Rivers Keep Many Secrets:
Bronze and Iron Age Depositions of Human Remains
In River and Settlement Contexts
MA Thesis of Andrea Pintar
University of Amsterdam/VU University
Prof. Dr. N.G.A.M. Roymans, VU
Dr. S.J. Kluiving, VU
June 2016
1 | P a g e
Rivers Keep Many Secrets:
Bronze and Iron Age Depositions of Human Remains
in River and Settlement Contexts
Academic Master’s Thesis Of
Andrea Pintar
University of Amsterdam/VU University
Faculty of Humanities
Committee:
Supervisor: Prof. Dr. N.G.A.M. Roymans, VU
Second Supervisor: Dr. S.J. Kluiving, VU
June 2016
Amsterdam
2 | P a g e
Rivers Keep Many Secrets:
Bronze and Iron Age Depositions of Human Remains
in River and Settlement Contexts
MA Thesis of Andrea Pintar
University of Amsterdam/VU University
Contents
Acknowledgements 3
1. Introduction 5
1.1 Research Goals and Methodology 5
1.2 Sacred Rivers and the Underworld: 7
Water as a Liminal Boundary in Bronze and Iron Age Cosmography
1.3 Conceptions of Death: Excarnation and the Soul 9
1.4 Reading Rivers: Palaeogeography and Archaeological Interpretations 12
2. Wetland Contexts 15
2.1 England: Thames River 15
2.2 The Netherlands: Kessel/Lith at the Maas 20
2.3 Germany: Tollensetal (Tollense Valley) 24
2.4 England: Godwin Ridge, Great Ouse River, Cambridgeshire Fenlands 28
2.5 Threatened Heritage: Neolithic and Bronze Age 31
Alpine Lakeside Settlements Under Water
3. Dryland Contexts 33
3.1 Atlantic Scotland (Multiple sites) 33
3.2 Scotland: Broxmouth 38
3.3 Scotland: Covesea 41
3.4 Germany: Kirschbaumhӧhle (Cherry Tree Cave) 46
3.5 Czech Republic: Velim Skalka 47
3.6 Netherlands: The Terp Region (Groningen, Friesland) 48
4. Analysis of the Kessel Material 54
4.1 Identification of Skeletal Elements and Minimum Number of Individuals 54
4.2 Age/Sex Determination 54
4.3 Pathology/Trauma 56
4.4 C14 Dating and Strontium Isotope Analysis 57
4.5 Interpretations of Kessel/Lith 59
5. Interpretations of Mortuary Practices 60
5.1 Human Remains and the Life-Cycle of Communities 61
5.2 Veneration and Curation 62
5.3 Friend or Foe? Depositions Related to Conflict 63
5.4 Contextualizing Rivers in Mortuary Practices and Prehistoric Society 64
6. Conclusions 66
Bibliography 68
3 | P a g e
Acknowledgements
I dedicate this work to:
My mother, Gina and stepfather, Doug
And my grandparents, Joan and Bud
For instilling in me the drive to seek out knowledge,
And not be ashamed to love what I do.
My supervisor, Prof. Dr. Nico Roymans, VU, whose guidance and support
Motivated me to go beyond what I needed to do, and even what I thought I could
accomplish.
My second reader, Dr. Sjoerd Kluiving, VU, for his contributions
To help me make this work a success.
My best friend of over 20 years,
Rachael.
My amazing friends, especially Aidan, who never stopped
Believing in me.
In memory of my father
John Robinson (1951-2013).
Many thanks to:
Dr. Liesbeth Smits, UVA
For sharing her valuable time, knowledge and expertise with me.
The Verhagen and Stolzenbach Families
Whose dedication to archaeology
Made this and many other projects possible.
4 | P a g e
I hereby declare that this dissertation is an original work, written by myself alone. Any
information and ideas from other sources are fully acknowledged in the text, notes and
bibliography.
