Quaternary International: Raiko Krauß, Elena Marinova, Hanne de Brue, Bernhard Weninger

Quaternary International xxx (2017) 1e18
Contents lists available at ScienceDirect
Quaternary International
journal homepage: www.elsevier.com/locate/quaint
The rapid spread of early farming from the Aegean into the Balkans via
the Sub-Mediterranean-Aegean Vegetation Zone
Raiko Krauß a, *, Elena Marinova b, c, Hanne De Brue d, Bernhard Weninger e
a
Institute for Prehistory, Early History and Medieval Archaeology of the Eberhard Karls University Tübingen, Schloß Hohentübingen, Burgsteige 11 D-72070,
Germany
b
Center for Archaeological Sciences, KU Leuven, Celestijnenlaan 200E, b.2408, B-3001 Leuven, Belgium
c
Royal Belgian Institute for Natural Sciences, Department Quaternary Environments & Humans, Rue Vautier 29, B-1000 Brussels, Belgium
d
Geography Research Group, Department Earth & Environmental Sciences, KU Leuven, Celestijnenlaan 200E e P.O. box 2409, B-3001 Leuven, Belgium
e
Institute for Prehistory, University of Cologne, Weyertal 125, D-50931 Ko €ln, Germany
a r t i c l e i n f o a b s t r a c t
Article history: Close examination of the geographic position of Early Neolithic settlements in SE-Europe shows that the
Received 26 April 2016 oldest sites are almost exclusively situated in some very specific biogeographic areas. These earliest
Accepted 16 January 2017 Neolithic settlements are all concentrated in a region that Pavle Cikovac calls the Sub-Mediterranean-
Available online xxx
Aegean (SMA) biogeographic region. It covers the northern and north-western edge of the Aegean,
including Thessaly, Greek Macedonia and Greek Thrace, and extends further into the Balkans, but only
along the valleys of the Vardar (Axios), Struma (Strymon) and Mesta (Nestos) rivers. Surprisingly,
although Thrace is the closest landscape to Anatolia, it does not contain any sites of the earliest phase of
the Neolithic at all beyond the narrow zone of the northern Marmara. In the present paper we explain
this remarkable situation in terms of the natural environment in this particular region of the Southern
Balkans. To begin, we propose that the lack of oldest Early Neolithic settlements in Thrace is related to
the extreme microclimate of this region. As shown by modern vegetation analogues, Thracian oriental
hornbeam-downy oak forests are exposed to stronger continental influence with frosts in the winter and
average temperatures during the coldest months that are ca. 2e3 C lower than those in the Central
Balkans that have Sub-Mediterranean vegetation. In general terms, what we may expect is that the
earliest Neolithic groups would first appear in regions with similarly mild conditions, on a yearly average,
to those in the Mediterranean. Such mild conditions are indeed present in the SMA biogeographic region.
On the other hand, before moving further to the north along the north-south oriented river systems of
the Central Balkans, the Neolithic economy based on agriculture and stockbreeding would first have to be
adapted to the relatively harsh winters in the Balkans. In consequence, it would have been possible to
apply the new Neolithic lifestyle in the neighbouring areas of Thrace, Walachia, Dobrudzha and the
Carpathian Basin only after a certain period of adaptation. Available 14C-data show that the adaptation
period is identical to the time-span of Rapid Climate Change (RCC: 6550-6050 calBC) as defined in
previous studies.
© 2017 Elsevier Ltd and INQUA. All rights reserved.
1. Introduction spectrum of highly aesthetic pottery products e spread with the

