The Oxford Handbook of Neolithic Europe
Page 44
HOUSES
As briefly mentioned above, the size and shape of the lake-dwelling house are fairly standardized throughout the entire Circumalpine region, from the Neolithic to the Bronze Age. What, on the other hand, vary considerably are the building techniques, which are adapted according to the various types of houses (Menotti 2012). Not all houses within the village had the same function; they in fact span from simple habitations to storage buildings, and even ‘cult’ houses.
Thanks to dendrochronology, archaeologists can now reconstruct the lake-dwelling house biography quite precisely, from its initial construction to its destruction/abandonment, including repairs and possible extensions. The remarkable level of preservation of some houses (and indeed entire villages) allows us to reconstruct or redefine domestic space by identifying internal spatial divisions, task- and gender-specific spaces, as well as socio-economic aspects of lacustrine communities. Recent research has not only more clearly identified different households related to different tasks within a single village community, but has even advanced the possibility that different houses may belong to a single household (Doppler et al. 2011).
Typological classification
Lake-dwelling houses of the Circumalpine region can have either elevated floors (e.g. on stilts), or be built directly on the ground. Although not necessarily constructed in the water, pile dwellings (Fig. 15.2) are probably the most common type, and they are found throughout the entire Alpine region, including the southern part (Menotti 2001a, 2012). A typical example where both house types (stilts and on the ground) were adopted in the same village is that of Fiavé, northern Italy (Perini 1987).
FIG. 15.2. Reconstruction (scale 1:1) of two Neolithic lake-dwelling houses (left: Hornstaad-Hörnle, Germany; right: Arbon-Bleiche 3, Switzerland).
(Photograph: AATG, Urs Leuzinger, www.archaeologie.tg.ch).
Houses with floors constructed directly on the humid ground (also called Packwerkbauten) are more common in marshland or peatbog areas, for instance at the settlements of Wauwilermoos (Switzerland), on Lake Feder (Germany), and in the Ljubljana Marsh (Slovenia). Ground-joint beams (Schwellbauten—see Gachnang-Niederwil, Switzerland), or simply cross-logged floors (Schwellbalkenbauten—see Thayngen-Weier, Switzerland) are adopted in peatbog houses, where well-insulated floors are built directly on the ground (Guyan 1967).
Habitations on stilts are, on the other hand, more common around lakes, and they are usually built by driving the piles directly into the ground (or the soft lake marl) or by securing them with perforated wooden plates (Menotti 2012, 136). Piles supporting the elevated floor and those of the house frame are often combined; in some cases, different kinds of wood are used (see Arbon-Bleiche 3, Switzerland—Jacomet et al. 2004). The block-building construction technique (in both lake and marshland areas) appears in the Neolithic, but it is not until the Bronze Age that it becomes widely used.
All the above-mentioned building techniques are found throughout the Neolithic, but they do not follow a specific diachronic chronological development, as opposed to the Bronze Age, when some techniques are preferred at different specific times. For instance, elevated houses on stilts were favoured in the early Bronze Age, whereas the block-building technique was more common in the late Bronze Age; and it is indeed during this period that the latter construction method develops into the so-called ‘plank-pillar’ building technique (see Ürschhausen-Horn, Switzerland—Gollnisch-Moos 1999).
There are also cases where more than one construction technique was used, for instance the Neolithic lake village of Hornstaad-Hörnle, Germany (Dieckmann et al. 2006) and the Bronze Age lacustrine settlement of Greifensee-Böschen, lake Greifen, Switzerland (Eberschweiler et al. 2007).
Different buildings, different functions
Already in the 1950s it was proposed that different houses within the village might have had different functions, such as storage buildings and animal stalls. Yet it was not until recently that lake-dwelling archaeologists started to challenge this theory. Whilst a few storage buildings have been identified in some villages (e.g. Arbon-Bleiche 3, Lake Constance, and Sutz-Lattrigen Südwest, Lake Biel, Switzerland), the presence of animal stables is yet to be proven (Ebersbach 2002). What has been noticed, however, is that animals were sometimes kept within the inhabited house itself (e.g. house 1 at Pestenacker – Schönfeld 2002).
New studies on the different functions of the single buildings have recently advanced the idea that some structures within lacustrine settlements might even have been used as ‘cult’ houses. One of the best examples is the settlement of Marin (Lake Neuchâtel), where a particular building was constructed on top of an artificially made hill (within the village), situated at the end of a path over 100m long coming from outside the village (Honegger 2001).
