by Chris Fowler
Thorpe, I.J.N. 2009. Becoming Neolithic in southern Britain. In Glørstad, H. and Prescott, C. (eds), Neolithisation as if history mattered, 23–63. Lindome: Bricoleur Press.
Tilley, C. 1996. An ethnography of the Neolithic. Cambridge University Press.
Tresset, A. and Vigne, J-D. 2007. Substitution of species, techniques and symbols at the Mesolithic–Neolithic transition in Western Europe. In A. Whittle and V. Cummings (eds), Going over: The Mesolithic–Neolithic transition in north-west Europe, 189–210. Proceedings of the British Academy 144. Oxford: Oxford University Press.
van der Schriek, T., Passmore, D.G., Stevenson, A.C., and Rolão, J. 2007. The palaeogeography of Mesolithic settlement-subsistence and shell midden formation in the Muge valley, Lower Tagus Basin, Portugal. Holocene 17, 369–385.
Vanmortfort, B. 2008. A southern view on north–south interaction during the Mesolithic– Neolithic transition in the Lower Rhine area. Analecta Praehistorica Leidensia 40, 85–97.
Visset L., L’Helgouac’h J., and Bernard J. 1996. La Tourbière submergée de la pointe de Kerpenhir à Locmariaquer (Morbihan), étude environnementale et mise en evidence de déforestations et de pratiques agricoles néolithiques. Revue archéologique de l’Ouest 13, 79–87.
Whittle, A.W.R. 2000. ‘Very like a whale’: menhirs, motifs and myths in the Mesolithic–Neolithic transition of northwest Europe. Cambridge Archaeological Journal 10, 243–359.
Whittle, A.W.R. 2007. The temporality of transformation: dating the early development of the southern British Neolithic. In A. Whittle and V. Cummings (eds), Going over: the Mesolithic–Neolithic transition in north-west Europe, 377–398. Proceedings of the British Academy 144. Oxford: Oxford University Press.
Whittle, A.W.R., Barclay, A., Bayliss, A., McFadyen, L., Schulting, R., and Wysocki, M. 2007. Building for the dead: events, processes and changing worldviews from the thirty-eighth to the thirty-fourth centuries cal. BC in southern Britain. Cambridge Archaeological Journal 17, 123–147.
Whittle, A.W.R., Bayliss, A., and Healy, F. 2011. Gathering time: dating the Early Neolithic enclosures of southern Britain and Ireland. Oxford: Oxbow.
Woodman, P. and McCarthy, M. 2003. Contemplating some awful(ly interesting) vistas: importing cattle and red deer into Prehistoric Ireland. In I. Armit, E. Murphy, E. Nelis, and D. Simpson (eds), Neolithic settlement in Ireland and western Britain, 31–39. Oxford: Oxbow.
Zvelebil, M. and Rowley-Conwy, P. 1986. Transition to farming in northern Europe: a hunter-gatherer perspective. In M. Zvelebil (ed.), Hunters in transition, 67–93. Cambridge University Press.
PART III
NEOLITHIC WORLDS AND NEOLITHIC LIFEWAYS
Houses, Habitation, and Community
CHAPTER 12
SETTLEMENTS IN SOUTH-EAST EUROPE
PÁL RACZKY
INTRODUCTION: TELL AND NON-TELL SETTLEMENTS
NEOLITHIC open-air settlements in south-east Europe fall into two groups. One consists of a single, usually horizontal layer, the other comprises superimposed remains from several habitation layers. Whilst horizontal settlements occur throughout Europe, the distribution of ‘settlement mounds’ rising above the natural surface of the landscape is limited to the Near East, the Balkans, and the south-eastern part of the Carpathian Basin (Childe 1950, 38–39; Wace and Thompson 1912; Gimbutas 1974, 19–25, 29–33; Kalicz and Raczky 1987, 14–19; Chapman 1989, 1997a, 158–162; Raczky 1995; Whittle 1996, chapters 3–5; Bailey 2000, 156–177; Steadman 2000; Gogâltan 2003; Link 2006, 7–14; Rosenstock 2005, 2006, 2009, Raczky and Anders 2008, 35–37; Anders et al. 2010).