incredible speed of some 2000 km in less than 200 yrs from the
Following 6100 calBC, a complex Neolithic package of new ideas, Aegean through the Balkans into the southern Carpathian Basin.
capabilities and technologies e comprising some of the most Although such quasi-instantaneous Neolithisation of SE-Europe is
important cultivated cereals and legumes, a characteristic set of well-known to archaeologists (Guilaine, 2000/2001; Zvelebil, 2001;
domesticated animals, a developed settlement infra-structure, an Clare and Weninger, 2014), it is in complete contradiction to so-
advanced technology in grinding-stones, as well as a diverse called Wave-of-Advance models (Ammermann and Cavalli-Sforza,
1984), which instead propose the existence of a demographically
motivated, much slower (by a factor 10) average speed of ~1 km/yr
* Corresponding author.
for the Neolithic move. By whatever mechanism, for example
http://dx.doi.org/10.1016/j.quaint.2017.01.019
1040-6182/© 2017 Elsevier Ltd and INQUA. All rights reserved.
Please cite this article in press as: Krauß, R., et al., The rapid spread of early farming from the Aegean into the Balkans via the Sub-
Mediterranean-Aegean Vegetation Zone, Quaternary International (2017), http://dx.doi.org/10.1016/j.quaint.2017.01.019
2 R. Krauß et al. / Quaternary International xxx (2017) 1e18
random walk or directional, and by whatever carrier, for example of appears, all these rapid and long-distance movements must have
ideas instead of people, this Wave-of-Advance approach to been underway, at the latest in the Neolithic mind, the very
Neolithic dispersal has been recently refuted once again, this time moment the climatic amelioration following the end of RCC must
by application of the very method on which it was supposedly have become apparent even to the most convinced Neolithic anti-
founded, i.e. radiocarbon dating (Weninger et al., 2014). In terms of climate skeptic (Weninger et al., 2014). Indeed, even today is
the historical reality of scientific rationality quite telling, the there much pride and interest in travelling both geographically and
observed rapidity of Neolithisation is equally in conflict with the environmentally extreme distances in the shortest of time, and we
opposite theory, i.e. an entirely autochthonous development of the may assume the same applied to prehistoric adventurers/scientists.
complex package by the local Mesolithic ‘substrate’ (cf. Srejovi c, What is more, even superficial analysis of available Neolithic 14C-
1973: 176e180). Recently, the continuing discussion (Benz, 2000; data informs us that the transition from the Mesolithic to the
€
Ozdo an, 2011) has been extended to include the question
g Neolithic not only in the Aegean and adjacent parts of SE-Europe,
whether the spread of Neolithic lifestyle from its core zones in the but in essentially all circum-Mediterranean regions was accompa-
Levant and upper Mesopotamia to the West (Braidwood, 1952; nied by an abrupt transition in Economy, but equally so in the
Aurenche and Kozłowski, 1999) could have its background in cli- Neolithic Mind. In our view we may readily assume, even when
matic variability (Clare, 2010, 2016; Weninger et al., 2009). lacking direct confirmation from empirical data (e.g. historical
Whatever approach we favour, the Neolithisation of SE-Europe sources), that the Neolithic expansion would be accompanied by
is incomprehensible without an understanding of the de- some entirely new concepts in thinking, in particular in terms of the
velopments within its many source regions. As is well-known, the semantics of what makes short and long geographic distances.
area of the northern Fertile Crescent represents the main single Similarly, we may assume corresponding changes in the assumed
region in which the most important cultivated cereals such as size of the world, both in metric terms and all the more in mental
einkorn wheat, barley and emmer are known to have grown wild, categories. For all we presently know the begin of this specific
and also where the predecessors of the domesticated sheep and mental transition dates to the very begin of the RCC-period (6550-
goat were present (Watkins, 2007). In view of recent palaeogenetic 6050 calBC). Similar changes in societal and mental landscapes, but
studies (e.g. Scheu, 2012; Geo €rg, 2013; Bollongino et al., 2015; for the much earlier transition from hunter gathering to farming
Scheu et al., 2015) even the theoretical possibility that wild forms lifestyles, were noted by Bar-Yosef (2011) as being due to the impact
of at least cattle and swine could have been locally domesticated in of the Younger Dryas.
SE-Europe now seems highly improbable. Nonetheless, from the
obviously to be assumed long-distance transfer of animal, plant, 1.1. Maritime networks and land routes in the Aegean and SE-
and human DNA (Hofmanova et al., 2016) from the Near East to SE- Europe
Europe, and this means a (probably stepwise) movement of many
real molecules over many thousands of kilometres, we should not The Neolithic dispersal from the Near East into the Aegean can
immediately conclude that the underlying cultural trajectories for be attributed to the existence of long-distance coastal networks
the dispersion of animal and plant DNA must e by necessity e have that are known to have been operating throughout the entire
actually taken a ‘long time’. Formulated differently, it is only eastern Mediterranean already during the preceding Epi-
seemingly logical to associate, but actually not at all justified to palaeolithic/Mesolithic and/or Pre-Pottery periods (Reingruber,
equate apriori (i.e. before having looked), a ‘long’ geographic dis- 2011; Мaniάth2, 2014; Horejs et al., 2015). The same applies to
tance of thousands of kilometres with the ‘long’ time it may take to the successful transfer of taurine cattle from the Levant to Cyprus
travel from the Near East to Europe, such that any movement from across the open sea, which is documented already at the very onset
one place to the other will by necessity (?) require hundreds of of the Holocene (e.g. Zeder, 2008), as well as to the wide distribu-
breeding generations. To avoid such flawed thinking, evidently still tion of Melian obsidian in the Aegean during the Mesolithic (e.g.
today underlying the Wave-of-Advance approach, we must be very Bergner et al., 2009; Reingruber, 2011; Mili
c, 2014). It would appear
careful in distinguishing between the temporal and the geographic natural that from some time onwards these pre-established marine
scale, and this applies in particular to what we imagine are long/ networks would be used to transfer other components of the
short times, and what are long/short distances, in relation to what Neolithic package, including its sedentary lifestyle, grinding-stone
such concepts could have meant during the Neolithic. Firstly, technology, farming and animal husbandry, as well as the trans-
among modern archaeologists, there is general agreement that the mission of pottery products. Let us not dwell on the altogether
initial domestication of plants and animals within the Fertile secondary question, as goes for the material artefacts, whether it
Crescent is best described as a retracted process, and to which we were the products themselves or rather their modes of production
may adequately assign millennium-scale time-ranges (see Riehl that had to be transferred. If we instead undertake a comparison
et al., 2015). However, looking closer at the 14C-database it be- between the different fields, it is in our view the adaptation to the
comes apparent that similarly slow (~millennium-scale) dispersion continental conditions in SE-Europe with its markedly cold winter
processes are really only obvious in cultural terms for certain months that must have represented the greatest challenge to the
components and sub-regions of the Pre-Pottery Neolithic, and even newcomers.
then only on broadly superficial scales, and surely not for the de- This already follows from an examination of the different
velopments during the much later Pottery Neolithic. At the latest landscapes (terrestrial, coastal, marine) that the newcomers
around 6000 calBC, and according to available data actually encountered along the routes from the Near East. What we call here
beginning some 500 yrs earlier, the geographic range of the the natural route, is to move (or sail) along the southern coast of
Neolithic lifestyle rapidly increases, and this is perhaps most visible Turkey, then turn into the Aegean, and finally decide which of the
in the stepwise expansion of Neolithic lifestyles out of North many islands, coastal floodplains and mountainous regions is the
Mesopotamia to the East (~6000 calBC), out of the Levant into NE- most preferable. The existence of this southern (coastal) route is
Africa (~6000 calBC), as well as in westward direction to the Aegean confirmed, albeit only indirectly, by the location of the two major
(~6550 calBC), but where it not only arrives abruptly, but also leaves settlements on Turkish West Coast, that is Çukuriçi and Ulucak,
again abruptly (~6000 calBC) both in northern direction towards both of which represent the Neolithic in its very earliest stage, and
the Pannonian Basin, as well as along the Adriatic coast, and finally each of which has essentially the same foundation date of
(but only ~ 200 yrs later) to reach the Iberian Peninsula. As it 6550 ± 30 calBC (Weninger et al., 2014). It is the same date as
R. Krauß et al. / Quaternary International xxx (2017) 1e18 3
established for the earliest documented use of the Franchti Cave on Islands (Malinowski, 1922 [1979]). With the choice of title for his
the Peloponnese (Weninger et al., 2014). book, Malinowski refers intentionally to the myth of the Argonauts,
In parallel to this southern (coastal) route, one would at first which is set on the Aegean-Pontic coast. During the Neolithic,
anticipate that the Neolithic might also have spread simultaneously archaeologically tangible material goods from the Aegean such as
out of NW-Anatolia, utilizing the entirely unproblematic crossing obsidian and also Spondylus shells could have served as exchange
both of the Dardanelles and of the Bosporus, first into neighbouring goods, as was the case in Melanesia at the beginning of the 20th
Thrace and northern Greece, and from there further into the Bal- century. Moreover, although archaeologically intangible, along
kans. Yet, precisely the Thracian Plain is one of the areas on the with the trading goods there would be transfer of the manifold
Balkan Peninsula, in which Early Neolithic settlement has been immaterial social, economic, and religious aspects that are of
shown to have taken place particularly late (Nikolov, 1990; importance for all exchange systems (Malinowski, 1922 [1979]:
Todorova, 1990). It is noteworthy that archaeological research in 115e141), and which we may assume are also conditional for pre-
Thrace not only began very early, but always had special focus on historic societies. Such aspects can be recognised, in particular, in
the Neolithic period. Within Bulgaria, Thrace can be considered as the close analogies between ceramic styles during the SE-European
one of the best explored landscapes. In the area of the Marmara Sea, and Anatolian Neolithic, and which can be interpreted as more than
which represents a borderland between Anatolia and SE-Europe, a just a reflection of personal contacts or preferences. In geographic
marked Neolithic had already developed several centuries before terms, we can be sure that the Cyclades, the Troad, the islands of the
€
the spread of agriculture and livestock-raising to Europe (Ozdo an,
g Sporades and the Turkish and Greek littorals of Thrace would all
2014). In contrast in Thrace, there are no known Early Neolithic have participated in maritime exchange networks during the
finds at all that would date earlier than Karanovo I. But further to Neolithic period in Northwest Anatolia and the Aegean (Horejs
the west, the earliest Neolithic cultures north of the Aegean are et al., 2015).
definitely well-documented in the river valleys of the Vardar
(Axios), Struma (Strymon) und Mesta (Nestos). In view of the much 2. Two different developments in the Marmara region
closer proximity of Thrace to NW-Anatolia this situation is puzzling
and requires further explanation. According to a model recently presented by Mehmet Ozdogan,€
In general terms, a comparison of the Balkan Neolithic with during the Early Neolithic the cultural development in the Marmara
West Anatolia reveals a distinct cultural gradient. Whereas in West region can be divided into two clearly distinguishable spheres
Anatolia solid clay houses were built with stone foundations, those (Fig. 1): a western sector along the coastal strip of the Aegean, and
of the Balkan Neolithic are constructed more simply. Similarly, the an eastern sector with sites concentrated almost exclusively within
material culture in the Balkan sphere seems quite reduced in the present-day limits of the mega-city of Istanbul and its sur-
comparison to that in West Anatolia: the pottery in the latter region roundings, as far south as Lake Ulubat (Ozdo € an, 2014: 41e47).
g
is clearly of greater quality and multifaceted. Despite these differ- Although in both regions the Neolithic can be deduced from com-
ences, an understanding of the pottery development in West mon origins in southwestern Anatolia, they nonetheless developed
Anatolia enables us to recognise the moment at which the basic slightly different “Neolithic packages”, which are expressed by
impulse of Neolithisation also commenced in SE-Europe. There are €
“different technologies” and “distinct identities” (Ozdo an, 2014:
g
altogether three innovative elements in pottery production that are 41e43). Technological differences are manifest in the bullet-core
characteristic for the oldest pottery of the Balkans, and all three technique and the use of obsidian as raw material for stone-
elements are recognisable at the West Turkish sites of Ulucak, here chipping in the East, and the so-called Karanovo-blade technique
for the first time in Ulucak Va, as well as at Çukuriçi Ho€yük, for the in the West. Distinct identities are expressed by the presence of
first time in Period VIII. The three pottery elements are, firstly, extra-mural burials in the East, which are however completely
Impresso decoration; secondly, depictions of ungulates applied in missing in the West. Characteristic for the Western Marmara are
relief on the vessel's surface; and thirdly, white-on-red painted certain aspects of ceramic decoration such as red-slipped surfaces
pottery. Although white painting on pottery is likely to be some- and mat-impressed bases (Ozdo € an, 2014: Fig. 12). What is com-
g
what older in West Anatolia than in SE-Europe, once introduced it mon to both regions is the occurrence of anthropomorphic clay
appeared more frequently. Also older on the Turkish West Coast is figurines, although they are rare in the East.
the tradition of the red slip on pottery, but this is an omnipresent €
According to Ozdo an (2014: 2f.) the most obvious reason for
g
element which also emerges with the Neolithisation of the Balkans. the observed cultural variability in the two regions would be major
One crucial difference, however, is the distinct bonding of the differences in the underlying formation processes already during
West Anatolian Neolithic with the Mediterranean Sea, in contrast to the Early Neolithic. Whereas the Neolithisation in the Eastern
the Neolithic spread into the Balkan-Carpathian sphere, which Marmara could be seen as terminate moving of a Neolithic popu-
must have progressed over land. On the land routes it would appear lation into an already inhabited area, which would require merging
natural for the rivers to play an important role, and boats were of the incoming farmers with local groups of hunter gatherers, the
surely important for navigation on long stretches of rivers. Yet, the Neolithisation process in the western Marmara could rather be
bond with a river valley is not comparable to maritime contact over seen as a movement of farmers in successive waves into areas void
the open sea. Indeed, the very connection of agriculture and animal of any previous population. Note that, in both regions, the respec-
husbandry with sea travel over long distances would appear par- tive judgement is based on intense and systematic surveying. For
adoxical: at first view, the Neolithic form of economy requires a the Western Marmara there is a complete lack of any data that
sedentary way of life, whereas sailing rather implies the mobility of might indicate the existence of a pre-Neolithic population.
human communities. Whether or not there existed groups with A view of the Marmara region as seen from the Aegean reveals
different forms of livelihood, including communication specialists further west-east differences. The settlement of Aktopraklık B, for
who may have interacted solely for economic reasons, cannot be instance, displays a semi-circular, closely built plan of rammed clay
determined at the present state of research. Whatever the case, the houses with interior posts supporting the roof (Karul and Avcı,
very specific natural environment of West Anatolia with its offshore 2013: Figs. 12e15, 17). Such constructions are also documented in
Aegean islands offers optimal preconditions for a complex ex- Otzaki-Magoula, in Sesklo and Achilleion, in Thessaly (Miloj ci
c,
change system based upon maritime travel, as described by Bro- 1983: Figs. 2e3; Alram-Stern, 1996: Fig. 29; Theocharis, 1973:
nisław Malinowski with reference to inhabitants of the Trobriand Fig. 179, Gimbutas et al., 1989: 32e68). Documented further east
€
Fig. 1. Different assemblages of the Neolithic in Eastern (blue) and Western Marmara (red). (according to Ozdo an, 2014). 1 Çardakaltı, 2 Aşag
g ı Pınar, 3 Bulgar Kaynag
ı, 4 Alpullu, 5
Toptepe, 6 Hoca Çeşme, 7 Ugurlu. 8 Kaynarca, 9 Coşkuntepe, 10 Ag _
açlı, 11 Domalı, 12 Yarımburgaz, 13 Yenikapı, 14 Haramidere, 15 Fikirtepe, 16 Pendik, 17 Ibonu Rampası, 18 Ilıpınar,
19 Menteşe, 20 Barçin, 21 Musluçeşme, 22 Aktopraklık. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
are framework clay buildings (wattle and daub), for example, in 1990). This is as follows. Densely settled areas are always located
Barcin Ho €yük (Gerritsen et al., 2013: 6e7, 9) and in Menteşe in river valleys flowing into the Aegean as well as in the Central
(Roodenberg and Alpaslan-Roodenberg, 2013: Fig. 9). Initially, Balkans, but with almost no sites located in the areas east and west
Ilıpınar also has wattle-and-daub structures (Roodenberg and of such concentrations. Still today, the Struma valley appears to
Alpaslan-Roodenberg, 2013: Fig. 3). Then, as of layer VI, the char- represent the major route for the Neolithisation of the Balkan re-
acteristic architecture of row houses arranged in a semi-circle and gion. Indeed, starting from the Aegean, the Struma and Vardar river
built of air-dried mud bricks was adopted (Roodenberg and valleys seem to be the only two corridors of any importance for the
