Picturing the Merovingian Landscape

(1)

Picturing the Merovingian Landscape

Studying changing perspectives on the landscape surrounding

Maastricht from the late Roman to early Carolingian period

(400 – 800 AD)

(2)

(3)

(4)

Cover image: Autumn landscape of (Dutch) Southern Limburg (after www.inhetlandvankalk.nl)

(5)

Picturing the Merovingian Landscape

Studying changing perspectives on the landscape surrounding Maastricht from the late Roman to early Carolingian period 400 – 800 AD

Bas van de Weerd, BA Studentnumber: 1139851

RMA thesis Archaeology, 1046WT Supervisor: Prof. Dr. F.C.W.J. Theuws

Research Master Transformations of the Roman World

Leiden University, Faculty of Archaeology Leiden, 14-05-2018

(6)

(7)

Preface

Writing this thesis has been a long and sometimes difficult process, as the people mentioned in this preface will have noticed. Without the support of many people, this thesis would not have been possible and I would like to thank them accordingly. As mentioned, various people have aided me in this process to bigger or lesser degrees. Unfortunately, the following part did not offer space to thank everyone individually. This does not mean that your help was forgotten and I would like to thank everyone which has helped me through this process in any degree.

First, and foremost, I would like to thank my thesis supervisor, Prof. Dr. Frans Theuws. Especially the last months with regular meetings, and your enthusiasm on the subject left me inspired after every meeting and provided great angles of approach which kept the morale to an acceptable point during long weeks of writing.

Secondly, I would like to thank all the people that aided in checking paragraphs, chapters, punctuation and my horrible spelling. Lots of correcting was completed due to these people. In this line I would like to thank Fenno Noij, for checking argumentation and text structure, Iris van den Brink and Ruben Modderman, for checking English Grammar and Spelling, Femke Lippok for checking the German summary and Beatriz Gomez de Silva for checking the French summary.

Thirdly, I need to thank all the people with which I spent the highly necessary breaks for coffee or beer. This group includes, but is not limited to Fenno Noij, Iris van den Brink and Elke van der Heijden.

Last, my gratitude goes to my girlfriend, Sterre Houtzager, who helped me through a process of more than a year by preserving my mental health, providing much needed affection and probably prevented a burn out.

(12)

(13)

1. Introduction

One of the most basic and obvious types of research is the creation of an comprehensive overview of the basic elements of an archaeological society, for instance where people live, work and bury their dead. Up to now, this kind of research is practically lacking in early medieval archaeological research on the continent, apart from some exploratory research (e.g. Theuws et al. in prep.). These kinds of studies tend to remain relatively superficial or auxiliary in representing available data and its distribution.

It is the goal of this thesis to go beyond the gathering of an overview by analysing regional relations between different elements in the Merovingian society. Regional overviews have been created in many cases to serve as the background of the research. For example, an indication of the cemeteries in an area. However, these do often not include different factors and do not study the mutual effect of these factors. Moreover, many times the effect of landscapes, and the effect on landscapes was not analysed. Even though the discipline of Merovingian archaeology would benefit even more from a bigger (inter-)national or macro-regional overview, this is beyond the scope of this study.

In this introduction, the research questions of the thesis are introduced, followed by a characterisation of the subject and research area. The following chapters present a historical overview of the area and the theoretical framework for the interpretation of the results. Based on these, a methodology is introduced and its results presented, followed by an interpretation. This thesis ends with a discussion on the value of the results, answers to the research questions and propositions for future research goals.

1.1. Research questions

In order to complete a regional analysis, the landscape should be used as the central subject of the research rather than a supportive one, onto which the research

(14)

will be presented. 1_{This landscape perspective is very typical for the Dutch} archaeological tradition that evolved from an ecological determinism. However, it should be taken into account that the applied theoretical framework is dependent on the research questions asked. Research questions of a cultural character, for instance on the process of Christianisation in the area, can and should not be answered using ecological explanations. One of the most prominent advantages of this is that it overcomes a dichotomy between natural and cultural causes of societal development and combines these in one landscape perspective (see chapter 3.1). It is the intention of this thesis to test my ideas by analysing a part of the middle Meuse valley, a region that is of great importance in the early medieval period. The analysis will concentrate on the arguably most important centre in the valley: Maastricht. Therefore the primary research question of this thesis is as follows:

“How can the landscape in the Merovingian period be characterised in the large region surrounding Maastricht?”

In order to get an idea of how to describe the aforementioned landscape, it is necessary to analyse how different characteristics of the surrounding landscape influenced the Merovingian society. To achieve this, a number of subsidiary questions are raised. It is, for instance, important to ask why certain areas were preferred for specific uses and which natural and cultural factors play a role in the choice of location.

The role of ‘towns’ in the research area is of great importance for the middle Meuse valley. It is therefore necessary to analyse what exactly made the Merovingian town a town. What was its function on the regional level, but also the local level? What was the town’s connection with the hinterland (and does the town even have a ‘hinterland’)?

The last element that is prominently present in the Merovingian landscape are the physical remains of the Roman Empire. Roman towns, villae, cult places and roads fill the landscape (Theuws 2014, 3), an element towards which people of the Merovingian period had to have an attitude, either consciously or unconsciously. How

(15)

did these ruins influence the way the people dealt with the landscape? And how did the Merovingians deal with the Roman ruins?

1.2. Research area

The research area stretches about 75 kilometres both on the east-west and the north-south axis and centres around the area of southern Limburg in the Netherlands (see fig. 1). The area spans from the city of Hasselt (Belgium) in the west to the city of Jülich (Germany) in the east, and from the location where the Rur River flows into the Meuse in the north to the Belgian town of Spa in the south.

The choice of this research area was based on the presence of an interesting development, namely the shift of the major power centre in the region. In the Roman period, Tongres was the municipium of the civitas Tungrorum. Moreover, in the later Roman period, this city was the episcopal centre of the same civitas (Baillien 1979, 5-6). After 405 AD almost nothing is left of this town, losing its episcopal power to Maastricht in the early Merovingian period, which in its turn loses this position gradually to the town of Liège in the late seventh or early eighth century (Panhuysen 2005, 30-33). From the early Carolingian period, the palace and vicus of Aachen became an important centre of power as well as the most important palatium of the Carolingian dynasty, reaching its peak during the reign of Charlemagne. Due to the constant shift of power between these four towns, this area is highly suitable for the analysis of different landscapes through the course of time.

As can be expected in archaeology, the choice of this research area comes with some complications (which will be discussed at a later point). However, these complications should not limit the choice of research area. It should ideally be determined by the research questions asked, not by the analytical problems a specific area provides.

