by Chris Fowler
In the Aegean at no later than 6500 BC some elements of the Neolithic appear (Perlès 2001, 94). Domesticated wheat and sheep are known from Knossos on Crete and perhaps some sites in Thessaly and the Argolis, whilst a few contemporary sites still indicate a Mesolithic way of life. Obviously, networks linked small fishing and foraging communities to the farmers of Anatolia or the east Mediterranean. The distribution of Melos obsidian indicates such a network between Anatolia and the Aegean (Fig. 3.1). In spite of these new plants and animals, fundamental changes in economic and social life did not occur. Rather, they represent the integration of new customs, items, and subsistence strategies into traditional daily routines and practices via the wider interaction networks of coastal societies.
FIG. 3.1. Early Neolithic networks in south-eastern Europe: obsidian from Melos and Tokai alongside major clusters of Neolithic mainland settlement before 6100 BC
(drawing by Holger Dieterich, Kiel).
A hundred years later this had changed (Reingruber 2008). Thrace and the Marmara region produce settlements where fishing and farming were the main elements of the subsistence economy, and throughout Thessaly, Macedonia, and the Marmara region the entire package of Neolithic innovation appears (Perlès 2001, 121). In small patches of the landscape, both fertile and near the sea, farming communities established a new way of life, in some cases through the acculturation of foragers (e.g. on the islands), in other cases through the immigration of whole farming communities. This change of lifestyle was associated with new occupation patterns across landscapes, best illustrated by the occupation and inland colonization of Thessaly by Neolithic groups. The ecological conditions of the fertile basin favoured small-scale agriculture. Neolithic occupation started around 6300 BC with a clear spatial (political) division of the plain by various communities, which established permanent settlements. As a result, settlement mounds (tells) were established, the land division supported equal numbers of inhabitants at each site, and the population density rose to about 30 persons per square kilometre during the early Neolithic. Without further analyses, the characterization of these Thessalian processes is difficult. Nevertheless, two arguments speak against clear links to the local Mesolithic. First, there are no typological and economic similarities. Second, the absence of late Mesolithic remains points to a sparse foraging population before the arrival of Neolithic communities.
The ‘newcomers’ (but where exactly did they come from?) introduced a new economy, a new social order with new political institutions, and a new lifestyle. Small rectangular houses of mud-brick, brick clay, or timber-and-daub were home to single households and these were agglomerated in small villages of up to 100 inhabitants (Müller 2007), often forming impressive tells through their continuous use. This repeated occupation of one place for domestic purposes shows that settlements had become places of memory and tradition. Domesticated sheep, goat, cattle, and emmer and einkorn wheat were herded or planted in increasingly open landscapes. Different cooking and storing techniques are indicated by the introduction of both coarse pottery and nicely decorated fine pottery. The household mode of production is bound into village life and communal endeavours like field systems and the herding of the flocks. The deposition and recycling of waste demands the spatial organization of activity areas (at least within the settlements). Social organization is more complicated, perhaps even stratified. Some individuals were buried within the settlements and the majority of burials are without grave goods. House altars and figurines signal a well-established sacred sphere (Chapman and Gaydarska 2007; Hansen 2007). While from an archaeological viewpoint these societies are not linked with foraging precursors, connections to the Anatolian Neolithic exist in material culture and social organization. Similarities in ornamentation, in the subsistence economy, and of arranging villages and local environs back the idea of Anatolian people moving in. In this sense, the introduction of the ceramic Neolithic into south-east Europe has to be seen as an innovation, founded on the movement of small groups of mobile individuals into new environs. This seems to be true in spite of the lack of aDNA evidence.
Instead of a ‘Neolithic wave’ which reached Europe and ‘flooded’ south-east Europe, further Neolithization took place in stages (e.g. Banffy 2004). It was detained in northern Macedonia until 6100 BC for reasons as yet unknown. Interaction, communication, and mobility took place within the Aegean-Anatolian sphere, but not further north. Yet from around 6100 BC it took only a few generations for the Neolithization of most of the Balkans, the Carpathian Basin, and other regions of south-east Europe (Biagi et al. 2005; Bocquet-Appel et al. 2009).
