The Later Aeneolithic in Southeastern Author(s): H. Arthur Bankoff and Frederick A. Winter Source: American Journal of Archaeology, Vol. 94, No. 2 (Apr., 1990), pp. 175-191 Published by: Archaeological Institute of America Stable URL: http://www.jstor.org/stable/505948 Accessed: 26/02/2009 20:31

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=aia.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

Archaeological Institute of America is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Archaeology.

http://www.jstor.org The LaterAeneolithic in SoutheasternEurope H. ARTHUR BANKOFF AND FREDERICK A. WINTER

Abstract European economy and society that, if not as striking Changes in the European economy and society during as the beginnings of agriculture that marked the start the later Aeneolithic (late fourth to third millenniaB.C.) of the preceding , were equally fundamental were fundamental in the of cul- determining trajectory in determining the trajectory of cultural change in tural change in Europe for at least the next 1500 to 2000 Europe for at least the next 1500 to 2000 years (or years. Recent research shown that this period, rather until well into the /La Tene We than being a short transitionbetween the "Neolithic"and Age). the "BronzeAge," begins earlierand lastslonger than has believe that recent research has shown that this pe- been traditionallythought. Thus, the sociocultural,eco- riod, rather than being a short transition between the and materialtransformations observed the end nomic, by "Neolithic" and the "Bronze Age," begins earlier and of the Aeneolithic may be seen to be the result of gradual lasts longer than has been traditionally thought. Thus, changes over a considerableperiod of time, rather than the and material transfor- necessitating explanations involving unique dramatic sociocultural, economic, events such as migrationsor invasions.* mations observed by the end of the Aeneolithic may be seen to be the result of gradual changes over a INTRODUCTION considerable period of time, rather than necessitating In southeastern Europe the two millennia that be- explanations involving unique dramatic events such gin at about 4500 B.C. form a period that does not fit as migrations or invasions. into the classic Three Age system. What one calls it is Beginning in the early fourth millennium B.C., implicitly based on how one approaches it. If tech- southeastern Europe underwent major changes in nology, especially metallurgy, forms the basic criterion settlement and subsistence economy that marked a to differentiate this period from the one before, then very significant break from the Neolithic.' These one opts for Age (or the fancier "Chalco- changes included the abandonment of many long- lithic"). The same emphasis is implicit in the more settled sites, an apparently more dispersed settlement common designation "Aeneolithic" (Lat. aeneus = of pattern, and a greater dependence on animal husban- copper or bronze). Considered as a bridge between dry. The socioeconomic patterns associated with these the Neolithic and the Bronze Age (an attempt to hold changes form the basis for the ensuing Bronze Age. onto the Three Ages), it can be considered a "transi- It is closer to reality to dispense with the traditional tion period." This paper concentrates on the latter labels altogether, and to treat the developments in part of this time span, the late fourth to late third southeastern Europe during the fourth and third millennia B.C., the later Aeneolithic, Chalcolithic, or millennia B.C. together, although they encompass por- Copper Age, depending on one's terminology, and tions of the traditional Aeneolithic and Bronze Age. the centuries immediately following it, the beginning Recognizing that southeastern Europe is not com- of what is traditionally referred to in southeastern monly the focus of Western European and North Europe as the Early Bronze Age. American archaeological studies, we will begin with a In fact, the period from about 3200 to 2300 B.C., review of the general geography and cultural se- no matter what it may be called, saw changes in the quence.2

* This is a revised version of a paper presented at the are rare and generally limited to a specific Columbia University Seminar on the Archaeology of the region. Perhapsthe most accessibleare the relevantchapters Ancient Near East, the Mediterranean, and Europe. We in CAH III, 1 (1982) and R.K. Evans and J.A. Rasson, "Ex would like to thank Prof. Edith Porada for giving us the BalcanisLux? Recent Developmentsin Neolithic and Chal- opportunity to speak, and for her helpful comments and colithic Researchin Southeast Europe,"American Antiquity suggestions.We would also like to thank Petar Glumac,Tim 49 (1984) 713-41. Sections of J. Coles and A.F. Harding, Kaiser,and Bernard Wailesfor their astute advice. Remain- TheBronze Age in Europe(London 1979), and R. Tringham, ing errors are our own. Hunters,Fishers and Farmersof EasternEurope, 6000-3000 T. Champion, C. Gamble,S. Shennan, and A. Whittle, B.C. (London 1971) contain good, if now somewhat dated, PrehistoricEurope (New York 1984) 154; A. Sherratt, information. A more detailed chronology of the area is "Ploughand Pastoralism:Aspects of the SecondaryProducts discussedin R.W. Ehrich and H.A. Bankoff, "Geographical Revolution,"in I. Hodder, G. Isaac,and N. Hammondeds., and ChronologicalPatterns in EastCentral and Southeastern Pattern of the Past (Cambridge 1980) 261-306. Europe," in R.W. Ehrich ed., Chronologiesin Old World 2 Surveysor review articlescovering the later part of the Archaeology2(Chicago, in press). Aeneolithic and the beginning of the Bronze Age in south- 175 AmericanJournal of Archaeology94 (1990) 176 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA94

Fig. 1. Map of archaeologicalsites in southeasternEurope

GEOGRAPHIC BACKGROUND west, the in the center, and in the east. Geographically, our focus will be the middle Dan- Although physiographically (and until 1920, a the ube river basin, a drainage that includes the "Carpa- politically) unity, Carpathian Basin has been broken into thian Basin," as well as the Moravian Corridor, the historically up many re- the the Alpine tributaries of and Slovenia, and gions including , Transdanubia, Little the the southern tributaries of the and the Alfold, (Great) Alfold, Transtisia, Transyl- vania, and Slovakia, all of which lie within the Car- (such as the ), which flow northward from the pathian arc. Balkan ranges of Bosnia and 1).3 This (fig. Across the Danube to the south loom the mountains river system drains an area of almost 730,000 m2 of of Sumadija, Serbia proper, broken the southeastern What is most about only by plain Europe.4 striking of the Morava, which flows northward into the Dan- this area, when it is viewed in to the comparison ube between and Pozarevac. Almost im- Middle East and circum-Mediterranean regions, is the mediately to the east of the Morava plain, from the temperate climate and the year-round free accessibil- left bank of the , the Djerdap region to water. ity of the Danube begins. Here the river, which has at the different subdivisions of this Looking larger heretofore run easily at the southern edge of the level the Basin" area, "Carpathian includes Pannonia in the loess country of the Pannonian Plain, violently cuts

3 R.W. and 4 Ehrich, "Geographical Chronological Pat- G. Hoffman, A Geographyof Europe(New York 1953) terns in East ,"in R.W. Ehrich ed., Chronol- 522. ogiesin Old WorldArchaeology (Chicago 1965) 403-58. 1990] THE LATER AENEOLITHIC IN SOUTHEASTERN EUROPE 177 through a system of alternating narrow defiles and the way to the south; the Sava, leading westward to slightlywider basinsextending for more than 150 km the Alpine forelands; the Danube, the primaryroute from on the Yugoslav side downstream to to and from CentralEurope; the Tisa, running down Turnu Severin on the Romanian bank. This part of from and Slovakia, whose plain into the river is sometimes referred to as the "," the metal-rich mountains of western Transylvania; although the term more properly belongs to a specific and the Tamis, connecting to the rich lands of the gorge near Sip. Today considerably widened and CarpathianBasin. To the east acrossthe Moravaflood tamed by the lakes formed behind two dams (Djerdap plain (8-12 km wide in its lower course) lie the copper- I just upstream of , Djerdap II at Prahovo), -bearing mountains of East Serbia. To the west, this was formerly the wildest part of the river, and easily accessiblealong the many small Moravatribu- formed the nearly impenetrable boundary between taries, lie the loess-coveredrolling hills of Sumadija.6 the Danube's middle and lower courses. Here the These waterwaysform the primary natural routes of river originallydescended from 70 to 40 m above sea communication,either by offering navigablepassages level, a far steeper gradient than anywhereelse along for water traffic or broad river valleys and easy avail- its length. Whereas the width of the Danube above abilityof water for humans or animals.These riverine the gorge sometimes exceeded 1800 m, in the nar- routes continued to be importantthroughout prehis- rower defiles of the Iron Gates it shrank to less than toric and historic times, as can be seen even today by 150 m.5The river depth varied considerably,at points inspectionof the routes of majorroads and railroads. reaching 20 m, with potholes eroded to 50 m below The river and tributarystream terraces also provide the river'ssurface. At other places sand and rock bars favored environments for settlements, often with ac- blocked the shallower portions. Thus, passage cess to arable river-bottomland and woods, as well as through this part of the river by water was extremely upland pastures. The importance of these rivers and hazardous. their valleys in the clarificationof the culture history The river gorges in this area are actually narrow and processes of change in southeastern European valleys, rather than perpendicularlycut canyons, the prehistoryis a corollaryof their importanceas routes steep sides of which slope to the water'sedge. In the linking different natural regions and different local narrow defiles the Danube fills the valleys from side archaeologicalcultural sequences. to side, and the sheer walls may reach up to 610 m CULTURE HISTORY above the original river level. Land passage along the river was difficult or impossible; the trails through With the geographical focus thus delineated, our this region, if they existed, would have run along the next task is to outline the general cultural sequence ridgeline as do the modern roads. In the wider basins of the middle Danube. We should stress at this point the river again hugs the southern (Serbian)side, while that we are talkingabout traditionallydefined archae- the Romanianbank is usually less steep. ological cultures, identified predominantlyon the ba- This, then, is a general and compressedgeographic sis of pottery styles and, to a lesser extent, on other and topographicpicture of the Middle Danube drain- aspects of materialculture.7 One of the most serious age. It includes regions within present-day Hungary, shortcomingsof the archaeologyof southeastern Eu- ,, and, to a lesser extent, . rope is the paucityof data from intensive, systematic, The authors'research has been concentratedin the field survey.8This makes the estimationof actual site of the Velika (Great) Morava.This river flows numbers and population density extremely suspect. northwardssome 170 km through centralSerbia from Culturalareal distributions,however, are more easily the junction of the Juzna (Southern) Moravaand the worked out on the traditionalbasis of non-intensive, Zapadna (Western) Morava at Stalac. Its confluence non-systematicsurvey and the plotting of character- with the Danube at Smederevo is one of the most istic stray finds.9 For archaeologistswho do not spe- important natural crossroads of Europe. Here, five cializein this area the unfortunate practiceof labeling major waterwaysconverge: the Morava,opening up cultural entities or groups on the basis of site names,