The valley of the River Vesdre has previously been researched in a similar way, focussing on the relations between different sites in the region (van Ruymbeke 2005, 104-105). Unfortunately, it was not possible to obtain the data of this research.

(16)

Figure 1: research area (after: De Grote Bosatlas 55th edition)

1.3. Chronology

The goal of this thesis is to investigate the way people deal with the landscape in the Merovingian period (450-750 AD). However, a view on this subject would be insignificant without a context to pair it with. In this light, the described situation will continuously be compared to the late-Roman and the early Carolingian situation. This

(17)

will also allow us to see whether the perspective on the landscape is changing or continuous over a long period of time

1.4. Relevance of the subject

As mentioned earlier in this chapter, one of the main goals of this study is the creation of a regional overview of the Merovingian landscape, something barely appreciated in Merovingian research up to this point. The creation of such an overview is of course not the purpose itself, but the usefulness of such a general overview and a distribution map can provide substantial insight into the way the landscape was perceived in Merovingian times. Moreover, it might prove a useful case-study in showing the strength of regional landscape research.

If proven effective this method might lead to a better understanding of the different elements in the Merovingian landscape and how, together, they formed the Merovingian societal lay-out that archaeologists excavate. This thesis explores the relation between various elements. For instance: the relation between settlement and cemetery, church and cemetery, or towns and their natural environment, to name just a few. In a sense, it creates a conceptualisation of the regional landscape.

Many of the problems this thesis deals with are consistent with the Dutch research agenda of archaeology, commissioned by the Dutch governmental service on cultural heritage (RCE 2016, 13-17). This research agenda poses 117 archaeological questions to guide Dutch archaeological research for the coming years. These 117 questions have been summarised in 23 research themes. In total, nine of these themes are, to a greater or lesser extent, dealt with in this research. These themes are for instance: the transition from the Roman to the early Merovingian period, Christianisation, development of towns, and dynamics in land use (appendix 1, for the complete list). This correspondence shows that the Dutch national archaeological service acknowledges the lack of sufficient comprehending of the themes dealt with in this thesis and indicates that these kinds of studies are required.

(18)

This thesis continues with providing the environmental and historical background of the research area as well as a theoretical framework. These form the basis for interpreting the result of the analysis, as described in the methodology. This thesis will conclude with a discussion of the usefulness of this methodology and an answer to the research question posed in this chapter, as well as recommendations for further research in this field.

(19)

2. Environmental and Historical background

This chapter gives a general overview of the context of the research area. The first paragraph provides this from an environmental point of view, the second one gives a general overview of the history of this area. The different maps that show the environmental situation of the research area can be found in appendices 2 to 5.

2.1. Environmental Background

The environment of the research area is characterised by two main elements. The first of these is the presence of elevation differences. The southern part of the research area is characterised by the Ardennes middle mountain range. The Ardennes form an environment of erosion, which provides the material for deposition on the alluvial plains downstream, starting approximately around Maastricht. The river Meuse has also changed the elevation in these areas. Through processes of erosion the Meuse river left higher plateaus, like the terraces of Dutch Limburg and the Kempen plateau.

These major rivers, mainly the river Meuse and Rur, form the second main element in this research area. These rivers, and also their tributaries, play a primary role in the geomorphological formation of the landscape in the research area. They are the main actors in the process of erosion and deposition in the area and form both natural borders and connecting routes through the landscape. In order to facilitate a more detailed description of the natural environment, the research area has been divided in multiple geographical regions (fig. 2). A description of these areas follows below.

2.1.1.The Ardennes

The forest of the Ardennes forms the southern part of the research area. The Ardennes is an old middle range mountainous area that has been subject to erosion ever since its geological formation. A number of different kinds of geomorphological formations are present at the surface of this area. In the most southern part of the

(20)

Figure 2: different regions in the research area mentioned in the basic environmental description.

research area, schist and quartzite are the main geological element. The northern parts of the Ardennes are characterised by the presence of limestone, chalk, shale and

sandstone, among other types of lithology (EDGI-database2_{). Large parts of the area are}

currently covered with pine forests, which prevents the area from further weathering and eroding.

Luvisols3_{and cambisols}4_{are the main soil types in this area, the first of these is a}

very suitable soil for crop cultivation, while the latter is less suitable for this purpose

(Strahler and Strahler 2006, 363-395). The only exceptions are the histosols5_{in the}

natural reserve of the ‘Haute Fagnes’. These peaty soils are heavily saturated and

2_{Online database and tool on European geology (www.europe-geology.eu)}

3_{Soils with clay particles in the B-horizon, well drained soils with good fertile properties.} 4_{Soils with a beginning soil formation and thin horizons, not (yet) very fertile.}

(21)

therefore unsuitable for both agriculture and civil engineering purposes (Strahler and Strahler 2006, 363-395; ISRIC 1989).

Many tributaries of the Meuse (The Vesdre, Ourthe, Hoyeux and Amblève), the Rur (Inde and Weser) and other minor streams run through this mountainous area. These form transportation routes through the landscape of the Ardennes, not only for people or goods, but also for sediment. Many of the river and stream valleys are

covered in colluvium6_{that is transported by the rivers towards the north, but these}

valleys themselves are likewise subject to sedimentation (Strahler and Strahler 2006, 363-395; ISRIC 1989).

2.1.2. The Condroz

The Condroz is a region along the promontory of the Ardennes middle mountain range and on the southern banks of the Meuse River. Only the most north-eastern tip of the Condroz-region is located within the research area. This undulating landscape is formed by ridges of varying lithological formations, for instance chalk, limestone and conglomerates (EDGI-database). Like the Ardennes mountain range, this region contains highly fertile luvisols.

2.1.3. The Hesbaye

The Hesbaye region is located between the Meuse river in the south, and the Demer river and the Kempen plateau in the north. The Hesbaye forms a transition from the limestone and shale just north of the Meuse (EDGI-database), to the lower alluvial plains of northern Belgium and the Netherlands. This is a shifting landscape between the Meuse and Demer rivers. The southern part, between the Jeker and Meuse rivers, is an undulating landscape, currently known for its high fertility, due to the luvisols that cover this region (Strahler and Strahler 2006, 363-395; ISRIC 1989).

6_{Colluvium is a package of sedimentation that eroded from the tops of hills downwards to the}

(22)

The northern part of the Hesbaye, located between the rivers Jeker and Demer, is a sedimentation environment. This is a relatively low-lying area between the southern Hesbaye hills and the plateau of the Kempen and mostly consists of aeolian sand deposits, sedimentary rock and alternating river deposits (EDGI-database, viewed

11-10-2017). The soils in this area consist of luvisols in the south and a mixture of podzols7_and

luvisols, otherwise known as podzoluvisols, in the north (Strahler and Strahler 2006, 363-395; ISRIC 1989).