Place,
Date,
Signature
5 | P a g e
1. Introduction
Rivers are important agents in the landscape. They shape the earth through geomorphological processes of erosion and sedimentation, and provide water, food, irrigation for crops, and recently, sources of energy. They have been subjects of geological and geographical studies to help us
understand past environments. But what can rivers provide us in relation to the study of past human communities and behavior? The subject of watery areas as archaeological hotspots is under debate. Many archaeologists believe artefacts recovered from rivers have been eroded and have lost their context, particularly when recovery is a result of dredging activities, which destroy the stratigraphic layers required for dating. But can it be argued that the river is in fact, itself a context? By studying assemblages of animal bones, pottery, metal objects, and human remains, we can deepen our understanding of the people living in these areas, their life practices, and events that may have affected them. We can look for patterns of deposition behavior that may be similar or dissimilar among other populations across time and space.
The goal of this research is to highlight the relevance of wetland archaeology in the discussion of prehistoric mortuary practices, and the importance of excavating, preserving and studying these archaeological remains, which are at risk of being lost forever. The importance of bioarchaeological research is that it involves interdisciplinary approaches to understand past and present populations1, and can bridge the divide, at least perceived in Europe, between the
humanities, social and natural sciences. Knudson and Stojanowski argue that identity is not solely comprised of origins, migration histories or biological connections, and that osteological indicators such as cranial and dental modifications, dental and other pathologies can illustrate patterns of human social behaviors, dietary choices or traditions and human activities. They believe that interpreting social life in the past can advance our understanding of the issues that we are experiencing today.2 Studying mortuary activities from wet- and dryland sites gives us the opportunity to make archaeology relevant to social anthropology of the modern world. 1.2 Research Goals and Methodology
Within this paper, I will compare several published studies on depositions of human remains in two contexts: rivers—which includes material recovered from the water, the riverbed sediment and the
1 Knudson/Stojanowski, 2008: 398. 2 Knudson/Stojanowski, 2008: 398-399.
6 | P a g e
banks—and settlements. Through this work, I refer to ‘wetland’ and ‘dryland’ sites. The parameters for ‘wetland’ sites are those where water is a dominating factor in the landscape, such as a fenland, estuary or riverbank. By contrast, ‘dryland’ sites are those which I consider to be areas less affected by lacustrine or fluvial forces, thus the majority of settlements in these areas are situated on dry soil. However, I will mention two caves, which may or may not have at one point contained pools of water. These will be discussed within the dryland contexts because from my point of view, they share more similarities with these sites than those in or near fenlands, estuaries and rivers. The study area is Britain, predominately the east and west coasts of Scotland, the Thames river and Fenland areas in England, the Netherlands, specifically at the Maas river and the northern Terp region, Germany, and the Czech Republic. I selected this geographic area to give a wide range of examples; examining mortuary treatments in only one or two of the areas would provide too limited a dataset to make comparisons. The time range of this study is the Early Bronze Age through the first centuries AD, and it is important to note that the Scottish Iron Age overlaps with the continental Roman Period.
The questions I sought to answer with this research are: what are the mortuary activities for prehistoric populations in the geographic areas studied? What can these depositions of human remains tell us about Bronze and Iron Age society in Northwestern Europe? What information can rivers provide in regards to these activities? To answer these questions, several methodologies will be employed. The first half of this work will examine current hypotheses regarding deposition patterns, and seek to investigate whether there are demographic trends for certain burials or mortuary
treatments. Such factors could be an individual’s age, sex, status within the community, and physical traits, such as deformities, disease or injury. Analysis of the assemblages is also important, as it depicts what bones are recovered, their preservation, and any characteristics that can indicate activities of the individuals in life, and treatment of their remains after death. After reviewing these cases, I will address any distinct patterns in deposition and possible interpretations.
