Changing Customs

(1)

1

Ester Schraven

(2)

2

Acknowledgements

This study has been conducted in order to complete the study Archaeology and Prehistory at the University of Amsterdam. This master thesis is written as a part of obtaining a master’s degree, and becoming an archaeologist.

The tutors for this thesis were Marleen Termeer and Dr. Stijn Heeren, who have both assisted me in an excellent way during the process of this research. I hereby want to thank them very much for their support. I also wish to express my gratitude to Jos van Tiel who was an immense help in structuring the data in the tables used for this research.

It has been a great pleasure to work on this thesis and it has enriched my knowledge of the burial customs practised in the area where I grew up, almost 2000 years before I was born. I hope to benefit from this knowledge in my future, when working on research projects which concerns sites from this period.

Ester Schraven, 25-07-2013

(6)

6

1 Introduction

When death comes this means the final moment in the life of a person is reached. For archaeologists however, the death of a person from the past can be the start of an archaeological investigation. The way a community deals with the death of a member can leave archaeological traces which potentially hold much information about the deceased and the community he or she belonged to. Maybe the person who died had made some arrangements for his or her own funeral, but often choices will be made by the survivors. . The treatment of the body is an important element of the burial customs of a group.

This study is about changes in the burial customs in the first three centuries AD in Germania Inferior. In order to establish these customs, burials from different burial grounds are studied with regards to the types of graves that were used, who were buried, and what, if any, grave gifts they received. The Roman province of Germania Inferior was chosen because in the period under investigation interaction of cultures took place here on an increasing scale, for example when veterans returned to their indigenous communities after having served in the Roman army and would have had intense contact with Roman culture. Because of this type of interaction, but also for example trade with the Roman army, the limes zone of the province offers a suitable area and period for studying the degree to which this increased contact influenced the dynamics in the burial

customs. The interaction will have had its effect on all people living in the area. But would there be an impact on the burial customs as well? The research question of this study therefore is:

“Do the burial customs in the selected sites from the province of Germania Inferior show significant changes over time?”

In order to get an answer to the research question, first the frame within which this research is executed will be presented. After presenting the main research questions it should become clear that in order to understand possible changes and their causes an explanation involving romanisation should be considered. Paragraph 1.2 will therefore introduce the debate surrounding romanisation. Only when the elements from this debate which are important for this research are better

understood, changes in the burial customs that are encountered at the sites can possibly be related to influences caused by an increased contact with Roman culture. After this, in section 1.3 the literature and data used for this research are clarified and the selection process is specified, by explaining the reasons behind the selections made for the burial sites, graves and grave gifts.

In chapter 2 the burial grounds are introduced. The burial sites are placed into context by providing background information about them with regards to their size, location, character and some extra information is provided if this was available. After each introduction the data of the site follow, which concerns the prevailing grave types, presence or absence of ditches surrounding the

(7)

7 grave and the grave gifts. All of these elements are studied on their appearance and possible

relations to the sex and age of the deceased. A short analysis of the data is provided at the end of each site discussion.

In the third chapter the features which were investigated per site are compared to the other sites in order to gain a broad insight in the burial customs from the research period and area. The next step is to investigate if there are changes in the burial customs over time and if that is the case, what type of changes. To what factors do they seem to be related and can an explanation be found for them? The answers to the research questions should provide an image about the dynamics regarding burial customs at the ten sites chosen from Germania Inferior in the 1st to the 3rd century AD. Those results are presented in a concise manner in the conclusion at the end of this research.

(8)

8

1.1 Framework

This part of the study will cover the frame, starting with an explanation of the research questions and the important terms used in this research. This leads to a short review of the romanisation debate, in particularly of the aspects which are relevant for this study. The last part of this chapter consists of paragraphs which discuss the context of the data.

1.1.1 Research questions and terminology

There are several possible traces which can be left behind by burial customs with regards to the way a community dealt with their dead. All of these traces can help to construct a bigger picture of the community to which the deceased belonged. If a grave is found with no human remains present a specific way of handling the dead can still be understood from this grave by its shape, the presence of a ditch surrounding the grave or the grave gifts. More data can be gained when human remains are present, which can show if the deceased was cremated or inhumated, and when cremation was practised if the human remains were sorted from the pyre remains. If studied carefully the remains can even be used to determine the biological sex and age of the person. Possibly personal

possessions were given as grave gifts, which than can present a first idea about objects which could have been favourite items of the deceased, or show his or her preferences. On the other hand can items be related to customs which exist in a community and are important elements of rituals. All these elements are important to study in order to get to an answer for the research question; “Do the burial customs in the selected sites from the province of Germania Inferior show significant changes over time?”

To provide an overview of features like the grave type, ditches surrounding the grave and the grave gifts, and to study possible relations of each feature with the age and/or sex of the deceased the following sub-questions are used:

-What are the prevailing grave types in the 1st-3rd Century AD for the selected sites?

-Are there differences in the grave types between the sites under research, possibly related to the character of the site?

-Are the grave types related to age and/or sex of the deceased? -Are there differences in the prevalence of the ditches surrounding the graves?

-Are there differences in the prevalence related to the character of the site? -Are the types of ditches related to age and/or sex of the deceased?

(9)

9 -What are the prevailing grave gifts in the burials under research?

-Do the grave gifts show differentiation between the sites, possibly related to the character of the site?

-Are there grave gifts preferred or even reserved for people of a certain age and/or sex?

In order to understand possible changes which occur over time the nature of these changes is studied, by answering the next set of sub-questions:

-What are the changes in the burial customs?

-When do changes show? Only after the establishment of the province or do they happen earlier on as well?

-Do they occur in a single site, a region or all throughout the research sites?

-Are the changes related to the grave types and thereby indirectly in the way the human remains were treated?

-Can the changes be related to the site type?

-Are there changes in the grave gifts found from the period under research?

-If there are changes in grave types, surrounding ditches or grave gifts are they universal or can a relationship with sex and/or age be established?

-Can the changes possibly be related to external factors? Is there a possible cultural influence from an increasing contact between different cultures interacting with each other at the frontier zone?