5 Great Britain: Naval Intelligence Division,Jugoslavia 8 See J. Chapman'schapter on the regionalsetting of (Geographical Handbook Series, BR 493), vol. I: Physical Ge- Selevacin R. Tringham and D. Krstic(supra n. 6). 9 ography(London 1944/1945). See Ehrich (supra n. 3); also, R.W. Ehrich, "Culture 6 See R. Tringham and D. Krstic, , A Neolithic Areas and Culture Boundaries Through Time: Tier 3," in Villagein Yugoslavia(Los Angeles, in press) for a review of E.-J. Rowlett and R. Rowlett eds., Horizons and Styles; Studies the geology of Sumadijawith relevant literature. in Early Art and Archaeology (G6teborg 1987). 7 See Evans and Rasson (supra n. 2) 716. 178 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA 94 which often includes defining the same culture by intensification are general in Europe at this time.16 In different names as one crosses modern national bor- southeastern Europe habitation sites include both sin- ders, discourages casual interest. What follows is a gle stratum villages and multi-stratum tell settlements. simplified sequence. During this period tell occupation continues in the The later Neolithic period in the middle Danube East Balkan area at sites like Polyanitsa,17 where drainage is characterized by sites of increasing size, houses within a square palisade become more complexity, and stability. Within Serbia and its envi- crowded through time, while in the West , rons, extending from Macedonia in the south through including the , the flat settlements of the Vojvodina in the north, this is the Vinca period, the later Vinca (Vinca C-D or Vinca-Plocnik) also perhaps best known for its attractive figurines and provide some evidence for a more nucleated village burnished pottery.'0 Vinca and related cultures also plan.18 occupy much of the interior of the Carpathian Arc. As the Aeneolithic progresses, cultural diversity Contemporaneous Salcuta-Krivodol pottery is cur- increases. Vinca traditions persist immediately along rent in the mountainous regions of the Southern the Danube and in interior Serbia down into Mace- Carpathians." In southeastern Transylvania and Mol- donia.19 In , the eastern Carpathian regions, davia, as well as further east, Cucuteni farmers and the lower Danube, new sites of the Cernavoda settled,'2 while in the lower Danube, the contempo- culture are found.20 In the Morava valley and east- raneous and Vinca-related culture is Gumelnita, par- ward into Bulgaria, the contemporaneous cultures alleled in eastern Bulgaria by the culture which in its can be placed into the Bubanj-Hum sequence, named latest stages is known from the famous cemetery at after two sites on the Morava near Nis.21 Both of these Varna.'3 eponymous sites, Bubanj and Velika Humska Cuka, Some Southeastern European archaeologists be- present stratigraphic problems, neither preserves the lieve that these cultures generally represent the con- entire cultural sequence of the so-called Bubanj-Hum tinuation and expansion of the farming cultures of periods, and neither site has ever been fully pub- the earlier Neolithic, which were established in the lished. The derived chronology is based partially on area several millennia before.14 Others still cling to vertical stratigraphy and partially on comparative ce- the notion of a second Neolithic colonization or mi- ramic typology. With that in mind, we can note that 5 gration. All agree that during the earlier Aeneolithic Bubanj-Hum is divided into several numbered period, settlements continued to grow in size and phases, which continue into the Early Bronze Age. number. New settlements were also founded on sec- Bubanj-Hum Ia is the Serbian variant of the middle ondary areas around the periphery of the older set- Aeneolithic Salcuta-Krivodol-Bubanj Complex,22 tled regions. These trends of expansion and which is widespread in the central Balkans. Sites of

10 M. Vasic,Preistoriska Vinca 1-4 (Beograd 1932-1936); 49, 1978); H. Todorova,Kupferzeitliche Siedlungen in Nord- J. Chapman, The Vinca Culture of South-East Europe: Studies ostbulgarien(Munich 1982). 18 in Chronology, Economy, and Society (BAR-IS 117, i and ii, Chapman (supra n. 10); Tringham and Krstic (supra 1981); M. Gimbutas, The Goddesses and Gods of Old Europe: n. 6). Myth and Cult Images2 (Los Angeles 1982); Vinca u praistoriji l9 M. Garasanin, Praistorija na tlu S.R. Srbije ( i srednjemveku (Belgrade 1984); Tringham (supran. 2). 1973) 65-114. " D. Berciu, "Les nouvelles fouilles de Salcuta (Rou- 20 N. Tasic, "CernavodaIII i Boleraz nalazi u jugoslov- manie) et le probleme des groupes Bubanj (Yougoslavie)et enskom Podunavljui problem hronologkogodnosa kultura Krivodol(Bulgarie)," in J. Bohm and S. DeLaet eds., L'Eu- bakarnog doba u karpatsko-podunavskimoblastima," Bal- rope la fin de l'age de la pierre (Prague 1961) 125-34. canica6 (1975) 9-22. 12 L. Ellis, The Cucuteni-Tripolye Culture: A Study in Tech- 21 M. Garasanin,"Neolithikum und Bronzezeitin Serbien nology and the Origins of Complex Society (BAR-IS 217, 1984). und Makadonien,"BerRGK 39 (1959) 1-130; M. Garasanin, 13 M. Gimbutas, "Gold Treasure at Varna," Archaeology "The Stone Age in the Central Balkan Area," CAH III, 1 30 (1977) 44-51; I. Ivanov, "Les fouilles archeologiquesde (1982) 75-135; "The EneolithicPeriod in the CentralBalkan la necropole chalcolithiquea Varna,"Studia Prihistorica 1- Area,"CAH III, 1 (1982) 136-62; "The Bronze Age in the 2 ( 1978) 13-26; A.C. Renfrew, "Varnaand the Social Central Balkan Area,"CAH III, 1 (1982) 163-86; M. Gar- Context of Early Metallurgy,"Antiquity 52 (1978) 199-203. asanin, "Considerationssur la transition a l'age du bronze 14 See N. Tasic and S. Dimitrijevic,"Uvod," Praistorija dans les regions centralesdes Balkans,"Godisnjak (Sarajevo) Jugoslavenskih Zemalja III (Eneolit) (Sarajevo 1979) 13. 21 (1983) 21-26; M. Garasanin,"Grupa Bubanj-Hum III," 15 See M. GaraSanin,"Zur chronologischen und kulturel- Praistorija Jugoslavenskih Zemalja 4 (1983) 719-22; Arheo- len Wertung der Bubanj-Funde," Nachrufsschrifte fur Vla- loski lokaliteti Bubanj i Velika Humska Cuka (Nis 1983). 22 dimir Milojcic (Mainz 1979). Garasanin (supra n. 19); N. Tasic, "Bubanj-Salcuta- 16 Champion et al. (supra n. 1) 133. Krivodol kompleks," Praistorija Jugoslavenskih Zemalja 3 17 H. Todorova, The Eneolithic in Bulgaria (BAR-IS suppl. (1979) 87-117; B. Brukner, "Der Forschungsstand des 1990] THE LATER AENEOLITHIC IN SOUTHEASTERN EUROPE 179