The northernmost Neolithic settlement mound is the tell of Polgár-Csőszhalom (Hungary), located by the northern reaches of the Tisza river, very close to the famous obsidian sources near Tokaj (see references in Raczky and Anders 2008). Artificial settlement mounds in Europe are between 2.5 and 10m high, their counterparts in south-west Asia 5–50m. Tells may be of conical or flattish shape, with horizontal extents varying between 0.1 to 10ha, but reaching up to 20ha in the Near East (Chapman 1989, 36–38, fig. 2; Rosenstock 2005, 222–224, fig. 1a–b; Menze et al. 2006, 322, 325, fig. 10–11).
Depending on local languages, these often attractive landscape features are called ‘tell’, ‘hüyük’, ‘tepe’, ‘magoula’, ‘tumba’, ‘mogila’, ‘mǎgura’, ‘település-halom–lakódomb’, or ‘Siedlungshügel–Wohnhügel’ (Chapman 1997a; Rosenstock 2005, 2006). The term ‘tell’ (mound) was first used in a European context by Ferenc Tompa (1937, 47) for settlements of the Tisza culture on the Great Hungarian Plain and has since become generally accepted in the archaeological literature (Gogâltan 2003, 222–223). In south-east Europe, tell-like settlements may be at least 1 to 2.5m thick, and possess at least two superposed habitation layers. Single-layer, horizontal settlements are usually characterized by a deposit only 25–50cm thick, although some have fills up to 1m thick (Kalicz and Raczky 1987, 14–16; Link 2006, 10–14; Gogâltan 2003, 223–224).
Depending on local research traditions, different forms of phasing and terminology (Neolithic, Eneolithic, etc.) are used for the first tell-building cultures in south-east Europe. In absolute terms, the beginnings of tell sites date to around 6700/6500 BC, and their end to approximately 4000/3700 BC. Tells thus existed over c. 2500–3000 years during a period when food-producing economies emerged over a wide area between Greece and Hungary.
THE PHYSICAL AND SOCIAL CONDITIONS OF THE FIRST TELLS
Tell settlements mainly consist of stratified debris from clay houses, constructed using various techniques (pisé, mud brick, wattle-and-daub, etc.; Aurenche 1981, 42–72; Naumann 1971, 43–51; Stevanović 1997, 341–345; Rosenstock 2005, 228–233; Piesbergen 2007, 20–32). These structures frequently burnt down, but were systematically reconstructed on the same spot. Construction deposits were thus created by intentional levelling, alongside the daily accumulation of refuse. Settlement mounds hence represent long-term, planned activity. In addition to natural erosion, there is also evidence of conscious landscaping through the systematic removal of rubble and the remains of earlier houses. Physically, the resulting tell ‘body’ was created through a complex creative process including broad-based communal effort and resulting in a regionally significant topographic monument. The mound itself may hence be seen as the material manifestation of a community, its coordinated activity and communal life (Evans 2005). The apparently undisturbed 2,500 years of tell development in south-east Europe suggest long-term economic and social stability, which partly inspired Gimbutas’ (1974, 17–19) idea of a collective identity labelled the ‘Civilization of Old Europe’, presuming the evolution of an urban-type system of institutions.
Evidently, the systematic activity of tell creation had a feedback effect on those who built these sites and, over the long run, contributed to the cohesion of communities and their inseparable, complex system of economic, social, sacral, and symbolic norms (Chapman 1997a; Kotsakis 1999; Bailey 1999; Tringham 2000b; Evans 2005; Gheorghiu 2008).
EARLY RESEARCH HISTORY OF TELLS IN EUROPE
Tells in Europe have long fuelled the imagination of modern villagers and treasure hunters, whilst their sequence of superimposed strata offered relative chronologies for artefactual assemblages. Unsurprisingly, the first more-or-less scholarly excavations in Europe also targeted these mounds. One tell of key importance was investigated by Vasić (1932–1936) near Vinča in Serbia between 1908 and 1934. He found periodically renewed adobe houses built on wooden frames, plastered open-air fireplaces, and refuse from Neolithic daily life, down to a depth of 9.5m.
