otok- Neolithic period, belongs to the eastern parts of settlements in Croatia. The lack of absolute Mandekov vinograd; 5 Privla- the country and in awider cultural frame, to the dates, unfortunately, is not only present in the ka-Gradina; 6 Herrmannov vi- Balkan or Southeast European Neolithic. At a case of tell settlements: almost the identical si- nograd; 7 Č epin-Tursko number of sites we witness ahabitation layer of tuation is visible in the earlier Neolithic period, groblje; 8 Knež evi vinogradi– the Starč evo culture in the Early and Middle Neo- prior to the formation of tells. There, compared Osnovna š kola; 9 Slavč a; 10 Krč avina-NoviPerkovci; lithic periods, while the Late Neolithic is marked to the number of excavated sites, we can also 11 Ivandvor; 12 Ravnjaš ; by the Sopot culture: its cultural expression is re- witness alow number of radiocarbon data. In 13 Vidovci-Glogovi; presented by alocal technique of dark-bur- the past the chronology of the Neolithic Period 14 Bapska-Gradac; nished pottery. The origin of the Sopot culture is in Croatia founded solely upon pottery typology, 15 Klokoč evik. closely related to the Vinč aculture. It is consid- ered as apost-Starč evo culture production strongly influenced by Vinč apottery.1 The first archaeological surveys as well as the first excavations in Croatia were carried out at sites of the Sopot culture during the 19th cen- tury,2 yet only afew of them yielded data that can be used in amodern archaeological science. Despite modern dating approaches, in which the accuracy and availability of the radiocarbon methods have been significantly increased, Eastern Croatia as apart of technological and cultural trends of SoutheastEuropean Late Neo- lithic still reflects vague cognition. The main rea- son for this situation is the low number of sys- tematic research in which radiocarbon analysis has been performed as astandard procedure. It is worth stressing that Eastern Croatia repre- sents one of the western fringes of the tell-settle- ment phenomenon ( Fig. 1 ), where one can find stratigraphic accumulations of the late Middle 1 Dimitrijević 1968, 53– 59;Dimitrijević 1979a, 267– 269, 3 Yerkes et al. 2009, 1075. 273– 285, 293– 298. 4 However, he was the first Croatian archaeologist to use 2 Victor 1870;Celestin 1897. the radiocarbonmethod (Dimitrijević 1968, 92). 143 ArchäologieinEurasien 31 – 09_Buric – Seite 143 – 27. 7 . 15 Neolithic and Copper Age between the Carpathiansand the Aegean Sea mainly through the work of S. Dimitrijević . 5 nental part of Croatia is Sopot,8 located on the Based upon pottery typology and artistic devel- outskirts of the town of Vinkovci.9 In the past opment of the Sopot culture, his sequence for twelve years ( 1996– 2008)the site was systema- the Late Neolithic was made at several key sites: tically excavated by the Vinkovci Town Museum, Klokoč evik, Otok, Sopot and Bapska.6 Therefore, giving us the greatest number of radiocarbon some aspects of their chronology will be dis- data from any single site in the Croatian Neo- cussed in this text. lithic.10 Based on the 14Cdata, the time span of the excavated part of the settlement ( 376 m 2 ,on Briefly, the fact is that it is hard to reach any ser- the southwest plateau of the site) is between ious scientific conclusions about the relations 5050 and 4040 calBC ( Tab. 1 – 2 ).11 between the Late Neolithic sites and their abso- lute chronology in Eastern Croatia, that is, of According to the director of excavations, the course, within the framework of modern archae- radiocarbon results from the Sopot tell can be di- ological science. Fragmented and short-term test vided into three major groups (excluding the one excavations, which were the main methodologi- belonging to the Early Neolithic Starč evo group): cal approach in the last 50 years (and in many cases still are), are partly the reason for the Group 1 :The oldest dates from the settlement. above