2.1.4. The plateau d’Herve

The description of the plateau of Herve is very comparable to that of the Hesbaye, but some important differences should be noted. This area is located between the rivers Vesdre and Meuse and the Belgian-Dutch border. The landscape of the plateau of Herve, a promontory of the Ardennes Mountains is maintained in the hilly landscape of the Meuse terrace of southern Dutch Limburg. Furthermore, the plateau of Herve mainly consists of cambisols, while luvisols are less prevalent, which means they are presently not suitable for crop cultivation (Strahler and Strahler 2006, 363-395; ISRIC 1989).

2.1.5. Dutch Limburg

The southern part of the Dutch province of Limburg is characterised by the most northern promontory of the Ardennes which was largely carved out by the Meuse River. Sudden changes in climate and tectonic uplift caused the Meuse to carve out this environment in different areas, leaving steep terraces in the landscape (Berendsen 1997, 10-25). From the border between the Netherlands and Belgium, the Meuse changes in character from a mountainous rain river to a meandering lowland river (de Bakker 1990, 167-176). The Meuse deposited large amount of clay on the adjacent alluvial plains, but due to the steep terraces, the higher parts of the region of Limburg have been lacking

7_{Podzols are soils in sandy soils with a high amount of downwards leaching of mineral, producing}

(23)

sedimentary deposition. Aeolian sands and silts were deposited in this area after the last Ice Age (Berendsen 1997, 10-12; EDGI-database). These areas were subject to the formation of fertile luvisols (Strahler and Strahler 2006, 363-395; ISRIC 1989). Since the land reclamations in the later medieval period, the higher plateaus have started eroding and created a thick colluvium that filled the lower valleys in this area. This should be noted while looking at the distribution of early medieval finds in the landscape. These areas will see a negative bias towards location choice due to the absence of escavated early medieval archaeology, as these are likely to be buried under a thick layer of colluvium. The most southern part of Dutch Limburg has largely been untouched by the erosion of the Meuse River and shows characteristics comparable to the plateau of Herve. This region contains large areas of chalk in the subsoil as well as the aeolian sands and silts (EDGI-database, viewed 11-10-2017).

2.1.6. The Belgian Kempen region

The Kempen region is almost completely made up of cover sands, apart from some fluvial sediments (Berendsen 1997, 26-41; EDGI-database). The formation of the Kempen plateau is comparable to the formation of the terraces of the Meuse River in the Dutch province of Limburg. The steep ridges of the plateau bear witness of fluvial erosion by the Meuse in the east and the Demer in the south (Paulissen 1970, 167-212). The region consists of podzols which are relatively unfertile (Strahler and Strahler 2006, 363-395; ISRIC 1989), especially on the plateau due to a low water table. Typically, villages in this region are therefore mostly located in stream valleys (Paulissen 1970, 167-212).

2.1.7. The German Vor-Eifel

In a geological sense, the German Eifel is practically the German name for the Ardennes. This is very clear on the geological map, as it shows a continuation of lithological formations. The German Vor-Eifel (also known as Rur-Eifel) consists of shale and some limestone in the northern parts around Aachen and a tiny area of granite just

(24)

north of the town of Simmerath (EDGI-database). The soils consist mostly of cambisols, similar to the Ardennes (Strahler and Strahler 2006, 363-395; ISRIC 1989).

2.1.8. The Rur-valley

The development of the Rur-valley and its tributaries (especially the Wurm and the Inde) have formed in a similar fashion as the Meuse valley. Just south of Düren, the river transforms from a rain river into a lowlands meandering river. These lowlands have been cut out in terraces through the ages, similar to the situation of the Meuse River, although less pronounced. The same goes for its tributaries, the Wurm and Inde rivers. The area is covered in aeolian sands and silts (Plum 1995, 11-18), forming luvisols, and some podzols are present locally, for instance in the area around Heerlen-Übach and north of Heinsberg. East of Jülich lies an area containing cambisols (Strahler and Strahler 2006, 363-395; ISRIC 1989). In recent times a large part of this area around Aldenhove has been reshaped due to the intensive ignite mining in this area. This is especially the case in the area between Aachen and Jülich. Current mining areas are still visible on the elevation maps used in this thesis. These mining activities have caused recent diversions of rivers and partial flattening of the landscape. However, this has also created an area with a very high archaeological research intensity. This is something that has to be taken into account when analysing the data.

(25)

2.2. Historical background

In addition to the geological, soil and other environmental processes, this region has been subjected to formation in a historical sense. Human interactions with the landscape, prior to and after the Merovingian period, have left a historical depth in the landscape. This is an important factor in the formation of an attitude towards the landscape, as people can associate with or distance themselves from the historical heritage. In order to characterise the historical depth of the landscape, it is necessary to establish a short historical overview of the research area. The development of the important towns, like Aachen and Maastricht, in the research area is presented in the following section.

2.2.1. The pre-Roman period (before 1 AD)

Already in the Iron Age, people in the research area interacted through the Meuse and other routes leading to northern France. Through their exchange networks, for instance, they adopted the Marne-type pottery (early-middle Iron Age) from the middle Meuse region, showing interaction and exchange through these channels. People lived in small agricultural villages and focused on the production of cereals, for example wheat, which became particularly interesting for the Roman legions in a later period. These small communities created communal cemeteries in between villages, cremated their dead and buried the ashes in urns under small hills (Louwe Kooijmans et al. 2005).

2.2.2. The early Roman period (1AD – 300AD)

During his campaigns in Gaul, Caesar conquered the region of Belgica secunda around 55 BC. At that time it was inhabited by a Gallic tribe called the Eburones, later named the Tungri (hence the name of Tongres). After it was conquered, the Romans placed a legionary camp in the region which grew to become the town of Tongres (fig. 3) over the course of the first century AD (Nouwen 1997, 31-86). In the second century, the city became the capital of the equally named civitas, upon which the city commenced

(26)

Figure 3: Topography of Roman Tongres. In red: early Roman period, black: middle Roman period, green: late Roman period, blue: natural landscape (after: Vanvinckenroye 1971, appendix 1).

building its first city wall (Creemers et al. 1999). The surrounding area of Tongres, the Hesbaye, was used extensively for agricultural purposes from this time onwards. The fertile areas became filled with villae, especially along important infrastructure.