this complex are known from Oltenia and western the Aeneolithic continues, first the Tiszapolgar,26 and Bulgaria, central and southern Serbia, down to Ma- then Bodrogkeresztur cultures develop.27 It seems cedonia and Albania at the same time as latest Vinca. likely that they arise from the earlier localized Vinca Bubanj Ib (not Bubanj-Hum, since remains of this of the region. They are best known from extensive phase are only found at Bubanj) is equivalent to the flat inhumation cemeteries,28 whose graves contain Oltenian Cernavoda-Renie II culture. Following this the occasional copper artifact, as well as pottery. there is a break in the occupation at Bubanj. During The Late Copper Age or Chalcolithic of Pannonia this time when Bubanj was not occupied, later Salcuta and the Hungarian Plain conventionally begins with (IIc-IV) pottery is current in Oltenia and Cernavoda the Baden culture.29 Various chronological systems I in . The gap in the Bubanj ceramic assem- have subdivided this culture's development into two,30 blage between Bubanj Ib and Bubanj II occurs at the three,31 or five32 phases. Regardless of the number of time when Cernavoda III/Boleraz pottery appears in defined subphases, the earliest, the Boleraz phase, the north, in other words, during the initial phases of marks a clear break with the Neolithic-derived past. the Baden pottery period.23 Bubanj remains unoccu- The ceramic assemblage is characterized by channeled pied during the time when classic Baden pottery is or fluted decoration, some finger-impressed bands, found on the Hungarian Plain, of which more infra. handled jugs, and wide conical bowls.33Moreover, the Occupation of at least part of the site of Bubanj (E. first evidence for paired-ox traction in Europe dates Plateau II/IIa) is next attested during the later Baden- to this phase. Boleraz is almost identical in material period. Bubanj-Hum II is a local Serbian culture to Cernavoda III, a middle Aeneolithic cul- variant of Oltenian late Cotofeni (Cotofeni III), which tural group whose sites have a middle to lower Danube is probably equivalent to the latest Aeneolithic Vuce- valley distribution.34 The mixture of Cernavoda III/ dol culture on the Hungarian Plain.24Finally, Bubanj- Boleraz with autochthonous traditions is considered Hum III pottery has close ties to the EBA horizon of by many to be the basis for the later Aeneolithic Armenochori and the late Macedonian Early Bronze Baden, Kostolac, Vucedol, and Cotofeni cultures,35 Age, as well as with that of the earliest Bronze Age which cover the whole middle Danube drainage. cultures of the Carpathian Basin to the north.25 The changes in material culture that are associated North of the Danube, on the Hungarian Plain, as with the beginning of the later Aeneolithic in south-

Aneolithikumsin Ostjugoslawien,"Atti del X simposiointer- ica 16-17 (1985-1986) 7-16. nazionalesullafine del neoliticoe gli inizi dell'etddel bronzoin 28 I. Bognar-Kutzian,The CopperAge Cemeteryof - Europa(Verona 1982) 77-97. polgar-Basatanya( 1963). 23 29 V. Nemejcova-Pavukova,"Zur Ursprung und Chron- J. Banner,Die PecelerKultur (Archaeologica Hungarica ologie der Boleraz-Gruppe,"Badener Symposium 1973, 297- 35, Budapest 1956); N. Kalicz,"Die Peceler (Badener) Kul- 316; V. Nemejcova-Pavukova,"Nacrt periodizacie Badenskej tur und Anatolien,"StArch 2 (1963); Symposiumiiber die Kulturya jej chronologickyhuztahov k juhovychodnejEu- Entstehungund Chronologieder BadenerKultur ( rope,"SlovArch 29 (1981) 261-96; Tasic (supra n. 20). 1973); V. Nemejcova-Pavukova,"Beitrag zu Kennen der 24 N. Tasic, "Cotofenikultura," Praistorija Jugoslavenskih PostBoleraz-Entwicklungder Badener Kultur,"SlovArch 22 Zemalja3 (1979) 115-28; P. Roman, CulturaCotofeni (Bib- (1974); P. Romanand I. Nemeti, CulturaBaden in Romania. liotecade Arheologia26, Bucharest 1976); P. Roman, The (Bibliotecade Arheologia31, Bucharest 1978). Late CopperAge CotofeniCulture of South-EastEurope, BAR 30 S. Dimitrijevic,"Prilog stupnjevanju badenske kulture suppl. 32 (Oxford 1977); T. Bader, Epoca bronzuluiin u sjevernojJugoslaviji," Arheoloski radovi i raspraveJugoslav- Nord-VestulTransilvaniei: cultura pretracica si tracica (Bu- enskeakademije znanosti i umjetnosti2 (1962). charest 1978). 31 N. Tasic, Badenskii vucedolskikulturni kompleks u Ju- 25 Garasanin (supra n. 21); N. Tasic ed., Kulturen der goslaviji(Belgrade 1967). Friihbronzezeitdes Karpatenbeckensund Nordbalkans(Bel- 32 E. Neustupny, "Die Badener Kultur,"Badener Sympo- grade 1984). sium 1973, 317-52. 26 I. Bognar-Kutzian,The Early CopperAge Tiszapolgar 33 A. Sherratt, "The Pottery of Phases IV and V: The Cultureof the CarpathianBasin (ArchaeologiaHungarica n.s. Early Bronze Age," in A.C. Renfrew, M. Gimbutas,and E. 48, Budapest 1972). Elstereds., Excavationsat SitagroiI (MonumentaArchaeolo- 27 A. Sherratt,"The Development of Neolithic and Cop- gica 13, 1986) 442. Settlement in 34 per Age the Hungarian Plain. Part I: The N. Tasic,JugoslovenskoPodunavlje od indoevropskeseobe RegionalSetting," OJA 1 (1982) 287-316; A. Sherratt,"The do prodoraskita (Belgrade 1983); N. Tasic, "Die Cernavoda Development of Neolithic and Copper Age Settlement in III-Kultur und der Zerfall friiher aneolithischerKulturen the Great Hungarian Plain. Part II: Site Survey and Settle- des JugoslawischenDonauraums," Godisnjak (Sarajevo) 21 ment Dynamics,"OJA 2 (1983) 13-41; on chronology rela- (1983) 27-35. tive to 35 other Aeneolithic groups see N. Tasic, "Neue Daten Tasic (supra n. 31) with literature;also Tasic (supra n. iiber das relativ-chronologischeVerhaltnis der fruhen aneo- 34). lithischenKulturen imjugoslawischen Donauraum," Balcan- 180 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA 94

eastern Europe may be seen not only in the ceramic ginally productive from a traditional archaeological assemblages,36 but in metallurgy as well.37 In the ce- standpoint, because of poor preservation, few fea- ramics, there is a general similarity among all the tures, and a paucity of diagnostic artifacts. Data from cultures of this period: a preference for grey polished such sites, however, may provide as much or more fine wares that differ markedly from the earlier information about the changes in subsistence and painted ware assemblages of the same regions. An- lifeways during the later Aeneolithic as those from other commonality among these ceramic assemblages, the less typical but more accessible larger settlements. despite the confusion of local cultural names, is the A good example of such a site from this period is new prevalence of cup and small jug shapes. The Novacka Cuprija in the Morava valley, near Smeder- behavioral correlates of these vessels have been the evska Palanka in (fig. 1), which was subject of some discussion. They may be evidence for excavated in 1980 as ajoint Yugoslav-American proj- the first widespread use of milk.38 When taken to- ect.42 Descriptions of the project and its results have gether with the first appearance of wheeled vehicles, been published elsewhere.43 Three of the trenches at it is only slightly facetiously that Andrew Sherratt Novacka Cuprija contained Bubanj-Hum III type ma- referred to this later Aeneolithic period as a time terial from the very beginning of the Early Bronze characterized by "drinking and driving."39 Age, found in undisturbed context for the first time in the lower Morava Valley. AENEOLITHIC/EARLYBRONZE AGE SUBSISTENCE: Three (Pits 1, 2, and 3) of dimensions NOVA6KA pits differing (UPRIJA and depths, and traces of a ditch in which was found Excavated habitation sites of the latest Aeneolithic/ one of the smaller pits (Pit 3), were discovered in these Early Bronze Age transition period are relatively rare; trenches. As far as can be determined from the ce- closed contexts that can provide some information ramic typology, all of these features (i.e., the three about subsistence are even rarer.40 On the basis of pits and the ditch) were contemporaneous, and prob- what is still a corpus based largely on unsystematic ably represent the remains of a single partially sub- collection and survey in many regions, it would appear terranean architectural feature or house, or the that site sizes and types were more variable than in subterranean parts (cellar or storage space) of an the earlier parts of the Aeneolithic.41 Many of the sites above-ground structure. The clear delineation of are shallow and ephemeral, possibly representing these closed contexts enhances the importance of the population that was more dispersed over the land- analysis of the botanical, faunal, and artifactual re- scape. Most of these sites have been judged only mar- mains.

