by Chris Fowler
Sommer. U. 2006. The Linearbandkeramik settlement of Hanau-Klein Auheim (Hesse): a specialised mining settlement? In G. Korlin and G. Weisgerber (eds), Stone Age–Mining Age, 187–194. Bochum: Deutsches Bergbau-Museum.
Soudsky, B. 1969. Étude de la maison néolithique. Slovenska Archeologia 15, 5–96.
Stäuble, H. 1997. Häuser, Gruben und Fundverteilung. In J. Lüning (ed.), Ein Siedlungsplatz der Ältesten Bandkeramik in Bruchenbrücken, Stadt Friedberg/Hessen, 17–150. Bonn: Habelt.
Stäuble, H. 2005a. Häuser und absolute Datierung der Ältesten Bandkeramik. Universitätsforschungen zur prähistorischen Archäologie 117. Bonn: Habelt.
Stäuble, H. 2005b. Tradition und Moderne im bandkeramischen Hausbau. In J. Lüning, C. Frirdich, and A. Zimmermann (eds), Die Bandkeramik im 21. Jahrhundert: Symposium in der Abtei Brauweiler bei Köln vom 16.9.-19.9.2002, Beilage 10. Rahden: Marie Leidorf.
Stäuble, H. and Lüning, J. 1999. Phosphatanalysen in bandkeramischen Häusern. Archäologisches Korrespondenzblatt 29, 169–187.
Stehli, P. 1994. Chronologie der Bandkeramik im Merzbachtal. In J. Lüning and P. Stehli (eds), Die Bandkeramik im Merzbachtal auf der Aldenhovener Platte, 79–191. Cologne: Rheinland Verlag.
Strien, H-C. 2005. Familientraditionen in der bandkeramischen Siedlung bei Vaihingen/Enz. In J. Lüning, C. Frirdich, and A. Zimmermann (eds), Die Bandkeramik im 21. Jahrhundert: Symposium in der Abtei Brauweiler bei Köln vom 16.9.-19.9.2002, 189–197. Rahden: Marie Leidorf.
van Berg, P. 1989. Architecture et géométrie de quelques villages rubanés récents du Nord-Ouest. Helinium 29, 13–41.
van de Velde, P. 1979. On Bandkeramik social structure: an analysis of pot decoration and hut distributions from the central European Neolithic communities of Elsloo and Hienheim. Analecta Praehistorica Leidensia 12, 1–242.
van de Velde, P. 2007a. On the Bandkeramik features. In P. van de Velde (ed.), Excavations at Geleen-Janskamperveld 1990/1991. Analecta Praehistorica Leidensia 39, 71–90.
van de Velde, P. 2007b. On the Neolithic pottery from the site. In P. van de Velde (ed.), Excavations at Geleen-Janskamperveld 1990/1991. Analecta Praehistorica Leidensia 39, 99–128.
van de Velde, P. 2007c. The Bandkeramik settlement. In P. van de Velde (ed.), Excavations at Geleen-Janskamperveld 1990/1991. Analecta Praehistorica Leidensia 39, 223–244.
van de Velde, P. 2007d. The Neolithic houses. In P. van de Velde (ed.), Excavations at Geleen-Janskamperveld 1990/1991. Analecta Praehistorica Leidensia 39, 21–70.
Veit, U. 1993. Burials within settlements of the Linienbandkeramik and Stichbandkeramik cultures of central Europe. Journal of European Archaeology 1, 107–140.
Waterson, R. 1990. The living house: an anthropology of architecture in south-east Asia. Oxford: Oxford University Press.
Weiner, J. 1995. Bogenstab- und Pfeilschaftfragmente aus dem altneolithischen Brunnen von Erkelenz-Kückhoven: ein Beitrag zur Bogenwaffe der Bandkeramik. Archäologisches Korrespondenzblatt 25, 355–372.
Whittle, A. 1996. Europe in the Neolithic: the creation of new worlds. Cambridge: Cambridge University Press.
Whittle, A. 2003. The archaeology of people: dimensions of Neolithic life. London: Routledge.
Wüstehube, V. 1993. Frühneolithische Grubenhäuser? Neue Überlegungen zu einem alten Problem. Germania 71, 521–531.
Zimmermann, A., Meurers-Balke, J., and Kalis, A. 2005. Das Neolithikum im Rheinland. Bonner Jahrbücher 205, 1–63.
CHAPTER 15
LAKESIDE DWELLINGS IN THE CIRCUM ALPINE REGION*
FRANCESCO MENOTTI
INTRODUCTION
WHAT is today known as the ‘lake-dwelling phenomenon’ in the Circumalpine region was triggered by the serendipitous discovery of a submerged prehistoric village on Lake Zurich in 1854. The question as to where and how the lake-dwellings were constructed triggered an incandescent dispute known as the Pfahlbauproblem (pile-dwelling dispute), not resolved until fairly recently (Menotti 2001b). It has now been agreed that there were various types of lake settlements, and their typological distinction depended upon environmental and cultural factors.