The sudden establishment of early Starčevo-Criş-Körös communities saw the introduction of the Neolithic way of life in fertile core areas. Whereas material culture still shows no links to foraging groups, the variation of sites and cultural environs gets more diverse. In some areas the creation of tells is observed, in others not. Short-lived domestic sites with shifting households shape more and more of the domestic sphere. The composition of lifestock gets more varied—beside sheep and goat, cattle husbandry becomes more important. The variation of cereals also increases. Other aspects of material culture demonstrate dissimilarities. Ceramic design still followed shapes and decoration patterns already visible in the Aegean-Anatolian area, but regional diversity was increasing (e.g. Schubert 1999). Within generations, regional styles had developed which might be seen as proxies of local and regional communication and interaction spheres.
The transformation of the south-east European landscape during the Neolithic is clearly linked to questions about the degree of mobility within and between existing networks and the creation of new networks. The development of diverse Starčevo-groups with a similar record of the regional material culture as from c. 5800 BC underlines the adaptation of local environs and the forming of local and regional communication spheres. Cross-cultural links are difficult to detect. Apart from the introduction of houses and objects similar to those in the south, independent developments are obvious. This is linked to inner colonization and the establishment of new groups. Both depend very much on the demographic development of societies, thanks to successful adaptations of local and regional environs as well as successful social organization. The development of late Neolithic societies in south-east Europe is a clear continuation of these processes.
The flow of items and resources would have been necessary for production and reproduction, and indicates the interaction spheres of early and late Neolithic communities in south-east Europe. For instance, the early and late Neolithic communities of the Carpathian Basin were dependent on raw materials which were available in the Bükk mountain range or the Bihar mountains, sometimes more than 100km away (Raczky et al. 1996; Kaczanowska and Kozłowski 1994). Local communities probably arranged expeditions, such as to the Tokay obsidian sources, where material was collected for use elsewhere. Similarly, flint material was only available in remote areas, but was nonetheless transported to the main settlements. Thus, the necessities of the subsistence economy and of tool production were responsible for some kind of mobility to areas surrounding the Carpathian Basin. In some cases, hundreds of kilometres had to be overcome to gain such materials, furthering contact with other groups and other societies.
The long-distance exploitation of raw materials may have been complemented by another form of mobility on a comparable scale—long-distance transhumance. Whilst there are many methodological difficulties with identifying such transhumant activities, early Neolithic Adriatic Impresso societies are a good example.
Impresso—an example of transhumance
The Adriatic Basin is mainly formed by the Dalmatian eastern Adriatic coast below the Dinaric limestone mountains and by the Italian coastline below the Apennines. Early Neolithic communities with a distinct type of impressed pottery are found here in both Dalmatia and Apulia. These transadriatic societies were integrated into a network of interaction: they were connected to local late Mesolithic foragers, including those i
n the Dinaric Alps, but also to early Neolithic groups, namely Starčevo-Criş-Körös communities (e.g. Müller 1994, 205–227; Spataro 2002), in the central Balkans. Yet the Adriatic Impresso is a purely Mediterranean phenomenon, and as such contrasts with continental development. The infiltration of Impresso pottery patterns into the Balkans via the Neretva route highlights the differences between the two: it resulted in Impresso-decorative patterns on pots with Starčevo organic tempering, hence reflecting the integration of Adriatic elements into an otherwise purely Starčevo production sphere.
There are more than 50 sites with Impresso pottery, ranging from the Trieste Karst at the border of northern Italy, to the Albanian river Mat in the south. Whilst open settlements cluster in fertile terra rossa plains distant from the sea, cave sites are distributed at altitudes of 1,000m in the Dinaric mountain range. The mountain range offered the potential for gathering or hunting, but also for herding in plateau areas. In contrast, the coastal zone displays Flysch valleys between limestone ridges, covered by regosols or mineralised terra rossa areas. Near rivers and the rare karstic springs, it was of great potential for farming, including cereal cultivation. The coastal islands possessed huge marine resources (Müller 1994, 50–71).