House biography
Thanks to dendrochronology combined with experimental archaeology, it is now possible to reconstruct the biography of a lake-dwelling, from its construction to its abandonment. Although, as pointed out above, the duration of a house varies according to the location and the wood species used in the construction, it has been commonly agreed that a building in the wetlands would have lasted 10–15 years. Assuming that the house was constructed straight after felling the trees needed as building material, the presence of the cambium (the bark ring) allows archaeologists to work out when the construction process took place and how long the building itself lasted. The presence of different cambium dendro-dates within the same building (and within the house’s lifespan) is clear evidence of repairs, which were fairly common almost everywhere in the Circum-Alpine lake-dwelling tradition.
Once the house is lived in, it becomes a living entity in itself, and develops along with the people who occupy it. As a result, the building undergoes a dynamic and continuous change throughout its existence, which is reflected not only by periodical repairs, but also through transformations and extensions due to the family’s demographic expansion, or economic necessities (e.g. the need for more storage, or/and working space). The presence of lake marl deposits within the site stratigraphy (indicative of lake level transgressions), as well as evidence of burning (intentional or accidental conflagration), might also tell us how and why a house was eventually abandoned (Menotti 2012, 273).
The household: definition of domestic space
Notwithstanding the exceptional level of preservation, recognizing the internal division of lacustrine houses has always been rather challenging, and this is mainly due to the complex site formation processes that an archaeological site (in this case a house) undergoes, from its abandonment to its discovery. Not only are archaeologists faced with the natural distortion of the archaeological record which occurs in the post-abandonment context (especially concerning collapsed elevated house floors, where movable artefacts, as well as permanent hearths, are unrecognizably displaced from their original position—see Dieckmann et al. 2006), but also with the alteration of the site prior to being deserted (e.g. a planned abandonment, which implies a systematic removal of intentionally selected objects—Cameron and Tonka 1993). It is consequently incorrect to assume that all the artefacts found in situ reflect activities carried out whilst the house was still in use. A distinction between, on the one hand, depositional sets and areas, and, on the other hand, activity sets and areas is therefore needed to understand the role depositional processes played in creating the archaeological record (Schiffer 1987).
Internal division, social organization, and identity
Even when archaeological remains allow identifying the internal division of a house, investigating social organization is not an easy task. One of the most common ways is to use spatial analysis tools to identify and define spatial cells in the material culture. A few successful cell-mapping techniques have been developed in the past three decades, with syntax and grammar methodologies being the dominant ones (Hillier and Hanson 1984). Using artefact distributions and associations (to identify particular human activities) is another long-appli
ed methodology in spatial definition (Binford 1981; Carr 1984). Archaeologists have even applied statistical approaches, with special emphasis on Multiple Correspondence Analysis, whereby different data are analysed at the same time, producing multidimensional potential correlations between the artefacts (Brinkkemper 1993).
Addressing spatial patterning is of course important, but understanding the social meaning of a household is a different matter. One has to study the diachronic and synchronic contexts in which the habitation is found before being able to understand social meaning. According to Rapoport (1990) for instance, the built environment (setting) offers people ‘cues’ as to a possible behaviour, which is appropriate to a specific situation dictated by the setting itself at a particular time. There are, therefore, various and specific ‘systems of activities’ occurring in particular ‘systems of settings’. Hence, the built environment gains or loses multiple and changeable meanings according to specific situations. It is furthermore understood that the built environment imposes schemes of social organization on people, and this organization of space reflects and generates social structure, triggering a discursive relationship between spatial organization and social structure (Bourdieu 1977). A large range of perspectives in a variety of social and material contexts has to be taken into consideration when defining the household, because habitations not only reflect social structure, but they determine it (Shanks and Tilley 1987).
There are a few important factors that have to be considered within the lake-dwelling built environment: a house location within the settlement, the layout of the settlement, and the size of the houses. It is, for instance, common to interpret architectural homogeneity as a reflection of social homogeneity. However, it has been shown that a specific village layout (rows of houses and grid layouts, for example) requires an institutionalized form of central leadership (Stöckli 1990); and the same can be said for the internal divisions of the houses. As pointed out above, due to site formation processes, finding an intact house floor that clearly shows the internal divisions of a lacustrine house on stilts is almost impossible. Yet, if the house was built directly on the ground, the floor is more likely to be preserved and internal divisions are clearly visible. In this case, the entire space is divided into two (sometimes three) little rooms, although they may only be separated by a thin wall or divider. Whilst size and layout change, the content of the house remains fairly standardized; a permanent item that always appears though is the fireplace (sometimes we have more than one, and even vaulted ovens). The floor consists of small logs or planks, sometimes covered with a stratum of clay. Walls were either made of thin planks, horizontal logs (block-construction), or wattle, and subsequently daubed (Fig. 15.3). Because of the limited internal space of the house (c. 8x4m = 32 m2), and the failure to identify sleeping places (e.g. bedrooms), it has been assumed that the rooms were multifunctional; for instance, working spaces during the day might have become sleeping areas at night. We even have cases, especially in wetland houses built on the ground, of small animal stables included within the already limited space of the house (see Schönfeld 2002). The accentuated accumulation of organic material, such as animal bones, food remains, and coprolites, found in some areas of houses with elevated floors, suggest that the remains could have been dumped under the houses through waste trapdoors. Finally, the orientation of the house does not seem to have had a crucial importance within the Neolithic lacustrine communities, although it has been noticed that in some cases it changed drastically (90°) from one occupational phase to the other (see Sutz-Lattrigen Rütte, Lake Biel—Hafner and Suter 2004).