36 Kalicz (supra n. 29). projectwas granted by the NationalScience Foundation,the 37 H. Miiller-Karpe,Handbuch der VorgeschichteIII: Kup- NationalEndowment for the Humanities,the NationalGeo- ferzeit(Munich 1974); H. Todorova,Die kupferzeitlichenAxte graphic Society,and the PSC-CUNYResearch Award Pro- und Beile in Bulgarien(Munich 1981). gram of the City Universityof New York. 38 See Sherratt(supra n. 1) 275-82. 43 H.A. Bankoff, D. Krstic, M. Vukmanovic, and F.A. 39 Sherratt (supra n. 1) 263-66; S. Piggott, The Earliest Winter, "Praistorijskilokalitet 'Novacka Cuprija',"Zborik Wheeled Transport, From the Atlantic Coast to the Caspian Sea radova NarodnogMuzeja 12 (1986) 17-62; H.A. Bankoff (Ithaca 1983). and F.A.Winter, "Brooklyn College-Beograd Narodni 40 Muzej Some palaeobotanicaland faunal analyses have been Excavationsat NovackaCuprija 1980,"ArchNews 10 (1981) from this time published approximate period. Of interest 9-12; H.A. Bankoff and F.A. Winter, "The MoravaValley for comparison are the analyses of the Kostolac plant re- Projectin Yugoslavia:Preliminary Report, 1977-1980,"JFA mainsfrom Gomolava(W. van Zeist,"Ugljenisani biljni ostaci 9 (1982) 149-64; H.A. Bankoff and F.A. Winter, "The na nalazigtu Rad vigeslojnom Gomolava," VojvodjanskihMu- LowerMorava Valley Project," in D.R. Kellerand D.W.Rupp zeja23-24 [1974-1978], especiallytable 1); also the material eds., ArchaeologicalSurvey in the MediterraneanArea, BAR- from Ezero (G.I. Georgiev, N.J. Merpert, R.V. Katincarov, IS 155 (1983) 203-205; H.A. Bankoff and F.A. Winter, and D.G. Dimitrov, Ezero: Rannobronzovoto Selisa (Sofia "Excavationof SmederevskaPalanka, Yugoslavia," National 1979). (German resume 535-43). Indirect arguments for GeographicResearch Reports 18 (1985) 131-42; H.A. - changed agriculturalstrategies are given in Championet al. koff, F.A. Winter,and H. Greenfield,"The Culture n. History (supra 1) 156-62. of the Lower MoravaValley, Yugoslavia,"Current Anthro- 41 et al. n. Champion (supra 1) 162. pology21 (1980) 268-69; H. Greenfield, The Paleoeconomy 42 The excavationsat Novacka Cuprijawere jointly con- of the CentralBalkans (Serbia) I: A ZooarchaeologicalPer- ducted by Brooklyn College of the City Universityof New spectiveon theLate Neolithic and BronzeAge (ca. 4500-1000 York and the National Museum of Belgrade, Yugoslavia. B.C.) (BAR 304, 1986); H. Greenfield, "SummaryReport Excavationswere directed H. by Arthur Bankoff and Fred- on the Vertebrate Fauna from Novacka Cuprija,"Zbornik erick A. Winter (Brooklyn College) and Dusan Krstic and radovaNarodnog Muzeja 12 (1986) 63-74. MirjanaVukmanovic (National Museum). Funding for the 1990] THE LATER AENEOLITHIC IN SOUTHEASTERN EUROPE 181

During the 1980 season at Novacka Cuprija, one Early Bronze Age. Grains include all three species of month was spent on intensive recovery of plant re- , as well as and millet. Examples of all mains from exposed contexts.44 A further period was the cultivated legumes also were found. In this regard, spent on analysis and identification, as well as on it is noteworthy that the fruits, which would probably examination of the local flora with a view to compar- have been collected in smaller quantities than the ative work. Among the contemporary flora, most rel- grains, and consumed soon after their collection, are evant to Novacka Cuprija were isolated stands found only in Pit 2. They are probably therefore in representing the original climax vegetation found in primary context. A certain (i.e., indefinite) amount of the Morava valley. These stands are dominated by oak this botanical material was connected with the activi- interspersed with occasional elm and ash, while ma- ties within the feature itself (when it existed in its ple, hawthorn, and wild cherry plum form the un- entirety), while the remainder would have been de- derstory of shrubs. This is, of course, the situation to posited within it at the time of its demolition or de- be expected on heavy clay-laden soils in Europe such struction, or as a result of activities in the immediate as are typical for the Morava valley. More locally, the vicinity a short time thereafter. north-south valley that runs just to the west of the site The neighboring smaller pit (Pit 1) probably served is the only exception to what must have been botani- as a storage place for grain (in the EBA) as is indicated cally a relatively homogeneous environment. The val- by the amount of grain (emmer and einkorn) found ley bottom, an area with springs and southward within it. Neither legumes nor fruits occur in the running water, is dominated by marsh species. The samples. The remains of freshwater mussels, rabbit ground is wet even in summer, and there is a possi- bones, and one human bone in this pit show that it bility that standing water existed in the past. was used secondarily as a garbage pit. Another indi- The soil extracted for processing was found to be cation of this pit's use for garbage was the great suitable for flotation.45 The upper levels of deposit presence of flint debitage (70% of all debitage found had a very high humus content and took much longer at the site). Pit 3, to the southwest of Pit 2, containing to break down, while the lower levels had a soil that material chronologically indistinguishable from the readily dispersed in water, thus releasing charcoal preceding pits, was almost without botanical remains. fragments easily. Eighty-six samples were taken dur- The shallower part of the pit, in the shape of a ditch ing the season, and these represented approximately or channel, contained the remains of at least one very 3.75 tons of deposit. After processing, samples were large vessel (possibly for holding food or liquid), dried and sorted. In addition, casts were made from which had formerly stood, most probably, along the plant impressions in daub to recover evidence of cer- very edge of the pit. Outside of the aforementioned tain plant material not recovered by flotation. pottery, this pit also contained a pendant worked of The plant materials recovered (Table 1) include dog's tooth and a fragment of a bone needle. various cereal grains with occasional spikelet frag- The faunal material from Novacka Cuprija has ments in sufficient quantities to establish the presence been treated at length elsewhere.46 As might be ex- of different species of wheat: einkorn (Triticum mon- pected, almost all (97%) of the animal bones found at ococcum), emmer (T. dicoccum), and bread wheat (T. the site come from domesticated species. Cattle re- aestivum). Three species of cultivated legumes (vetch, mains were the most numerous, followed by sheep or peas, and lentils) occur in these pits, and three edible goat and pigs. It is not possible to tell whether the succulent fruits (blackberry, cherry, and plum) that cattle were being raised for milk or for meat produc- may have been gathered from the surrounding veg- tion; at the neighboring Vinca site of Selevac, the etation. Non-cultivated plants include weeds and rud- latter has been proposed (for a period some 2,000 erals, though of more interest is the carbonized wood, years earlier than at Novacka Cuprija).47 Aside from representing several different species and throwing these domestic animals, bird bones and freshwater light on the ancient climax vegetation. mussel shells indicate other sources of animal protein In the largest pit (Pit 2), which is what remains of utilized by the Early Bronze Age inhabitants. the structure (or its subterranean portion), was found Artifactual material from the pits includes pottery, the greatest number of botanical remains from the lithics (both finished tools and waste), and worked

44 This section incorporates, with the permission of the Techniques," Mid-ContinentalJournal of Archaeology1 author, much of the report on the plant remains from No- (1976) 77-100. vacka(uprija by George Willcox. 46 Greenfield(supra n. 43). 45 P.J. Watson, "In Pursuit of PrehistoricSubsistence: A 47 A. Legge in Tringham and Krstic(supra n. 6) ch. 5. Comparative Account of Some Contemporary Flotation 182 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA 94

Table 1. Botanical Samples la. Occurrence of Grains

Context No. Sample Triticum Hordeum Panicum samples wt monococcum dicoccum aestivum sp. miliacum (kg) (einkorn) (emmer) (bread) (barley) (millet)

Pit2 27 1811 47 3 5 24 3 Pit 1 8 500 124 82 1 3 2 Pit 3 3 300 1 1 1 3

Totals 38 2611 172 86 7 30 5

lb. Occurrence of Legumes

Context No. Sample Vervilia Lens Pisum samples wt (bitter culinarus sativum (kg) vetch) (lentil) (pea)

Pit 2 27 1811 10 93 9 Pit 1 8 500 - - - Pit 3 3 300 1 2 3

Totals 38 2611 11 95 12

lc. Occurrence of Fruits

Context No. Sample Cornus Rubus Cornus Prunus samples wt sanginea fruticosus mas cerasifera (kg) (dogwood) (blackberry) (cherry) (plum)

Pit2 27 1811 1 4 9 3 Pit 1 8 500 Pit 3 3 300 -

Totals 38 2611 1 4 9 3

Id. Trees (Presence/Absence)

Context Quercus Ulmus Corylus Carpinus Crataegus Fraxinus (oak) (elm) (hazel) (hornbeam) (hawthorn) (ash)