The most important road in northern Gaul, from Bavay to Cologne connects Tongres with the German Limes. This road was used to connect the highly fertile lands of the Hesbaye to the limes in order to supply the troops. At this point, the town of Maastricht (fig. 4) was developed, first as a vicus, later in the form of a castellum for the protection of the important bridge across the Meuse River. Besides the crossing of the road to Cologne and the river Meuse, Maastricht is also the starting point of the road towards Nijmegen along the western banks of the Meuse. It is suggested that a secondary road in that direction led from Maastricht along the eastern banks (Panhuysen 1996, 28-31). The presence of a southern road in the direction of St. Peter’s

(27)

Figure 4: Topography of Roman Maastricht. 1: walls, 2: reconstructed walls, 3: hypothetical walls, 4: foundations, 5: gravel road, 6: Roman road Bavay-Cologne, 7: Meuse banks confirmed, 8: Meuse banks reconstructed, 9: early 1st_{century wooden}

construction (Panhuysen 1996, appendix 3) .

hill has been suggested as well as an early medieval road towards Trier, called the via

mansuerisca.

Even though the existence of some form of habitation next to the bridge over the Meuse has been documented in literary sources, a find layer from the first century AD is absent in most parts of the Maastricht at the western bank of the Meuse, although, a fair amount of imported pottery from this period has been found. At the eastern side of the river, some farmsteads from the second half of the first century were found. In the second and third centuries AD, Maastricht grew from a small settlement next to the bridge to a moderate sized Roman vicus (Panhuysen 1996, 33-51).

Aachen was conquered by the Romans shortly after the areas in Belgium (fig. 5). It was quickly recognized for the quality of its hot springs and used by Roman soldiers stationed at the German limes. The Roman name for Aachen,

Aquae Granni, refers to those

properties. A larger scale settlement, though, has mostly formed on the market hill from the first century AD (Cüppers 1982, 1-76).

(28)

The site of Liège remains relatively empty during the Roman period. In the second century AD, a large villa is constructed at the present Place Saint-Lambert, the location of the later episcopal church of the town. It has been debated whether there was a vicus present at the location of the bridge crossing the river Meuse, at Herstal, on the trajectory of the road from Tongres to Arlon. It has also been suggested that the

vicus was located on the other side of the river, as more Roman tombs were excavated

in Jupille than in Liège itself (Degbomont 1990, 137-142). Remains of a vicus have never been found though.

The remainder of the area can probably be characterised as rural area, with many villae, local villages and natural vegetation, intersected by Roman infrastructure.

(29)

2.2.3. Late Roman development (300 - 450AD)

The late Roman period is characterised by an increased struggle to maintain the

limes and hold on to the border provinces. In an attempt to hold on to power, foederati

were stationed on the Roman side of the border. In the Civitas Tungrorum, where the research area is located, large numbers of Frankish foederati were deployed. At the same time, the defences of cities and castella were adapted and many villae were already deserted at the end of the third century. Many cities close to the Roman border show additional responses to the growing unrest in the form of fortification of the towns’ defences.

Tongres clearly shows an example of these updated defences. After the town had shrunk following its heydays in the second and early third centuries, the number of citizens had diminished and the fourth-century defences covered a much smaller area (fig. 2). However, it is clear that continuous habitation is present in the city until at least the start of the fifth century, though possibly in very small numbers (Vanvinckenroye 1975, 44-71; Vanderhoeven 2011, 128-138). From the later fourth century onwards a Christian group may have been present in Tongres. This is also the phase of the construction of a basilica, which was erected inside the city walls (Vanderhoeven 2011, 128-138). Whether this basilica functioned as a Christian cult place is still under debate, although the presence of an episcopal seat in Tongres seems to suggest that this was indeed the case.

Like Tongres, the town of Maastricht had shrunk considerably by this time. The defences of the fourth century show the use of a much smaller plot of land on the western bank of the Meuse at the bridge. These defences consist of thick walls with towers on the corners, surrounded by a ditch (fig. 6). The construction of these walls exemplifies two main developments. Firstly, the apparent necessity of heavy defences around cities indicates that there were many threats during this period. Secondly, it was apparently considered to be important to hold on to the location of the Meuse river bridge.

The dead were buried along the Roman road to the west of the castellum. According to legend, St. Servatius was buried in this cemetery around 384 AD and

(30)

Panhuysen (1996, 51-73) suggests that a burial chapel was erected at the location of his grave. It should be noted however that both these statements are still under discussion. It is not known whether Servatius was actually buried there, let alone that a burial chapel was erected on his grave (Theuws and Kars 2017, 38-39) This supposed chapel later grew to become the basilica of St. Servatius (Panhuysen 1996, 51-73). However, the early history of the St. Servatius Complex is still unclear and heavily debated.

The Gallo-Roman villa at the Place St. Lambert was, like most villae in the area, deserted in the late third century AD, and aside from some single finds from the fourth or fifth century, no real reoccupation of the site is recognized until the sixth century (Otte 1992, 245-249).

The developments seen in the above-mentioned towns are much less clear in Aachen. It has been suggested that Aachen, as any other town, was fortified in the fourth century, although clear evidence of this development has not been found (Cüppers 1982, 11-14). A process that

has been observed though, is Christianisation. One of the Roman bathhouses, die Münsterquellen, lost its function in the later fourth century possibly in the early fifth century. In its place, according to Cüppers (1982, 32-38) a small Christian church was constructed from the remains of the Roman bathhouse. The rest of the bathhouse was reconstructed into living quarters, supposedly those of the Christian community. This first church was erected on the location of the later chapel of the palatium of Aachen (Cüppers 1982, 32-38).

Figure 6: Late Roman Castellum at Maastricht. 1: walls, 2: walls reconstructed, 3: walls hypothetical, 4: foundations, 5: Roman road Bavay-Cologne, 6: Meuse banks,

(31)

2.2.4. The Merovingian period (450 - 750AD)

After Roman authority lost control over the northern areas of the former empire, local rulers gradually bound warriors to themselves in a struggle for power. The Merovingian dynasty was one of the most successful in this regard. Childeric, possibly a former Roman general, was the first to solidify the Merovingian dynasty in the area of

Belgica Secunda8_{. His successor, Chlodovech}9_{the first, greatly expanded the kingdom he} inherited from his father towards the Rhine in the east and into Visigoth territories in the south. Around 500 AD he converted to Christianity the official religion of the Merovingian kings from then on (Dierkens 1996, 183-191). This conversion made bishops part of the power structure of society.