In order to advance with this research some key concepts need to be explained. Burial customs is one of the very important terms in this research which should be clearly defined. In this study those customs can consist of multiple features. First there is the manner in which the human remains treated, were they inhumated or cremated? A next step is the matter of how the remains were collected and deposited, which can point at certain specific actions that were undertaken. An example is collecting the cremated remains, putting them in an urn and filling the grave with clean sand, which would make a type A grave. A different custom concerns the act of constructing a demarcation around the grave, like ditches surrounding the graves. Every choice made regarding the treatment of the body of the deceased, from rare inhumations, to more common collection of the cremated remains and placing them in a ceramic container, is of importance in establishing the grave type. Which grave types are prevalent in the sites under research and a more detailed explanation of them can be found in appendix 1. Another example of a burial custom is providing the deceased with grave gifts, like ceramics. Grave gifts are in this research those items which are intentionally put in a

(10)

10 grave during the burial.1 This can concern gifts that were burned on the pyre with the deceased, or placed in the graves along with or directly after the cremated remains were put in the grave.2 Finally, another act which makes up a burial custom is the eating and drinking at burials, which can be established to have taken place when animal bone remains were put in the grave.3

The focus of this research is on changes which might be explained as a result of the meeting and interaction of two cultural traditions in a border zone. Whether or not changes can be regarded as meaningful will be discussed in paragraph 1.2, which deals with the debate on romanisation. The type of changes meant here are those changes which will lead to a deviation from a norm. Such a deviation can be the appearance of previously unknown grave types in a specific community, which were common in other areas or communities. Such a change can point to a degree of interaction between groups of people with different cultural backgrounds.

When discussing changes in a part of the Roman Empire often the term romanisation is used. This is, like “Roman” a very important term. What is understood to be Roman, or romanised, is a difficult matter, and cannot be explained in a single sentence. In order to provide a better insight in the matter about what is or can be seen as Roman the next paragraph will explain in brief the romanisation debate and highlight elements of that debate which are important for this study. The terms Roman and romanisation will also be explained there.

1.2 Short introduction

In order to discuss the romanisation debate first a clear definition for the key terms surrounding this debate has to be provided. In this research Roman culture is understood as Woolf4 presents it, which is “the range of objects, beliefs and practices that were characteristic of people who considered themselves to be, and were widely acknowledged as, Roman.” For the term romanisation again Woolf5 provides a good definition: “a convenient shorthand for the series of cultural changes that created an imperial civilization, within which both differenced and similarities came to form a coherent pattern.” With the term acculturation, which will often be used in relation to romanisation, the exchange of elements of two cultures is meant. This is opposite to the idea that one culture takes over another, and thereby imposes its ideas and customs on every aspect of society upon the other. For the term Roman, we can look at the term Roman culture, and understand someone belonging to

1_{Magnée-Nentjes/Enckevort 2010, 67.}

2

Idem 2010, 67.

3_{Ibidem, 73 mentions the actions of providing grave gifts for the deceased and the organising of meals at} funerals.

4

Woolf 1998, 11. 5

(11)

11 this culture, and not just who feels him or herself as belonging to that culture, but is also

acknowledged and recognized by others to belong to that culture as a Roman. Now that these key concepts are made clear it is time to review the development of the debate and the relevant parts of the debate for the study of burial customs.

1.2.1 The general debate

Haverfield defined romanisation in a different manner6 than his predecessors who had no attention for the role of the native communities in this process.7 He was the first to pose that a degree of mutual contact between Romans and native cultures at the frontier zones had to take place. He felt however that Roman civilization was superior to the civilizations outside the Roman Empire and as the empire expanded, so did the spread of the customs and culture of the “superior” Romans.8 The frontier zones marked the boundaries of the Empire, in which Roman culture was firmly set and in which there was, as Haverfield9 calls it, “the safety of civilization”, against the outside world full of the barbarians and their chaos.10 This view implied a strong dichotomy with strict boundaries and no diffusing of cultures. If this was indeed the case, then it follows that there were cultural objects, habits and rituals which would have also greatly differed between the civilized Romans and the wild Barbarians. From this colonial view the idea has developed that romanisation stands for something different altogether, a two-way traffic in the exchange of culture. The theory that Romans were better, supreme even, than other cultures, could not withstand some of the strong critiques which were expressed.11 Romanisation was no longer seen as a linear development away from their own culture towards the better and higher valued Roman culture.12

This colonial idea of how Roman culture spread and the nature of romanisation are discussed by many authors, among which Millett13 and Webster.14 The post-colonial theorists felt that thinking about one culture taking over another was problematic because of the assumed one way direction in which the interaction would take place. This contributed to the downfall of this theory and the rise of

6

Haverfield 1912. Haverfield and his Romanisation theory are also discussed by a wide range of other authors, among which are Millett 1990b, and Hingley 1996.

7_{Heeren 2009, 3 discusses this, in which he refers to Mommsen 1885, 71-106, who had an important role in}

the early theories about the Romanisation. 8

Haverfield 1912, 10. He speaks of the Roman Empire as the civilized world, implicitly stating that the outside world was not civilized and mentions the “wild chaos of barbarism”, again implicitly stating Roman civilization being neatly ordered and making the Roman civilization a “better” civilization.

9

Idem 1912,10. 10_{Ibidem, 10.}

11_{Webster 1996 mentions the development of the theories on Romanisation, but Hingley 1996 also discusses}

the two way process, thereby leaving the old theories behind him.

12_{Heeren 2009, 9 mentions that this view in his discussion of the term Romanisation. Webster 1996 writes in}

more detail about this development. 13

Millett 1990a and 1990b. 14

(12)

12 a post-colonial theory.15 In this theory there was more attention for the story of the indigenous people, as a result of the insight that the conquests of the Romans were not all such a peaceful and positive experience for the natives who were submitted to them, nor did the natives would have always surrendered to romanisation without revolting. Millett added to this that romanisation was not just a process which was happening, but that the indigenous people had a more active role in it, and wanted to become part of this Roman world.16_{One of the problems dealing with this new insight} was that the degree to which the indigenous people stood up for themselves and their own culture was not always understood.17 In other words it seemed that although the acknowledgement of the native people not being helpless against the Romans, and had their own wills, wishes and demands had started, the level to which this took place was not always fully recognized.