situation. However, since Croatia’ sinde- The house marked SU 23 (SU stands for ‘ strati- pendence in early 1990s, especially during the graphic unit’ )would be the oldest one belonging last decade, many rescue excavations have ta- to the Sopot culture on this site ( 5212– ken place in continental Croatia. Asubstantial 4711 calBC). The collapsed walls of this house number of them dealt with sites of the Neolithic are dated to 5212– 4703 calBC (charcoal sam- and Copper Age periods. The excavations were ples).12 triggered mainly by the construction of new high- ways (in Đ akovo and Osijek area). Yet, then Group 2 :The middle phase of the excavated set- again, we note with regret that within the frame- tlement is dated to 4322– 3798 and 4337– work of this research there is only arelatively 3982 calBC (several house floors and ditch fills). small number of 14Canalyses, and even fewer analyses have been published to the present Group 3 yielded dates ranging from 4443– day.7 Furthermore, the majority of the dates were 3824 calBC.13 Attention must be drawn to two obtained by conventional radiometric 14Canaly- dates that originate from the same excavation sis instead of the Accelerator Mass Spectrome- unit (SU 183a): samples Z-3868 ( 6295 ± 135 BP) try ( AMS) method. As will become visible from and Beta-230033 ( 5760 ± 40 BP), both on char- the data presented in the text following below, coal. As Krznarić - Š krivanko already pointed out, the standard deviation for results around both samples diff er considerably in radiocarbon ± 100 years or more yields alonger time span of age: 6295 ± 135 and 5760 ± 40. 14 The youngest dates, which is no longer acceptable for an accu- house from the Sopot tell excavated so far is rate chronology according to the current metho- the one labelled SU 11,with the date 4443– dological standards in archaeology. Considering 3980 calBC (Z-2827,see Tab. 1 – 2 for details). the fact that asubstantial number of these dates were published locally and/or in Croatian, we The majority of the radiocarbon data from Sopot will use this opportunity to present all of them originates from wood (charcoal) samples; there- here, regardless of the radiocarbon method by fore, we must take arelatively high percentage of which they were obtained. ‘ old wood e ff ect’ into account. Only four (Late Neolithic) dates are from human teeth (Beta- 251913, 15 Beta-251912,Beta-251908,Beta- Flat settlements and tell sites 25190716). of the Late Neolithic in Croatia with 14C-dates 8 Sopot (tell settlement) 45° 15' 51. 72''N 18° 46' 4 . 78''E 9 Dimitrijević 1968;Dimitrijević 1979a; Krznarić - Š krivanko 2006, 11– 19,Burić 2009a. The type-site of the most important and most 10 For an extensive list of the articles about the type site, widespread Late Neolithic culture in the conti- see Krznarić - Š krivanko 2011. 11 Krznarić - Š krivanko 2011, 210– 211.Inthis paper only the dates that derive from closed contexts (pits, houses 5 Dimitrijević 1966;Dimitrijević 1968;Dimitrijević 1969; etc.) are presented. Dimitrijević 1971;Dimitrijević 1974;Dimitrijević 1979a. 12 Krznarić - Š krivanko 2011, 212. 6 Dimitrijević 1968, 30– 31;Dimitrijević 1979a, 273. 13 Krznarić - Š krivanko 2011, 219. 7 As far as we know there are some radiocarbon dates col- 14 Krznarić - Š krivanko 2011, 210, 218. lected during 2005– 2006 excavations from Kaznica-Rutak 15 The date is not included in Table 2 :itdoes not originate and Debela š uma (both in the area around the sites No. 10 from aclosed context. and 11 on Fig. 1 ), but they are still unpublished and una- 16 The date is not included in Table 2 :itdoes not originate vailable. from aclosed context. 144 ArchäologieinEurasien 31 – 09_Buric – Seite 144 – 27. 7 . 15 M. Burić – Late Neolithic Absolute Chronology Dubovo-Koš no (flat settlement) Lab.-No. Sample type/ContextBPcalBC ( 2 σ ) During the year 2000,arescue excavation was Z-2752 Charcoal,house floor, 1 .