Alpaslan-Roodenberg, 2013: Figs. 4e7). As it appears, the Mar- spread of the Neolithic to the North. Given this communication
mara region became integrated in the Aegean Neolithic in a step- system, one might expect that the further any given site is located
by-step manner. Peculiar to the east of this region is the burial of away from the Aegean, the less it would compare with sites in West
deceased in cemeteries. This practice is not attested anywhere else Anatolia and in Greece. Yet, the exact opposite applies.
in the entire Mediterranean area at this early time. The unexpectedly strong relations that exist between Anatolia
and the Early Balkan Neolithic over large distances are perhaps best
3. The circumvention of Thrace illustrated by the following two known examples. First, there is no
difficulty at all in linking the pottery inventory of Kova cevo in the
Viewed in a wider perspective, it appears that the Neolithic of Struma river valley with the material from Hacılar in the southwest
the Eastern Marmara is not just loosely connected to the earliest Anatolian lake district (Lichardus-Itten et al., 2006: 86e88; Brami
Neolithic in the Balkans. The Eastern Marmara represents almost a and Heyd, 2011: 182f.). Second, excavations at Dzhuljunica in
barrier between the development of the Neolithic in Anatolia and northern Bulgaria have yielded pottery that can also be found at
the Balkans. It is not that relationships did not exist between several Early Chalcolithic (in Turkish terminology: after ~6000 BCE)
Anatolia and the Balkans; instead, whenever contacts were estab- sites in central West Anatolia, including Ulucak, Ege Gübre and
lished, these contacts always took place while bypassing the Mar- Çukuriçi (Krauß et al., 2014: 58f.). In both cases, the pottery simi-
mara region. What is also conspicuous, even though the number of larities are quite plausibly linked to the river systems of the Central
known Early Neolithic sites has increased significantly, the general Balkans. What is even more conspicuous, however, is that for
picture of the spread of Early Neolithic settlement into the Balkans Dzhuljunica there does not exist any direct connection via NW-
is still as described by Vassil Nikolov already in 1990 (Nikolov, Anatolia. For this route, which would indeed have been the
shortest (via the Black Sea coast), the Eastern Marmara acts as a assuming that they do actually exist but remain covered from view,
complete barrier for interactions of any kind. Furthermore, already or that a significant rise in Black Sea level would have flooded the
a brief glance at the distribution map of the so-called earstuds former coastlines. Indeed, little is known about the palaeogeo-
(labrets or bucrania idols: a find category clearly restricted to the graphic development of the Thracian Plain, although landscape and
oldest phases of the Early Balkan Neolithic), illustrates the soils in SE-Europe and in particular in Northern Greece have un-
bypassing of the whole Marmara region (Fig. 2). Since there are not dergone significant changes since the Neolithisation (Ghilardi et al.,
many finds of this specific category in West Anatolia, it is actually 2008; Lespez et al., 2000; Lespez et al., 2013; Krahtopoulou and
questionable whether these artefacts originated from areas east of Veropoulidou, in press). Nevertheless, taking the conspicuous lack
the Aegean, as one might assume. Whatever we infer for these of earliest Neolithic sites in Thrace as a present fact, in the following
idols, the general picture is that on their way into the Balkans the we put forward and evaluate the hypothesis according to which the
earliest farmers not only bypassed the Marmara region but also spread of earliest Neolithic settlement into the Balkans can be
circumvented Thrace. As mentioned above, this is all the more linked to and is partly explained by the existence of some specific
astonishing, given that inner Thrace is the landscape situated bioclimatic and environmental (vegetation) zones.
closest to Anatolia, and since there are no sites of the earliest phase
of the Neolithic there at all. 3.1. Neolithisation along the SMA Vegetation Zone
To explain this specific find situation we could now argue that
major sedimentation rates in the wide alluvial plain of the Marica A look at the situation of the Early Neolithic find spots in SE-
(Meriç/Evros) and Tundzha Rivers have deeply buried the sites, Europe reveals that the oldest sites are almost exclusively spread
Fig. 2. Spread of symbolic figurines (earplug, labret or bucranium idols) from the Aegean to the Carpathian Basin. Examples from Rakitovo (after Matsanova, 1996: Table 9). Ev-
idence of the find places see Krauß, 2016, 2017. 1 Nea Nikomedeia, 2 Vashte €mi, 3 Podgorie, 4 Rakitovo, 5 Vaksevo, 6 Komarevo, 7 Skalsko, 8 Hotnica, 9 Beljakovec, 10 Dzhuljunica, 11
Orlovec, 12 Koprivec, 13 Kremestice, 14 Jasenovik, 15 Ornice-Makresani, 16 Drenovac, 17 Blagotin, 18 Divostin, 19 Grivac, 20 Banja-AranCelovac, 21 Rudnik, 22 Minine Vode, 23
Bukovce, 24 Usce Kamenickog Potoka, 25 Knjepiste, 26 Velesnica, 27 Dubova-Cuina Turcului, 28 Lepenski Vir, 29 Kucajna, 30 Drmno, 31 Dobanovci-Ciglana, 32 Idvor, 33 Donja
Branjevina, 34 Magareci Mlin, 35 Zmajevac, 36 Foeni-Salaş, 37 Timişoara-Fratelia, 38 Miercurea Sibiului, 39 Ocna Sibiului, 40 Gura Baciului, 41 Vrsac-Kozluk, 42 Major Uzunovo, 43
Zappeio, 44 Sesklo, 45 Achilleion, 46 Argissa Magoula, 47 Soufli Magoula, 48 Hacılar.
over a specific biogeographic area (Fig. 3). The earliest Neolithic barriers in the southern Balkans, some reaching more than 2000 m,
settlements are concentrated in a region that has been described by were an obstacle to the further spread inland of the influence of the
Pavle Cikovac as the Sub-Mediterranean-Aegean (SMA) biogeo- Mediterranean climate (cp. Cikovac, 2002: 25). Correspondingly,
graphic region (Cikovac, 2002: 22e25). This particular region in- sub-tropical flora is found only in the Thessalian basin, in the broad
cludes the northern and north-western edge of the Aegean, coastal zones of Macedonia, and in the three aforementioned river
including Thessaly, Greek Macedonia and Greek Thrace, and ex- valleys. Among the flora to name here are foremost thermophile
tends further into the Balkans only along the rivers Vardar, Struma species such as the semi-evergreen Macedonian oak (Quercus tro-
and Mesta. The region further encompasses the actual Vardar zone, jana), the hop hornbeam (Ostrya carpinifolia), the Oriental horn-
as well as northern Pelagonia in the Republic of Macedonia beam (Carpinus orientalis) and the manna ash (Fraxinus ornus)
(F.Y.R.O.M.). In Bulgaria it is limited to the narrow river banks of the (Cikovac, 2002: 25).
Struma and Mesta rivers, which open towards the south to the The SMA biogeographic region is further defined as the area of
northern Aegean in a canyon-like landscape. The high mountain the natural spread of the European olive tree (Olea europaea)
Fig. 3. Northern border of the natural spread of olive tree, fig tree and wild grape. Hatched is the SMA biogeographic region (after Cikovac, 2002: 22e25).
(Zohary et al., 2012: map 15), which in a remarkable way covers duration of continental winter with week-long periods of frost
those areas that are part of the maritime networks of the circum- would have presented a serious hindrance to crops. Thus, in com-
Aegean Neolithic described above. Of course, the olive tree was bination, it can be assumed that the earliest agriculture would have
cultivated at a significantly later time. However, we are concerned been initially restricted to areas with correspondingly optimal cli-
here specifically about those areas that have optimal conditions for matic conditions for growing and harvesting cereals. Hence, leaving
agriculture, and for which we propose the theory that these areas the Mediterranean zone in a northern direction would have meant
overlap with the distribution area of the typical Mediterranean being confronted with shorter growing seasons and later frosts,
crop plants. The distribution area of the wild olive tree extends which would have been a real threat for a subsistence based on
along the narrow coastal zone of western Anatolia, in Macedonia annual crop cultivation. This would readily explain why the earliest
only on the Chalkidiki peninsula, in the eastern part of Thessaly, Neolithic in the Balkans is only to be found in areas that are located
and further to the Greek mainland south of the Pindos Mountains, within the range of the SMA vegetation zone, where they are not
and finally of course the Aegean islands. The distribution area of the threatened by harsh winters.
wild fig tree (Ficus carica) extends only a little further to the North Regions with a marked continental winter climate are also likely
(Zohary et al., 2012: map 17). It encompasses the southern edge of to have had an unfavourable effect on animal husbandry, although
Turkish Thrace, the broad coastal zone of the northern Aegean as far not an existential threat. In the course of a typical year, from late
as the Rhodope Mountains, and in principle the entire Greek spring until autumn, pastures in the high mountains of the
mainland, with the exception of the Pindos. This given, the original southern Balkans offered an ideal pasturage. However, this situa-
distribution area of the olive and fig tree apparently circumscribes tion is seriously threatened by cold winters. For this very reason, a
the area which was ideal for an agricultural way of life in the early seasonal transhumance has been long practiced in this region, even
Neolithic period. Correspondingly, although based on a still limited until historical times, which entails protecting the grazing animals
number of radiocarbon dates one can conclude that the areas of the during the winter in shelters that are located either in the mountain
Greek mainland within the distribution area of the olive and fig tree valleys, or else in the milder coastal zones of the Mediterranean
had already become Neolithic in the 65th century calBC (Beuermann, 1967). Hence, when proceeding from existing mari-
(Reingruber, 2008: 613e616; Мaniάth2, 2014), most probably time networks of the Aegean, it is the SMA biogeographic region
initiated from West Anatolia. that would have been best suited for the Neolithic manner of
The natural distribution of the wild grape (Vitis vinifera), whose economy. The further expansion would then have naturally fol-
cultivation is known to have begun at some time in the 4th mil- lowed the SMA zone in northern direction, along the valleys of the
lennium calBC (Zohary et al., 2012) and which belongs to one of the Vardar, Struma und Mesta rivers. In order to live in regions on the
most important cultivated plants in the region, can also be other side of the mountain barriers of the southern Balkans, based
considered a geographically delimiting factor for the Aegean solely on Neolithic economic activities, meant that the Neolithic
Neolithic. Although its cultivation did not begin in the Early package had to be adapted to conditions of the comparably harsher
Neolithic, our theory according to which the oldest Early Neolithic and longer winters.
in the Balkans emerged from zones with ecological conditions
similar to the Aegean, can possibly be traced by studying the dis- 3.2. The Early Neolithic in the Struma (Strymon) river valley
tribution of this wild form of grape. In addition to the Mediterra-
nean sphere in a broader sense, the original distribution area of the The fact that the Struma valley represents one of the main routes
wild grape also encompasses the entire littoral of the Black Sea involved in the Neolithisation of the Balkans was realised compa-
(Zohary et al., 2012: map 16). In SE-Europe the line of its north- rably late in research, and this is particularly due the lack of sites
ernmost distribution follows the coast of the Black Sea, runs along the north Aegean coastline (Andreou et al., 2001; Urem-
somewhat centrally through the Thracian Plain towards the West Kotsou et al., 2014: Fig. 1). In many cases even today the under-
and then follows the north-eastern ridge of the Dinarids as far as standing of the dissemination of the Neolithic via the Struma valley
the rim of the Slovenian Alps. The distribution of wild wine thus is still largely based only on the comparison of ceramic types and
embraces the Dobrudzha, Eastern Thrace and the Rhodopes, all of their decoration (Brami and Heyd, 2011; Kraub, 2011). This view is
Macedonia as far as the northern Vardar zone, Albania, clearly augmented by the distribution of Neolithic clay stamps,
Montenegro, western Bosnia and Dalmatia. In contrast to the nar- which also indicate the importance of connections reaching from
row distribution area of the olive tree, for wild wine only a sporadic Pisidia in south-west Anatolia to Thessaly, and which include the
congruency with the distribution of the oldest Early Neolithic great river systems Struma and Vardar that exit from the Central
period can be discerned. In the following we address the question Balkans into the northern Aegean (Lichter, 2011: Fig. 1). Interest-
as to the underlying cause of this curious situation. ingly, this picture of the distribution of Neolithic clay stamps is
To begin, let us take a step back and have a closer look at the missing in the Eastern Marmara, an absence that in turn speaks for
Neolithic economy in its initial period: as is well-known, the the barrier function of that region.
Neolithic was developed some 10.000 yrs ago in the Fertile Cres- Of particular interest then is the question as to the extent of any
cent, and expanded as far as the Aegean Sea only much later, in the structural mutuality that might have existed between the Struma
course of the 7th millennium calBC. In principle, however, the valley and southern regions, aside from the ceramic technology
Neolithic economy was a form of husbandry that could be con- employed. Discussion of this issue, however, is hindered by the
ducted in many different geographic regions. All that is necessary is general lack of Early Neolithic finds in the north Aegean coastal
a landscape suitable for the cultivation of steppe grasses, and for zone, a lacuna that is most probably associated with shifts in the
pasturing ungulates. Open landscapes were initially preferred for coastline and heavy sedimentation (Lespez et al., 2013). Diagnostic
settling, where crop cultivation and pasturage were possible architectural contexts from the earliest Neolithic in the Struma
without any laborious clearing of trees. The main challenge for the valley are presently only known from Kova cevo. Even at Kova cevo,
earliest farmers was to protect the sowings, and it is precisely this the Early Neolithic settlement context has been severely damaged
aspect that is most likely to have led to the sedentary lifestyle that through erosion, and also as a result of earthquakes and later in-
is so characteristic for early farming. A moderately warm climate trusions (Lichardus-Itten et al., 2002: 107f.). The remaining, free-
and sufficient precipitation would have been favourable for the standing houses display a square ground plan with walls
growth and ripening of the early kinds of cereals. The longer measuring 5e6 m in length and floors made of stamped mud. The
standing walls are constructed of posts-and-framework covered by implication of the genetic data is that the movement of herds of
a layer of mud/daub (Lichardus-Itten et al., 2002: 108e114; Figs. 5 cattle in the Balkans, even on local scales, is likely associated with
and 2; 6,1e2). This kind of house construction can be traced back to groups of people immigrating from widely differing regions.
structures in central West Anatolia, although the characteristic Namely, the spread of the Neolithic occurred far too rapidly, and the
stone foundations are completely absent in houses in the Struma number of accompanying cattle was far too large, implying that
valley. This particular absence can hardly be explained by the lack these movements can be explained solely by assuming the exis-
of accessible materials; indeed, there is a plenitude of stones in the tence of some loose networks of people with different economic
area. A post construction with several rooms is known from bases. Even if pre-Neolithic populations did exist in this area, and
Balgarcevo II; it displays features of an independent Balkan type of on demographic scales for which we have absolutely no evidence,
house, which at least cannot be traced back to the South they had no attestable agency in mediating the Neolithic cultural
(Pernicheva-Perets et al., 2011: 71e82). elements. Our conclusion, based on archaeological, 14C-radiometric
The discovery of skeletal remains of children within the settle- and on cattle-genetic data, is that the agents of the Neolithisation of
ment of Kova cevo (Lichardus-Itten et al., 2002: 116; pl. 9,1e2; 10,1) the Struma valley must have been groups of people who were
together with the absence of cemeteries outside of the settlements already fully familiar with Neolithic economy, who were long
are indicative of a burial custom that is possibly similar to that experienced with plant cultivation, and who were accompanied by
practiced in central West Anatolia. What we can at least say is that their herds of animals.
burial customs at Kova cevo differ distinctly from the rituals found
in the Eastern Marmara, where several cemeteries are known. 3.3. Neolithisation of the Vardar (Axios) river area
The spectrum of slaughter remains in pits in Kova cevo reflects
an inventory containing large amounts of domestic animals and The few buildings that have been discovered in the Vardar zone
only a small proportion of wild game, which is common for the display a remarkably solid construction. The settlement plan of
Early Neolithic. The majority of domestic animal remains comprises Tumba MaCari shows rectangular structures measuring up to 80 m2
sheep/goat, amounting to 65%, followed by ca. 21% domestic pig. in floor area (Kanzurova and Zdravkovski, 2011: 140f.). Inside the
Significantly, the scholars studying the animal bones have identi- houses containers of round or square plan were firmly installed
fied good comparisons of this inventory not only with neighbouring (Kanzurova and Zdravkovski, 2011: Fig. 11; 13), which until now
sites in the Balkans, but also with the material found in the set- have singular appearance in the oldest Early Neolithic in the Bal-
tlements of Nea Nikomedia and Achilleion in Greek Macedonia and kans. Such installations are best traced to the south and, thus,
Thessaly (Lichardus-Itten et al., 2002: 126; cf. Halstead and provide hints of contacts between the Vardar zone and the Aegean,
Isaakidou, 2013; see also their contribution to this volume). Finds at least with the immediately bordering area of Greek Macedonia.
of Glycimeris and Spondylus support a clear association with the Prototypes could have been settlements such as Nea Nikomedia
Aegean; they appear already in the Early Neolithic layers in Kova- and Servia (Perle s, 2001: Fig. 9.1; Alram-Stern, 1996: Fig. 38). It is