But there is more to the debate on romanisation. Six important issues of this debate will be addressed here. Firstly, although there seems to be a broader understanding of the fact that romanisation was a process of acculturation, a two way exchange of elements from the different cultures in which new elements could be formed, there are still several factors unknown or not well understood. Hingley, among others, deals with the fact that although there is now acknowledgement for the fact that we are dealing with a process of acculturation, the general view regarding

romanisation is still one of passive natives who used elements of Roman culture to try and belong to that culture.18 The understanding of acculturation as the exchange of elements from different cultures, rather than one culture dominating another is very important when researching the contact between cultures. Yet I feel the way Hingley sees it, that there is still too little understanding of this two way process, is too negative and somewhat obsolete. We have to keep in mind that Hingley made this comment back in 1996. I feel that 18 years later, while working on this research, there is a bigger awareness of the effects cultures meeting at the frontiers had on each other and that this is taken into account in modern day research. Hingley feels that of the term romanisation only certain aspects have been redefined.19 The post-colonial school of thought has indeed led to an awareness of the negative impact the term romanisation can carry.20 And although a new more neutral term, free of possible negative connotations is an attractive goal to aim for, the current lack of such a term21 does not need to be a problem for this research. Because there is no new term that has been

15

Webster 1996, 6.

16

Heeren 2009, 4 referring to Millett 1990b. 17_{Idem 2009, 7; Hingley 1996, 42.} 18_{Hingley 1996, 41.} 19 Idem 1996, 41. 20_{Heeren 2009, 5.} 21

Idem 2009, 12-14 discusses other terms that have been suggested, but none of them have been regarded as

suitable replacements for the term Romanisation. He uses the word Romanisation in his study, but feel redefining is needed, which he also offers at pages 15-16.

(13)

13 established to replace ‘romanisation’, and the term itself will not be the most important aspect when conducting research into the development of cultures and their customs, the term romanisation will still be used in this study.

A second point that is justifiably discussed regarding romanisation concerns the reasons why the Roman Empire expanded in the first place22 and the degree of force that accompanied this expansion, through which Roman culture got spread over the new areas. Although indeed the reasons behind the expansions of the Empire are not clearly understood at all levels, this part of the discussion is not relevant for this study23, as this study limits itself to burial traditions and their (dis)continuity over time. The expansion itself does matter, as with the expansion Roman soldiers got spread over wide areas bringing with them their culture, and in that way often either initiating contacts of intensifying already existing contacts with native cultures. It is often assumed the initial contact happened between the upper class of the rural communities and the Romans. This view might have been influenced by the fact that the elite has been the focal point in archaeology for so long.24 The neglect of the lower class, the class par excellence to revolt, has led to a gaping hole in the knowledge about the degree of force used to romanise cultures.25 The better researched elites are after all the ones of which it is often assumed that they spread a Roman culture.26 Although this fact can and should not be ignored, this study tried to avoid this issue by presenting a picture which is as complete as possible regarding the degree of romanisation among different groups, by

incorporating elite burial grounds as well as rural sites. This however does not resolve the question of force, which should not be ignored. Yet this should not cause problems for this research either, as the force behind the expansion does not concern one of the main issues of this research, namely the burial customs and their development over time.

A third point of the debate to be mentioned is the lack of a null point to measure

romanisation against.27 This is brought up by Reece and is an interesting point for this study which deserves some explanation. According to him material culture exchange was already going on before the troops came to the provinces.28 This lead to a very minor scale of romanisation going on all over Europe, far before the contacts increased due to expansion of the empire. So effectively there is no completely Roman-free world to compare the degree of romanising in other cultures of a later period

22

Heeren 2009, 6-7 mentions some possible reasons of the expansion of the Empire that were discussed in the romanisation debate.

23_{Millett 1990a writes more about the possible reasons behind the expansion.}

24

Heeren 2009, 10-11 discusses this elite perspective in more detail.

25_{Hingley 1996, 42-43.} 26 Idem 1996, 41. 27 Reece 1990, 33. 28 Idem 1990, 32.

(14)

14 against.29 This then leads to the problem that according to Reece a good picture of how romanisation happened is not possible. He states that the only way to study romanisation properly is by looking at geographically different sites which were operating at the same moment in time.30

The point about world free of Roman influence not existing is possibly slightly exaggerated, but should be keep in mind when studying the developments regarding the use of elements from Roman culture. However the fact that these developments cannot be studied through time seems not justifiable to me. Even if there is no null point, still an increase or decrease in specific objects appearing at a site can provide an image of processes going on with regards to acculturation. Millett has a strong case when he discusses what can be used as indicators of romanisation.31 He states that meaningful changes which were set off or influenced by other cultures should be sought in the economical basis of society.32 Changes in the basis of subsistence or sudden change in the scale of those activities can be seen as consequences of acculturation. Not just the agriculture, which is a good indicator according to Millett, but also material culture, can show such changes, when there are new additions to the spectrum which cannot be related to any other previously used objects.33 Yet, even if non-local materials are found one has to keep in mind the quantity versus the time span those materials come from in order to understand the degree of change34, or more specific for this study, the degree of romanisation. If during a long time span the “new” objects are found in relatively low numbers, the degree of romanisation would not have been high but possibly the objects would have been more valuable due to their scarcity, so this might be an indication of the wealthier people romanising more than the people from the lower levels of society.35 However the number of materials found might equally suggest something about a site’s position or connections in a network of trade, or a variation in the level of demand.36

A fourth point that is stated, among others, by Reece is the way how Roman culture is often seen as a standard package.37 Becoming Roman would mean one would adapt his or her lifestyle to fit to this package. Woolf discusses the theory that “the” Roman culture is not a fixed package.38 He goes on to state that being an active participant in the debate of what being Roman was meant was just as important in becoming Roman as it was to fit the package that at a certain moment in time 29_{Reece 1990, 32-33.} 30 Idem 1990, 31. 31 Millett 1990b, 9-12. 32_{Idem 1990b, 9.} 33_{Ibidem, 10, 12.} 34 Ibidem, 30. 35_{Ibidem, 30.} 36 Ibidem, 30. 37

Reece 1990, 31 but also Hingley 1996, 42. 38

(15)

15 was seen as Roman.39 He also discusses the idea that romanisation happens in degrees.40 This

diversity in what was seen as Roman culture should be kept in mind when conducting a research like this one.