FIG. 15.3. Internal division of a wetland Neolithic house at Ehrenstein (Schussenried culture), Germany. 1) Wooden floor directly constructed on the ground; 2) Stratum of clay laid on the wooden floor; 3) Oven; 4) Fireplace; 5) Wattle wall daubed with clay (Modified from Schlichtherle and Wahlster 1986).
Task-specific spaces
Thorough artefact distribution analyses (including statistical approaches, e.g. Multiple Correspondence Analysis) carried out on well-preserved lacustrine settlements have allowed archaeologists to distinguish task-specific areas not only within a single household, but in the entire village. One of the best examples is the Neolithic lake-dwelling of Arbon-Bleiche 3 on the Swiss shore of Lake Constance, where specific tasks seem to have been performed only in particular houses, or even in specific parts of the village. For instance, some households were more specialized in hunting activities, and others in cattle breeding. Even domestic animals seem to have had a specific division within the village, with cattle kept more in the northern part of the site and pigs in the southern part. The same can be said for fishing; fishermen in the northern part specialized in shore net-fishing, whilst the fishermen in the southern part practiced more offshore fishing from boats (Menotti 2012, 276). A division of labour was also noticed in pottery production; in fact, ceramic stylistic studies have revealed that one single person was probably in charge of pottery production for a few houses (De Capitani et al. 2002). On the other hand, stone and flint tools were more or less equally produced by each household.
The concept of community: beyond single households
Recent research has shown that houses had distinct similarities that could have grouped them together (Doppler et al. 2013). Is it therefore correct to assume that one single unit is necessarily a single household? Despite the fact that similar distributional patterns shared by some houses may be interpreted as social practice of single houses acting in an autarchic way, recent studies have advanced the possibility that the inhabitants of different houses might have been in fact related to a single household (Doppler et al. 2011).
In considering the possibility that the household goes beyond the single house and/or social unit, one has to take into consideration the whole settlement (layout, typology, location, and terrain morphology) to understand the multiple-residential household concept. As previously mentioned, the typical layout of a lacustrine or marshland Neolithic village would normally have consisted of rows of small houses clustered together and divided by narrow alleys; the remaining small areas are interpreted as ‘communal space’. Although all houses could be considered as ‘self-sufficient’ (they all had hearths and various activity areas), a large number of daily tasks were carried out in the communal space (Fig. 15.4). However, this ‘communal space’ was likely to be influenced by unpredictable (or predictable, e.g. seasonal) environmental variability (flooding), which would have restricted accessibility. Therefore, the link between houses did not only depend on artefact distribution and social practices, but also on the location of the dwelling, which would have determined the availability of ‘communal space’ throughout the year.
FIG. 15.4. Miniature model (scale 1:32) of houses and surroundings of the Arbon-Bleiche 3 Neolithic lake-dwelling settlement.
(Photograph: AATG, Daniel Steiner, www.archaeologie.tg.ch).
CONCLUSIONS
Understanding and interpreting the social complexity of the Circumalpine region Neolithic lacustrine communities is far from simple. Decoding their social complexity is a fascinating journey that starts from the settlement and its surroundings, and ends inside a single house. A ‘simple’ internal division of a small house is as important as settlement layout, size, and location within the landscape. The seemingly ‘simple’-looking material culture does not necessarily mean lack of social complexity, or an expression of unstratified society. Systematic analyses on the internal division of houses and their contents are germane for a better understanding of the people who built them. The house is a living entity that not only reflects but determines social structure, and it is only by studying its biography that the social organization and identity of its inhabitants are fully understood. This is surely a long and time-consuming task, but, thanks to their outstanding level of preservation, it is not too presumptuous to say that the lake-dwellings of the Circumalpine region definitely have great potential!
NOTE
1.All d
ates quoted for the Circumalpine settlement sites in this chapter are therefore obtained by dendrochronology, unless otherwise stated.
REFERENCES
Affolter, J. 2002. Provenance des silex préhistoriques du Jura et des régions limitrophes. Neuchâtel: Musée cantonal d’archéologie.
Arbogast, R., Jacomet, S., Magny, M., and Schibler, J. 2006. The significance of climate fluctuations for lake level changes and shifts in subsistence economy during the Late Neolithic (4300–2400 BC) in central Europe. Vegetation History and Archaeobotany 15, 403–418.