The second part of this research is actual analysis of human remains recovered by amateur archaeologists from the river Maas at Kessel/Lith. This osteological study will attempt to determine approximate age at death, sex, pathology and trauma of the individuals, thereby proving that riverine material can yield a significant quantity of information which can be used in combination with other archaeological, historical and geological information to interpret sites. Additional information regarding radiocarbon dating of the material and the significance of isotope analysis on bones deposited in water at Kessel is currently in preparation by Lisette Kootker at the VU University Amsterdam. Her analysis further supports the importance of recovering, studying and preserving riverine material and combats negative, or at best, ambivalent attitudes towards such archaeological
7 | P a g e
remains. By connecting the role of fluvial environments to the deposition practices described in the first half of this work, I hope to illustrate the valuable place riverine archaeology occupies within the discourse on prehistoric local and supralocal communal, cult and mortuary practices, as well as conflict contexts, and why the overall study of prehistoric society is important for us today.
1.2 Sacred Rivers and the Underworld: Water as a Liminal Boundary in Bronze and Iron Age Cosmography
This research covers several millennia of mortuary treatments of human remains in settlement and river contexts. Before discussing the numerous patterns of depositions, it is important to understand possible interpretations of prehistoric people’s attitudes towards their environment, including potential uses of certain elements in ritual and religion. This can help put mortuary practices in a clearer context, since life and inevitably, death, played a major role in most, if not all, societies throughout the world. Oestigaard describes water as “a medium which links or changes totally different
aspects of humanity and divinities into a coherent unit; it bridges paradoxes, transcends the different human and divine realms, allows interaction with the gods, and enables divinities to interfere with humanity.”3 Rivers have
been seen as natural geographical borders, which connect political power and religious belief, allowing the living and the dead to connect with the spiritual world.4 Oestigaard believes that rivers, springs, wells, bogs, and swamps among other sources of water are divided into two categories regarding use: water for purification and preparation for the afterlife, and water to bring a rich and bountiful harvest.5 However, a third category is the use of water as a liminal boundary between actual geographical locations and the division between the worlds of the living and the dead.
The first and third categories form part of the interpretation of the depositions in the
Thames, Ouse, Maas, and Tollense rivers. In the cases of Kessel and the Tollensetal, the rivers served as territorial boundaries, which could be a factor in the conflict between and ultimately death of significant numbers of people at sites near the Maas and Tollense rivers. While simple battle clean-up has been suggested as a possibility for the number of human remains recovered from these rivers, it is also possible that the dead were deposited with their ritually bent weapons in this liminal place between the earthly realm and the underworld. At the Thames, human remains have been found in conjunction with many Bronze Age metal objects, and while further dating is necessary, and selective deposition is not the only activity occurring at the site, it seems likely that ritual may have played a role in a certain number of skulls and other skeletal elements deposited in the river. Depositional
3 Oestigaard, 2011: 38.
4 Brück, 2011: 389; Oestigaard, 2011: 39-40. 5 Oestigaard, 2011: 39, 42.
8 | P a g e
patterns along the Ouse River in the Cambridgeshire fenlands further supports the idea that water was a key element in Bronze and Iron Age mortuary traditions.
Bronze Age, and probably, Iron Age people divided their world into three parts: the sky, which was considered a place of light and life, symbolized by wheels, the sun and horses; the earth as the domain of the living; and the underworld, a watery place of death symbolized by moons, ships and water birds.6 The numerous metal objects recovered from the study area rivers, especially at confluences, are further supported by excavations at the intersection of the rivers Havel and Spree at Berlin-Spandau. Here, large numbers of Bronze Age swords, axes, spearheads, rapiers and daggers, mostly of northern German and southern Scandinavian origin as well as some imports, were
recovered, and were deposited over a period of about a century. In addition to these objects, some of which had been unused at the time of deposition, it was discovered that a platform had been
constructed over the Spree, which allowed those participating in the offerings to be separated from and visible to their audience.7 Iron Age people too, used water for propitiatory offerings. Significant quantities of Iron Age metal objects have been recovered from rivers in Britain and Ireland,
especially the Thames, Witham and Bann, as well as the Maas in the Netherlands. Sites such as La Tène (Switzerland) and Llyn Cerrig Bach (Wales) have also produced depositions of human remains, prompting some researchers to suggest that human offerings were made through sacrifices.8
In addition to rivers, caves, some of which may have contained pools of water, were also places that could have served as liminal boundaries between earth and the underworld, and this cosmological idea could partly provide an interpretation for the depositions of human remains. For example, Sculptor’s Cave at Covesea (Scotland), which will be discussed in more detail later, is situated at an inaccessible inlet on the shore of Moray Firth, and is easiest to reach by sea, which could have increased its appeal for ritual use.9 Kirschbaumhӧhle (Cherry Tree Cave, Bavaria), also discussed later, is a perfect example highlighting the longevity of such practices in watery and dark places, preceding the Iron and Bronze Ages. Depositions of human remains here spanned millennia, including the Late Neolithic (ca. 2860-2660 cal BC), Early Bronze Age (ca. 1900-1750 cal BC), and the Hallstatt or Early La Tène period of the Iron Age (ca. 760-410 cal BC).10 Ritual destruction of objects has also been a pattern seen in river and land sites. A Bronze Age hoard at Crévic, Meurthre-et-Moselle, northeast France, yielded an axe and spearhead which had been bent and battered with 6 Brück, 2011: 393. 7 Brück, 2011: 389. 8 Joy, 2011: 410-411. 9 Brück, 2011: 390. 10 Seregély, 2014: 18.