A fifth point in this paragraph concerns the low speed of the process. The average indigenous person would not have felt very much of the process of romanisation as the changes simply took more than a life time for noticeable changes to settle.41_{If there were changes however, they seldom} came alone. When elements were taken from one culture to another, often the ideas behind the objects would travel with them. On the other hand, objects that were incorporated in a new culture could very well have become detached from their original culture, and would therefore not be meaningful in a research of acculturation.42

A sixth and final point to be mentioned here is that changes in customs can happen without conscious decisions being taken, but that they are the result of gradual developments. One always has to wonder which changes would have happened regardless of the contact and which changes can indeed be appointed to contact with or influences from other cultures.43 To go even further on this, Woolf rightly mentions that when there are no signs of acculturation, like romanisation, this does not mean there was no Roman presence, knowledge of the culture or contact. Lack of markers of

acculturation might just as well indicate a possible choice to not incorporate elements from Roman culture, an outing of resistance or neglect.44 And even if the material culture changes, this does not mean that the indigenous people disappeared.45 In other words, people from communities native to the area who started to use material culture fitting a Roman culture did not necessarily become Roman, they were still who they were before, but adopted new materials to use. This is an important point to keep in mind during the final outcome of this research.

1.2.2 Romanisation in the burial ritual

This section focuses on those aspects of the debate on romanisation which are relevant for this study of burial customs. The prevailing type of graves as well as the ditches surrounding the grave and the number and nature of gifts are studied in order to relate them to an indigenous community or to recognize types that can relate to what is accepted as “Roman”.

39_{Woolf 1998, 11.} 40_{Idem 1998, 6.} 41 Ibidem, 7-8. 42_{Ibidem, 14.} 43 Millett 1990b, 9. 44 Woolf 1998, 12. 45 Cooper 1996, 92.

(16)

16 But what is it that can be seen in the burial customs that can be regarded as Roman, or that can indicate Mediterranean origins? It was often assumed that objects on their own could be used as indicators of a Roman identity, as was for instance the case with terra sigillata objects.46 It is now understood that such a thought is too simplistic and an object does not necessarily indicate a person’s identity.47 This means that mobilia which were previously used as indicators of a Roman identity are no longer suitable markers to establish a degree of romanisation, especially when they can be used for different functions than the function they had in the culture they originate from. Only specific objects can be used as markers of Roman culture and only when they come from specific contexts. In order to establish the presence of romanisation other elements, like the burial customs can be of help, as they are, in contrast to adaptations of materials and objects used in everyday life, among the last to be changed.48 Proving a degree of awareness of the origin of objects can be very difficult, but when items specifically related to Roman burials customs are found in burial sites of communities who were native to the area before it became a province, they seem to be a stronger indicator of romanisation of the burial customs than any other object linked to Roman culture. Objects from daily life which are clearly linked to Roman culture, like writing equipment or items related to personal care and hygiene are also incorporated in this study as markers of romanisation, as those type of objects were very likely never used in any other manner as for their intended purpose and are known from graves which are affiliated to a Roman culture.49 They are also known as objects which express a Roman identity.50 This is contrary to for example tableware which might be Roman in nature but which could have been used for secondary purposes unrelated to their original use and culture. Due to this possible wider use of these items they cannot be used as definite markers of romanisation in the burial customs.51 If this research can show that there are changes in the burial customs which can be related to a Roman influence this would indicate that people were indeed romanising their burial customs.

When it comes to treatment of the dead cremation was the prevailing custom in the early and mid-Roman period of North West Europe.52 This is the entire period covered by the graves used

46

Aarts/Heeren 2011, 8-9. 47

Idem 2011, 9.

48_{This is known from situations of colonization, where people stick to their burial customs at a moment when}

new cultures are introduced. Ziskowsky 2007, 4 and Sjöqvist 1973, 68. Admittedly the situation with regards to Romanisation is not like colonisation but it seems plausible that holding on to one’s own traditions with regards to burials also happened in the dynamic period when the new Roman culture advanced and the contact with the native culture intensified, as a Roman presence became stronger in an area which was beforehand occupied by native communities.

49_{Aarts/Heeren 2011, 236.} 50 Idem 2011, 135, 253. 51 Ibidem, 236. 52 Smits 2006, 7; Magnée-Nentjes/Enckevort 2010, 17.

(17)

17 in this research, which are from the period 19-270 AD.53 Cremating remains of the deceased was seen as a typical Roman custom54, yet this practise was also prevalent at the native communities occupying this region.55 Although the placement of the cremated remains in an urn is not mentioned to be specifically Roman this action can be an indicator of Roman burial practices according to Enckevort/Magnée-Nentjes 2007.56 Because of the fact that cremation is not exclusively Roman, this can however not be regarded as a marker of Roman culture. In this research the specific grave type known to relate to the Roman army, the so called bustumgraves57, are seen as a more reliable marker of Roman culture or presence, as urn burials also occurred in native sites.58 Graves which are also known to be typical Mediterranean in style are those constructed with Roman tegulae.59

Ditches surrounding graves are known from both rural and urban sites and stem from the pre-Roman period, but are used in the Roman period as well.60 On their own they can therefore not be seen to represent a specific tradition with any certainty. They are investigated however to determine if they indeed appear at all types of sites, and if their presence changes over time.

Examples of grave gifts that are related to Roman burial customs are incense burners and lamps61, which represent elements of light and fire62 and are important elements in Roman burial traditions.63 Coins are often also seen as a likely indicator of Roman burial traditions, as the deceased received coins for crossing the river Styx in the afterlife.64 Eating and drinking ceremonies which are often held throughout the year in memory of the one passed away at the grave are another custom which was practised.65 However during these activities items like the coins could be accidentally lost and eventually ended up in the grave, but were never meant as intentional gift.66 For this reason coins are not used as markers of Roman culture in this research, unless it is certain they are

53

Magnée-Nentjes/Enckevort 2010, 17. 54_{Idem 2010, 17.}

55

Ibidem, 17 but also Enckevort/Thijssen 2000, 52.

56

Enckevort/Magnée-Nentjes 2007, 45. Although this is stated by these authors in this publication, this contradicts the fact that for example the rural site at Tönisvorst, which is not regarded as reflecting Roman traditions, also had a lot of burials of cremated remains in an urn.

57

Smits 2006, 156; Magnée-Nentjes/Enckevort 2010, 69. A further more detailed description of such a grave can be found in Appendix 2.

58

Hendriks/ Magnée-Nentjes/Smits 2008, 128. 59

Toynbee 1971, 101.

60_{Haalebos 1990, 200. Smits 2006, 73, 110 mentions the surrounding ditches as an element of indigenous}

groups, but at the burial site of Nijmegen Museum Kam straat, which is military urban in nature, there is also one grave which has such a structure present.