cevo and served as raw material for making ornaments (Lichardus- also noteworthy that the structures in Tumba MaCari do not
Itten et al., 2002: 127f.). Analyses of the many botanical remains compare well with contemporary buildings in Thessaly at all, since
found at Kovacevo show the normal spectrum of Early Neolithic the latter houses were commonly constructed with mud bricks and
cereals, including einkorn wheat, emmer, barley and hard wheat in some cases have wall pillars in the interior to support the roof.
(Lichardus-Itten et al., 2002: 127). These results speak neither in Hence, for Thessaly, it seems that closest connections existed not on
favour of nor against a connection of Kovacevo with any old any land route, but across the Sea, directly with central West
Neolithic region in the south, however, the weed assemblages from Anatolia. This also becomes apparent when we take a closer look at
the site indicate clearly affiliation to the Eastern Mediterranean the building plan in Anza II, where one house is documented in the
(Marinova, 2006: 56e62). settlement as having wooden post-and-framework walls that are
When studying the natural environment of the Struma valley, it erected upon a stone foundation (Gimbutas, 1976: pl. 3); this
becomes clear that the spread of the Neolithic into the Balkan area building strongly resembles the solidly built structures in central
occurred not simply by a switch from maritime to terrestrial West Anatolia. In the case of Anza II, since it is one of the oldest
communication, but that ship navigation in general was no longer known house contexts, this manner of construction can hardly be
the preferred manner of transportation. Marine communication is traced back to a local origin. Nevertheless, on the upper reaches of
typically quite rapid. However, even on land, the speed of the the Vardar River some elements otherwise typical for the maritime
movement is apparently so high (~2000 km in less than 200 yrs) networks of the Aegean are not present; it would appear as if they
that it is impossible to temporally and geographically resolve these ‘did not arrive’. For example, almost no obsidian tools are attested
pathways, even with high-resolution 14C-site chronology, by any there (cp. Gimbutas, 1976: 257e282, cf. also Mili c, 2014). The
kind of random-directional demographic wave-modelling presence of the kinds of cereals and legumes common to the entire
(Weninger et al., 2014). From a purely archaeological standpoint, Aegean and Balkan zone is not surprising, nor is the clear domi-
Neolithisation from the northern coastal zone of the Aegean as far nance among domesticated animals of sheep/goat over cattle and
as the Danube River has the appearance of representing an entirely swine (Gimbutas, 1976: 300e331). Close contacts with the mari-
sudden and abrupt process, that lacks any large temporal depth, time networks in the Aegean are again attested by artefacts made of
and that took place during the course of the 62nd century calBC. Spondylus, which appear in comparably large numbers already in
With respect to the probable existence of different communication the oldest settlement layers in Anza (Gimbutas, 1976: 242e251; cf.
paths for various components of the Neolithic package, Amelie also Bajno czi et al., 2013).
Scheu's palaeogenetic examination of domestic cattle interestingly Turning now to the manner in which such partially local con-
revealed a genetic similarity between cattle in Kovacevo and tacts are expressed in terms of pottery styles, it is important to
Neolithic cattle in their (far away) source areas in Iran, that was ca. recognise that a widely similar spectrum of forms developed
twice as high as between Kova cevo and Koprivec, although both throughout the entire zone of the Southern Balkans, at least during
sites are situated relatively close to each other (Scheu, 2012: 106). the advanced Early Neolithic. This spectrum extends indeed all the
Local aurochs did not play a role in the genetic characteristics of way from the Western Balkans to the Black Sea, and points to some
Neolithic cattle; indeed, apparently Neolithic cattle were imported quite intensive contacts at long distances, during the first half of the
directly from the Southeast (Bollongino et al., 2008). A further 6th millennium calBC. Nonetheless, these contacts only formed
after the actual phase of Neolithisation, and are founded upon land- of Krajnici. The sites have low to moderate elevations, mainly below
bound communication between individual groups of people. 300 m, with a maximum elevation of 617 m asl for Krajnici. With
Considering the rugged topographic texture that is geographically respect to the slope aspect, sites do not show any preferential
sectioned by some extremely high mountains, foremost the orientation. The natural vegetation, finally, is dominated by oak
Southern Balkans between the Albanides, Baba Mountains, Voras forests and alluvial forests. Most of the Early Neolithic sites belong
(Kajmakcalan), Belasica, Rila, Pirin, the Rhodopes and the Balkan to ecotones (i.e. transitional biomes) between lowlands and
Mountains themselves, the intensity of such contacts is indeed mountains (150e500 m asl), or wetlands and open oak forests
astonishing. Ultimately, a greater mobility must be assumed in this (0e150 m asl), which are particularly rich in various natural re-
area already in the 57th century calBC, at the latest, and which sources and are situated in the wider zone of the thermophilous,
developed fully independently of the “old” maritime networks of mixed deciduous, broad-leaved forests (Fig. 5). In fact, both the
the Aegean. Only the Vardar zone and Pelagonia, via the plain of the completely developed Star €ro
cevo-Criş-Ko €s-Karanovo IeII, as well
Aliakmon, were influenced by the primary neolithisation in the as Anzabegovo-Vrsnik and the early Impresso complexes are all
62nd century calBC. The point of departure therefore seems to have restricted to the SMA zone. Nevertheless, although this bio-
been the SMA vegetation zone, which in this case encompasses the geoclimatic zonation provides a general framework for the later
entire river valley. By contrast, the neolithisation of the zone of the expansion of the Neolithic, it does not explain the absence of the
Macedonian lakes of Prespa and Ohrid further west evidently earliest sites in Thrace and other regions of the Balkan Peninsula. In
occurred only later, in the first centuries of the 6th millennium the following and last section of our paper we address this question
calBC. But again, like in the Struma valley, the initial impulse to- in more detail.
wards neolithisation started off from the Aegean maritime net-
works and is clearly distinguishable from the terrestrial routes, 3.5. Mediterranean versus continental climate
above all in the Vardar zone (Fig. 4). Important substantiation of
this process, aside from pottery comparisons, is provided by the One main result of the GIS-studies, described above, was to
architectural contexts in Anza II and Tumba MaCari as well as the identify the strong regionality that underlies the spread of farming
plenteous use of Spondylus from the Aegean to make ornaments. in SE-Europe, in particular in its very earliest phases. The approach
Regarding the question of agency in the Neolithisation of the is now extended with the aim of correlating the archaeological data
Vardar zone and Pelagonia, the same obstacles arise as already (mainly: settlement distributions, pottery-stylistic mapping) with
encountered in the neighbouring Struma valley, to the East. It is the environmental and climatic data, first in terms of modern
difficult to imagine that the first herds of sheep, goats and cattle analogies for the hypothetical palaeo-vegetation, and then by
could have reached inland without guidance by shepherds, and this comparison with the environmental conditions that were actually
includes their movement as exchange goods. The appearance of a prevailing in SE-Europe during the Early Neolithic, based on a
comprehensive Neolithic package constituted by cultivated plants, compilation of terrestrial and marine palaeoclimatic records. Con-
domesticated animals, advanced pottery production, and house cerning the southern parts of SE-Europe, the present situation is
construction is explainable only through the immigration of a that droughts are more important factors than frost, whereas
population that was practicing an already fully Neolithic economy. moving to the north, frost (especially in early spring) plays a much
Although this population must have arrived by the Mediterranean more decisive role. The reconstruction of the temperatures during
Sea, the relatively short period of time during which the Neolithic the period following 6050 calBC shows that in SE-Europe the
further disseminated via the land route, including the expansion winter temperatures were typically almost at present-day values
from one fertile (but initially forested) flood-plain to the next and summers were slightly cooler (Davis et al., 2003: 1711).
through narrow river valleys and mountainous terrain, is indeed Therefore, as goes for the periods following the earliest Neolithic,
noteworthy. There are no traces in Macedonia of any Mesolithic the climatic conditions in SE-Europe appear to have been more or
settlement. Whether such settlements existed or not, and in our less comparable with today. However, this does not apply to the
view the second assumption is the most likely, they played no part, previous period (~6550-6050 calBC), for which there exists a
or at least not any archaeologically discernible part, in the primary wealth of strong evidence for the regular occurrence of polar air
spread of the Neolithic in this area. outbreaks during late winter or early spring, both in modern times
(Saaroni et al., 1996) as well as in prehistoric periods (Marino et al.,
3.4. Examination of the expansion hypothesis 2009; Pross et al., 2009; Rohling et al., 2002). We will come back to
the meteorological discussion of corresponding RCC winter/spring
Finally, let us now re-examine the expansion hypothesis in the conditions below, and for the moment put further focus on the
light of the results put forward above, which indicate that the palaeobotanical aspects of SE-European winter and summer
geographical position of the earliest Neolithic settlements shows conditions.
strong relations to certain bioclimatic properties of the present Olives, representing a modern crop but which can be used as
natural environment. To gain more insight into the bioclimatic sensitive indicator also of the Mediterranean palaeo-climate, do not
regulation of archaeological site locations, we have used GIS soft- endure the harsh northern Greek and Bulgarian winters, and from a
ware to extract the environmental characteristics of these sites comparative regional perspective this is due to the existence of a
based on available European geospatial data, i.e. the high- strong climatic and vegetational gradient that separates these re-
resolution SRTM digital elevation model (Jarvis et al., 2008), ESDB gions from the milder Aegean coastal regions. Naturally, therefore,
soil data (Panos, 2006) and EuroVegMap vegetation maps (Bohn both the crops and associated methods of food-processing intro-
et al., 2003). The investigated parameters include elevation above duced from the Mediterranean and Near East into these northern
sea level, aspect of the slope, soil texture and drainage, proximity to regions of the Balkans would have had to be adapted to such cold-
major rivers, potential natural vegetation, and several climatic winter conditions, hence, we now perceive these adaptations as
variables. The results indicate that the majority of sites is located on representing an important agricultural transition that must have
well-drained soils with fine to medium textures, and situated occurred somewhere between the eastern Mediterranean and the
within walking distance of rivers. In quantitative terms, more than Balkans during the very earliest phase of the Early Neolithic. The
50% of the sites are situated closer than 6 km to the nearest major most suitable corridor for such an environmental adaptation to
river, with an exception maximum distance of ca. 20 km for the site take place smoothly is the area of the northwestern Aegean
Fig. 4. Neolithisation of SE-Europe via the SMA vegetation zone. Red: before 6050 calBC; orange: before 5900 calBC. 1 Çukuriçi Ho € yük, 2 Dedecik-Heybelitepe, 3 Yeşilova, 4 Ulucak, 5
Ege Gübre, 6 Musluçeşme, 7 Aktopraklık, 8 Barcin Hüyük, 9 Menteşe, 10 Ilıpınar, 11 Pendik, 12 Fikirtepe, 13 Istanbul-Yeniko € y, 14 Yarımburgaz, 15 Toptepe, 16 Hoca Çeşme, 17 Argissa,

