The Neolithic Diaspora in Europe

Peter Bogucki
School of Engineering and Applied Science
Princeton University
March 25, 1997

The spread of farming across Europe between 8,000 and 5,000 years ago was the result of two processs: the migration and dispersal of farmers and the adoption of crops and livestock by indigenous foragers. In some regions, one of these processes clearly predominated, while in others is it not clear which one played the major role. This paper will discuss the geography of agricultural dispersals in Europe and identify some signature characteristics which can be used to differentiate the above two processes, based on recent research in several parts of Europe.

Introduction: Agricultural Dispersal in Europe

Agricultural communities were established across Europe between 8,000 and 5,000 years ago. The best evidence currently available to archaeologists indicates that two different processes were involved: the colonization of new habitats by populations of farmers and the adoption of agriculture by indigenous foragers (Bogucki 1996). Despite efforts of archaeologists to clarify which of these processes was active in a particular region, there is still considerable regional debate between those who favor colonization and those who argue for in situ development. Even when the discussion in a particular region appears to have reached closure, such as in central Europe, someone will often suddenly take a contrarian position and re-open the debate (e.g., Whittle 1996).

This paper will examine the question of the Neolithic diaspora in Europe, with special reference to central Europe. I have chosen to use the term "diaspora" deliberately, for I am persuaded that the dispersal of people, and not simply of cultivated plants and domesticated animals, was a key element in the rapid establishment of farming communities throughout Europe. Even when indigenous foragers chose to adopt agriculture, the appearance of farming communities in adjacent regions was often the conduit which brought them the founder crops and animals that eventually transformed their economy.

The presence of domesticated wheat, barley, sheep, and goat unequivocally provides the vector for the Neolithic diaspora in Europe. Since these species exist in their wild form only in the Near East, their dispersal was clearly from the southeastern corner of Europe to the northwest and to the west. Domestic livestock, of course, can move on their own, and the role of feral animals in expanding the range of domesticated species in Europe cannot be overlooked (Bogucki 1995). On the other hand, domestic plants require human agency to be carried across the landscape, so the presence of wheat and barley by about 3500 BC in the British Isles was the result of many human interactions over the 3500 kilometers between the Levant and the English Channel.

The Case for Colonization in Central Europe

In several parts of Europe, it has been argued that colonization by agricultural peoples was the process by which farming communities were established. In central Europe, this position is very much the orthodox view (Bogucki 1996), whereas in other parts of Europe it competes somewhat more evenly with the indigenous development hypothesis. Recently, van Andel and Runnels (1995) have argued that the plains of Thessaly were settled by immigrant farmers and that in general population movement played a key role in the introduction of agriculture to southeastern Europe. At the other end of the continent, Joao Zilhao (1993) has argued that the onset of agro- pastoral economies in Portugal is linked to the arrival of small groups of settlers, which contrasts markedly with the conventional wisdom in this region that indigenous foragers adopted agriculture. Zilhao takes the position that the mixing of deposits in the caves and rockshelters where the traces of the earliest farmers in the western Mediterranean are found gives a greater impression of continuity than is warranted.

In contrast to these views are the positions taken by what Ammerman (1989) has called the "indigenists". For example, Dennell (1992: 91) maintains that "early farming in Europe always occured in areas where there were already hunter-forager communities. These cannot be regarded as irrelevant to the pattern of agricultural expansion." Although I disagree with the inclusion of the word "always", it is true that foragers were indeed widely dispersed throughout pre-agricultural Europe. But it does not follow that the development of agricultural communities was always an in situ process. Alasdair Whittle has recently disputed the notion that colonization was a major factor in agricultural dispersal in Europe (Whittle 1996). He prefers instead to see the introduction of food production as an in situ development resulting from its adoption by indigenous peoples virtually everywhere in Europe.

With regard to the colonization vs. indigenous development question, it is helpful to consider the criteria laid down by Irving Rouse in 1958 when evaluating whether population movement is a better explanation for change in the archaeological record than in situ development:

  1. Identify the migrating people as an intrusive unit in the region they have penetrated;
  2. Trace this unit back to its homeland;
  3. Determine that all occurrences of this unit are contemporaneous;
  4. Establish the existence of favorable conditions for migration;
  5. Demonstrate that some other hypothesis, such as independent invention or diffusion of traits, does not better fit the facts.
    David Sanger (1975: 73) added a sixth rule:
  6. Establish that all cultural subsystems are involved and not just an isolated one (such as burial practice).

Let us examine these criteria, which still seem valid 40 years after Rouse first articulated them, with regard to the first farmers in north-central Europe, an area which reaches from the middle Danube valley where Hungary, Austria, and Slovakia come together, north to the lower Vistula and Oder rivers, and west to the Paris Basin, an area of approximately 750,000 km2. The period between 5500 and 5000 B.C. saw the establishment of farming-based communities throughout this area in several phases.

