In this chapter I wish to return to the main themes that have been explored in this study: landscape, the construction of local social identities, and the distribution and representation of claims to land. In several ways it is intended as a synthesis. In the preceding chapters, an analytical separation was made between several social levels and between spatial scales.The house and household were treated in relative isolation from the local community and settlement territory, and local communities were not viewed in relation to each other until the chapter on micro-regional habitation and land use patterns. It could be argued that these distinctions were to some extent also meaningful in the past, as different social groups were shown to define themselves through different means and on different geographical scales.The aim of this chapter, however, is to focus on the interaction between social levels and their differentiated relationships with land and landscape.

Another artificial separation was made between social and cultural dimensions on the one hand and ecological and economic possibilities and constraints on the other. While chapters 3 and 4 placed a stronger emphasis on the former, the focus in chapter 5 was mostly on subsistence economy, demographic trends and ecological change. This has been a useful analytical distinction for the sake of argumentation and clarity, but it has little to do with real life. In the introduction I argued for a theoretical perspective that is sufficiently sensitive to the need to look at how people in the past actively interpreted and ordered the world around them, but that does not disregard the influence that ecological stimuli may have had in instigating social and cultural change. Up to this point in this study I have made few systematic attempts at combining them.While not claiming to overcome dichotomies between material and ideological or between nature and culture, I will attempt in this synthesis to identify the interplay between the different dimensions of human interaction with the landscape.

In order to keep this synthesis to a manageable size, I will largely work from the interpretations and conclusions reached earlier, without repeating in full the arguments that underlie them.Where appropriate, I refer to the relevant sections in the preceding chapters.

The Meuse-Demer-Scheldt (MDS) region is a Pleistocene coversand plateau of approximately 250 kilometres (east-west) by 120 kilometres (north-south) (fig. 2.1). It covers the modern-day province of Noord-Brabant and the sandy parts of Dutch Limburg in the Netherlands, and the provinces of Antwerp and Limburg in Belgium.To the north of the sandy landscapes of the MDS region lies a broad zone with Holocene Meuse and Rhine sediments.The western edge of the study area is formed by the delta region of these rivers and the Scheldt river.To the south and east of the MDS region lie the loamy sand and loess regions of Belgium, southern Dutch Limburg and the German Rhineland.Together these features define the MDS region as a geographical entity.

With respect to geographical situation, long-term structure of the agrarian economy, and potential for archaeological study, the MDS region is comparable to the other coversand landscapes or Geestlandschaften that form a series along the south coast of the North Sea. In this chapter I will present the main issues that set the parameters for archaeological research in a coversand landscape in general and the MDS region in particular.These are the geological and geomorphological situation, the general structure of the landscape in premodern times, and the history of archaeological investigations in the study area.

ASPECTS OF GEOLOGY AND GEOMORPHOLOGY

The highest part of the MDS region is situated in the southeast and rises a little over 100 metres above sea level. From there the terrain gently drops down to about sea level at the northwestern border.The main geomorphological element in most of this region consists of sand deposits that were laid down under cold and dry conditions during and after the last Ice Age. They overlie older aeolian sands and riverine sand and gravel sediments. The coversands are mostly between half a metre and one and a half metres thick and consist of fine to coarse sands, sometimes with an admixture of loam.They form low, elongated ridges following a general southwesterly to northeasterly direction. In the northwestern border area peat and marine clays are the dominant element. Even though strictly speaking this border area is part of the MDS region, it will not be taken into account in this study; very little is known about the prehistoric habitation of this area.

Meuse-Demer-Scheldt Region. Distribution of Urnfields

AN ANTHROPOLOGICAL PERSPECTIVE ON HOUSES AND HOUSEHOLDS

INTRODUCTION

To say that a house is more than a physical shelter against the elements is simply pointing out the obvious. Houses are in many ways at the heart of social and cultural life, both in non-modern societies and in present-day, western society; they ‘constitute culturally significant space of the highest order’. In contemporary northwestern Europe houses are embedded in a web of diverse notions including home, family, privacy, investment, status and the like. The significance of houses in other societies may well be based on wholly different ideas, but houses are never socially or symbolically neutral. Le Roy Ladurie notes about the 13th century Pyrenean village of Montaillou that

…the best way to understand Montaillou is to abandon temporarily the problems of social stratification within it and go straight to the basic cell which, multiplied a few dozen times, went to make up the village.This basic cell was none other than the peasant family, embodied in the permanence of a house and in the daily life of a group co-resident under the same roof. In local language this entity was called an ostal; and in the Latin of the Inquisition files it was called a hospicium or, more often, a domus. It should be noted that the words ostal, domus and hospicium all and inextricably mean both family and house.The term familia is practically never used in the Fournier register. It never crosses the lips of the inhabitants of Montaillou themselves, for whom the family of flesh and blood and the house of wood, stone or daub were one and the same thing.

The ethnographic literature is rich with examples that show how houses are invested with social and symbolic meanings in a great variety of ways.The identification of a house with its inhabitants is a recurrent element in the ethnographic literature (particularly but not only in societies that lack a commercial house market), but it is only one possible aspect. Houses can be reflections and structuring features of socio-cosmological orders. They can provide the context for a particular ‘dwelling-habitus’ consisting of cultural ideas and values that structure daily life in and around the house.