61_{The lamps are often oil lights, Magnée-Nentjes/Enckevort 2010, 73.}

62_{Haalebos 1990, 203.}

63

Koster 2010, 261, but Enckevort/Thijssen 2000, 56 also mentions objects related to light and/or fire as non-native elements in the burial customs.

64

Magnée-Nentjes/Enckevort 2010, 73. Again Enckevort/Thijssen 2000, 56 supports this. 65

Idem 2010, 73.

66

(18)

18 intentionally deposited, for example when they are burned due to their presence on a pyre. Remains of the meals might be seen as a testimony of such ceremonies which were held at the funeral and are for that reason studied in this investigation. Other gifts which are related to the expression of a Roman identity and are therefore useful for this study concern for example writing equipment and (glass) bottles for ointments or perfumed oil.67

A material that according to Koster68_{shows a more direct contact with the Italic area is} amber, because in that area there were trading centres famous for this material and places where the crude amber was reworked. She states that the presence of multiple amber objects in a grave from the elite burials in Nijmegen shows the contact with that area, and a liking of Roman culture.

In order to investigate changes in burial customs, markers of such customs of the native communities had to be established as well. For the indigenous communities under research several aspects have been regarded as native. These are weapons69 and metal, often iron, grave gifts.70 When we see changes in them it is valuable to investigate if the increasing contact with a Roman culture could have caused such changes.

Now that for the different cultures markers of burial customs are discussed it is time to focus on other problems to be dealt with before a degree of romanisation can be established.

One such problem which had to be avoided was comparing sites of different periods and by that not being able to make justified comparisons. Woolf discusses the fact that studying romanisation can turn out to be difficult due to the often problematic dating of material.71 Because all the selected cemeteries have a firm dating this problem was easily avoided. Often the phases were even refined which made it possible to date the material in the graves, as well as the customs that were practised, to the time they were used for the last time at that specific site. Another benefit of the burial fields is that if romanisation was going on at the sites which were in use for a longer period, the changes should show, despite the relative low speed of the process of romanisation.

Just as we now understand the different possibilities of the way the contact took place, the debate on romanisation has also brought the understanding that there is a bigger story behind an

67_{Unguentaria, aryballos and balsamaria are examples of such flasks related to a Roman identity.}

68

Koster 2010, 262. 69

Idem 2010, 256 discussed the use of weapons as indigenous grave gifts. She also discusses the use of grave chambers as being an indigenous feature in the burial customs. She however investigated indigenous graves of very rich elite which cannot be compared one on one to the other graves from less rich people used in this research, due to this difference in wealth. Bridger 1996, 166-167 also mentions weapons as a common grave gift in Gallo-Roman graves.

70

Bridger 1996, 321, 324. Bridger mentions that sites which are known to be romanised often have a lower appearance of metal/iron objects than sites which are not or barely romanised.

71

(19)

19 object, that people using specific objects possibly would not have known its background, and the fact that just using such an object does not necessarily make one Roman at all.

As paragraph 1.2 should have made clear the debate on romanisation has a long history and covers a broad spectrum. Relevant for this research are the insights that romanisation concerns acculturation, and is made up of levels and degrees, and that what is perceived as Roman is dynamic and changes over time. Romanisation is not a linear process from underdeveloped natives towards becoming the “better” Roman. Roman culture in bits and pieces was already penetrating a wide area across Europe but the conquests which caused an intensification of contact must have influenced this process. The acculturation got more intense as more people from different cultures came into contact with Romans on the ever advancing borders of the Roman Empire. There are certain objects which are understood to be Roman in nature which can help to provide an indication of the process of romanisation, specifically in the burial customs. We should however keep in mind that those key objects only show degrees of this process and in this research only the burial customs are

investigated and this will not show the entire process of a community romanising over time. Interpreting the degree or romanisation felt by people can be a challenge but we can make a firm attempt by keeping in mind the meaning certain objects carry, never forgetting the wider context they originate from.

1.3 Gathering the data

In this paragraph the available literature and data are discussed. To understand how the data set is put together the method and selection process are explained. The reasons for the choice of the specific grave sites and an explanation of the grave types are provided at the end of this paragraph.

1.3.1 Literature and data

The literature used to gather the information about the burial sites and their graves concerns site reports, containing catalogues of burials. The quality of the publications varied considerably due to the different backgrounds of the authors, the Fidelis excavation for example is written by amateur archaeologists, often with global descriptions of the objects found whereas the site of Tönisvorst is published with great detail for each category of material that was encountered.

For the study of human remains the information provided in the catalogue was used, despite the fact that sometimes more specific research results were available. These consisted of physical anthropological research results, but more than once they proved to be inconsistent with the data from the catalogue. Because of the desire to have the most consisted data possible the decision was

(20)

20 made not to use the results from the physical anthropological study but instead use the catalogue data.72

The ten selected sites produced 1008 graves which could be used in this research. The time span covered by the sites spans a dynamic period for the research area, with the consolidation of the province. Hingley mentions that in the British province of the Empire during the period between 43AD- 410 AD the self-image of the people would have changed drastically under influence of the increased acculturation which took place.73 This might have also been the case for Germania Inferior, in which the degree of acculturation in the first three centuries also intensified. The effects of this acculturation are possibly reflected in the burial sites, which all fall within this period, albeit more in the first part of this period. With these data explained the data processing is next to be clarified.

1.3.2 Method, the reasoning behind the data

A certain amount of change in customs can be expected to take place over such an extended period of time, but abrupt, drastic or striking changes ask for further explanation. When this is encountered it is worth investigating if those changes can be related to the contact between the two cultures in the frontier zone, which became more intense over time. Important in this process is to understand that one culture is not taking over another, but elements can be used from each other’s cultures. If this happens and brings about changes detectable for archaeologists, this still does not mean the people became fully Roman. This process than ‘merely’ shows that influences of the other culture show. That is what this research is about, not pinpointing people to have become Roman but if, at any level, they adopted elements of what at the moment of adoption was seen as Roman culture according to their own traditions. What is meant by “at the moment” of adoption is that Roman culture was not a static culture, and the interpretation of this culture was therefore also dynamic.74 This issue has been dealt with more extensively in paragraph 1.2.