Rouse's criteria support the case that the establishment of early farming communities in this area was the result of colonization rather than a local development:

  1. There is no in situ development of pottery in north-central Europe. Pottery manufacture appears fully-developed within this area with clear antecedents outside the region in Hungary.
  2. The house forms, settlement patterns, and burials of the first farmers of this region are completely different from any preceding ones in the region (as poorly-known as the latter are);
  3. The chipped stone tool types of the first farmers are also markedly different from preceding microlithic forms;
  4. The key domestic plant species and two of the major livestock species sheep and goats have no native, wild, conspecific forms in central Europe and were introduced from southeast Europe and ultimately from southwest Asia;
  5. The pottery, house forms, and settlement locations are extraordinarily uniform in their general appearance, although with minor local variations over time, from Slovakia to eastern France;
  6. The warm mid-Holocene climate was especially agreeable to cultigens of Near Eastern origin, and the river systems of central Europe presented excellent corridors for the movement of people and livestock;
  7. The most remarkable aspect of the Neolithic diaspora in central Europe is its speed; in the course of 500 years, and probably less, farming communities were established between Ukraine and eastern France.

The central European evidence, then, conforms to Rouse's criteria for demonstrating population movement: the earliest agricultural communities are intrusive, have a clear source outside the region, appear across central Europe in a very short span of time, and took advantage of favorable conditions for migration. The clincher, however, lies in its conformity with Sanger's additional qualifier. All cultural subsystems technology, subsistence, settlement, mortuary practice are involved.

Colonizing Logic

Once it is established that the introduction of agriculture to central Europe was fundamentally the result of colonization through population dispersal, certain aspects of the archaeological record assume a new significance. These can be termed the "colonizing logic" (Beaton 1991) of the early agriculturalists. Two characeristics of the archaeological record deserve special attention in this connection: the preference of the earliest farmers in this region for very specific habitats to the exclusion of virtually all others and a degree of technological conservatism manifested in pottery, stone tools, and especially houses.

In a comparison of the Pleistocene colonizations of the Americas and Australia, Beaton (1991) discussed the notion of what he refers to as "megapatches" and their implications for the study of colonization and migration. By using the term "megapatch", Beaton means that colonists may not have a sufficiently fine-grained view of the variability contained by virgin territory and see it at a very coarse level of resolution. As a result, a characteristic of a colonizing population might be that it will very clearly demonstrate a preference for gross habitat types, and within these gross habitat types they might locate their settlements in very similar places. Only over time would the colonizers distribute their sites in a variety of habitats as the resolution at which they perceive their environment becomes finer and finer.

I believe that this discussion has relevance for understanding the spread of early farming in Europe. A signature of colonization, it could be argued, would be the focus on one gross habitat type to the virtual exclusion of all others, no matter how similarly attractive or productive the others might eventually prove to be. Within that gross habitat type, moreover, the location of settlements in a specific landscape zone to the exclusion of others would also be a signature characteristic of a colonizing population.

By contrast, a settlement pattern which exhibits the use of many different types of gross habitat types and where settlements are located in many different parts of the landscape would be more characteristic of a continuation of an existing pattern of foraging settlement, even if there were slight changes in the settlement pattern. The exploitation of many different habitats would suggest a fine-grained familiarity with the environment that might be expected of an indigenous population rather than the coarse-grained perception of a colonizing population.

In central Europe, the concentration by early farmers on a limited number of gross habitat types is obvious. In the upland loess basins, early farmers chose to settle almost exclusively along the valleys of the smaller tributaries of the major river systems, which would have been a the most fertile and moist habitats in the loess belt. These also would have been the areas of natural breaks in the vegetation, what Verboom (1977) has called "zones of weakness" in the primeval forest. Many other areas which separate the loess basins, where early agriculture certainly could have been practiced, were simply ignored by the early farmers and were not colonized until later in the Neolithic.

When the early central European farmers entered lowlands of the North European Plain, there was a similar concentration on gross habitat types. The lowland analogs of the upland loess basins in northern Poland are patches of ground moraine which are engraved with sub-glacial meltwater valleys. The earliest farming settlement in northern Poland occurs almost exclusively along these so-called tunnel valleys, These zones would also have been the natural breaks in the forested lowland environment, since they would have been part of the hydrological network during the moist Atlantic period. Intervening areas, despite their suitability for agriculture, are not occupied until later in the Neolithic.

This concentration by the early farmers on a limited number of habitats contrasts markedly with the pattern observed in other parts of northwestern Europe. In the Rhine-Meuse delta, for example, early farming sites are known from a variety of habitats and come in varying sizes and functions (Louwe Kooijmans 1993). Such variety appears to reflect a much finer-grained knowledge of the environment than the concentration on a very specific habitat type in the loess zone and in northern Poland. Similarly in Ireland the early farmers used the landscape very widely (Green and Zvelebil 1993), which appears to reflect the residual knowledge of the terrain from indigenous foraging peoples.