Although bishops after St. Servatius moved the seat of the bishopric to Maastricht somewhere around the sixth century, the bishop’s seat was officially still located in Tongres. The bishop’s seat had officially been moved to Maastricht before 535 AD at its latest (Baillien 1979, 5-6). The bishop lived in Maastricht though, meaning that Tongres might not have fulfilled more than a ceremonial function. Whether the occupation of Tongres shows continuity into the early medieval period remains relatively unclear. Only a few archaeological remains date to this period, among which only one inhumation on a late Roman cemetery. However, written sources speak of a meeting between two bishops in Tongres around 660 (Baillien 1979, 5-14).

While Tongres diminished in the Merovingian period, Maastricht grew far beyond the walls of the fourth-century castellum towards the cemetery and the church of St. Servatius in the west and another cemetery in the north (fig. 7). The eastern banks of the Meuse, present-day Wijck, were also gradually inhabited. This development made Maastricht a central place in the administration of the diocese and together with the pilgrimage for St. Servatius, this attracted many new inhabitants. At the same time the Basilica of Our Lady, inside the castellum, must still have been in use (Panhuysen 2011, 67-79). The cemeteries along the road to the west of the settlement near the St. Servatius complex show a total of 818 certain and 468 possible Merovingian inhumation

8_{The region of Hesbaye (see fig. 2) in the research area is part of Belgica Secunda as well.} 9_{more commonly known by his Romanized name, Clovis}

(32)

graves (Kars 2011, 95-100), and another 40 inhumations to the northern area (Panhuysen 2011, 67-79). In the year 703-705, St. Lambert, at that time the bishop of Maastricht, was murdered at Liège. At the site of his death, a martyrium was constructed. Since this event, the bishops started to reside in Liège and at one point during the eighth century, the episcopal seat was moved there (Panhuysen 2011, 67-79).

After a period of abandonment, a Merovingian village had grown in Liège from the seventh century onwards. Only a few finds can be traced to the period between the Roman abandonment and the seventh century. It is known that in the late seventh

Figure 7: Merovingian Maastricht, 1: Roman road Bavay-Cologne, 2: early medieval street network, 3: early medieval settlement, 4: early medieval cemetery, 5: the rivers Meuse and Jeker (Panhuysen 2011, 70)

(33)

century, the bishop of Maastricht, St. Lambert, builds a domus in Liège on the ruins of the Roman villa (Stiennon 1984, 295-297). As stated above, after the death of St. Lambert around 705, the episcopal seat gradually moved to Liège in the course of the eighth century, after his remains were transported to Liège (Kupper 1984, 21-22).

A similar problem regarding the Merovingian occupation as compared to Liège occured in Aachen. After the fall of the Roman Empire, the number of finds from Aachen drops, although recent excavations have shown this might have been less pronounced. The early Merovingian phase is merely represented in the form of sherd finds. In the seventh century, a cemetery is situated on the site of the later Carolingian Cathedral (Schaub 2011, 119-127).

2.2.5. The early Carolingian period (after 750 AD)

At the end of the seventh century, the Merovingian dynasty experienced heavy power struggles of which other aristocratic families took advantage and acquired more influence. Among these was the dynasty of the Pippinids in the middle Meuse valley, which secured a powerful position at the Merovingian court for its descendants. One of those, Pippin III (the Short), eventually definitively overthrew the Merovingian king in 751 and acquired the title of king, starting the rule of the Carolingian dynasty. The son of Pippin the Short, Charlemagne, expanded the kingdom to its largest extent. He made the pope crown him emperor in 800, making him the first emperor in western Europe since the Roman era (Scheiffer 2014, 25-27). Their centre of power was located in Aachen.

In the early Carolingian period, Tongres was considered to be part of the crown lands of the Carolingian dynasty. It was presumably Pippin the Short who partially donated the lands of Tongres to the Basilica of Our Lady in Tongres. The other part was donated to the bishopric of Liège. It is therefore presumed that the Basilica never halted their service during the Merovingian period (Baillien 1979, 15-20).

The Carolingian phase in Maastricht is very ambiguous and scholars do not agree whether the town had grown from the complex of St. Servatius or from the Merovingian settlement at the location of the former Roman castellum of the fourth century. The

(34)

curve of the later city wall around the St. Servatius complex, however, seems to indicate a development from a rounded castrum-like area that is possibly located at this site. This area might have been connected by a road towards the river Meuse and the old Roman bridge. At the same time, some sort of market village developed at the other side of the river (the current location of Wijck) (Theuws 2005, 87-122). Even though the bishopric had moved away from the complex of Our Lady in Maastricht to Liège, Maastricht was still an important religious centre of pilgrimage for St. Servatius and the complex started to accumulate possessions of land from the ninth century onwards (Panhuysen 2011, 78-79).

The bishops of the Carolingian period were seated in Liège. The martyrium erected for St. Lambert became the new Episcopal church at this time. The excavations at the Place St. Lambert uncovered the Carolingian church from this period, as well as a number of other buildings from the period (Otte 1992. 248; Otte 1988, 250).

Because it was assigned to be one of the most important palatiae of Charlemange, Aachen goes through many changes during the Carolingian period. The most important aspect is, of course, the construction of said palatium (fig. 8). Additionally, an Aula Regia was constructed, in the north, on the current market square. This Aula Regia was fit for public matters of the king and was connected to the cathedral of Aachen by a long hallway. The construction of the cathedral, on the location of the Merovingian cemetery and Roman Christian cult place, lead to the relocation of the cemetery towards the southeast. The exact location and character of the settlement itself are still quite unclear (Müller and Schaub 2014, 246-253).

2.3. The research area in short

This chapter has provided a short overview of the area being researched in this thesis. It demonstrates the unique natural position of the area on the transition of the Ardennes to the Meuse floodplains. Moreover, this area contains four towns that played major roles in the region. This region has seen many shifts in power relations during the first millennium AD, which shaped the landscape as we know it today.

(35)

Figure 8: Reconstruction of the palatium of Aachen. In the left is the Aula Regia connected to the chapel (right) by the hallway (Ristow 2014, 232).

(36)

(37)

3. Theoretical Background

In order to answer the research questions posed in the first chapter, it is necessary to elaborate on certain relevant theoretical subjects. These include concepts that deal with different aspects of society, for instance, economical and socio-cultural aspects in relation to the landscape.

Since we are dealing with a relatively broad research area, there is a strong necessity for concepts that help to theoretically structure space. This needs to be done in a way that the concepts can be used in a sensible fashion, which helps to analyse the chronological and spatial development of the hinterland of Maastricht.