18 Otzaki, 19 Sesklo, 20 Achilleion, 21 Servia, 22 Nea Nikomedeia, 23 Podgorie, 24 Veluska Tumba, 25 Cuka, 26 Vrsnik, 27 Anzabegovo, 28 Grncarica, 29 Tumba MaCari, 30 Kovacevo,
31 Vaksevo, 32 Nevestino, 33 Sapareva Banja, 34 Dzhuljunica-Smardes, 35 Poljanica-Platoto, 36 Orlovec, 37 Koprivec, 38 Gradesnica-Malo pole, 39 Jelasnica, 40 Kucajna, 41 Ornice-
Makresani, 42 Blagotin, 43 Divostin, 44 Grivac, 45 Banja AranCelovac, 46 Lepenski Vir, 47 Padina
(Thessaly) and further, in the lower river valleys of Vardar, Struma appeared in regions that have, amongst others, a higher average
and Mesta. temperature of the coldest month, approximately similar to that in
Looking at the modern vegetation in this region, we see it is the Mediterranean. Such conditions are most likely to be present in
dominated by Albanian-Macedonian-Greek, mixed Oriental the Sub-Mediterranean biogeographic region. Once established (for
hornbeam-downy oak forests, that are clearly of sub- bioclimatic reasons) in such regions, the further expansion to the
Mediterranean character and which are indeed representative for North would naturally have followed the north-south oriented
mild winters (Fig. 6 and Table 1). But now looking further to the river systems of the Central Balkans.
East, beyond the fertile plains of northern Greece, the question In support of these qualitative model expectations, finally, we
arises as to why the earliest Early Neolithic e when moving North e have organised the data such that it is capable of making further,
did apparently not pass through Thrace? Instead, as it appears from quantitative predictions. To this purpose, we have used the same
data assembled in Table 1, the expansion to the North only took environmental and climatic parameters as introduced above. Using
place along the north-south oriented river systems of the Central the method of Single Linkage Cluster Analysis (SLCA) which is in-
Balkans. Again, a comparison of the modern vegetation in both tegrated in the PAST-statistics package (Hammer et al., 2001), we
regions provides us with an important key to addressing this obtained the results shown in Fig. 7. SLCA is a nearest-neighbour
question. Today, the Thracian Oriental hornbeam-downy oak for- type of analysis in which the clusters are shown in a dendrogram
ests are well-adapted to the prevailing strongly continental influ- for which the similarities and distinct splitting levels are defined
ence with regular winter frosts and an average temperature of the according to the smallest distances between the paired elements.
coldest month ca. 2e3 C lower than in the Sub-Mediterranean The distance matrix is computed by using Pearson's correlation
regions of the Central Balkans. Indeed, the first Neolithic groups coefficient as similarity measure, whereby the sample values and
Fig. 5. Distribution of the thermophiluous mixed deciduous broad leaved forests in the study region in comparison with the locations of the Neolithic sites (black triangles, site
reference see Fig. 7) considered. Extract of the Map of the Natural Vegetation of Europe, Scale 1:2,500,000 (Bohn et al., 2003).
their standard deviations are first centred by subtracting the mean, Branjevina, i.e. sites located in the flat regions of the southern
and then scaled by dividing by the standard deviation. An advan- Carpathian Basin, and, finally, the two Transylvanian sites Ocna
tage of the method is that it can be used with non-normalized data Sibiului and Gura Baciului. Due to a general lack of high-resolution
14
and can trace the trend in the matrix. The most important result of C-site chronologies in southern Europe, and notably also in The-
this study, in our view, is to demonstrate the existence of four ssaly, at the present state-of-research it is impossible to reliably
clearly separated clusters of archaeological sites. Put differently, interprete the chronological background of the first three site-
each site-cluster is now uniquely distinguished (only) in terms of clusters. What is presently only clear is that the fourth major
the associated environmental and climatic parameters. The first cluster (with sites from N-Bulgaria, the Central Balkans, and the
cluster contains the Thessalian sites (Achilleion, Argissa, Otzaki). Carpathian Basin) apparently covers Neolithic sites (we should
The next brings together the Macedonian site of Nea Nikomedeia better state: settlement environments) that were (exclusively)
with Hoca Çeşme (situated at the mouth of Marica River). The sites established later than ~6100 calBC i.e. following the end of the
of Anzabegovo and Vrsnik in the Vardar river valley falls in one Hudson-Bay-outflow (Weninger et al., 2009).
cluster with the sites of Kova
cevo and Krajnici (both situated in the
Struma river valley). These first three clusters are clearly distinct 3.6. The position of Thrace within the rapid climate corridor
from the fourth major cluster, that is formed by the Early Neolithic
sites in northern Bulgaria (Dzhuljunica, Koprivec, Orlovec and These results fit well with a variety of chronological observa-
Poljanica-Platoto), but which also contains sites in the Central tions, both in terms of pottery similarities as well as 14C-ages. For
Balkans (Lepenski Vir, Blagotin), along with Foeni-Salaş and Donja example, in Thessaly, the earliest (known) Early Neolithic
Fig. 6. Distribution of the Sub-Mediterranean to Meso-Mediterranenan oak forests (after Bohn et al., 2003) relevant for the study and location of the earliest Neolithic sites (black
triangles, site reference see Fig. 7).
settlement appears to begins after 6400 calBC (Reingruber, 2008: recent archaeological studies based on pottery synchronisms be-
316f.) and the pottery has an age similar to that of Hacɩlar in Ana- tween Anatolian and Greek sites (e.g. Çilingiroglu, 2010; Brami and
tolia (Brami and Heyd, 2011). Unfortunately, discussion of Heyd, 2011; Reingruber, 2011), and recent new 14C-ages for the
comparative stratigraphy sensu strictu (i.e. the combination of northern Greek sites of Mavropygi (Karamitroi-Mentessidi et al.,
architectural stratigraphies, pottery synchronisms, 14C-ages, and 2013; Мaniάth2, 2014) and Paliambela (Мaniάth2, 2014) provide
cultural relations) is strongly complicated by the almost complete further insights into the cultural developments within the Aegean,
lack of 14C-ages on stratified, short-lived samples in all our study inside the long RCC-interval.
regions (Thessaly, Greece, Anatolia). Nevertheless, in combination, What can be stated already now, at higher dating resolution
Table 1
Main climatic and environmental constrains for the vegetation types shown on Fig. 4.
Albanian-Macedonian Thracian
Average annual temperature: 12-13 C 10-12 C