The other dimension of the colonizing logic of the earliest farmers of central Europe which deserves attention lies in their conservative and strikingly remarkably standardized technology. If there had been a mass conversion to agriculture of indigenous foragers over the 750,000 square kilometers of central Europe, then one would expect almost immediately to see considerable variation in house forms, pottery, and other aspects of culture. Instead, the initial pattern is strikingly homogeneous. Houses and pottery from Slovakia look like those from the Paris basin. The expected regional variation does not set in until several centuries later, when local populations develop distinct material identities.

In particular, the use of a standardized, modular form of construction the longhouse of the early farmers is consistent with a colonizing population using proven technology to facilitate its expansion. These longhouses, up to 45 meters in length, were the largest buildings in the world at this time. They were multipurpose structures, providing shelter for humans and livestock as well as storage and working space. Despite some regional variation in construction details, the dimensions and general organization of the longhouses is remarkably similar throughout central Europe.


The evidence still strongly indicates that the establishment of agricultural communities in central Europe was largely the result of the colonization of this region by migrating farming peoples. Recent indigenist arguments notwithstanding, the case for population movement seems even more compelling in light of the application of Rouse's Rules and of the identification of the colonizing logic reflected in the choice of habitats and in technological conservatism.

The suggestion is offered here that it might be possible to look for similar signatures of colonization in other parts of Neolithic Europe. In particular, the plains of Thessaly studied by van Andel and Runnels (1995) also seem to be good candidates for the colonization argument. Here, too, there is an overwhelming preference for a specific, but rare, habitat floodplains and backswamp levees and populations which can be traced back to their Anatolian origins who settled in areas where antecedent forager populations were slight. On the other hand, if early agricultural settlements are found in many different types of habitats, as they appear to be in Ireland, then the colonization hypothesis would appear to lose force in favor of the indigenous-adoption position.

A more realistic picture of the Neolithic diaspora in Europe is to consider it to have been a mixture of various waves, currents, and eddies of people, animals, and plants. In some areas, an influx of migrating farmers swept with it any sparse local foraging populations as it deposited its own agricultural communities. Elsewhere, early farming communities appeared as isolated pioneer outposts in a multicultural landscape of foragers and farmers. Finally, in many regions the indigenous foragers eventually found it advantageous to incorporate domestic plants and animals into their diet and in the process became farmers themselves.


Ammerman, Albert J.
1989 "On the Neolithic transition in Europe: a comment on Zvelebil and Zvelebil,"
Antiquity 63: 162-165.

Beaton, John M.
1991 "Colonizing continents: some problems from Australia and the Americas," in The
First Americans: Search and Research
, Tom D. Dillehay and David J. Meltzer,
eds., pp. 209-230. Boca Raton: CRC Press.

Bogucki, Peter
1995 "The Linear Pottery culture of central Europe: conservative colonists?" in The
Emergence of Pottery. Technology and Innovation in Ancient Societies,
K. Barnett and John W. Hoopes, eds., pp. 89-97. Washington: Smithsonian
Institution Press.

Bogucki, Peter
1996 "The spread of early farming in Europe," American Scientist 84: 242-253.

Dennell, Robin
1992 "The origins of crop agriculture in Europe," in The Origins of Agriculture, C.
Wesley Cowan and Patty Jo Watson, eds., pp. 71-10. Washington: Smithsonian
Institution Press.

Green, Stanton W., and Marek Zvelebil
1993 "Interpreting Ireland's prehistoric landscape: the Bally Lough Archaeological
Project," in Case Studies in European Prehistory, Peter Bogucki, ed., pp. 1-29.
Boca Raton: CRC Press.

Louwe Kooijmans, L.P.
1993 "The Mesolithic/Neolithic transition in the Lower Rhine Basin," in in Case
Studies in European Prehistory,
Peter Bogucki, ed., pp. 95-145. Boca Raton:
CRC Press.

Rouse, Irving
1958 "The inference of migrations from anthropological evidence," in Thompson, R.H.
(ed.) Migration in New World Culture History, pp. 63-68. Tucson: University of
Arizona (Social Science Bulletin 27).

Sanger, David
1975 "Culture change as an adaptive process in the Maine-Maritimes region," Arctic
12(2): 60-75.

van Andel, Tjeerd H., and Curtis N. Runnels
1995 "The earliest farmers in Europe," Antiquity 69: 481-500.

Verboom, W.
1977 "Lines of weakness in the forests, and early human settlements," ITC Journal
1977-3: 531-44.

Whittle, Alasdair
1996 Europe in the Neolithic: the Creation of New Worlds. Cambridge: Cambridge
University Press.

Zilhao, Joao
1993 "The spread of agro-pastoral economies across Mediterranean Europe: A view
from the Far West," Journal of Mediterranean Archaeology 6: 5-63.

Return to: Peter Bogucki's home page | Princeton University's home page