This chapter will present a theoretical discussion on the broader concept of landscape archaeology as applied paradigm. The pros and cons of this approach will be discussed, followed by a characterisation of towns in the early medieval situation. The chapter will conclude with a discussion on the appropriation, reuse and recycling of older elements present in the early medieval landscape.

3.1. Landscape archaeology

Landscape archaeology is a branch of archaeological thought that aims to combine the spatial elements of landscapes with its physical components and immaterial values. This branch developed simultaneously in multiple scientific fields through the shared focus on the understanding of landscapes, not only in case-studies, but especially on a theoretical level. The term landscape itself has remained the subject of continuous theoretical discussion and ambiguous practice since it has come in use and different scholars use the term in different ways (Olwig 1996, 630-631).

(38)

3.1.1. The history of landscapes

The term ‘landscape’ started being used during the Renaissance in relation to

the practice of ‘landscape painting’ in the Low Countries where this particular type of painting came into fashion (Doneus 2013, 40-45). The landscape in these paintings was not necessarily the physical surroundings in which humans act, but rather the view of a single person on this environment. The use of this concept as an aesthetic value was then picked up by social geographers, who used the concept to add a layer to the land that accounts for the personal (aesthetic) values that humans perceive and create when they engage in these landscapes. Thus, landscapes were no longer a background for the picture, but rather the subject. Sometimes these landscapes were not realistic, but imagined or exaggerated to accentuate a message sent by the painter (see fig. 9).

The use of the word landscape changed through time and translation as well. In present-day society, the word still holds varying meanings in different languages. The English word ‘landscape’ stands for what one can see from a point of perspective. This stresses that a landscape is a sensation, something a person experiences at one point in time. The German word ‘Landschaft’ on the contrary is a more administrative term, meaning a well-defined area of land, which stresses the spatial extent of a landscape. Lastly, the Dutch word ‘landschap’, means a piece of land that holds special meaning towards the construction of personal, communal, political or locational identity. The

(39)

Dutch meaning therefore stresses the long-term subjective relation between humans and their environment that constitutes such landscapes (Olwig 1996, 630-631). Landscapes are thus in its basis highly personal sets of cultural, religious and economic values projected on the surroundings. It should, however, be noted that the scale of this view can vary. When zooming-out, a personal view can become the view of a family, a village, a cultural group, religious group, a group of people with the same profession or any other type of community.

There is an objective counterpart, to landscape, the subjective personal view on the surroundings, called the environment. This is the totality of measurable physical properties of the surrounding. These can, for example, be geological properties, soils, air,

rivers, fauna, flora and elevation.10

3.1.2. A landscape archaeology paradigm?

Ever since archaeologists have been interested in the past, they have been interested in the spatial span and distribution of this past. However, this spatial element used to be merely the background against which the archaeology was found (Ashmore and Knapp 1999, 1). From the late seventies and early eighties of the twentieth century onwards, the concept of landscape gradually gained more attention in archaeology and the term started to make its appearance in the major archaeological journals (David and Thomas 2008, 28-30).

Processual archaeologists of the seventies and eighties used this concept in a rather different fashion then their colleagues in social geography. According to Processual archaeologists, the landscape was an economic factor, which determined the degree to which humans could thrive in their environment and landscape (Anschuetz et

al. 2001, 170-171). Humans feed and live off the landscape and can even alter their

landscape (David and Thomas 2008, 28-30). Landscapes, therefore, were the basis on which humans lived their lives and it provided the key variables for the habitability of a

(40)

certain location. Thus, landscapes were considered ideal for creating models of how people used this landscape.

One of the most influential archaeologists within the processual line of thought was Lewis Binford. His theoretical essays, as well as his fieldwork studies, show a strong desire to model human actions in systems, based on the optimised strategy in a certain environment. A clear case of this systemic modelling is seen in his ethnoarchaeological research of the Nunamiut native communities (Binford 1978). Binford tracked the migration patterns of these inhabitants of Alaska and analysed their food gathering strategies. His conclusion was that the Nunamiut migrated to locations where food was most abundant in a particular time of the year. They systematically followed the same cycles every year, from each of these seasonal camps smaller task camps were set out in areas with more production opportunities (see fig. 10) (Binford 1983, 357-366). This modelling is typical for the systemic approach of the processual archaeology, as the processual archaeology based itself

primarily on empirical research and testability, the lack of which was their main critique on ‘traditional archaeology’ (Clarke 1972; Binford 1989, 12-23). Reher (1977, 13-40) presents a clear case of this line of thought in a study in which he argues that change within socio-cultural hunter-gather groups is primarily induced by climate change, which caused the hunter-gatherers to adapt their system to the new environmental situation. This Darwinian reasoning shows the easily adaptive nature of human beings to their ever-changing environment. Not just

(41)

human biology, but also culture was seen as an evolutionary system of multiple variables, which formed a social construct implemented to ensure group survival

(Binford 1972, 89-97).

When this systemic theoretical approach is applied in landscape archaeology we can expect to see results comparable to the Nunamiut study of Binford (1978). A suitable tool for this kind of research is GIS analysis, although the processual archaeologists did not have access to this technique. GIS provides an easily-accessible and manipulative digital landscape in which one could experiment with the many variables while providing a readable interface (Chapman 2006, 129-140). GIS analysts have tried their best to keep up with the theoretical framework of landscape archaeology, but recent advances in especially phenomenological approaches (see Tilley 1996) has proven to be a challenge for computer simulations. It has been especially challenging to command the program to show human emotions that correspond to the landscape in which the agent is dealing. Moreover, phenomenology states that one’s experience in a landscape is heavily personally determined. This personal experience is not easy to recreate in a GIS-environment. This causes GIS-analysis to be a tool, which is easily applicable in environmental archaeology, but theoretically unsuitable for post-processual, especially phenomenological, approaches in landscape research.

The role of the landscape in processual research is one of an environmental prerequisite for human survival. Humans are dependent on the environment and what it provides them in terms of sustainability. Environments (not to be confused with landscapes) were to be understood as the preconditions and limitations of human lives in the broadest possible sense. It should however be noted that these processual studies were mainly conducted on hunter-gatherer societies. One might wonder whether this could be applied by more complex civilisations. According to the processual paradigm, human societies will always adapt to their environmental circumstances and try to profit from these as efficiently as possible. In other words, a human being is to be understood as a Homo oeconomicus. The focus of research on models and systems in the processual archaeology came from a functionalistic movement.