Average annual precipitation: 550-600 mm 590-600 mm
Average temperature of the warmest month: July 23e25 C July 20e22 C
Average temperature of the coldest month: January 1e2 C January 2 to 0 C
Local climate or other climatic peculiarities: Warm, summer-dry climate, in the north Subcontinental-Sub-Mediterranean transition,
continentally influenced climate with frosts in the winter
Fig. 7. Results of the SLCA applied on the environmental parameters of the considered Neolithic sites. Environmental parameters given on Table 2.
(decadel scale), is that there exist some quite remarkable syn- to connect the observed geographic distribution of Early Neolithic
chronisms between the earliest arrival of the Neolithic at a number sites within the Sub-Mediterranean-Aegean Vegetation Zone, both
of large sites on the west coast of Turkey (Çukuriçi Ho €yük, Ulucak, with the observed 500 year delay in Neolithic dissemination, and
and only slightly later at Barcin) and its earliest arrival on the with the extreme climatic conditions that are known to be pre-
Peloponnese (Franchthi Cave) at around 6550 calBC. vailing during the 8.6e8.0 ka calBP interval of Rapid Climate
This arrival is synchronous (within error limits of ~100 yrs) with Change (RRC). As illustrated in the RCC climate records (Figs. 8e9),
the onset of RCC-conditions in the Eastern Mediterranean. On the from the climatological perspective, the very existence of the SMA
other hand, the departure from the Aegean, and in particular the Vegetation Zone is related to its geographic position simulta-
further dispersal of the Neolithic in northern direction through the neously in close proximity to the warm Aegean, as well as within
Central Balkans into the Pannonian Basin only occurs after 6050 the geographic corridor for influx of cold air masses from the polar
calBC i.e. immediately (1e3 generations) following the end of RCC regions into the Aegean during RCC-times (Weninger and Harper,
conditions (Weninger et al., 2014). 2015). Of archaeological interest, the RCC-winds have their main
Notwithstanding remaining problems of ‘climate determinism’, occurrence during late winter and early spring, at which times we
that for some of our colleagues are still under discussion (e.g. may expect the farming communities to have highest cold-climatic
Berger et al., 2016), in our view it is now only a small remaining step vulnerability. Geographically, the RCC-corridor for polar air begins
Table 2
Environmental parameters for the sites from the used GIS database compiled from the data on available European geospatial data, high-resolution SRTM digital elevation
models (Jarvis et al., 2008), ESDB soil data (Panos, 2006) and European vegetation maps (Bohn et al., 2003).
EuroVegMap SRTM SRTM Distance ESDB soil ESDB soil Average temperature Climate
vegetation aspect elevation to major texture drainage of the coldest month
class (m a.s.l.) rivers (m)
Achilleion G57 SE-SW 286 5672 Medium dry 1 to 2 C Warm, summer-dry climate, in
the north continantally
influenced
Argissa U35 NE-SE 70 5080 Fine dry 4 to 9 C long-term summer droughts in
the east Balkan; areas with
continentally tinged
Mediterranean climate
Otzaki U38 NE-SE 78 1208 Fine dry 7-12 C warm, Mediterranean with
mild winters and long dry
seasons in summer
Nea Nikomedeia U18 NE-SE 7 1063 Medium dry 0 to 2 C mild, moderately continental
tinged, summer-warm climate
Hoca Çeşme U36 SE-SW 10 2193 Very fine dry 0 to 1 C summer-warm, continentally
tinged climate
Kova
cevo G57 SE-SW 493 15,138 Fine dry 1 to 2 C Warm, summer-dry climate, in
influenced
Vrsnik G57 SW-NW 282 4684 Coarse No Data 1 to 2 C Warm, summer-dry climate, in
influenced
Anzabegovo U35 NW-NE 323 8846 Medium No Data 4 to 9 C long-term summer droughts in
the east Balkan; areas with
continentally tinged
Mediterranean climate
Krajnici U18 SW-NW 617 20,142 Fine dry 0 to 2 C mild, moderately continental
tinged, summer-warm climate
Dzhuljunica F67 NW-NE 66 1063 Medium fine dry 1 to 2 C Warm, moderately continental
climate
Poljanica-Platoto F67 NE-SE 241 15,501 Medium fine dry 1 to 2 C Warm, moderately continental
climate
Orlovec F68 NE-SE 305 5800 Medium fine dry 1 to 3 Warm, moderately continental
climate
Koprivec G22 SE-SW 176 13,829 Medium fine dry 0 to 3 C Moderately continental and
continental-sub-Mediterranean
transition climate
Blagotin G19 SW-NW 259 9415 No Data No Data 2 to 0 C warm, dry summers
Lepenski Vir G8 SE-SW 391 2594 Coarse No Data 2 to 1 C warm and dry microclimate on
sunexposed slopes in the south
Donja Branjevina F47 NW-NE 79 13,789 Medium fine No Data 0 to 2 C
Foeni-Salaş P32 SE-SW 76 5964 Medium fine dry 7 to 3 C warm and dry; annual climatic
water deficiencies on more
than 150 days; high frequency
of winds and strength
Ocna Sibiului G3 SW-NW 535 13,652 Medium dry 5 to 4 C opposing local climatic
conditions from cool to
moderately dry on northern
slopes to hot to dry on steep
south exposed slopes
Gura Baciului G16 NW-NE 429 19,431 Medium humid to 4 to 1 C In late summer an enduring
very humid drought of several weeks is
possible
in regions north of the Himalaya, runs through the North Pontic from recognition that Thrace is one of the main terrestrial regions
steppe regions, crosses the Lower Danube, and runs from there into for entry of RCC winds into the north-east Aegean. This hypothesis
northern Greece. Following orographic channeling through the is further indicated by the unusually high frequency and severity of
Balkans, the cold winds then proceed southwards across the polar outbreaks that are known to occur in the north-eastern
Aegean Sea (cf. Weninger et al., 2014). In combination with results corner of the Mediterranean basin (Saaroni et al., 1996). Alto-
put forward above, we may now expect the geographically complex gether, it can hardly be a ‘chance’ coincidence that the Neolithic
orographic channeling of cold air masses through the Balkans to disseminated both into and out of the Aegean exactly in phase
contribute to the observed large micro-climatic vegetational vari- (within given error limits) with begin and end of RCC-conditions.
ability. Major changes both in terrestrial and marine ecosystems Already by ca. 5900 calBC, Neolithic settlement can be confirmed
during the 8.6e8.0 ka calBP RCC-interval period are already now in all landscapes of the Balkans, as far North as the southern Car-
documented in a large number of archives and this motivates pathian Basin, and even in Thrace for the first time.
further studies as to the micro-regional vegetational impact of RCC.
Our present studies may provide guidance for such studies, in that 4. Conclusions
Thrace appears to be the region most strongly affected by RCC. This
is presently indicated by the archaeological data, and follows also In the present paper we introduce a possible explanation for
Fig. 8. Selected terrestrial and marine palaeoclimate records for the eastern Mediterranean and southeast Europe that show Holocene Rapid Climate Change (RCC). Greenland
records are added to provide atmospheric RCC-timing at higher temporal resolution. (A) Greenland GRIP ice-core d18O, on GICC05 age model, as proxy inter alia (in the following
abbreviated: proxy.) for North Atlantic Air Temperature, showing Hudson-Bay outflow 8.2e8.0 ka calBP (Grootes et al., 1993); (B) Sofular Cave (41.4166 N; 31,9333 E) d13C
(Fleitmann et al., 2009) proxy for rainfall amount during the 8.6e8.0 RCC-interval (Go € ktürk et al., 2011); (C) Steregoiu (47.8138 N; 23.5318 E) Vegetation-based mean annual
temperature of the oldest month (Feurdean et al., 2008); (D) Tenaghi Philippon (40.9733 N; 24.2236 E) tree pollen proxy for climatically induced vegetation change during the
Hudson-Bay interval (Pross et al., 2009); (E) Lake Prespa (40.8833 N, 21.0222 E) Total Organic Carbon content (wt %) proxy for organic accumulation and lower temperatures
(Panagiotopoulos, 2013); (F) Lake Ohrid (41.0419 N, 21.7190 E) Weight Percentage of CaCO3 as proxy for climatically induced environmental changes (here: cooling) (Holtvoeth
et al., 2010); (G) Benthic Foraminiferal Oxygen Index (BFOI) for North Aegean marine core SL 148 (39 7533 N, 24.0966 E, 1094 m water depth)proxy for deep-water ventilation due
to changes in precipitation (Kuhnt et al., 2007); (H) BFOI for south Aegean marine core SL123 (35.7550 N, 27.5550 E, 728 m water depth) proxy for deep-water ventilation due to
changes in precipitation (Kuhnt et al., 2007); (I) BFOI for south Aegean marine core LC21 (32.67,417 N, 34.6500 E, 892 m water depth) proxy for deep-water ventilation due to
changes in precipitation (Schmiedl et al., 2010); (J) Sea Surface Temperatur for marine core LC21proxy for strength of upper water column cooling due to RCC-winds (Rohling et al.,
2002); (K) Gaussian smoothed (200yrs) GISP2 nss [Kþ] proxy for the Siberian High (Mayewski et al., 1997); (L) High-Resolution GISP2 nss [Kþ] proxy for the Siberian High
(Mayewski et al., 1997).
certain geographic preferences that are apparent in the pathways observations. Starting point is the observation that the farmers/
used by Neolithic communities during their initial colonization of herders initially avoid the foundation of settlements in Thrace.
the Aegean and southern Europe. Our studies are based on GIS Although to some extent speculative due to missing data (and in
analysis of environmental factors related to site distribution, and particular in lack of high-resolution 14C-ages for the majority of
on palaeo-climatological observations aside with modern analo- study sites), our approach is nevertheless aimed at achieving a
gies, both in combination with a wide variety of archaeological deeper understanding of the processes underlying the spread of
Fig. 9. Map showing location of paleoclimate records in Fig. 8.
the oldest Early Neolithic out of the circum-Aegean regions into biogeographic region), also existed at the time of the 6550-6050
the Balkans. The modelling is based upon comparisons of the calBC RCC-event, then we can simultaneously explain both the
modern natural environments in the southern Balkans and the long (~500 year) delay of the Neolithic in the circum-Aegean re-
Aegean with the (assumed) corresponding prehistoric settlement gions, as well as the extremely rapid (max 200 yrs) and long-
patterns. With this approach, a number of close similarities be- distance (~2000 km) further dissemination of early farming,
tween regional microclimates and archaeological patterns become immediately following the end of the RCC-interval. This further
apparent, and in particular the vegetational patterns provide an Neolithic expansion to the North followed the north-south ori-
explanation for the observed absence of oldest Early Neolithic ented river systems of the Central Balkans. In order to live in re-