The finds from its superimposed layers made the Vinča tell a yardstick for the Balkan Neolithic and, to some extent, Copper Age cultural development. Its phases have been alternatively labelled Vinča I (Tordos) and Vinča II (Pločnik) or A to D (Chapman 1981; Schier 1997). Gordon Childe compared this site to the tell of Troy in western Anatolia, and—on the basis of its finds—hypothesized a cultural/chronological connection between the two sett
lements at the beginning of the third millennium BC (Childe 1929, 34–35; Renfrew 1976, 42–52). This was one of the cornerstones of his historical model of ‘ex oriente lux—from the east the light’, whereby ethnic groups originating in the Near East crossed the Aegean and penetrated the Vardar and Morava river valleys before reaching the Danube region (Childe 1939, 1950, 36–57). In the German literature, Fritz Schachermeyr (1953) popularized the same idea as ‘vorderasiatische Kulturtrift—Near Eastern culture flow’.
During the first half of the twentieth century, researchers linked south-east European mounds to this population movement, which supposedly took place in several waves and resulted in the colonization of Thrace, Macedonia, and the Lower Danube region (Gaul 1948, 49–79; Childe 1950, 41–42, 51–53).
TELLS IN SOUTH-EAST EUROPE AND THEIR LOCAL STRATIGRAPHIC SEQUENCES
Tell research followed the development of different national archaeologies. In Greece, Sesklo has long served as a reference point for studies of settlement structure and relative chronology (Tsountas 1908; Theocharis 1973, 68; Kotsakis 2006). Excavations during the 1950s revealed stratigraphic sequences at the magula of Argissa, Otzaki, Arapi, Agia Sofia, and Pevkakia, which together created a coherent diachronic system for refining regional Neolithic chronologies (Milojčić 1960). In Bulgaria, chronological phases I through VI at Karanovo have framed the standard chronology for the Neolithic and Copper Age (Georgiev 1961; Vajsová 1966, 5–8; Todorova 1981; Hiller and Nikolov 2000), whilst stratigraphies at Vidra, Gumelniţa, Sǎlcuţa, and Hârsova—to name but a few—became the chronological standard in Romania (Berciu 1961, 82–86, 158–166; Comşa 1974, 32–33; Mantu 2000; Gogâltan 2003). Tells, especially those at Hódmezővásárhely-Gorzsa, Öcsöd-Kováshalom, and Berettyóújfalu-Herpály, also helped establish relative chronologies for the Great Hungarian Plain (see references in Tálas and Raczky 1987).
These examples illustrate how tell stratigraphies in south-east Europe became almost exclusive yardsticks for reconstructing Neolithic development. Syntheses and interregional comparisons were attempted using parallel phenomena and the presence of ‘import–export’ artefacts (Treuil 1983, 13–114). Consequently, the south-east European Neolithic was summarized in unified chronological tables presenting a lato sensu cultural system in a clearly visualized format (e.g. Ehrich 1992; Parzinger 1993, Beilage 1–5; Todorova 1998, tables 1–3; Bailey 2000, fig. 1.3).
With the radiocarbon revolution, settlements along the interface between the Aegean and the Balkans gained pivotal significance. By the 1970s, the stratigraphic sequence at the Sitagroi tell in east Macedonia had shown that local early Bronze Age type finds similar to material from Troy were deposited in layers above strata of the Vinča-Gumelniţa culture in the Balkans (Renfrew 1970, 295–308): this provided evidence that the Vinča-Gumelniţa cultural complex was older than that from Troy, lending credibility to previously contested radiocarbon dates which, contradicting Childe’s ideas, had placed the beginning of the Vinča tell 2,300 years before the emergence of urban development at Troy at around 3000 BC. Evidently, the Vinča settlement was not established by emigrants from Troy, who hence did not colonize the Balkans (Renfrew 1976, 101–109).