There is some evidence for the differing function of Impresso sites (Müller 1994, 50–71). Early Neolithic open settlements are concentrated in agricultural core areas like Istria and central Dalmatia, and more specifically on the border of the terra rossa plains near a water source. By contrast, most cave sites are distributed on karstic soils in the high mountain range, but interestingly near deeper valleys, which in some cases possess the potential for agricultural activities. Bone analysis confirms the distinction between open settlements and what may be upland herding camps. According to different authors (e.g. Mlekuz 2005; Schwartz 1988), domesticated sheep and cattle, along with a respectable percentage of domesticated pig bones, were found on the open site of Smilčič, whereas at the cave sites of Gospodška pecina and Odmut domesticated pig bones are nearly absent, suggesting their use as herding places (in the case of the huge dominance of domesticated animals) or even hunting camps (in the case of the dominance of wild animals) by ‘lowlanders’.
Thus, the subsistence economy of Impresso communities adapted to the potential of the eastern Adriatic environs: there are permanent agricultural settlements in coastal areas and base camps in the higher Dinaric range for transhumant herding purposes (and perhaps fishing camps on peninsulas and the islands, too). As the transhumant sphere of Impresso groups overlaps with the site catchments of Castelnovian Mesolithic groups, signs of social interaction between them are hardly surprising (Fig. 3.2): there is goat (Odmut), cattle (Crvena stijena), and pottery from Mesolithic sites, albeit with no matching Mesolithic elements on Impresso sites. There may have been a special kind of interaction between foragers and farmers. If we look closely at their distribution, late Mesolithic sites are absent from the core areas of Impresso open settlements, and only found in regions with Impresso ‘base camps’ for hunting and herding. Communication between communities with two totally different economic systems most probably only took place in certain areas.
FIG. 3.2. Transhumance and communication: the Dinaric example of early Neolithic farmers and Mesolithic foragers. Boundaries of agricultural core areas are marked by the dotted line and transhumant activities by the broken line.
(drawing by Holger Dieterich, Kiel).
The inter-relation between the continental Starčevo and the Adriatic Impresso sites is less important. There are significant differences in the temper material and decorative patterns of the two ceramic traditions and figurines do not occur in Dalmatia (Müller 1994, 212). Only the communication route along the Neretva produced a special situation in Bosnia, with Odmut and Obre both yielding Impresso and Starčevo material. Here at least, the ideological differences between central Balkan communities and the Impresso groups—best demonstrated by the latter’s failure to use Starčevo ritual artefacts—were perhaps overcome by common economic interest.
MOBILITY AND DEMOGRAPHY: LATE NEOLITHIC BUTMIR
Mobility is also highly dependent on factors such as population size and population density. If local environs are not large enough to satisfy subsistence demands, parts of a population have to become mobile for at least some of the year in order to gain access to other ecological zones. Group size and mobility is also closely linked to marriage behaviour and personal inter-relations: with villages limited to about 200–300 inhabitants, exogamy was required for the population to successfully reproduce.
Group size and population pressure are likely to have played a key role in the development of agricultural core areas during the Neolithic and Chalcolithic. Yet interpretation is problematic, with models of emigration and immigration dependent on estimated population growth, the intensity of interaction, and social classification or identity. Even where we know all these variables, a detailed reconstruction of demography as a proxy for mobility remains difficult (Müller 2007). Given the lack of proper archaeological data and problems with spatial analysis, demographic reconstructions are rare. Nevertheless, research on late Neolithic settlement size and population density in the Butmir occupation of the Bosnian Visoko basin of the central Balkans (Fig. 3.3; see Hofmann et al. 2007) exemplifies the possibilities and the consequences of such efforts for the discussion of mobility and movements.