When processing the data the differences in phasing used by the authors of the site reports caused a wide range of phases to exist. Some sites were dated and had no specific phases to divide the time span, like the Fidelis research from Cuijk, Hengstberg, Schebbelaarseveld, Noviomagus and the Sperwerstraat, whereas others had up to eight phases, like Tiel and Tönisvorst. In order to see the continuity or changes over time the decision was made to establish three new main phases and to reclassify all the phases from each burial site into the new phases. These phases should help

72_{This is for example the case in the Koster 2010, 64 where table 2 present different results than the catalogue,}

and the physical anthropological report provides only more general remarks on the sex of the deceased, so information per grave is only taken from the catalogue.

73

Hingley 1996, 43. 74

Woolf 1998, 10. Woolf his work shows that he is aware of the overall dynamics of Roman culture and what being Roman meant but on this specific page he underlines that statement.

(21)

21 present a clearer image of the situation regarding burial customs and possible changes in these customs over time. The distribution of the graves according to the new phases is shown in tables 1 and 2.

Table 1 New phases, their time span and the number of graves per phase.

Phases Time span covered Number of graves from this phase

1/2 3

2/3 3

? Studied but phases unknown 51

- No phase mentioned 22

Table 2 Graves which could not be appointed a phase.

The two period divisions that have been chosen are 85 AD and 150 AD. The first marks the moment the province Germania Inferior was formed, around 85 AD.75 The second phase starts roughly 64 years later. With the division of the time span the aim was on the one hand to obtain three phases of rather equal length, whilst on the other hand also consider the distribution of the graves. The

balance has been found in the division as presented above, although the last phase is longer than the other two. Dividing this phase even further was not desirable because the distribution of the grave numbers. Besides that, when wanting to investigate possible changes in burial customs through time there should be a decent amount of time per phase, as those types of changes are not expected to happen within in very short period of time. Reducing the time span of the second phase further in order to get a more even distribution of the grave numbers would lead to the risk of studying graves from a single generation only. As mentioned in paragraph 1.2 of this study, if changes happen as a result of acculturation, they will likely need more than a single lifetime to take place76 so a further division is not desirable. Graves which could not be ascribed a specific phase could not be used to investigation of changes in the burial customs over time but have been looked at when investigating the grave types, ditches surrounding the grave or grave gifts.

For the research method I decided not to follow Millett’s material cultures based approach, nor did I dismiss the option to look at the development through time, although I am aware of the

75

Carroll 2001, 15. 76

Woolf 1998, 7-8.

Phases Time span covered Number of graves from this phase

1 10 AD -85 AD(75 years) 261

2 86 AD – 150 AD (64 years) 392 (of which two not entirely sure belonging to this period) 3 151 AD – 270 AD ( 119 years) 276 (of which one not entirely sure belonging to this period)

(22)

22 possible problems that can follow from the method followed in this research.77 In my approach I rather choose to avoid the problems mentioned by Reece and Millett in paragraph 1.2, still being able to compare different sites on the level of developments through time and between different settlement types, without making groups of material culture. The geographical spread of the sites under research here can indeed provide a varied image, but when this is kept in mind, this can also show insights in the development of settlements of the same nature which are spatially separated. A bias can occur in the outcome of a study due to the degree of research that has been done in a certain area and the variation in the level of the publications that are available.78 This issue has been dealt with by using sites from several areas and which are all well published, albeit that there are differences in style. The matter of the development though time presenting a biased image is dealt with in the paragraph on romanising, the section on Reece’s null point. I believe developments indeed happen regardless of the process of acculturation, but that certain changes like for example the introduction of specific

objects will stand out enough to be reliable markers of the process of romanisation.79

1.3.3 The sites

The first selection of the sites was based upon the fact that they are all located in the same province. They are all located in the frontier zone, there are only relatively small differences in the distance to the border, as can be seen in

77_{Reece 1990, 31presents critique on trying to compare material culture through time, as the original}

assemblages will have changed over time and cannot be compared to the assemblage from the starting point of one’s research. His answer to this is that sites can only be compared at the same time. Millett 1990b, 12 sees a problem in looking at sites from geographical distinct areas at the same time, as this geographical different background can cause for differentiation in the material culture which has nothing to do with acculturation, and can even influence the process of acculturation in an uneven manner. He therefore compares only specific objects from the material culture assemblage to each other.

78

Millett 1990b, 12. 79

Objects that can carry such information are for example the previously mentioned lamps and writing equipment.

Figure 1 Map showing the sites used for this research. Nijmegen and Cuijk have been enlarged in order to show the exact locations of the sites in or near those cities in more detail.

(23)

23 figure 1, the site distribution map. What sets them apart from each other, and formed a second selection criterion, is their different characters. This makes the burial grounds suitable for

comparisons in order to see if there are similarities in burial customs in connection to their character. The characters from the sites in this study are rural, urban and military. These characters can help to achieve, together with the bigger context to which the burial sites belonged, a further understanding of the choices made with regards to the treatment of the deceased. For instance grave types that are usually found in military context would make sense when they are found in a site nearby a

settlement which is military in nature, but would seem rather out of place in a site which is far away from any military settlements. The same goes for gifts which relate to the army. When dealing with a burial site nearby a military camp it can be expected to come across militaria in graves belonging to this site, but when the same objects are found in a rural site far away from any military site some further investigation is needed to understand how they would have gotten there, and what their meaning would have been for the people who buried somebody with such specific equipment or objects.

Besides these features the sites are spread in such a way that comparisons can be made between sites close to the same city, or spread far from each other. For example, Tiel and Tönisvorst both are of rural nature but since their geographical distance from each other it is worth looking if the nature of the burials is the same or that the regional difference is reflected in the burial customs. The sites that are chosen are shown in table 3.