THE REPRESENTATIVE VALUES OF TELLS AND HORIZONTAL/EXTENDED SETTLEMENTS
Increasingly, then, tells became a primary source of information for settlement history. Indeed, compared with tells, horizontal/extended settlements and cave sites (similarly numerous in the region) were often underrepresented or neglected in large thematic summaries. Tell distributions in certain regions often meant that an entire culture was considered a ‘mound culture’ (cf. the Bulgarian Mound Culture; Gaul 1948, 79–207).
Until the 1990s, this research bias hindered the development of a balanced view of tells and horizontal settlements across south-east Europe. Yet, many tells were connected to adjacent horizontal settlements forming an organic unit with the mound (Bailey 1999, 2000, 174–177). At Sesklo, for instance, Theocharis reconstructed a large (almost 10ha) horizontal settlement around the tell, fortified by a stone wall. He recognized that the tell and its external horizontal settlement probably represented a complex, acropolis–polis settlement structure (Theocharis 1973, 68, fig. 178) in which the acropolis had a special function relative to the outer settlement, the scene of daily life (Kotsakis 1999, 69; 2006, 209–218). This showed that previous tell-centred settlement histories for south-east Europe had led to the generalization of special phenomena erroneously seen as ‘representative’ on a broad scale (e.g. Todorova 1982).
Following these developments, and fostered by planned excavations, large and small horizontal settlements were observed in association with tells from various Neolithic and Copper Age periods. At Podgoritsa in northern Bulgaria, the association between the tell and external ‘non-tell’ features could be demonstrated as a general south-east European phenomenon (Bailey 1999, 2000, 175, fig. 5.8). Recently, geophysical surveys and large-scale excavations have revealed numerous examples showing this ‘symbiosis’ between tell and horizontal/extended settlements, including Paliambela in Greece (Kontogiorgos 2010), Pietrele and Uivar in Romania (Hansen et al. 2006, 4–8, abb. 5–7; Schier 2009, 222–224), Okolişte in Bosnia (Müller et al. 2011), and Öcsöd-Kováshalom, Berettyóújfalu-Herpály, Polgár-Bosnyákdomb and Polgár-Csőszhalom in Hungary (Raczky and Anders 2008, 2010).
Actually, Chapman had already outlined the importance of external spaces for early food production. Tells were densely covered by houses, and within their limited spaces it would have been impossible to cultivate plants and keep animals to feed the population. Consequently, most subsistence activities must have taken place in the wider environment (Chapman 1989, 34–39). Moreover, external spaces had to be shared following principles of the structured communal economy valid inside the tell to permit sustainable, long-term sedentary agriculture. According to Hodder (1990, 83–87), the house-centric world of tells corresponds to the domestic, domus, surrounded by the wild (agrios). These two spatial spheres represented a complementary dualistic relationship throughout the Neolithic.
There is a diversity of physical relationships between tells and contemporaneous horizontal settlements across Neolithic south-east Europe (Chapman 1981, 1997a; Kotsakis 1999, 2006; Bailey 2000, 174–177; Halstead 2005), with stratified mounds forming an increasingly complex settlement structure impacting on horizontal sites adjacent to the tell and beyond (Chapman 1998, 113–118; 2010; Halstead 1999; Raczky and Anders 2008, 2010). Along the edge of the Balkan tell distribution area, Makkay (1982, 104–164) and Sherratt (1982) identified such a complex system, consisting of a tell and its numerous small, horizontal satellite settlements, in the Tisza culture of the southern Great Hungarian Plain (see also Parkinson 2006, 139–143). In the northern Great Hungarian Plain, only horizontal Tisza culture settlements are known, illustrating how dualistic settlement characteristics vary between geographical zones within this culture (Kalicz and Raczky 1987, 14–19; Makkay 1991; Raczky 1995).