Pit 2 P P P P A A Pit 1 P P P P A P Pit 3 P P P A A A 1990] THE LATERAENEOLITHIC IN SOUTHEASTERNEUROPE 183 bone. Pottery comprises by far the most numerous Late Vinca site of Selevac, some 20 km away, reveals and varied class of artifacts.Utilized and shaped bone some striking continuities over the approximately tools were found in all three pits belonging to this 2,000 years that separate the two sites.48The palaeo- complex. Again, the assemblageis typologicallyquite botanical remains from Selevac indicate that, as at homogeneous, both among the pits and among the NovackaCuprija, einkorn was found more frequently units of each individualpit. Awls or needles, probably than emmer wheat. Bread wheat (T. aestivum),which used for perforating hide, basketry,or woven mate- was absent from the Selevac samples, perhaps as a rial, were the most numerous bone implements. In all result of sampling error, occurs in small quantitiesin cases the implement is worked from a sliver of long the Novacka Cuprija inventory. Barley occurs much bone, and has a polished point. The larger examples more frequently at Novacka Cuprija,although there from Pit 1 may be leather awls; one shows file marks is no indicationof whether it was hulled (as at Selevac) on the point, and rotary wear as from use in drilling or naked, two-rowor six-row type. Broomcorn millet is visible on another. Three "spatulas" made of (Panicum miliacum) is unknown from Selevac, al- thinned and polished rib bones were also found in Pit though found at other contemporaneous sites, and 1. A similar-shapedpiece from Pit 2 had a perforation might have been used for making leaven or a fer- in one end, possiblyfor use as a pendant. Other items mented drink at NovackaCuprija.49 In both sites, peas that appear more than once are dog-tooth pendants and lentils are represented, and similar fruits were and bird bones possiblyused to apply impresseddec- gathered. The absence of bitter vetch is anomalous at oration to pottery. Other single implements such as a Selevac,compared with other sites of the time. scapular piece from Pit 1 and a polished horn piece Artifactcategories other than ceramicsconfirm the from Pit 2 may have been used for potteryburnishing. impression of similaritybetween the two sites. Bone Aside from plant and animal remains, activities artifacts, although fewer in types and numbers at connected with food preparation at the site are at- NovackaCuprija than at Selevac,appear to have been tested by the presence of grindstone fragments in Pit constructedand used in the same fashion.50Novacka 2. These fragments indicate that the grindstoneswere Cuprijacontains no new types of bone tools, and lacks made of metamorphicrock, rectangular,some five to the antler inventory of Selevac. This may be due to seven centimetersthick, and worn smooth by abrasion sampling error at the later site, since large numbers on one side. The source of the stone from which these of antler tools in finished and unfinished condition grindstones was made is unknown, but the known are known from other Serbian EBA sites such as sources of such metamorphicrocks point to a proba- Ljuljaci.51The paucity of antler, and of deer remains ble origin somewhere to the east of the Morava. in general at NovackaCuprija,52 might also reflect an Pit 2 had the most varied lithic assemblage of the actual drop in the deer population and in the impor- complex. Stone tool production or resharpening is tance of hunting around this site in the latest Aeneo- attestedby both stone chips and complete implements lithic/EBAperiod. (bladesand a smallernumber of denticulateartifacts), In general, then, subsistence activitiesas reflected as well as hammerstones.Polished stone axes are also in the faunal, floral, and lithic remains do not seem found in this context, as was a stone that may have to indicate any great differences between the Late been used for sharpening them. The total number of Aeneolithic/EBAat Novacka Cuprija and the Early chipped stone pieces recovered from all three pits is Aeneolithic Vinca site of Selevac. The most striking quite small (33) compared with the amount of pottery differences between the sites relate to the possible (over 4,500 pieces). The common presence of cortex social organizationaland population changes. At Se- on finished tools may indicate a dearth of good raw levac Chapmanposits a population range of 120-240 materialsand the concomitantuse of even the smallest people for the early period of the site, with the pos- and worst pieces of usable flint. Most probably the sibility of as many as 600-1,200 inhabitants during flint tools were fashioned on pebbles from the nearby the late phase.53Nothing at Novacka C(uprijagives rivers;there is no evidence that nodules were obtained any evidence of habitation by more than a few ex- from distant sources. tended families, comprising 50-100 people at most. A brief comparison of these data with those of the This small settlement size would appear to be a heri-

48 Selevacdata is fromF.S. McClaren and R.N.L.B. Hub- 51 M. Bogdanovic,Etnokulturna kretenja u centralnojSrbiji bardin Tringhamand Krstic(supra n. 6) ch. 6. u bakarnoi bronzanodoba (Belgrade 1983) 78-83. 49 J. Renfrew, 101. 52 See Greenfield 50 Palaeoethnobotany(London 1973) (supra n. 43). Selevacdata from N. Russellin Tringhamand Krstic 53 J. Chapmanin Tringhamand Krstic (supra n. 6) ch. 2. (supran. 6) ch. 14. 184 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA 94

tage of the Middle to Late Aeneolithic, if one can alogues.60 According to Sherratt,61 there is a hiatus in judge from the extent of Baden and Kostolacsettle- the Bulgarian sequence during the Cernavoda III/ ment at such sites as Gomolava.54Even large villages Boleraz period, while Early Ezero (Horizons XIII- of the earliest Bronze Age in the Morava drainage, VIII) is contemporaneous with classic Baden. The such as Ljuljaci,with some 29 excavated houses,55do Sitagroi sequence should connect the Morava valley not approach the nucleation of settlement seen at sequence more directly with the Aegean. The ceramic Selevac.At NovackaCuprija, and other sites in south- inventory from Sitagroi Va and Vb, especially the eastern Europe after the Middle Aeneolithic, smaller latter, resembles the pottery from the Novacka Cu- groups of people could (and in most cases did) live as prija pit complex quite closely. Again, this corre- well or better than those larger groups living at the sponds to Troy I or early II, or EH II in the Aegean, earlier settlements. At least some of the activitiesthat as does the single-handled cup from Aghios Kosmas had been the responsibilityof supra-familialgroups with Baden similarities. now were able to be taken care of by smaller house- It remains for us to put this material into an abso- hold groups. lute chronological framework. In the last 35 years, archaeologists have become ihcreasingly dependent CHRONOLOGY on radiocarbon dates for the creation of the temporal The relative chronology of the later Aeneolithic framework upon which the study of cultural devel- and earliest Bronze Age of the Morava valley, as opment rests. The effect of radiocarbon dating on exemplified by the assemblage from Novacka Cuprija, theory and interpretation in European archaeology can be tied into the stratified sequences available for has been most marked in Neolithic and Early Aeneo- this time period in southeastern Europe, especially lithic studies.62 The acceptance of a radiocarbon- the sites of Baile Herculane in Oltenia,56 Ezero in based higher chronology for the inception of these Bulgarian Thrace,57 and Sitagroi in northeastern periods allows a longer period of time for the devel- .58 This is shown in simplified form in Table 2. opment of agriculture and related Neolithic devel- Several points deserve special mention. Horizons opments. The high chronology in turn, has led to a VIII-III of Ezero can be related to Troy I and Po- reconsideration of the importance of diffusion from liochni "blue" and "green."59 They are thus "Early the Near Eastern/Aegean area in the cultural dynam- Bronze Age" in Aegean terms, although, as in the ics of continental Europe during the Neolithic and Morava valley sites, there is nothing in the metal Aeneolithic periods, and a recognition of this region's inventory or analyses to distinguish these levels from independence and vitality.63While radiocarbon dating the Aeneolithic or Chalcolithic levels on many other has not been without its opponents, its rejection can tells. According to Garasanin, Horizons XIII-VIII no longer seriously be considered.64 (and possibly VII-V), which he equates with Troy I, Despite these developments in radiocarbon dating, contain pottery with similarities to Cernavoda III/ and although not as true as it was 15 years ago, Boleraz, while Horizon III contains classic Baden an- chronological reasoning in southeastern Europe is still