Once the Pfahlbauproblem was no longer an issue, the focus of research shifted to chronology and occupation patterns. It was then established that the lake shores were occupied for more than 3,500 years (from the late fifth millennium to the second half of the seventh century BC), but, with the advent of dendrochronology, it was soon realized that occupation was far from homogeneous—periods of occupation alternated with hiatuses which were caused by either environmental or cultural (possibly even both) factors (Menotti 2001a, 2002, 2003).
A recent development in lake-dwelling research is the study of complex systems of occupational patterns of single settlements in particular areas and within specific time spans. Special emphasis is placed upon settlement size, location, typological characteristics, and biographic aspects (Menotti 2012). Settlements are now studied from a life history perspective, from construction to abandonment, with a thorough consideration of the functions of single houses, including sacred and profane components of their inhabitants’ lives.
THE PFAHLBAUPROBLEM
The sensational discovery of the Ober-Meilen prehistoric lacustrine settlement in 1854 was promptly examined by the Swiss naturalist Ferdinand Keller, who, by the end of that year, published Die keltischen Pfahlbauten in den Schweizerseen (Keller 1854), a detailed report which made the Circumalpine region lake-dwellings known all over the world. Following pioneering ethnographic studies, but basing his argument on very little scientific evidence, Keller concluded that those ancient dwellings were built on large platforms in the shallow water of the lake. Over time, new lake-dwelling sites came to light, and with them different kinds of house floors, apparently built directly on the semi-dry ground. Lack of scientific evidence, however, could not confirm that they were indeed lake-dwellings, and they were simply treated as a new type of wetland construction. It was not until the 1920s, thanks to the work of Hans Reinerth, that Keller’s dogma was contested. Basing his theory on new evidence from the lacustrine site of Sipplingen (Lake Constance), Reinerth (1925) argued that the lake settlements were indeed built on stilts, but they were only flooded periodically during the lakes’ seasonal fluctuations.
Reinerth’s theory of a semi-dry setting gained more and more credibility, but this was simply not enough for those in favour of a completely dry building ground. A decisive ‘attack’ was delivered by Oscar Paret in the early 1940s, when, following meticulous analyses, he provided indisputable evidence that the lake-dwellings of Hitzkirch-Seematt and Hochdorf-Baldegg (Lake Baldegg, Switzerland) were definitely built on dry ground (Paret 1942). However, Paret’s theory was not unanimously accepted by all Swiss scholars. Keller-Tarnuzzer (1944, 1945) and Pinösch (1947), for instance, strongly argued that the lake villages of Breitenloo, Arbon-Bleiche, and Burgäschi/Ost were definitely erected on stilts and surrounded by water. The fierce dispute continued until the Swiss scholars Emil Vogt (1951) and Josef Speck (1955) were able to obtain new data from Egolzwil 3 and Zug-Sumpf (Switzerland) respectively, showing that the ground where the lacustrine settlements were built was definitely dry at the time of occupation. Paret’s theory was eventually accepted and the 100-year jubilee of Keller’s pile dwellings in 1954 was ironically and sadly celebrated by denying their existence (Guyan 1955).
Fifty years later, the 150-year jubilee was celebrated in a much more relaxed atmosphere with all scholars agreeing that all the above-mentioned theories were right (Menotti 2004) (Fig. 15.1); there is not a single type of lake-dwelling construction within the Circumalpine region but several, and each of them was used according to people’s cultural needs and the surrounding environmental characteristics.
FIG. 15.1. Schematic drawing of the Pfahlbauproblem (the lake-dwelling dispute) (Modified from Menotti 2001b).
CHRONOLOGY: DISCONTINUITY VS. CONTINUITY
Once the lake-dwelling dispute was resolved, the focus of research shifted to chronology.
The remarkably well-preserved wooden remains often found in lake-dwelling settlements have allowed archaeologists to develop reliable dendrochonological sequences, which have placed the northern Circumalpine lake-dwelling phenomenon between the end of the fifth millennium BC and the seventh century BC1 (Menotti 2001a, 2004). However, there is also a marked discontinuity (Suter et al. 2005, 18), as periods of occupation alternated with periods of abandonment caused by either cultural or environmental factors, and, most probably, even a combination of both.