Name of Site Character of site Location

Hatert Rural Nijmegen

Tönisvorst-vorst* Rural Tönisvorst

Tiel-Passewaaij* Rural Tiel

Fidelis excavation Military-castellum Cuijk

Hengstberg Urban-military Ubbergen, near Nijmegen

Schebbelaarseveld Rural Hernen, township Wijchen

Noviomagus Urban Nijmegen

Sperwerstraat Urban Nijmegen

Molenberg Rural Wijchen

Nutricia terrain Military-castellum Cuijk

(24)

24 The total time span covered by this research is based on the time the sites were in use. The start of the earliest of the burial site, 10 AD in Hatert marks the beginning of the total time span, and the end of the burial sites which were used last, 270 AD in Cuijk Nutricia and Tiel mark the ending.80

1.3.4 Grave types

In order to investigate the effects of the increased contact with Roman culture burials can be

excellent study material as they often comprise of intimate rituals which are important in the culture one grows up in.81 Besides this important factor, burials can contain different aspects that can be compared very well to each other, so that a single burial site can provide material for comparisons on different levels. The grave types can reveal a preference for specific treatments of the body, but also parts of customs surrounding the way dealt with the cremated remains can be revealed, by for example the presence or absence of charcoal which indicates if the complete funerary pyre was deposited in the grave or if the human remains were sorted out. This handling of the remains has been an important factor in establishing the grave types. The grave types present are explained in detail in appendix 1, which also explains the conversion used to achieve a uniform set of grave types, but table 4 provides a short overview.

80

Tiel-Passewaaij has a longer use of the burial ground as after 270 AD the burial ground is not used for burials, but somewhere between 340-500 AD a single last burial, an inhumation, is placed here. Since this does not concern a burial from the time span under research here, Tiel is also regarded to have been in use until 270 AD. This is also the accepted ending of the use of the site as burial ground by the authors of the publication from which the data for this site originates. Aarts/Heeren 2011, 227-228.

81_{The intimacy of the burial customs speaks in part from the fact that these customs are among the last type of}

tradition which will be changed in a situation of contact between cultures, in which one culture is brought to an area where a community with its own traditions is already established. Although this is known from

colonization situations, and Romanisation is not like colonization, the fact that people hang on to their burial tradition in times when new cultures appear does remain. Ziskowsky 2007, 4 and Sjöqvist 1973, 68 both mention this.

(25)

25

Grave type

Short description of the grave

A Cremated remains sorted out, placed in an organic container in a grave pit, which was filled with clean sand. (knochenlager, crematierestendepot). This type is likeHiddink type A.82 B Cremated remains sorted out, placed in an organic container in a grave pit. The pit contained

remains from the pyre as well. (Brandschüttungsgrab mit knochennest /

crematierestendepot met verbrandingsresten). This is like Hiddink type B when no container is present.

C Burned bones with a part of the pyre and burned (ceramic) sherds in a grave pit. No container was present. Brandgrubengrab/verbrandingsrestengraf/brandgraf) This is Hiddinks type C-1.

C-2 Hiddink type C2, like type C above, but with too few remains, less than 100 grams, to interpret it as a grave. Not maintained as a subtype of C but converted to K, despite the small amount of cremation present.

D Sorted cremation remains placed in an urn of non-organic material (ceramics or glass for example). The grave pit is filled with clean sand. (Urnengrab/urnbegraving)

E Sorted cremation remains placed in an urn of non-organic material (ceramics or glass for example). The grave pit is filled with sand and remains of the pyre. (brandschüttungsgrab mit urne/urnbegraving met verbrandingsresten)

F Bustumgrave. The pyre is placed straight above the previously dug out grave pit, in which the pyre will collapse at a given moment. The remains are not sorted out.

G Inhumation, the body is not cremated but placed in a grave, optionally in a chest. H Unclear or unknown what has happened to the cremation.

I Sorted and strewed cremation in a grave without pyre remains.

J Sorted and strewed cremation in a grave which also contained pyre remains. (Brandschüttungsgrab)

K Grave pit without cremation, not interpreted as a grave.

Tabel 4 Grave types.

The main differences in the grave types are related to whether or not the remains were sorted, if they were placed in a container, the type of container that was used, and if remains of the pyre were included in the filling of the grave. This is an indication of the energy that was spent for the handling of

82

Hiddink 2003, 21-25, explains grave types A, B and C which are the references for the types A, B, C and C-2 (which has been converted to K) used in this research.

(26)

26 the cremated remains, as sorting out the cremated remains would take more time then leaving them together with the pyre remains. The presence of a ditch surrounding the grave also indicates that more energy was spent in constructing the grave. These ditches came in different shapes which each got their own number appointed, as can be seen in table 5.

Shape of ditches surrounding the grave

Corresponding number No structure present 0 Rectangular 1 Square 2 Circular 3 Oval 4 Keyhole 5 Irregular 6 Unknown shape 7

Table 5 Ditches surrounding the graves.

The differences in shapes of these ditches might be of importance in relation to age or sex, both aspects which are researched here. By studying the grave types and surrounding ditches the differences in burial customs within and between sites becomes clear and differences show in the amount of energy invested in the construction of the graves.

1.3.5 The selection process, how 1008 suitable graves came to be

With a large amount of data from ten sites to work with some selection was needed in order to separate the relevant data from the total data available. Because in each publication graves were given either a number or a character to denote which type grave was meant, but these were not necessarily compatible to each other, all the grave types were converted to a system used for this research. The same was done for the surrounding ditches. How this was done exactly is further explained in short in paragraph 1.3.4 and in detail in appendix 1. Structures which were not graves, or which could not be firmly established or recognized to be graves were left out of the dataset. Uncertain graves which have been ascribed type K, and graves which were later understood as (parts of) other graves were left out, just as invalid graves, which concerned traces which were initially thought to be graves but turned out not to be. Entries with no dating, no grave type and no cremation at all were also left out, and a single instance of a concentration of sherds without cremated remains has also been left out. Finally a selection of graves which had the remark “no

(27)

27 grave”, “no information” or “possibly not a grave” were not used because they could not (with any certainty) be established as being graves. Exceptions are those instances where there were clear indications a grave was encountered, like the presences of multiple grave gifts. This has been done in order to gain a data set which is as secure as possible with regards to the burial customs. Also residual material, intrusion or later depositions are not taken into account as a firm relation to the primary deposition cannot be made.

Ditches of several shapes surrounding the graves were incorporated as these form a part of the grave structure. They are seen as an important element of a grave to study as possible relations can be made from their presence at sites and whose grave had gotten a ditch. If a ditch was present this has been noted, along with its shape.