61 54 J. Petrovic,"Eneolithique moyen et tardifa Gomolava," A. Sherrattin Renfrew et al. (supra n. 58) 445. in N. Tasic and J. Petroviceds., Gomolava;Cronologie [sic] 62 E. Neustupny, "AbsoluteChronology of the Neolithic und Stratigraphieder vorgeschichtlichenund antikenKulturen and Aeneolithic Periods in Central and Southeastern Eu- der Donauniederungund Sudosteuropas( 1988) 39- rope, I," SlovArch16:1 (1968) 19-60; E. Neustupny, "Ab- 46. solute of the Neolithic and Aeneolithic Periods 55 Chronology Bogdanovic(supra n. 51) 62. in Centraland SoutheasternEurope, II,"Archeologicke Rozh- 56 S. Marinescu-Bilcu, Cultura precucuteni pe teritorul ledy21 (1969) 783-810; E. Neustupny, "Der Ubergang vom Rominiei (Institutulde Arheologie 22, Bucharest 1974); V. Neolithikumzum Aneolithikumund der Ausklangder Len- Dumitrescu,"The Prehistoryof Romania:From the Earliest gyel-Kultur,"Studijne Zvesti (Nitra) 17 (1969) 271-92; A.C. Times to 1,000 B.C."CAH III, 1 (1982) 1-74. Renfrew,"The Autonomyof the South-EastEuropean Cop- 57 Georgiev et al. (supra n. 40). per Age," PPS 35 (1969) 12-47; A.C. Renfrew,"The Tree- 58 A.C. Renfrew, "Sitagroiand the Prehistoryof South- ring Calibrationof Radiocarbon:An ArchaeologicalEvalu- East Europe,"Antiquity 45 (1971) 275-82; Renfrew, Gim- ation,"PPS 36 (1970) 280-311; A.C. Renfrew,Before Civi- butas, and Elster eds. n. 33). lization York 59 (supra (New 1973). Georgiev et al. (supra n. 40); A.F. Harding in A.G. 63 Renfrew 1973 (supra n. 62). Poulter, Ancient Bulgaria: Papers Presented to the Interna- 64 See Ehrichand Bankoff (supran. 2); Evansand Rasson tional Symposium on the Ancient History and Archaeology of (supra n. 2) 716; for an opposing view of the validity of Bulgaria, University of Nottingham, 1981 (Nottingham radiocarbondates see J. Bouzek, The Aegean,, and 1983). CulturalInterrelations in the SecondMillenium B.C. 60 Europe: Garasanin (supra n. 15). (Goteborg1985) 19, 244. 1990] THE LATER AENEOLITHIC IN SOUTHEASTERN EUROPE 185 based firmly in a seriational and typological method- fusion and acceptance of an artifact type or the ology that ties the area tightly to Central Europe and duration of its use adds an uncertainty factor to this indirectly to the historical sequences of the Near East. kind of cross-dating that may be almost as large as Any discussion of the absolute chronology of the that of the standard deviation in a radiocarbon date, southeastern European late Aeneolithic and Bronze although in the case of cross-dating, the uncertainty Age must begin, therefore, with an analysis of the is usually less explicitly stated and occasionally not basis for the absolute dates for our terminus ante even recognized. quem, the Bronze Age of Central Europe. The Central European Early Bronze Age Absolute Chronology of the Later Prehistoric Periods in In theory then, the Early Bronze Age metal inven- Central Europe tory and its Aegean analogues should give a date for Since the Central European Bronze Age falls within the beginning of the central and (by extension) south- the historic period in Egypt and the Near East, cross- eastern European Early Bronze Age. Traditional dated chains of synchronisms and externally and in- cross-dating between the Egyptian/Mediterranean ternally consistent stratigraphic sequences may be historical sequence and the central and southeastern connected with Egyptian historical dates.65 European Early Bronze Age places the inception of historical on several Egyptian dates are dependent this period to slightly after 2000 B.C.68 According to astronomically datable events recorded as occurring Gimbutas in specific regnal years.66 Theoretically, the historical Egyptian chronology can be tied into that of the Eu- "almostall metal artifactsused by the ... [cultures of the EBA of the middle Danubian southern ropean Bronze Age through correlations at three tem- Hungarian Plain] and imitated by their northern have porally distinct points: a) the Early Bronze Age metal neighbors analogiesor prototypesin the Near Eastbetween Egypt inventory; b) the Shaft Graves at Mycenae and their and northern Iran, the most numerous and closest par- connections with the Middle Bronze European Age; allels being along the Syrian-Palestiniancoast and on c) the horizon of new metal types marking the begin- Cyprus.These are: neck-ringswith rolled ends, curved- ning of the Late Bronze Age.67 It is this earliest cor- shank pins with knot heads called Cypriotepins, or with relation point that concerns us here. simple spiralor loop heads, sheet-metalbelt-plates with rolled ends and embossed wound The use of Egyptian dates for European Bronze decoration,cylinders of thin copper wire, double wire spirals, with Age events is, however, fraught with difficulties. No earrings flattened ends, bracelets,and double artifacts have ever been found in plain spiral spiral Egyptian prehistoric pendants."69 contexts in Europe outside the Mediterranean. Non- Aegean prehistoric European artifacts are not found Such metal objects typify the inventory of Rei- in Egypt. The Egyptian dates, therefore, are associ- necke's Bronze Al period in central Europe, as well ated with Egyptian assemblages that are in turn used as the earliest EBA graves in the south Hungarian to date other non-Egyptian assemblages in which Eu- cemeteries and the Moris culture of the Yugoslav ropean artifacts or artifact types considered to be .70 Similar metal types, including ring ingots related to them occur. This, of course, is the most characteristic of a developed phase of the Central problematic kind of extended cross-dating. Disagree- European EBA (Reinecke Late Bronze Al), occur at ment over the length of time necessary for the dif- the sites of Byblos, Ras Shamra, Hama, and Tell As

65 P. Astr6m, High, Middle or Low? (Acts of an Interna- B.C. (Hayes, op. cit., 183) or 1517 B.C. (Kitchen, op. cit., tional Colloquiumon AbsoluteChronology Held at the Uni- 42), while several Sothic and lunar dates allow the absolute versity of Gothenburg 20-22 August 1987) (Gothenburg dating of the reigns of the pharaohs of the 19th and 20th 1987). Dynastiesto within a decade throughout the 15th through 66 The earliestsuch astronomicallyfixed point in Egyptian 12th centuries B.C. history is 1872 B.C. (W.C. Hayes, "Chronology:Egypt to 67 M. Gimbutas,Bronze Age Culturesin Centraland East- End of Twentieth Dynasty,"CAH I [1970] 174) or possibly ern Europe(The Hague 1965) 33; See also Coles and Hard- 1830 B.C. (R. Krauss, Sothis- und Monddaten,Studien zur ing (supran. 2) 379; A.F. Harding,"Radiocarbon Calibration astronomischenund technischenChronologie Altdgyptens (Hil- and the Chronologyof the European Bronze Age," Archeo- desheimerAgyptologische Beitrdge 20, 1985) 73; see also K.A. logickeRozhledy 32 (1980) 178. Kitchen,"The Basicsof EgyptianChronology in Relationto 68 E. Neustupny, "Absolute Chronology of the Bronze the Bronze Age," in Astrom (supran. 65) 43-44. Documen- Age in Central Europe," Istrazivanja(Novi Sad) 5 (1976) tary evidence (listsof the pharaohsand the durationof their 111-16. reigns) allowsthe calculationthat 3114 B.C. was the absolute 69 See Gimbutas(supra n. 67) 32-33. date for the beginning of the First Dynasty. Early New 70 B. Brukner, B. Jovanovic, and N. Tasic, Praistorija Kingdomdates are pegged to an astronomicalevent at 1537 Vojvodine(Novi Sad 1974) 190. 186 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA 94

Table 2. Cultural Sequences/Chronology of Southeastern Europe

Middle Danube Lower Danube DATE HUNGARIAN VOJVODINA/ MORAVA/N.W. TRANSYL- OLTENIA E. DOBRUDJA B.C. PLAIN N. SERBIA BULGARIA VANIA BULGARIA

2000 Vatin Verbi- 2100 cioara Glina III 2200 Nagyrev Maros /Schneck- 2300 / Bubanj-Hum III enberg Late 2400 Mako Vucedol Ezero 2500 Vucedol/Zok 2600 Bubanj-Hum II 2700 2800 Late Baden Kostolac Middle Ezero 2900 Bubanj Ezero 3000 (EPlat II) 3100 Cotofeni 3200 Classic Classic IIIb/c 3300 Baden Baden Cotofeni Cotofeni 3400 II/IIb 3500 Early 3600 Boleraz Boleraz/ III Ezero? Cernavoda 3700 Cernavoda Cernavoda 3800 Bodrog- III I 3900 keresztur Salcuta IV Bubanj Ib 4000 Cucutefii B Salcuta IV 4100 Tiszapolgar 4200 (Varna) Gumelnita 4300 Salcuta III 4400 Tisza/ Vinca (Late) 4500 Herpaly Bubanj-Hum Cucutefii Salcuta Karanovo Ia/Krivodol (SE Tran) lib, IIc VI

on the eastern edge of the Mediterranean in contexts historically derived dates are not possible, one looks dated to the Egyptian X-XI Dynasty, about 2100 B.C. to physics for help in establishing the absolute dates. Quoting Gimbutas again for the traditional view of The radiocarbon chronology for the later Neolithic the chronology and cultural dynamics: period in southeastern Europe is reasonably well es- tablished.72 Dates for final Vinca levels cluster in the "There is no doubt that the ornaments mentioned just second half of the fifth . . . first were made in the Near East and then were millennium B.C. The dates for the Aeneolithic are but distributedto eastern central Europe, as they are dated succeeding period rarer, at a much earlier period in the Near East, many of the do exist (Table 3). In Oltenia, two dates place Salcuta prototypesreaching the middle of the third millennium lib and IIc, which should correspond to the end of B.C."71 the Bubanj-Hum Ia period in the Morava valley and to between 4425 and 4305 B.C. in calendar This is, as we said, a traditional and now outmoded Serbia,73 years. Cotofeni II/IIb at Baile Herculane has an av- view, stressing the passive role of Europe as opposed erage date of 3370-3040 B.C., while dates from this to the innovating active role of the Near East. site and Ostrovul Corbului give an average date of Southeastern Europe: Aeneolithic to Early Bronze Age 3160-2910 B.C. for Cotofeni IIIb/c or Cotofeni with As in the case of earlier Neolithic chronology, where latest Baden (Kostolac and Vucedol) elements. This