settlements in Thrace. As can be observed in modern times, the gions beyond the Aegean with an economy based on agriculture
Thracian Oriental hornbeam-downy oak forests are exposed to a and stockbreeding solely, the Neolithic package first had to be
much stronger continental influence than those measured for SMA adapted to conditions of the relatively harsh winters in the Bal-
vegetation in the Central Balkans, with extensive frosts in the kans. Only following this period of climatic and environmental
winter and average temperatures of the coldest month that are ca. adaptation, it was possible to apply the Neolithic lifestyle to the
2e3 C lower. The existence of such milder Sub-Mediterranean neighbouring areas, such as Thrace, Walachia, Dobrudzha and the
conditions in parts of the Balkans, although not widely acknowl- Carpathian Basin.
edged, could be one of the main reasons why the first Neolithic
groups initially appeared in these very specific regions, which
References
would have had all-year-round conditions quite similar to those
even on the Mediterranean coasts. Assuming that such conditions, € a
Alram-Stern, E., 1996. Die Ag €ische Frühzeit. 2. Serie, Bd. 1, Das Neolithikum in
as can be observed today in the Balkans (far-inland in the SMA Griechenland mit Ausnahme von Kreta und Zypern (Wien 1996).
Ammermann, A., Cavalli-Sforza, L.L., 1984. The Neolithic and the Genetics of Pop- Halstead, P., Isaakidou, V., 2013. Early stock-keeping in Greece. In: Colledge, S.,
ulations in Europe (Princeton 1984). Conolly, J., Dobney, K., Manning, K., Shennan, S. (Eds.), The Origins and Spread of
Andreou, S., Fotiadis, M., Kotsakis, K., 2001. The neolithic and bronze age of Domestic Animals in Southwest Asia and Europe, pp. 129e143.
northern Greece. In Aegean Prehistory: a Review. In: Cullen, T. (Ed.), American Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. Past: paleontological statistics software
Journal of Archaeology, Supplement vol. 1. Archaeological Institute of America, package for education and data analysis. Palaeontol. Electron. 4 (1), 9.
Boston, pp. 259e327. Hofmanova , Z., Kreutzer, S., Hellenthal, G., Sell, Ch, Diekmann, Y., Díez-del-
Aurenche, O., Kozłowski, S.K., 1999. La naissance du Ne olithique au Proche Orient ou Molino, D., van Dorp, L., Lo pez, S., Kousathanas, A., Link, V., Kirsanow, K.,
le paradis perdu (Paris 1999). Cassidy, L.M., Martiniano, R., Strobel, M., Scheu, A., Kotsakis, K., Halstead, P.,
Bajno czi, B., Scho € ll-Barna, G., Kalicz, N., Siklo
si, Zs, Hourmouziadis, G.H., Ifantidis, F., Triantaphyllou, S., Kyparissi-Apostolika, N., Urem-Kotsou, D.-Ch, Ziota, Ch,
Kyparissi-Apostolika, A., Pappa, M., 2013. Tracing the source of Late Neolithic Adaktylou, F., Gopalan, Sh, Bobo, D.M., Winkelbach, L., Blo €cher, J.,
Spondylus shell ornaments by stable isotope geochemistry and cath- Unterla €nder, M., Leuenberger, Ch, Çilingirog lu, Ç., Horejs, B., Gerritsen, F.,
odoluminescence microscopy. J. Archaeol. Sci. 40, 874e882. Shennan, S., Bradley, D.G., Currat, M., Veeramah, K.R., Wegmann, D.,
Bar-Yosef, O., 2011. Climatic fluctuations and early farming in west and east asia. Thomas, M.G., Papageorgopoulou, Ch, Burger, J., 2016. Early farmers from across
Curr. Anthropol. 52 (S4), S175eS193. europe directly descended from Neolithic Aegeans. PNAS 113 (25), 6886e6891.
Benz, M., 2000. Die Neolithisierung im Vorderen Orient. Theorien, archa €ologische http://dx.doi.org/10.1073/pnas.1523951113.
Daten und ein ethnologisches Modell (Berlin 2000). Holtvoeth, J., Vogel, H., Wagner, B., Wolff, G.A., 2010. Lipid biomarkers in Holocene
Berger, J.-F., Lespez, L., Kuzucuoglu, C., Glais, A., Hourani, F., Barra, A., Guilaine, J., and glacial sediments from ancient Lake Ohrid (Macedonia, Albania). Bio-
2016. Interactions between climate change and human activities during the geosciences 7, 3473e3489. http://dx.doi.org/10.5194/bg-7-3473-2010.
early to mid-Holocene in the eastern Mediterranean basins. Clim. Past. 12, Horejs, B., Mili c, B., Ostmann, F., Thanheiser, U., Weninger, B., Galik, A., 2015. The
1847e1877. aegean in the early 7th millennium BC. Maritime networks and colonization.
Bergner, M., Horejs, B., Pernicka, R., 2009. Zur Herkunft der Obsidianartefakte von J. World Prehistory 28 (4), 289e330.
Çukuriçi Ho € yük. Stud. Troica, Band. 18, 249e271. Jarvis, A., Reuter, H.I., Nelson, A., Guevara, E., 2008. Hole-filled SRTM for the globe
Beuermann, A., 1967. Fernweidewirtschaft in Südosteuropa. Ein Beitrag zur Kul- Version 4. Available from the CGIAR-CSI SRTM 90m http://srtm.csi.cgiar.org.
turgeographie des o € stlichen Mittelmeergebietes. Westermann, München 1967. Kanzurova, E., Zdravkovski, D., 2011. Latest archaeological research regarding the
Bohn, U., Gollub, G., Hettwer, C., Neuh€ auslov a, Z., Raus, T., Schlüter, H., Weber, H., neolithic period in the republic of Macedonia. In: Krauß, R. (Ed.), Beginnings e
2003. Karte der natürlichen Vegetation Europas/map of the natural vegetation New Research in the Appearance of the Neolithic between Northwest Anatolia
of Europe. Maßstab/Scale 1 : 2 500 000. Landwirtschaftsverlag, Münster. and the Carpathian Basin. Menschen e Kulturen e Traditionen. Studien aus den
Bollongino, R., Elsner, J., Vigne, J.-D., Burger, J., 2008. Y-SNPs do not indicate Forschungsclustern des Deutschen Archa €ologischen Instituts 1 (Rahden 2011),
hybridisation between european aurochs and domestic cattle. PLoS One 3 (10), pp. 139e155.
e3418. Karamitroi-Mentessidi, G., Efstratiou, N., Kozłowski, J.K., Kaczanowska, M.,
Bollongino, R., Burger, J., Powell, A., Mashkour, M., Vigne, J.-D., Thomas, M.G., 2015. Maniatis, Y., Curci, A., Michalopoulou, S., Papathanasiou, A., Valamoti, S.M.,
Modern taurine cattle descended from small number of near-eastern founders. 2013. New evidence on the beginning of farming in Greece: the Early Neolithic
Mol. Biol. Evol. 29 (9), 2101e2104. settlement of Mavropigi in western Macedonia (Greece). Antiquity 087 (335)
Braidwood, R.J., 1952. The Near East and the Foundations for Civilization: an Essay (zizertaoleRar, reecMew).
in Appraisal of the General Evidence (Oregon 1952). Karul, N., Avcı, M., 2013. Aktopraklık,. In: Ozdo€ an, M., Başgelen, N., Kuniholm, P.
g
Brami, M., Heyd, V., 2011. The origins of Europe's first farmers: the role of Hacılar (Eds.), The Neolithic in Turkey, vol. 5, pp. 45e68. Istanbul 2013.
and Western Anatolia, fifty years on. Praehistorische Z. 86/2 (2011), 165e206. Krahtopoulou A. and Veropoulidou R. Late Pleistocene-Holocene shoreline recon-
Cikovac, P., 2002. Soziologie und standortbedingte Verbreitung tannenreicher struction and human exploitation of molluscan resources in northern Pieria,
Wa €lder im Orjen Gebirge (Montenegro) (München 2002). Macedonia, GreeceJ. Archaeol. Sci.2016(in press), http://dx.doi.org/10.1016/j.
Çilingiroglu, Ç., 2010. The appearance of impressed pottery in the neolithic aegean jasrep.2016.02.003
and its implications for maritime networks. TÜBA-AR 13, 9e22. Krauß, R., 2011. On the ‘monochrome’ neolithic in southeast Europe. In: Krauß
Clare, L., 2010. Pastoral clashes: conflict risk and migration at the pottery neolithic (Hrsg), R. (Ed.), Beginnings e New Research in the Appearance of the Neolithic
transition in the southern levant. Neo-Lithics 1 (2010), 13e31. between Northwest Anatolia and the Carpathian Basin. Menschen e Kulturen e
Clare, L., 2016. Culture Change and Continuity in the Eastern Mediterranean during Traditionen. Studien aus den Forschungsclustern des Deutschen Arch-
Rapid Climate Change: Assessing Impacts of a Little Ice Age in the 7th Millen- €
aologischen Instituts 1 (Rahden 2011), pp. 109e125.
nium CalBC. Ko €lner Studien zur Pr€ ahistorischen Archa €ologie 7. VML Verlag Krauß, R., 2016. Zur Symbolik des Rindes im europa €ischen Neolithikum. In:
Marie Leidorf GmbH, Bonn. Bachvarov, K., Gleser, R. (Eds.), Southeast Europe and Anatolia in Prehistory:
Clare, L., Weninger, B., 2014. The dispersal of neolithic lifeways: absolute chronology Essays in Honor of Vassil Nikolov on His 65th Anniversary, vol. 293. Uni-
and rapid climate change in central and west Anatolia. In: Ozdo € an, M.,
g versita €tsforschungen zur Pr€ ahistorischen Archa €ologie, pp. 233e251. Bonn 2016.
Başgelen, N., Kuniholm (Hrsg), P. (Eds.), The Neolithic in Turkey, vol. 6, pp. 1e65. Krauß, R., 2017. Nordwestanatolien, Balkan und Karpatenbecken im diachronen
Istanbul 2014. Vergleich e kulturkontakt und Kommunikation vom 6.-4. Jt.v.Chr. In:
Davis, B., Brewe, S., Stevenson, A., Guiot, J., 2003. Data Contributors. The tempera- Schimmelpfennig, D., Link, T. (Eds.), Fokus Jungsteinzeit e Berichte der AG
ture of Europe during the Holocene reconstructed from pollen data. Quat. Sci. Neolithikum 5 in press; Kerpen-Loogh.
Rev. 22, 1701e1716. Krauß, R., Elenski, N., Weninger, B., Clare, L., Çakırlar, C., Zidarov, P., 2014. Beginnings
Feurdean, A., Klotz, S., Mosbrugger, V., Wohlfarth, B., 2008. Pollen-based quantita- of the neolithic in southeast Europe. The early neolithic sequence and absolute
tive reconstructions of Holocene climate variability in NW Romania. Palae- dates from D zuljunica-smardes (Bulgaria). Doc. Praehist. XLI, 51e77.
oclimatol. Palaeogeogr. Palaeoecol. 260, 494e504. Kuhnt, T., Schmiedl, G., Ehrmann, W., Hamann, Y., Hemleben, C., 2007. Deep-sea
Fleitmann, D., Cheng, H., Badertscher, S., Edwards, R.L., Mudelsee, M., Go €ktürk, O.M., ecosystem variability of the aegean sea during the past 22 kyr as revealed by
Fankhauser, A., Pickering, R., Raible, C.C., Matter, A., Kramers, J., Tüysüz, O., benthic foraminifera. Mar. Micropaleontol. 64, 141e162.
2009. Timing and climatic impact of Greenland interstadials recorded in sta- Lespez, L., Dalongeville, R., Noirel-Schutz, C., Suc, J.-P., Koukouli-Chryssanthaki, H.,
lagmites from northern Turkey. Geophys. Res. Lett. 36, L19707. http://dx.doi.org/ Treuil, R., 2000. Les pale oenvironnements du site pre historique de Dikili Tash
10.1029/2009GL040050. (Mace doine orientale, Gre ce). 2. In: Bulletin de correspondance helle nique, 124,
Geo €rg, Ch, 2013. Pala €opopulationsgenetik von Schwein und Schaf in Südosteuropa pp. 413e434 livraison.
und Transkaukasien. Menschen e Kulturen e Traditionen. Studien aus den Lespez, L., Tsirtsoni, Z., Darcque, P., Koukouli-Chryssanthraki, H., Malamidou, D.,
Forschungsclustern des Deutschen Arch€ aologischen Instituts 4 (Rahden 2013). Treuil, R. and Oberlin, C., 2013. The lowest levels at Dikili tash, northern Greece:
€
Gerritsen, F., Ozbal, R., Thissen, L., 2013. Barcın Ho €yük. The beginnings of farming in a missing link in the Early Neolithic of Europe. Antiquity 87: 30e45.