A next step concerned the selection of the grave gifts.83 In order to study possible changes in grave gifts obviously there had to be gifts, but those that were regarded as non-intentional

depositions were left out. Bone fragments from the cremation, or inhumation whenever that occurred, are not included in the total amount of grave gifts, as they obviously are not a grave gift. Charcoal has also been left out as a grave gift, since this usually concerns leftover material from the cremation pyre. Charcoal is however mentioned in the inventory per grave, as a single entry, for its relevance to the type of grave. Another category of objects that are non-intentional are nails, as they would have been part of other objects and not have been given as a gift on their own. Shoe-nails are a bit different because they point towards the fact that the deceased was cremated or buried with shoes accompanying them, and they can, in contradiction to general nails, be appointed to a specific object, a shoe. For these nails, but also for other materials a remark has been added with regards to the material being in burned or unburned condition. When there was no mentioning of the state of an object with regards to burned or unburned in the catalogue this has been taken as the state of the material being unburned. By looking at the specifications and seeing if the objects were burned or if they show signs of melting, an idea could be formed about whether they were on the cremation pyre with the deceased or if they entered the grave after the cremation was over. In case of the shoe nails this could indicate whether the people were cremated while wearing their shoes, and likely were cremated with clothes on. It has to be mentioned that the iron nails often decay beyond recognition or disintegrate completely. Therefore they are often missing or unrecognized. This however makes them an unreliable marker for the fact if shoes were present or not.

Another criterion which was set in order to get a grip on the data was the observation that when there is only one intentional grave gift, that grave is considered less meaningful for the research of change. When items which had a single entry in the inventory but consisted of more

83

(28)

28 pieces belonging to each other, like the shoe nails, they are still counted as one, and not used. This does not mean however that the graves with a single gift are neglected completely, as they can still provide an image of a prevailing grave gift or custom.

In the end the selection process has led to the following results: there are 1008 graves in total from the 10 selected burial sites. At least one grave gift was present in 94 graves, and 715 have 2 gifts or more. In 199 cases there was no gift present but other factors, like the presence of

cremated remains, helped in establishing the structure as a grave.

In order to answer the research questions all of the data had to be analysed. To facilitate this process the data has been structured in a table in Microsoft Excel. Using this table provided structure to the data and facilitated the detection of possible changes. The table itself is digitally available as appendix 3 of this research; the outcomes of the research questions are found by using the filters and are presented in this research. All the remarks in the table are taken from the extra information provided in the catalogues used.

A last note has to be provided regarding the set-up of the main table used for this research because the way of noting information from ten different publications down in one big table proved to be a complex matter. Some choices deserve explanation in order to make them understandable for the outsider. Further explanation about how the data was processed in the main table is available in appendix 2.

1.3.6 Approach

The objects that can be accounted for as meaningful in order to investigate romanisation, as

explained above in paragraph 1.2.2, are features like the oil lamps and indications of funerary eating and drinking parties. Also specific items from a Roman culture which could have not been used for other secondary purposes than their intended use are regarded as markers of romanisation. These items are stili and inkpots, as well as objects related to the bathing ritual, like wash spatula and strigiles. If those type of objects are present they can shed light on if, at any degree, romanisation took place. Because of the extensive data set used in this study not just changes which might be due to an increased contact with Roman culture can be seen however, also other changes which might not be related to romanisation of the burial customs become clear after studying the data.

The grave type is studied in order to see if there was a preference towards a specific type, and related treatment of the dead. By looking at the graves of which the sex of the deceased was known the possibility occurred to determine if there are specific customs regarding male or female burials and if they changed over time. In order to facilitate comparison between the sexes the notation of the sex was converted to a general “M” for male and “M?” for possible male. For females

(29)

29 the notation is as follows: “F” for female and “F?” for possible female. Both males and females which fell in the category uncertain, “M??” and “F??”, were not used for questions regarding a relationship between grave types, surrounding ditches or grave gifts and the sex.

For every site each grave is studied for its type, presence or absence of a surrounding ditch, as well as its contents, meaning the human remains and the gifts. This is done in order to extract as much data as possibly to be able to investigate as much elements of the burial customs as possible. When each site is analysed, the nature of possible changes which might have occurred is sought to be explained, in which the markers of romanisation can be of help as they might indicate an increased contact with a Roman culture. The possible changes in the burial customs are also

compared between the different sites in order to see if there are similarities in the developments on each site.

What should be mentioned is a final but very important point which certainly should be kept in mind when studying the outcome of the analysis is the degree to which objects survive.84 Some materials disintegrate more than others and other objects might not have been recognized due to their state of decay, which presents a pictures of absence of certain items, which might have once been a grave gift.

84

(30)

30

2 The case studies

For this study ten burial sites have been selected, in order to create a broad geographical range within the province of Germania Inferior. Besides the geographical variation the characters of the sites are also diverse in order to get an image as complete as possible about the customs with regards to the treatment of the dead for the period and area under research. In the next section a short overview of each of the burial sites is presented in order to be able to place the information from the graves in a proper context and to gain a better understanding when looking at the types of rituals and gifts that are known from the sites. Each site has one figure of a grave from the site in order to give an idea of what was encountered during the excavations.

First information on the settlement or area to which each burial site belonged is provided. This is followed by an overview which contains the dating of the burials, the number of graves that are known from the site as well as the grave types present and, when possible, the physical anthropological data of the individuals from the graves are given. If there was additional relevant information on the site this is provided in the last section of the site introduction. The order in which the sites are presented is the chronological order of the start of each burial site, with the exception of the Nutricia terrain in Cuijk, which is chronologically last, but is dealt with together with the Fidelis research.

After the introduction the research results of the burial sites are presented. The grave types, ditches surrounding the graves and the grave gifts have been analysed on their presence, and their prevalence through time and per site. A further analysis of the prevalence of these elements in relation to the people who were buried there is also provided. Graves belonging to phase 0 have not been included in these results as they could not provide information on a development through time. Grave types of which the type was likely but not completely certain, demarcated with a question mark, have been counted as belonging to that type, in order to gain more insight in the data, instead of having a wide spectrum of possible options per category. The types of surrounding ditches are further specified in appendix 1, but it is good to mention that type 0 means that no ditch was present. This is only used at sites which also had graves with a surrounding ditches. Graves of which the type was not established at all, denoted with a question mark or a dash85, have not been used in this analysis. Graves which could not be appointed to a single phase were also left out, in order to avoid counting them twice. When looking at specific elements related to age, children, or non-adults, are those deceased who were under 20 years of age, and adults are those above 20 years of age at death. If categories involved ages of 15-25 they have been interpreted as adults, just as 17-24 years.

85

A question mark relates to the fact that the graves appear to have been studied but a grave type could not be established. The dash represents graves of which no information on the type was provided at all.