71 Gimbutas (supra n. 67) 34. 73 Garasanin n. 19) 190; Garasanin1983 n. 72 (supra (supra Ehrich and Bankoff (supra n. 2). 21) 165. 1990] THE LATER AENEOLITHIC IN SOUTHEASTERN EUROPE 187 would be equivalent to Bubanj-Hum II in the Morava sequence in this area. Dates for the beginning of the valley sequence.74 Elsewhere Garasanin equated Co- Middle Bronze Age suggest that the transition from tofeni III with Bubanj-Hum Ib.75 Early to Middle Bronze Age in both central and south- Final Baden, contemporaneous with Baden/Kosto- eastern Europe cannot be far from 1850 B.C. These lac and Vucedol,76 elements are therefore found in dates accord well with the high chronology for central the Oltenian assemblage of within 200 years of 3000 Europe.81 B.C. In Hungary, this corresponds well to dates for GENERAL OBSERVATIONS Baden-Pecel with Kostolac elements from Oszentivan, and an Ocher Grave burial above a Bodrogkeresztur The chronology constructed on the basis of the settlement at Ketegyhaza. The same date is suggested radiocarbon dates now available for southeastern Eur- for Baden-Kostolac by a date from Pivnica in northern ope emphasizes the length of the Aeneolithic and the Bosnia, several dates from Hissar in , and for earlier inception of the Early Bronze Age. In other the transition from Baden to Kostolac at Gomolava in words, the radiocarbon chronology expands the du- Srem (eastern ).77 Further afield, a series of 27 ration of the earlier chronological phases, while basi- dates from Ezero78 date the Bulgarian horizons equi- cally leaving the later part of the Bronze Age its valent to Baden and Baden/Kostolac (Bubanj-Hum traditional length. Like the similar results of the ra- II) to between 3150 and 2900 B.C. Dates for Vucedol diocarbon-based absolute chronology of the Neolithic from Hrustovaka and Koprivicka Rijeka in northeast and earlier Aeneolithic periods, this has ramifications Croatia suggest that this period can be set between in terms of the interpretation of the evidence and 2900 and 2700 B.C. cultural dynamics. For example, if the radiocarbon Slag from a Baden context in Trench 40 at Novacka chronology is correct, the "Anatolian" and "Near East- Cuprija suggests that the use of arsenical bronze be- ern" metal types of the European Early Bronze Age gan by the late fourth millennium B.C.79 Although may be as early or earlier in Europe than in the traditionally Baden is considered to be late Aeneo- Levant. Considering the metallurgical history of the lithic, this is another example of the strong continuity southern Carpathians, this is not surprising. Extensive or even unity between the later Aeneolithic and the copper mines at Rudna Glava in East Serbia date to Bronze Age. If one cares about retaining the termi- the Vinca-Plocnik phase of Early Aeneolithic in the nology, it would not be inconsistent to include Baden late fifth millennium B.C.82 Pottery of similar date is as the beginning of the Early Bronze Age in south- also associated with the large mines at Aibunar, near eastern and central Europe. The radiocarbon dates Karanovo in Bulgaria.83 These early centers, as well from Novacka Cuprija suggest that the traditional as others in Romania and East Slovakia, provided the Early Bronze Age in the Morava valley (Bubanj-Hum raw material for the manufacture of the massive cop- III) began by the mid-third millennium B.C. The per axes and axe-adzes of the earlier European dates from Sitagroi IV, Va, and Vb provide confir- Aeneolithic.84 This tradition of copper smelting con- matory evidence.80 New evidence from the Morava tinued in the middle Danube region throughout the valley and from Vinca can be used to define a "proto- Aeneolithic, although the evidence is at present Vatin" phase of the Early Bronze Age in the Vojvo- scanty. The aforementioned copper slag from a dina, which should date to around 2000 B.C. This Baden context at Novacka Cuprija indicates the con- phase marks the end of the late Aeneolithic/EBA tinuity of a crucible smelting technology or technique

82 74 Garasanin (supra n. 19). B. Jovanovic,Metalurgija eneolitskog periodaJugoslavije 75 Garasanin(supra n. 19) 224; Garasanin,Nis 1983 (su- (Belgrade 1971); B. Jovanovic, "Rudarstvoi metalurgija pra n. 21) 167. eneolitskog perioda Jugoslavije,"Praistorija Jugoslavinskih 76 Tasic (supra n. 31); Tasic (supra n. 34); Nemejcova- Zemalja3 (1979) 27-85; B. Jovanovic, Rudna Glava: Naj- Pavukova (supra n. 23). starije rudarstvobakra na centralnomBalkanu (Belgrade 77 H.T. Waterbolk, "C14-Datirungen von Gomolava," in 1982). Tasic and Petrovic (supra n. 54) 119. 83 E.N. Chernykh,Gornoe delo i metallurgijav drevnejsej 78 H. Quitta, "Radiovugerodni dati i tri hronologiceski Bolgarii (Sofia 1978); E.N. Chernykh, "MetallurgicalProv- sistemi," Interdisciplinarni Izsledvanija (Sofia) 1 (1978) 12- inces of the Fifth and Second Millenniumin EasternEurope 24; H. Quitta and G. Kohl, "Neue Radiocarbondaten zum in Relationto the Processof Indo-Europeanization,"Journal Neolithikum und zur fruhen Bronzezeit Siidosteuropas und of Indo-European Studies 8 (1980) 317-36. der Sowjet-union," ZfA 3 (1969) 223-55. 84 A. Sherratt, "Resources, Technology and Trade in 79 P. Glumac, The Advent of Metallurgy in Prehistoric Early European Metallurgy,"in G. Sieveking et al. eds., South-east Europe (Diss. Univ. of California, Berkeley 1990). Problems in Economic and Social Archaeology (London 1976) 80 See Sherratt (supra n. 33) 443. 557-82. 81 See Neustupny (supra n. 62). 188 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA 94

Table 3. Selected Radiocarbon Dates from the Later Aeneolithic/Early Bronze Age of Southeastern Europe

Area Culture Site . no. Date B.P. Calibrated date (5568 hi) (MASCA/Michaels) (1 sigma range)

Oltenia Salcuta IIb Salcuta 2 GrN-1990 5475 ? 55 4425-4320 Salcuta IIc Salcuta 1 GrN-1989 5450 ? 55 4420-4120 Salcuta 1 GrN-1985 5450 ? 55 4420-4120 Hungarian Tiszapolgar Tiszapolgar- Bin-510 5870 ? 100 4960-4565 Plain Csoszhalom Bin-512 5775 ? 100 4745-4535 Bin-509 5575 ? 100 4550-4380 Tiszapolgar- Deb-348 5020 ? 180 3960-3650 Basatanya Deb-348? 5060 ? 170 3960-3755 Deb-122 4850 ? 150 3865-3495 Deb-361 5350 ? 190 4425-3890 Deb-214 4980 ? 140 3905-3650 Averages: Tiszapolgar-Csaszhalom 5740 ? 60 4785-4540 Tiszapolgar-Basatanya 5050 ? 75 3920-3765

Hungarian Bodrogkeresztur Tiszapolgar- Deb-5 4960 ? 130 3830-3645 Plain Basatanya Deb-4 4820 ? 140 3795-3485 Oltenia Cotofeni IIb Baile Herculane LJ-3533 4460 ? 80 3370-2970 Ostrovul LJ-3797 4520 ? 60 3375-3145 Cotofeni III Corbului LJ-3798 4360 ? 50 3170-2910 LJ-3799 4360 ? 60 3170-2910 Baile Herculane LJ-3535 4350 ? 60 3165-2905 LJ-3534 4360 ? 100 3180-2895 LJ-3536 4300 ? 60 3055-2885 Averages: Cotofeni IIb 4490 ? 50 3370-3040 Cotofeni III 4346 ? 30 3160-2910

Pannonia Baden Oszentivan Bln-476 4515 ? 80 3380-3035 Baden Vucedol Z-1446 4540 ? 100 3395-3025 Z-1617 4500 ? 100 3375-3020 Z-1619 4400 ? 100 3360-2910 Z-1618 4300 ? 100 3155-2870 Baden Gomolava GrN-13168 4380 ? 70 3175-2920 Average: Baden 4440 ? 40 3360-2995

which first appeared in the Morava valley at Selevac, senical copper. According to Petar Glumac, the ele- some 1,500 years earlier.85 As at Selevac, this slag mental analysis of the copper metal in the Novacka implies that the raw materials, in the form of , Cuprija slag accords well with that of a Baden axe were transported to be smelted in the river-valley sites found in a house overlain by a tumulus at Tri at some distance from the ore sources. The slag from Humke. The arsenic, , and impurities would Novacka Cuprija, unlike that from Selevac, is of ar- seem to indicate that the Baden people at Novacka

85 p. Glumac,"An ArchaeometallurgicalStudy of the Ma- Evidence for the Use of Lead in Prehistoric South-East terial from Selevac,"Zbornik radova Narodnog Muzeja u Europe,"Archeomaterials 2 (1987) 29-37. Beogradu11 (1983) 135-41; P. Glumac and J. Todd, "New 1990] THE LATER AENEOLITHIC IN SOUTHEASTERN EUROPE 189

Area Culture Site Lab. no. Date B.P. Calibrated date (5568 hi) (MASCA/Michaels) (1 sigma range)

Hungarian Ocher Grave Ketegyhaza Bin-609 4265 ? 80 3045-2790 Plain Bell Beaker Island Bln-1221 4235 ? 100 3025-2775 Q-1122 4170 ? 90 2930-2640 W. Balkans Kostolac Pivnica KN-145 4110 ? 160 2915-2530 Pannonia Kostolac Gomolava GrN-7371 4360 ? 60 3170-2910 GrN-7372 4445 ? 70 3360-2995 GrN-13167 4210 ? 60 2935-2785 Average: Kostolac 4280 ? 40 3040-2880