(42)

The anti-functional movement of post-processual archaeology, in response to the evolutionary determinism of processual archaeology has also found its way to the landscape archaeologists. With the introduction of the phenomenological approaches in landscape archaeology, research saw a return to the original meaning as applied in social geography. The landscape became something personal once again. Every single individual has a different view on the landscape, which is constituted partly by the physical and spatial elements of the landscape, but mostly by the immaterial values with which the person associates the landscape. (Doneus 2013, 25; Ashmore and Knapp 1999, 3; Tilley 1994; Ingold 1993, 156; Tuan 1979, 90, 100; Meinig 1979, 33). In my opinion this early postmodern view carries this a bit too far, in the spirit of modern individualism. It is impossible to recreate landscapes of every possible individual, a practice that does not seem to resolve any theoretical research questions. Moreover, individuals are always part of a group, a society or a nation.

As a response to the structuralist approach from processual archaeology, post-processual archaeologists tried to come up with new research methods that would be suitable for their changing view on the relation between humans and nature. According to this new movement, human action was not suitable to be captured in structures and models. A real Homo Oeconomicus did not exist and slowly alternative theories were developed in an attempt to capture the character of human behaviour and perception in relation to the landscape.

One of the most prominent theory that was developed within this movement was the use of the concept of phenomenology. This concept was derived from theories of the German philosopher Edmund Husserl (Kockelmans 1994), later followers like Martin Heidegger (1975 [1923]; 1975 [1927]) and the French philosopher Merleau-Ponty (1945). The basis of the theory focuses on human sensory feelings and personal experience, for the study of human behaviour. It was no longer acceptable to summarise human behaviour into models, as the experience of a phenomenon was considered highly subjective, personal and influenced by many different factors, among which descent, upbringing, cultural background, past experience, memory and sensory feelings (Merleau-Ponty 1945, 11). Or as Merleau-Ponty put it:

(43)

“La rationalité est exactement mesurée aux experiences dans lesquelles elle se révèle”11

(Merleau-Ponty 1945, 15)

Tilley (1994) was the first to use this concept in combination with landscape archaeology, the field in which it is still primarily used. In this approach, archaeological landscapes were to be studied from the view of the people that inhabited the landscape. It was about the relationship between ‘Being’ (Sein) and ‘Being-in-the-world’ (Dasein), meaning that the person is able to objectify him- or herself from the landscape in order to analyse it, through his or her perspective. This analysis is then used by said person to engage in and reflect on the landscape (Tilley 1994, 12). The human body, therefore, becomes a fundamental point of reflection between the world and the observing subject. The human body perceives the world, reflects on it and understands it in its own way (Johnson 2007, 193-202; Tilley 1994, 14).

Even though this phenomenological approach seems to bring us closer to the nature of perspectives, this method has some major drawbacks. The method can exclusively be used in a descriptive manner and is not suitable for explanation and analyses, because of its inherent subjectivity (Merleau-Ponty 1945, 11). Moreover, it also assumes a certain measure of human uniformity. While we are now trying to sense the experience of the landscape, the way that people might have experienced that same landscape could very well have taken different forms in different contexts. Not only will this have changed through time and space, but also through changing environments and seasonal weather changes, depending on gender, age, cultural differences and natural variations in the human body (Hamilton and Whitehouse 2006, 34; Brück 1998, 23-36).

Unfortunately, the aforementioned differences in interpretation of landscapes between paradigms leave the reader with an unsatisfying division between multiple ways of interpreting landscapes. In a response to this dichotomy Anschuetz et al. (2001, 160-161) made a case for a separate paradigm for landscape archaeology in order to come to a general consensus on the term landscape and how it should be used. The landscape was supposed to be more than just a simple relation between humans and

11_{Translation: Rationality can be measured exactly through experiences in which it reveals the}

(44)

environment, but rather a multi-layered complex of space12_{, environment and values} (Ashmore and Knapp 1999, 8; Doneus 2003, 46; Roymans et al 2009, 336).

As is outlined above, the use of landscape as a research subject within archaeology has gone through a number of major shifts: starting with a use of the concept that came from social geography, it shifted to a more systematic approach with the emergence of processual archaeology. With the shift to post-processual archaeology, the use of landscapes in archaeology also changed and became very personalised, due to the use of phenomenological approaches and ultimately leading to the proposal of a separate ‘landscape archaeology’ paradigm.

Whether landscape archaeology should actually have its own paradigm within archaeology might be one step too far, since the theory is comparable to other theories on human thing relations. In my opinion it should not matter whether that thing is something tangible, for example material culture, or something more abstract, for instance landscapes. However, the reasoning behind it addresses some current issues in the discipline of landscape archaeology. At this point, the first step in reading an article on the subject of landscape archaeology is finding out what the author actually means by the terms landscape archaeology and landscape. Every author has different interpretations, ranging from purely systematic to phenomenological approaches. Both sides have over-exaggerated their views on relation between humans and the landscape. On the one hand, the structuralist approach in which humans are merely an organism functioning in a large system. On the other hand the phenomenological approach which tends to strive towards a high individualistic human agency that is impossible to research at all. In my opinion, it is the grey area in between that requires attention. From a combination of these perspectives, it can be possible to extract the behaviour of human groups or communities in a landscape.

(45)

3.1.3. Defining Landscape

As discussed above, landscapes, as they are used in archaeology, are derived from social geography. One of the first scholars to use the term social landscape was the geographer Sauer (1925). He stated that cultural landscapes were fashioned from the natural environment by a cultural group (see fig. 11) (Sauer 1925, 46). This placed the cultural group as the main agent in the formation of the landscape, as opposed to nature being the agent in environments (Anschuetz et al. 2001, 164).

Figure 11: Sauer’s (1925, 46) model of the creation of cultural landscapes

Sauer’s work has been leading in the consolidation of the term landscape in

social geography and later archaeology. The concept of a relation between human and nature has been the basis of the term ever since, although the meaning of the word has grown more complex ever since and has come to stand for an overarching concept. The following quote describes this concept in all its complexity:

“Landschaft ist also das Ganze, das Plenum, der Kontext,Sie lässt sich beschreiben als strukturelles13_{Phänomen, geprägt durch die Raum-struktur, das}

heißt dialektisch zwischen Ort und Raum als Trägerin von Werten, beides verbunden und abgegrenzt durch den Übergang, die Grenze.”14

(Gramsch 2003, 49)

13_{I believe Gramsch does not necessarily say that landscape are unchangeable, but rather that}

the phenomenon of landscapes follows a standard structure.

14_{Translation: Landscape is also the whole, the entirety, the context. It can be described as a}

structural phenomenon, carried by the structure of space, in other words switching between place and space, as the carrier of values, both connected and demarcated by the transition, the border.