Magny (1995, 2004) for instance shows a plausible correlation between climate and lake-dwelling occupational patterns; periods of favourable climate coincide with periods of lake-dwelling occupation, whereas abandonment is the result of climate deterioration. At the same time though, Pétrequin and Bailly (2004) argue that the relationship between climate and lake shore occupation does not always work; there are in fact periods when the climatic conditions in the lacustrine environment were favourable, but the lake shores were not settled anyway (e.g. the Bell Beaker period). It is therefore vital to consider all the variables, which can be neglected in a simple environmentally deterministic research approach.
For instance, short-term climatic deteriorations in the first half of the thirty-seventh and thirty-sixth centuries BC had little impact on lake shore occupation. The one in the thirty-fourth century BC, on the other hand, was more pronounced, but some lakes (especially in the western part of Switzerland) still continued to be occupied. Interestingly enough, during a period of favourable climate (c. 3500–3450 BC), the lake shores were mostly deserted in the whole northern Alpine region (Hafner and Suter 2000). A similar situation, but at a much larger scale, is to be found towards the end of the Neolithic and in the first part of the early Bronze Age (c.2400–1800 BC) when, apart from very sporadic examples, the lake shores were not occupied.
This characteristic discontinuity in lake-dwelling occupation has always fascinated archaeologists, who not only have attempted to bridge the above-mentioned hiatuses caused by well-defined climate change (Menotti 2003), but also tried to give plausible explanations as to why the shores failed to be settled, or were even abandoned, during favourable climatic conditions (Pétrequin et al. 2002; Arbogast et al. 2006).
A long-standing issue, which has been shed light on only recently, is what caused the final and permanent demise of the Circumalpine region lake-dwelling tradition towards the end of the late Bronze Age (see Menotti, 2015).
ECONOMY
The lake-dwellers of the Circumalpine region were a mix of agriculturalists, pastoralists, and hunter-gatherers – fishing was also an important part of their economy, especially on larger lakes.
The Neolithic lake-dwellers obtained at least 50% of their total calorie requirements from cereals, the four most important ones being naked wheat, emmer, einkorn, and barley. Of course, in terms of quantity these cereals were not used homogeneously throughout the lake-dwelling phenomenon. For instance, naked wheat was quite popular between 4000 and 3500 BC, but then lost its importance towards the end of the Neolithic, when emmer became more widely used. Einkorn was adopted to a much lesser extent and its consumption was quite steady. Finally, the use of barley fluctuated a lot (more from 4000 to 3500 BC and toward the late Bronze Age). The same can be said for poppy and flax–both popular between 3500 and 2700 BC, but decreasing towards the end of the Neolithic (Jacomet 2004). Hunting also fluctuated quite considerably throughout the Neolithic, with increments coinciding, in some cases, with climate deteriorations (Schibler 2006).
Animal husbandry was also quite irregular and cattle especially were used in different ways in different periods. During the early phase of the late Neolithic (4300–3500 BC) for instance, their use was not economically efficient, whereas from 3500 BC onwards and until the end of the Neolithic (c. 2500 BC), the increasing development of agriculture coincided with an intensification and a more diversified exploitation of animals, with cattle husbandry more oriented towards economic profit (animal traction and dairy products) (Ebersbach 2002). Contrary to what was previously thought, fishing was very much part of the lake-dwellers’ economy; diverse fishing techniques were used not only throughout the entire Alpine region, but even within single villages (Hüster-Plogmann 1996, 2004).
Finally, trade also played an important role in the economy of lacustrine groups, and it was quite developed within the Alpine region and beyond, especially concerning flints (Affolter 2002).
SETTLEMENTS
Despite being classified within a single term (lake-dwellings), the lacustrine settlements of the Circumalpine region are far from homogeneous in size, location, and typology. Small to large-sized villages were built in the shallow water of lake morainic shoals, on land near the lake shores, or even within marshy or peatbog environments, in the proximity of water basins. Their distribution in the landscape was also rather irregular and depended on a variety of factors: chronological period, geographical location, lakeshore morphology, and demography (Menotti 2012). The biographic development of lacustrine settlements (from the construction of the first house to the final abandonment of the settlement) did not follow regular and/or standardized patterns either, and it was strongly influenced by temporal and spatial factors, possibly also related to demographic fluctuations. Finally, the possibility of repairing the single habitations, as well as the availability of new potential settling land nearby, determined the duration of the settlements before their permanent abandonment.