CULTURAL PATTERNS C. 6700/6500–5500 BC
Between 6700/6500 and 6000 BC, the first long-term settlements in the south-east European Neolithic were built in Greece, Macedonia, Kosovo, and Bulgaria. They mirrored contemporary sites in Anatolia, including tells (Whittle 1996, 37–59; Chapman 1997a; Tringham 2000a, 19–26; Bailey 2000, 39–55; Rosenstock 2005, 225–233; 2009, 102–106; Perlès 2005; Guilaine 2007). The best-known early tells are Argissa Magoula, Otzaki Magoula, Prodromos, Achilleion, Anzabegovo, Vršnik, Veluska Tumba, Tumba Madjari, Rudnik, Karanovo, Tell Azmak, Čavdar, Rakitovo, Slatino, and Kovačevo (Fig. 12.1). Tells cluster in the alluvial areas of major river valleys, and avoid less favourable environments in between. For example, several tells are known from the Larissa Basin in eastern Thessaly and along the Maritsa and Tundja rivers and their tributaries in Bulgaria (van Andel and Runnels 1995; Perlès 2001, 125–131; Nikolov 2002, abb. 2). Horizontal settlements of various levels of integration (from household through hamlet to village, in dynamic interaction) can oc
cur near these tells (Chapman 2008). Neolithic villages appeared with an explosive intensity in the Carpathian Basin around 6000 BC, possibly through the endemic diffusion of the Near Eastern ‘Neolithic package’ along the great river valleys (Tringham 2000a, 19–33, fig. 2.3; Biagi et al. 2005; Davison et al. 2006; Bocquet-Appel et al. 2009). However, there were initially no tells north of the central Balkans, where a more mobile way of life was served by less permanent, horizontal settlements. Houses built at a distance from each other enabled horticultural, small-scale household-level cultivation in the immediate proximity of buildings and permitted animal keeping nearby.
FIG. 12.1. The geographical distribution of Neolithic tell settlements in south-east Europe between 6700/6500 and 4600/4500 BC. Solid line: first phase of tell distribution, 6000 BC; dashed line: second phase of tell distribution, 5500 BC; dashed and dotted line: third phase of tell distribution, 5100/5000 BC; dashed and double dotted line: fourth phase of tell distribution, 4600/4500 BC.
Selected list of stratified (tell and tell-like) settlements and horizontal sites in south-east Europe. 1 Achilleion, 2 Anza, 3 Argissa, 4 Ariuşd, 5 Bapska, 6 Berettyóújfalu-Herpály, 7 Bernadea, 8 Bolgrad, 9 Cašciorale, 10 Čavdar, 11 Čoka, 12 Crnokalačka Bara, 13 Dimini, 14 Dolnoslav, 15 Drama, 16 Durankulak, 17 Elateia, 18 Ezero, 19 Fajsz-Kovácsdomb, 20 Gǎlǎbnik, 21 Goljano Delčevo, 22 Gomolava, 23 Gornja Tuzla, 24 Gumelniţa, 25 Hârşova, 26 Hódmezővásárhely-Gorzsa, 27 Hódmezővásárhely-Kökénydomb, 28 Karanovo, 29 Korintosz, 30 Kovačevo, 31 Kremikovci, 32 LepenskiVir, 33 Lerna, 34 Nea Makri, 35 Nea Nikomedea, 36 Obre I, 37 Okolište, 38 Otok, 39 Otzaki, 40 Ovčarovo, 41 Öcsöd-Kováshalom, 42 Padina, 43 Parţa, 44 Pavlovac, 45 Pepelane, 46 Pietrele, 47 Podgorica, 48 Poduri, 49 Polgár-Csőszhalom, 50 Poljanica, 51 Porodin, 52 Prodromos, 53 Rakitovo, 54 Rast, 55 Ruse, 56 Sava, 57 Sǎlcuţa, 58 Servia, 59 Sesklo, 60 Sitagroi, 61 Slatina, 62 Stara Zagora, 63 Suceveni, 64 Szegvár-Tűzköves, 65 Tǎrtǎria, 66 Teliš, 67 Tell Azmak, 68 Tumba Madjari, 69 Uivar, 70 Vadaštra, 71 Varna, 72 Varoš, 73 Vésztő-Mágor, 74 Vidra, 75 Vinča, 76 Vinica, 77 Vlasac, 78 Vršnik.