FIG. 3.3. Okolište, the Visoko basin, and the reconstruction of local and regional late Neolithic mobility: a. Okolište; b. Visoko basin with domestic sites and arable land (hatched); c. estimated population densities in late Neolithic middle Danubian area and Bosnia (isolines, person/km²) and suggested herding activities of the Visoko communities into unpopulated mountain ranges (arrows); d. the estimated herding area of late Neolithic Butmir agricultural core areas; e. the percentage of late Neolithic impressed pottery; f. the boundary between arrowheads (Adriatic) and sling shots (Danubian) in the Neretva-Bosna area.
(drawing by Holger Dieterich, Kiel).
Survey and excavation have focused on the settlement mound of Okolište, which lies 30km north-west of Sarajevo at the river Bosna. Within the three basins in which settlement is focused (‘Siedlungskammern’) are 34 sites with Butmir ceramics (5300–4700 BC). The Visoko basin has a size of 110 sq. km and lies 400–410m above sea level. It is formed by Pleistocene river terraces and para-brown soils. Miocene mountains up to 1,000m (marl and sandstone with rendzinas) encircle the basin. Survey and prospection have discovered further late Neolithic tells from the area.
The evidence here has certain advantages for calculating population density. It is possible to establish the organization and household density of the latest settlement horizon at Okolište, and the duration of occupation and approximate number of houses at other late Neolithic sites across the Visoko basin and around Kakanj (e.g. Peric 1995) can be calculated. Most of the Visoko sites came to an end around 4800–4700 BC.
The Okolište tell is 3.5m high and 7.5ha in size. During at least the latest settlement horizon it had geometrically organized house rows and a huge defensive system of three ditches, one rampart, and a palisade built during Butmir II (4800–4700 BC), all of them surrounding 3.5ha. The contemporaneousness of the houses is very probable, with their ceramics of Butmir IIb character and most of the C14 dates belonging to 4800–4700 BC. A minimum of 200 houses can be extrapolated from the excavated evidence. They were destroyed in a huge, disastrous fire.
Elsewhere, domestic sites like Obre II—excavated in the 1960s (Benac 1973; Gimbutas 1974) and recently investigated by geomagnetic survey—are 2–3ha in size with an internal layout comparable to Okolište, but lacking its defensive system. About 50–150 contemporary houses for each settlement can be calculated using the house number/site size ratio at Okolište. Consequently, we are dealing with no less than 700 houses during the Butmir II occupation of the Visoko basin. Small house size (60 sq. m) suggests a household of about five people. The Visoko basin would therefore hav
e a population of 3,500 at c. 4800 BC.
Botanical analyses from Obre II and Okolište indicate the growing of emmer, einkorn, barley, and millet (Kucan 2007; Renfrew 1974). Palaeozoological analyses prove the dominance of cattle at both sites (Benecke 2007; Bökönyi 1974, 66). Pedological and climatic data from the Visoko basin suggest it had similar agricultural productivity to central Europe (cf. Ebersbach 2003; Ebersbach and Schade 2004; Zimmermann 2002), and if we take into account several calorie requirement models, based on ethnographic archaeological research (Ebersbach 2002, 81, 107–121; Gregg 1988), 1.5ha of cultivated land and 50ha of pasture are necessary for a household of five people.
A total of 10.5 sq. km of cultivated land and 350 sq. km of pasture were required to feed the 3,500 inhabitants of the Visoko basin around 4800 BC. Figure 3.3 models settlement boundaries and the arable ‘territory’ of sites. It assumes an open landscape along the Bosna river without any non-arable areas between settlements. Only about 250 sq. km of pasture was within a day’s march of these sites, so some kind of mobile stock farming in the surrounding mountainous areas would have been necessary. Such practices (cf. Ebersbach 2002, 158) have been ethnographically documented for south-east Europe during different periods under similar conditions (Beuermann 1967).