the Marmara region. In: Ozdog € an, M., Başgelen, N., Kuniholm (Hrsg), P. (Eds.), Lichardus-Itten, M., Demoule, J.-P., Perni ceva, L., Grebska-Kulova, M., Kulov, I., 2002.
The Neolithic in Turkey, vol. 5, pp. 93e112. Istanbul 2013. The site of Kova cevo and the beginning of the neolithic period in southwestern
Ghilardi, M., Fouache, E., Queyrel, F., Syrides, G., Vouvalidis, K., Kunesch, S., Bulgaria. In: Lichardus-Itten, M., Lichardus, J., Nikolov, V. (Eds.), Beitra €ge zu
Styllas, M., Stiros, S., 2008. Human occupation and geomorphological evolution jungsteinzeitlichen Forschungen in Bulgarien. Saarbrücker Beitr€ age zur Alter-
of the Thessaloniki Plain (Greece) since mid Holocene. J. Archaeol. Sci. 35, tumskunde, 74, pp. 99e158. Bonn 2002.
111e125. Lichardus-Itten, M., Demoule, J.-P., Pernicheva, L., Grebska-Kulova, M., Kulov, I.,
Gimbutas, M., Winn, Sh, Shimabuku, D., 1989. Achilleion. A Neolithic Settlement in 2006. Kova cevo, an early neolithic site in south-west Bulgaria and its impor-
Thessaly, Greece, 6400e5600 BC (Los Angeles 1989). tance for european neolithization. In: Gatsov, I., Schwarzberg, H. (Eds.), Aegean
Gimbutas, M. (Ed.), 1976. Neolithic Macedonia as reflected by Excavation at Anza, e Marmara e Black Sea: the Present State of Research on the Early Neolithic.
Southeast Yugoslavia (Los Angeles). Schriften des Zentrums für Arch€ aologie und Kulturgeschichte des Schwarz-
Go€ ktürk, O.M., Fleitmann, D., Badertscher, S., Cheng, H., Edwards, R.L., meerraumes, vol. 5, pp. 83e94. Langenweissbach 2006.
Leuenberger, M., Fankhauser, A., Tüysüz, O., Kramers, J., 2011. Climate on the Lichter, C., 2011. Neolithic stamps and the neolithization process. A fresh look at an
southern black sea coast during the Holocene: implications from the sofular old issue. In: Krauß, R. (Ed.), Beginnings e New Research in the Appearance of
Cave record. Quat. Sci. Rev. 30, 2433e2445. the Neolithic between Northwest Anatolia and the Carpathian Basin. Menschen
Grootes, P.M., Stuiver, M., White, J.W.C., Johnsen, S., Jouzel, J., 1993. Comparison of e Kulturen e Traditionen. Studien aus den Forschungsclustern des Deutschen
oxygen isotope records from the GISP2 and GRIP Greenland ice core. Nature Archa €ologischen Instituts 1 (Rahden 2011), pp. 35e44.
366, 552e554. Malinowski, B., 1922. [1979]. Argonauts of the Western Pacific. Routledge and Kegan
Guilaine, J., 2000/2001. La diffusion de lagriculture en Europe: une hypothe se Paul Ltd. German Edition: Argonauten des westlichen Pazifik. Ein Bericht über
arythmique. Zephyrus 53/54, 267e272. Unternehmungen und Abenteuer der Eingeborenen in den Inselwelten von
Melanesisch-Neuguinea. Syndikat, Frankfurt am Main 1979. Resilience at the transition to agriculture: the long-term landscape and
Marino, G., Rohling, E.J., Sangiorgi, F., Hayes, A., Casford, J.L., Lotter, A.F., Kucera, M., resource development at the aceramic neolithic tell site of chogha golan (Iran).
Brinkhuis, H., 2009. Early and middle Holocene in the aegean sea: interplay BioMed Res. Int. 2015, 22.
between high and low latitude climate variability. Quat. Sci. Rev. 28 (27e28), Rohling, E.J., Mayewski, P., Abu-Zied, R., Casford, J., Hayes, A., 2002. Holocene
3246e3262. atmosphere-ocean interactions: records from Greenland and the aegean sea.
Marinova, E., 2006. Vergleichende pal€ aeoethnobotanische Untersuchung zur Veg- Clim. Dyn. 18 (7), 587e593.
etationsgeschichte und pr€ ahistorischen Landnutzung in Bulgarien. Dis- Roodenberg, J., Alpaslan-Roodenberg, S., 2013. Ilıpınar and Menteşe. Early farming
sertationes Botanicae. J. Cramer in Gebr. Borntraeger Verlagsbuchhandlung. communities in the eastern Marmara. In: Ozdo € gan, M., Başgelen, N., Kuniholm
Science Publishers (Stuttgart) nr.401. (Hrsg), P. (Eds.), The Neolithic in Turkey, vol. 5, pp. 69e91. Istanbul 2013.
Matsanova, V., 1996. Cult objects from the early neolithic site at the town of Saaroni, H., Bitan, A., Alpert, P., Ziv, B., 1996. Continental polar outbreaks into the
Rakitovo. Poro cilo o raziskovanju paleolitika, neolitika eneolitika v Sloveniji 23, levant and eastern Mediterranean. Int. J. Climatol. 16, 1175e1191.
105e127. Scheu, A., 2012. Palaeogenetische Studien zur Populationsgeschichte von Rind und
Mayewski, P.A., Meeker, L.D., Twickler, M.S., Whitlow, S., Yang, Q., Lyons, W.B., Ziege mit einem Schwerpunkt auf dem Neolithikum in Südosteuropa. Men-
Prentice, M., 1997. Major features and forcing of high- latitude northern schen e Kulturen e Traditionen. Studien aus den Forschungsclustern des
hemisphere atmospheric circulation using a 110,000- year long glaciochemical Deutschen Arch€ aologischen Instituts 4 (Rahden 2012).
series. J. Geophys. Res. 102, 26345e26366. Scheu, A., Powell, A., Bollongino, R., Vigne, J.-D., Tresset, A., Çakırlar, C., Benecke, N.,
Мaniάth2, G., 2014. Хrοnοlόghsh mε άnqraka-14 tun mεgάlun pοlitismikώn Burger, J., 2015. The genetic prehistory of domesticated cattle from their origin
allagώn sthn prοϊstοrikή Мakεdοnίa: prόs4atε2 εxεlίxεi2. In: Stefani, E., to the spread across Europe. BMC Genet. 16, 54.
Merousis, N., Dimoula, A. (Eds.), A Century of Research in Macedonia 1912-2012. Schmiedl, G., Kuhnt, T., Ehrmann, W., Emeis, K.-C., Hamann, Y., Kotthoff, U.,
International Conference Proceedings, Archaeological Museum of Thessaloniki, Dulski, P., Pross, J., 2010. Climatic forcing of eastern mediteranean ecosystems
22-24 November 2012: 205e222. during the past 22 000 years. Quat. Sci. Rev. 29, 3006e3020.
Mili
c, M., 2014. PXRF characterisation of obsidian from central Anatolia, the Aegean Srejovi
c, D., 1973. Lepenski Vir. Eine vorgeschichtliche Geburtssta €tte europa€ischer
and central Europe. J. Archaeol. Sci. 41, 285e296. Kultur. Erweiterte Ausgabe mit einem revidierten und erweiterten Origi-
Milojci
c, V., 1983. Otzaki-magula II. Das Mittlere Neolithikum. Die mittelneolithi- nalbeitrag von Hans Quitta (Bergisch Gladbach 1973).
sche Siedlung, Bonn. Theocharis, D., 1973. Neolithic Greece (Athens 1973).
Nikolov, V., 1990. Das Flusstal der Struma an der frühneolithischen Strasse von Todorova, H., 1990. Das Frühneolithikum Nordbulgariens im Kontext des ostbal-
Anatolien nach Mitteleuropa. In: Die Ersten Bauern. Ausstellungskatalog (Zür- kanischen Neolithikums. In: Die Ersten Bauern. Ausstellungskatalog (Zürich
ich 1990), pp. 63e69. 1990), pp. 71e76.
€
Ozdo an, M., 2011. An anatolian perspective on the neolithization process in the
g Urem-Kotsou, D., Papaioannou, A., Papadakou, T., Saridaki, N., Intze, Z., 2014. Pottery
Balkans. New questions, new prospects. In: Krauß, R. (Ed.), Beginnings e New and stylistic boundaries. Early and middle neolithic pottery in Macedonia. In:
Research in the Appearance of the Neolithic between Northwest Anatolia and Stefani, E., Merousis, N., Dimoula, A. (Eds.), A Century of Research in Prehistoric
the Carpathian Basin. Menschen e Kulturen e Traditionen. Studien aus den Macedonia 1912-2012 (Thessaloniki 2014), pp. 505e517.
Forschungsclustern des Deutschen Archa €ologischen Instituts 1 (Rahden 2011), Watkins, T., 2007. Der Naturraum in Anatolien. Ein Zusammenspiel von Klima,
pp. 23e33. Umwelt und Ressourcen. In: Siebenmorgen, H. (Ed.), Vor 12.000 Jahren in
€
Ozdo an, M., 2014. A new look at the introduction of the Neolithic way of life in
g Anatolien. Die € altesten Monumente der Menschheit. Ausstellungskatalog
Southeastern Europe. Changing paradigms of the expansion of the Neolithic Karlsruhe (Stuttgart 2007), pp. 37e47.
way of life. Doc. Praehist. 41 (2014), 33e49. Weninger, B., Harper, T., 2015. The geographic corridor for rapid climate change in
Panagiotopoulos, K., 2013. Late Quaternary Ecosystem and Climate Interactions in southeast Europe and Ukraine. In: Schultze, E. (Ed.), Neolithic and Copper Age
SW Balkans Inferred from Lake Prespa Sediments. Diss. Ko €ln. Available under. between the Carpathians and the Aegean, 31. Arch€ aologie in Eurasien, DAI,
http://kups.ub.uni-koeln.de/5518/. Berlin, pp. 475e505.
Panos, P., The European soil database (2006) GEO: connexion, 5 (7), pp. 32-33: The Weninger, B., Clare, L., Rohling, E., Bar-Yosef, O., Bo € hner, U., Budja, M.,
European Soil Database distribution version 2.0, European Commission and the Bundschuh, M., Feurdean, A., Gebel, H.G., Jo €ris, O., Linsta€dter, J., Mayewski, P.,
European Soil Bureau Network, CD-ROM, EUR 19945 EN, 2004 Mühlenbruch, T., Reingruber, A., Rollefson, G., Schyle, D., Thissen, L.,
s, C., 2001. The Early Neolithic in Greece: The First Farming Communities in
Perle Todorova, H., Zielhofer, Ch, 2009. The impact of rapid climate change on pre-
Europe (Cambridge 2001). historic societies during the holocene in the eastern mediterranean. Doc.
Pernicheva-Perets, L., Gre˛ bska-Kulow, M., Kolov, I., 2011. Balgarchevo. Prehist. Settl. Praehist. 36, 7e59.
I (Sofia 2011). Weninger, B., Clare, L., Gerritsen, F., Horejs, B., Krauß, R., Linst€ adter, J., Ozbal, R.,
Pross, J., Kotthoff, U., Müller, U.C., Peyron, O., Dormoy, I., Schmiedl, G., Kalaitzidis, S., Rohling, E.J., 2014. Neolithisation of the aegean and southeast Europe during
Smith, A.M., 2009. Massive perturbation in terrestrial ecosystems of the eastern the 6600e6000 calBC period of rapid climate change. Doc. Praehist. 41, 1e31.
mediterranean region associated with the 8.2 kyr BP climatic event. Geology 37 Zeder, M.A., 2008. Domestication and early agriculture in the mediterranean basin:
(10), 887e890. origins, diffusion, and impact. PNAS 105 (33), 11597e11604.
Reingruber, A., 2008. Die Argissa-Magula II. Das frühe und das beginnende mittlere Zohary, D., Hopf, M., Weiss, E., 2012. Domestication of Plants in the Old World,
Neolithikum im Lichte trans€ ag€aischer Beziehungen (Bonn 2008). fourth ed. Oxford University Press, Oxford.
Reingruber, A., 2011. Early Neolithic settlement patterns and exchange networks in Zvelebil, M., 2001. The agricultural transition and the origins of Neolithic society in
the Aegean. Doc. Praehist. XXXVIII, 291e305. Europe. Doc. Praehist. 28, 1e26.
Riehl, S., Asouti, E., Karakaya, D., Starkovich, B.M., Zeidi, M., Conard, N.J., 2015.

Quaternary International: Raiko Krauß, Elena Marinova, Hanne de Brue, Bernhard Weninger

Uploaded by

Copyright:

Available Formats

Quaternary International: Raiko Krauß, Elena Marinova, Hanne de Brue, Bernhard Weninger

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Quaternary International: Raiko Krauß, Elena Marinova, Hanne de Brue, Bernhard Weninger

Uploaded by

Copyright:

Available Formats

Quaternary International xxx (2017) 1e18

Contents lists available at ScienceDirect

1. Introduction spectrum of highly aesthetic pottery products e spread with the

Average annual temperature: 12-13 C 10-12 C

Fig. 9. Map showing location of paleoclimate records in Fig. 8.

You might also like