E. Balkans Ezero average (27 dates) Ezero A 4336 ? 25 3150-2900 Ezero B 4355 ? 25 3160-2910 W. Balkans Vucedol Hrustovaka Bln-564 4125 ? 80 2900-2615 Pannonia Vucedol Z-1637 4300 ? 100 3155-2870 Z-1621 4300 ? 100 3155-2870 Z-1447 4290 ? 120 3150-2865 Z-1453 4290 ? 120 3150-2865 Z-1624 4200 ? 100 2980-2655 Z-1449 4190 ? 120 2950-2650 Z-1454 4130 ? 120 2905-2625 Z-1622 4100 ? 100 2890-2540 Average: Vucedol 4215 ? 35 2935-2785

Macedonia EBA Sitagroi IV/Va 4380 ? 30 3170-2925 average (9 dates) Sitagroi Vb 3950 ? 35 2640-2390 average (6 dates) C. Europe Unetice Prasklice Bln-475 3845 ? 80 2415-2185 (EBA) Vikletice GrN-9378 3760 ? 35 2320-2135 Hungarian Nagyrev Dunaujvaros Bln-340 3735 ? 80 2330-1975 Plain Toszeg GrN-6653 3685 ? 35 2185-1970 Moris nn 3500 ? 50 1960-1725 Morava Bubanj-Hum Novacka Beta-14789 3910 ? 100 2635-2300 Valley III Cuprija BC-84 3590 ? 100 2160-1850 Beta-2574 3300 ? 90 1745-1520 Crkvina BC-68 3990 ? 60 2650-2520

Cuprija were smelting sulfide ores. The utilization mid-third millennium B.C. of the sulfide ores of central Europe begins The assemblages of the central European Early with the inception of the Bronze Age, now dated Bronze Age are dominated by several spectrograph- by the radiocarbon chronology to roughly the ically distinct metal groups,86 among which the so-

86 S. Junghans, E. Sangmeister,and M. Schroder,Metal- N. Schroder, Kupfer und Bronze in der friihen Metalzeit lanalysenkupferzeitlicher undfriihbronzezeitlicher Bodenfunde Europas(Berlin 1968). aus Europa(Berlin 1960); S. Junghans, E. Sangmeister,and 190 H. ARTHUR BANKOFF AND FREDERICKA. WINTER [AJA 94

called "Osenhalsring" metal,87 possibly from a European scene. The question of whether we can see Slovakian88or an Alpine source,89is known from any regular relationships between language change, hoards containing large quantities of these ring-in- population change, and material culture change has gots. Examples found in the Levant might indicate generally been ignored.95 If this later Aeneolithic pe- that both semi-finished raw materialsand European riod is as long as we have proposed, full of socioeco- finished products were finding their way in small nomic change, inter- and intra-group competition, quantities into the international market. Their con- and regional and inter-regional contact, then there is texts in the Levant could then corroboratethe radio- ample reason to suppose that language change may carbon dates by providing a terminus ante quem for have played an integral part in these processes. Given their occurrence in Europe. the long time over which the transformations of the The later part of the Aeneolithic in the Morava Aeneolithic took place, language shift is hardly an Valley, as in the rest of southeastern Europe, gives unreasonable expectation. We should note in this con- ample evidence of changes in materialculture, settle- text that Renfrew96 has recently proposed that the ment pattern, and perhaps subsistence. We have al- first Indo-Europeans, whose homeland he places in ready noted the new ceramic inventory and the eastern Anatolia, arrive in Europe by 6500 B.C. We disappearance of the earlier Neolithic tradition of have deliberately avoided mention of the whole Indo- painted wares. The large nucleated settlements also European problem, which appears to us (in its classic disappear. Most of the long-established tell settle- form) to be quite possibly insoluble. ments of the eastern Balkansare abandoned as well.90 It is, however, an example of a wider, more basic The settlements of the later Aeneolithic are usually problem. One of the most basic assumptions under- characterized as short-lived, shallow, disturbed, or lying archaeological work in southeastern Europe is ephemeral. It is not until well into the Early Bronze that the perceived rate of cultural or assemblage Age that larger apparently stable tell-like sites are change is of great importance, in most cases deter- again found on the Hungarian Plain and in Bulgaria. mining the way that change is explained. If the change The appearance of the horse, animal traction,plows, is perceived as slow and incremental, it is explained and carts, all certainly associatedwith changes in ag- as local autochthonous development; if perceived as ricultural technology and probably with socio-eco- sudden, it betokens the intrusion of a new group nomic changes,91 have been dated to this later (usually thought of as a new ).97 These Aeneolithic period.92As noted above in connection implicit assumptions about the relationship of rates with the new pottery types, Sherratthas claimed93that and causes of cultural change help to explain the raising cattle for milk and sheep for wool is also to be continuing primacy of chronological investigations in traced to this time. southeastern European archaeology. Traditionally,these changes and others including According to the chronology presented here, ap- inhumation under tumuli, pit graves, and corded proximately two millennia exist for the transforma- pottery, have been ascribed to invasions of nomadic tion of the Neolithic village-farming socioeconomic pastoralistsfrom the Russian steppes.94These hypo- pattern into a uniquely European temperate farm- thetical nomads assume more importance to many as stead pattern, with the concomitant changes in society the first Indo-European speakers to burst onto the and subsistence. Such changes involved not only the

87 H.T. Waterbolkand J.JButler, "Comments on the Use 93 A. Sherratt, "The Secondary Exploitationof Animals of MetallurgicalAnalysis in PrehistoricStudies," Helinium 5 in the Old World,"World Archaeology 15 (1983) 90-104. (1965) 227-51. 94 See Gimbutas(supra n. 67); M. Gimbutas,"The First 88 B. "K Bath-Bilkova, problemu puvodu hriver-Zur Wave of EurasianSteppe Pastoralistsinto Copper Age Eu- Herkunftsfrageder Halsringbarren,"Pamatky Archeologicke rope,"Journal of Indo-EuropeanStudies 5 (1977) 277; Tasic 64 (1973) 24-41. (supra n. 35); H. Thomas, "ArchaeologicalEvidence for the 89 H. Neuninger and R. Pittioni, "Fruhmetallzeitlicher Migrationsof the Indo-Europeans,"in E. Polome ed., Lin- Kupferhandelim Voralpenland,"ArchAustr Beiheft 6 (1963) guisticaExtranea 14 (Ann Arbor 1982) 61-86; N. Tasic, "Das 1-39. Problem der sukzessivenMigrationen wahrend des Aeneo- 90 R. in Dennell, Early Farming South Bulgariafrom the lithikums in Karpaten-Donautal-Balkan-Gebiet,"ArchJug VI to theIII Millennia B.C. (BAR-ISsuppl. 45, 1978). 22-23 (1982-1983) 15-20. 91 95 Championet al. (supran. 1) 156-62; D. Anthony,"The D. Anthony and B. Wailes, rev. of Renfrew (infra n. 'KurganCulture,' Indo-European Origins and the Domes- 96) CurrentAnthropology 29 (1988) 441-45. tication of the Horse: A Reconsideration,"Current Anthro- 96 A.C. Renfrew,Archaeology and Language: The Puzzle pology27 (1986) 291-313. of Origins(London 92 Indo-European 1987). Sherratt (supra n. 1). 97 Evansand Rasson(supra n. 2) 718. 1990] THE LATER AENEOLITHIC IN SOUTHEASTERN EUROPE 191 graduallyfelt effects of continued agriculturalexpan- the problem may go back to our traditionalarchaeo- sion, combined with the probable synergistic effects logical paradigm of the Three Age System, in which of increased animal husbandryand grazing, local fac- the Aeneolithic,which in fact consisted of 2,000 years tors such as increasing soil salinityon the Hungarian of change, has been too often perceived as a fleeting Plain,98and transport technology perhaps derived transitionalphase, a point in time, rather than a long from further east,99but also possibly the increased period in which technologicaland social changes that viabilityof the household as the basic unit of produc- were long in coming arrived at a point at which they tion.100Many of these changes may be related to the were archaeologicallyvisible. intensifying effects of sedentism on economic and social relations.'10As Anthony and Wailes put it,102 without declaring later prehistoric Europe "a migra- DEPARTMENT OF CLASSICS tion-free zone on theoretical grounds,"it seems to us BROOKLYN COLLEGE OF THE that the need to invoke migration to explain the CITY UNIVERSITY OF NEW YORK changes is a facet of our perceptual difficulty.Part of BROOKLYN, NEW YORK 11210

98 A. Sherratt,"Mobile Resources: Settlement and Ex- 100 Tringham and Krstic(supra n. 6). changein EarlyAgricultural Europe," in A.C.Renfrew and '10 T. Kaiserand B. Voytek,"Sedentism and Economic S.J. Shennan eds., Ranking,Resources, and Exchange(Cam- Change in the BalkanNeolithic,"Journal of Anthropological bridge1982) 13-26. Archaeology2 (1983) 323-53. 99Anthony (supra n. 91). 102 See Anthony and Wailes(supra n. 95) 444.