Catalogue of urnfields of the Late Bronze Age, Early Iron Age and Middle Iron Age in the MDS region. 1) number, see appendix 1; 2) name of village and location; 3) country, province (B: Belgium; N: Netherlands;An:Antwerp; Br: Brabant; Lg: Limburg; Gl: Gelderland); 4) premodern landscape (h: heathland, c: medieval arable land, o: other); 5) date of urnfield (based on recorded graves, LBA: Late Bronze Age, EIA: Early Iron Age, MIA: Middle Iron Age). If no date is given, the date of the cemetery cannot be specified further than Urnfield period (c. 1050 – c. 450 BC); 6) remarks (on location, systematic excavations, remarkable finds); 7) literature (RMO: archives Museum of Antiquities, Leiden; CAA: Central Archaeological Archive, BM: Bonefantenmuseum; see References for other abbreviations).

INTRODUCTION

RESEARCH QUESTIONS

Having discussed households and local communities in the previous chapters, in this chapter I will address settlement patterns and subsistence strategies in relation to changing environmental conditions. A related issue for attention concerns long-term demographic trends. My objective is to link the settlement territories that remained rather abstract in the previous chapter to their landscape context, the physical landscape, that is, of soils and topography, and of vegetation, agricultural potential and limitations.

Agricultural systems and environmental change are usually studied in specialised sub-disciplines of archaeology, using soils and botanical and faunal remains as their main sources of information. In contrast, I will address these issues primarily through a study of micro-regional and regional settlement patterns, partly because relevant ecological data are rather scarce for the MDS region, but also because the settlement record provides valuable evidence that has not been systematically integrated with ecological data. In this chapter, an integration of both categories of evidence will take place only to the extent that ecological data are available or can be extrapolated from neighbouring regions, but methodologically the main aim is to investigate the potential of settlement data for studies of subsistence and environmental change.

There is one essential assumption to be made here.This is that there is a spatial relationship between farmsteads and settlements on the one hand and arable lands on the other, and therefore that the locations of farmsteads and settlements are representative of the locations of the fields.This appears a fair assumption for the later prehistoric societies of the Northwest European Plain, in which all or practically all households depended on a mixed subsistence strategy of crop cultivation and animal husbandry, and agricultural potential had a major impact on settlement patterns.The scale of analysis, however, is important. The immediate environment of a farmstead or settlement (if the environment could be reconstructed at such a fine resolution) is not necessarily indicative of the local environment in which the fields are located. But in a perspective that focuses on regional and micro-regional patterns and long-term developments, the coarser resolution means that farmsteads and fields can be combined more confidently.

GENERAL THEME AND AIMS OF RESEARCH

In this study I will draw on a range of archaeological materials to present a history of the communities inhabiting the Meuse-Demer-Scheldt (MDS) region between the beginning of the Late Bronze Age and the beginning of the Roman period.The aim is to elucidate some of the major social and cultural transformations that occurred during that period, covering roughly the first millennium BC.While a number of different histories could be written about the region and period, this one takes the form that it does because of the central theme that lies at its core: the reciprocal and dynamic relationships between human groups and the landscape.

This is a broad and vague description for a research theme; one that without further elucidation can conjure up quite different things, from ecologically-determined ‘people-land’ relationships to conceptualised landscapes and mythical geographies. It clearly needs a more precise definition; for the time being, however, I will retain this broad description and gradually clarify it in the course of this introduction. Moreover, as will become clear, the inclusiveness suggested by the description is an essential feature of the perspective that I advocate.

As a first exposition of the theme that I refer to as ‘reciprocal and dynamic relationships between human groups and the landscape’, let me briefly present a historical situation which contains in condensed form many of the elements that lie at the core of the subject of this study. In his book Bad land. An American romance the travel writer and novelist Jonathan Raban describes the history of the homesteaders on the prairie of Montana in the United States. Attracted by the prospect of a tract of free land, people from Europe and the American east coast settled down on the prairie in the early years of the twentieth century. They found themselves in a vast open space, totally devoid of geographical features that could orient them.There was nothing there with which they could in some way identify, nothing to remind them of their native villages and towns. It was a landscape without history, or more precisely, without a history that they knew how to read.

SETTLEMENT TERRITORIES AND LOCAL COMMUNITIES

INTRODUCTION

In many languages there are multiple terms to describe basic groups of rural dwellings, units in which most of the population of agricultural societies lives. Some terms have a meaning that stresses the geographical, others the social aspects (for example settlement versus community), although typically a combination of social, spatial or other contents is implied (hamlet, village, parish etc.). These are often primary organising features – although never the only one – of the larger bodies of societies.They are central elements in the construction of people's identities, in the transmission of ideas and values, and in the organisation of subsistence strategies, to name a few features. Deservedly, they have received much attention from geographers and anthropologists. It is the same with archaeologists, for whom in many regions and periods the settlement is a basic unit of analysis. In this study too, the logical next step of analysis after the farmstead and household is the settlement and especially its associated local community.

But having specified the object of study of this chapter, a problem of definition immediately looms large. As described in the previous chapter, farmsteads in the study region tend to be dispersed over the landscape throughout most of the Bronze Age and Iron Age, and nucleated settlements only began to develop in the last centuries of the first millennium BC (see section 4.5). How does one then define a settlement archaeologically? Even in the rare cases in which it is possible to point to several farmsteads whose distance from each other is significantly less than the distance to other farmsteads, the problems are not solved.The scale of excavations never allows one to be certain that a representative sample of the total number of dispersed farmsteads has been excavated, and moreover, the lack of precision in dating single farmsteads is too great to be able to suggest which farmsteads were contemporaneous.While it is quite possible to envisage an Iron Age settlement in the social sense, its geographical component is difficult to grasp through archaeological methods.

It has been recognised for some time in Dutch regional archaeology that a more appropriate analytical focus is not the settlement, but the settlement territory.