Size
The often restricted excavation areas have prevented archaeologists from establishing the full extent of many lacustrine settlements. Nevertheless, three distinct categories, according to the number of houses found in the settlement, have been identified within the whole lake-dwelling tradition: small, medium, and large. Small-size settlements do not usually exceed eight houses; examples of such little villages are not very numerous, but for instance include those of Seeberg on Lake Burgäschi, Switzerland, and Dullenried, Lake Feder (Federsee), Germany. Medium-size settlements have up to 20 houses; some of the best known are Sutz-Lattrigen Kleine Station, Switzerland, Seekirch-Stockwiesen, Federsee (Schlichtherle 1995), and Pestenacker, Germany (Schönfeld 1995). Lacustrine settlements with more than 20 houses are regarded as large-size villages–they are more numerous than the previous two categories, but it has to be stressed that not all houses were fully excavated and their number has often been calculated according to small but systematically planned excavation trenches. Some of the best examples of large settlements are Sutz-Lattrigen Hauptstation (thirty-second century BC—Hafner and Suter 2000) on Lake Biel (Switzerland), the Zürich-Mozartstrasse dwelling (layer 3—the Horgen culture occupation—Stöckli et al. 1995) on lake Zurich (Switzerland), and the enormous village (more than 160 houses) of Sipplingen-Osthafen, Lake Constance, in Germany.
Settlement distribution in the landscape: the ‘Siedlungskammern’
How densely populated the Circum-Alpine region’s lakes were depended, as pointed out earlier, upon the chronological period of occupation. Sometimes there are only a few settlements scattered over a large area, and at other times contemporaneous villages are clustered within small territories, often sharing the same catchment area. Some of the best examples are to be found in the Zurich Bay (Lake Zurich) between the classic Cortaillod and the Horgen culture (4000–3000 BC), where three early Horgen settlements (Kleiner Hafner, Mozartstrasse, and Grosser Hafner) were settled simultaneously; or in the Auvernier Bay (Lake Neuchâtel), where as many as four lacustrine villages (La Saunerie, Les Graviers, Ruz Chatru, and Les Ténevières), scattered within a 500m radius, were occupied at the same time in the twenty-eighth century BC (Stöckli et al. 1995). Whether or not the settlements shared agricultural fields and/or woodland resources is hard to prove. What we know for sure is that the concept of Siedlungskammern similar to that of the Linearbandkeramik culture (LBK) (Schade 2004) cannot be identified within the Alpine region Neolithic lake-dwelling tradition (see also Ebersbach 2013).
Se
ttlement patterns, duration, and rotation
Although clear patterns of occupation within the Neolithic lake-dwelling tradition (especially concerning the village biographic layout) are sometimes difficult to distinguish, the so-called ‘explorative’ one (one or two houses constructed first, followed by a building ‘boom’ later) seems the most common. This occupational pattern became more and more evident from the thirty-seventh to the thirty-fourth centuries BC (Billamboz and Köninger 2008), with some of the best examples being the settlements of Sutz-Lattrigen Hauptstation (thirty-sixth century BC) and Sutz-Lattrigen Riedstation (thirty-fourth century BC), both on Lake Biel, Switzerland (Hafner and Suter 2000, 2004). This typical occupational pattern started to fade away at the beginning of the third millennium BC.
One of the currently most discussed topics within lake-dwelling research is the duration of settlements (i.e. how long they were occupied before being abandoned). Although experimental archaeology shows that a pile dwelling near, but not in, the water could have lasted about 10–15 years (Leuzinger 2004), recent dendrochronology studies argue for even shorter occupations, e.g. 5–8 years (Bleicher 2009). Duration and rotation of lacustrine settlements do not follow a regular pattern throughout the entire Circumalpine region, but vary considerably from place to place. For instance, although both regions (west and east) of Switzerland did have short-term occupations in the first part of the Pfyn culture (c. 3900–3800 BC), the lake villages in the west were occupied longer than those in the east (including Lake Constance). Curiously enough though, the already small number of long-lasting settlements in the central/eastern part of Switzerland decreased even further in the last phase of the Pfyn culture (c. 3600–3400 BC), disappearing completely on Lake Zurich. On the other hand, long-lasting villages continued to be built in the western part of the country throughout the entire Pfyn period. A quite different scenario occurred in the Lake Feder area (c. 50km north of Lake Constance), where patterns of occupation differed considerably from those in the proximity of the Alps. At the Federsee, there are, in fact, no second-generation dwellings, a site was never occupied twice (with the exception of Alleshausen-Grundwiesen), and (with the exception of Torwiesen II) the occupation of settlements seems to have been shorter than elsewhere (Bleicher 2009). Curiously, houses on Lake Feder do not seem to have undergone repairs, as opposed to those on Lake Constance, where lots of repairs were carried out even within a short period of time (see Arbon-Bleiche 3—Jacomet et al. 2004). Finally, a characteristic of the western Swiss Neolithic lake villages, which is not particularly evident in the eastern part of the country, is their synchronicity with other settlements in the same area or bay (see Sutz-Lattrigen and Auvernier—Hafner and Suter 2004).