(46)

Landscape is, therefore, an all-encompassing concept, described in a structural phenomenology that carries value in a demarcated area. However, this does not mean that landscape is a simple equation of the relation between humans and environment. If anything, landscapes are highly dependent on its context (Ashmore and Knapp 1999, 6-8).

Since a landscape is a projection of the mind on the environment, the perception of the landscape can vary from person to person (Tuan 1979, 90; 100). However, it does not seem relevant to the archaeological debate to recreate individual landscapes. Instead, it is more fruitful to look at landscapes of specific related people. On a higher scale, this can be seen as a group perspective, and a landscape can thus be seen as the world as it is seen by the communities that live in, perceive and interact with this environment (Ingold 1993, 156). Every generation is able to leave its imprint on the landscape, which can be read by younger generations who in turn make their own imprint on the landscape (Aston and Rowley 1974, 11). This means that the landscape is continuously being constructed and deconstructed, both intentionally and unintentionally, creating a multi-layered complex made up by many different physical and mental concepts, which work in different ways when applied to different scales (Roymans et al. 2009, 339).

This seems like quite a complex explanation for a seemingly simple noun, but in

order to capture all of the many aspects of the term, this has become a necessity. In order to overcome this problem, Anschuetz (et al. 2001, 160-161) wrote down four premises of landscapes that summarise this concept in an adequate way. First, landscapes are not synonymous to environments, since environments are merely part of the landscape. Secondly, landscapes are worlds that are created through a cultural process. Thirdly, these landscapes are the arena for all of a communities’ (inter-)actions. Lastly, landscapes are dynamic and continuously subjected to change, both physical and conceptual (Johnson 2007, 193-202).

Because of this dynamic state of the landscape, the concept has become

inherently linked to the use of landscapes in the formation of heritage, economy, property and even power relations. As mentioned before, it is possible for people to

(47)

leave their imprint in the landscape, intentionally or unintentionally (Aston and Rowley 1974, 11). This means that memories are implanted in the landscape through a mental-material interaction, which can be picked up or changed by later generations (Lewis 1979, 10-11; Kolen and Renes 2015, 25). In this way, the landscape becomes a medium to express and appropriate memory or even a medium to propagate a certain ideology (see the section on places, p. 47). In the right (or wrong) hands this medium can become a powerful tool, as history has shown in many examples.

In finishing this paragraph, I consider it necessary to stress that it has not been

the goal to position environment and landscape as each other’s opposite, as it might have come across. Landscape archaeologists do not deny the limiting agency of environments on human lives, but rather see the environment as part of the landscape. Nonetheless, human agency is just as (if not more) important in this picture. It is the relation between the environment and cultural influences that makes up a landscape.

3.1.4. The application of landscape theory in archaeology

Besides many different views on the concepts of landscape and landscape archaeology in different paradigms, the implementation of landscapes in archaeological research can be considered just as varying. The direct application of landscapes and the mentioning of landscapes only started to take place in the seventies of the twentieth century (David and Thomas 2008, 28-30). Even though this development started fairly recently, different kinds of interpretation were already in use at an earlier date, although not as specific.

It is the objectified form of the environment that is the most common research subject, especially in the discipline of geoarchaeology. In most cases, one could say geoarchaeologists research the environment instead of the landscape. Their research results in valuable information on the geology, soil, hydrology, climate, flora and fauna of the research area, sometimes linked to human actions. Nonetheless, landscapes in this research play the role of environment (Kluiving and Guttman-Bond 2012; Dincauze 2000).

(48)

The more theoretically orientated scientists in landscape archaeology use landscapes as a concept for relational aspects between different variables within the research area. The exact interpretation of these relations differs for each scientist’s interpretation of the concept of landscape, but the general line of thought is that this is an all-inclusive way of thinking about different elements, including the natural environment, cultural environment, humans, flora and fauna, ideas and values, thoughts, memories and socio-political constructs, to name just a few (Roymans et al. 2009, 339). This way of thinking defines landscape as an all-encompassing theoretical concept which can be applied to the archaeological reality.

Phenomenological approaches have started to be implemented in archaeology since the last years of the twentieth century and have seen a rapid development ever since (Tilley 1994). The ideas of the phenomenological approach are based on the philosophy of Kant and Heidegger, which see phenomenology as the method for the study of ontological questions (Heidegger 1975 [1923]; Heidegger 1975 [1929], 26-32). This means that the world and things that are in that world, however abstract they are, should be researched from the point of view of the person that lives in it. After Heidegger’s (1975 [1923], 285-290) terms this would mean ‘das warhnehmen von Sein

durch ein Dasein’. The being-in-the-world, Dasein, stands for an agent which is aware of

its own existence, Bewustsein, while being able to distance itself from the environment, and being able to reflect on the environment in which it is situated. Trying to find these reflections is the key to the phenomenological approach (Tilley 1996, 11-14).

The differences in the above-mentioned applications of landscape archaeology are characteristic for the absence of uniformity in this field of study and emphasise the importance for any author to clearly describe his or her use of the theoretical framework in research.

3.1.5. Other terminology

As shown in the previous part of this chapter it is often considered difficult for the reader to find out in what way the author uses the theoretical framework of

(49)

landscape archaeology and what is meant by the terminology. Therefore, I feel that it is necessary to make a clear listing of relevant terms and the way they are defined in this thesis.

Environments

As is the case for the word ‘landscape, the term environment has been used in many different ways in archaeology since it has come in use. Generally, it has been used to describe the surroundings in which people live their lives or in which communities thrive. This means that an environment is always seen as a requirement for human activity and a requirement for humans to sustain either themselves or their social group. Communities would be living off the land by exploitation of their environment.

The original use of the term environment in archaeology was based on this concept of human sustenance. The environment was studied in order to answer questions on the possibilities and limitations for human presence. To answer these kinds of questions the disciplines of environmental archaeology and geoarchaeology placed the focus on natural factors that influence this environment, for instance, geology, soil formation, climate, hydrology, topography, elevation, flora and fauna (Doneus 2013, 22). This data together provides a complete picture of the possibilities and limitations and can be used to judge location choices of humans. For instance: are areas suitable for habitation and can the soil provide for enough potential for agricultural use? Cultural considerations are not taken into account in environments.

In these disciplines, environments are seen as an element which influences humans in a passive way. Humans themselves are the active agents in this perspective. Agents use their environment in the way they see fit, within the limitations of said environment and are able to change it if presumed necessary. The environment was used as an economical background for human life (Anschuetz et al. 2001, 170-171).

This way of doing environmental archaeology, in a natural scientific tradition that creates models of sustainability and habitation possibilities, has typically come from

Picturing the Merovingian Landscape