9 | P a g e
considerable force, and bracelets whose terminal ends had been broken off. Brück suggests that these actions could symbolize the decommissioning of objects which were thought to contain a particular potency or social significance.11 Ritual bending and breaking of objects has been evidenced in the Kessel material, and also, could be a reason for the enormous quantities of La Tène glass bracelets (5000 fragments recovered from the Lower Rhine areas in Netherlands, Germany and Belgium)12, and I suggested this breaking could have symbolized the fluidity of Iron Age identity. Changes in the life situation of the wearer, perhaps led to the decommissioning of some of these items. These
fragments represent symbolic destruction, rather than accidental breakage and recycling because the size and shape of the pieces seem to resemble deliberately halved/quartered segments, rather than fractures resulting from accidental or in-situ damage. La Tène bracelet fragments were probably not recycled because there is evidence of fragments being reworked into small amulets to be worn on strings; objects that retain importance even in fragmented states were probably not seen as waste to be reused. The same can be said for ritually bent swords or broken ceramics. Why else were the metals not re-smelted or the ceramic used as grog in temper? Smashed drinking vessels were also found in Mycenaean burials on Crete and mainland Greece, and Mycenaean tombs at Pylos on the Peloponnese peninsula and Ilysos, Rhodes have also yielded bent and broken swords.13 The large geographical area in which the deliberate breaking of valuable items occurs, is evidence for widespread and long-lived Bronze and Iron Age traditions.
1.3 Conceptions of Death: Excarnation and the Soul
Many of the human remains found on settlements in Britain and parts of the Netherlands have displayed evidence that some deceased individuals were exposed to the elements before bodies or parts of bodies were deposited in foundations of structures, pits or wells. This has led many
researchers to suggest that in addition to cremation, excarnation by exposure was another funerary rite practiced in many Iron Age local communities. This section provides a short background into the process of excarnation by exposure and what traces this rite could leave in the archaeological record. After the basis of the practice has been explained, I will give a few interpretations put forth by researchers on the concept of death within prehistoric cosmology.
As previously mentioned, there are many examples of inhumations of disarticulated or partially disarticulated skeletons in domestic areas of roundhouses in Britain and terp houses in the northern Netherlands. By describing the process of secondary burial, Carr and Knüsel provide an
11 Brück, 2011: 396.
12 See Roymans/Verniers, 2010: 207. 13 Brück, 2011: 397-398.
10 | P a g e
explanation for these remains.14 The fact that secondary burial accounts for only an estimated 5% of Iron Age local populations in a given area, suggests that either only small numbers of bones from the deceased were reinterred on the settlement, or only select individuals in a population were accorded this treatment.15 Structural evidence that may support the excarnation rite includes rectangular four-post scaffolds, which have been previously interpreted as storage or a type of granary. However, almost all sites at which partially disarticulated skeletons have been found, contained these post-built structures, and even more striking is that at the site Danebury, pits containing disarticulated or partially-disarticulated bodies clustered around where the structures once stood.16
Outside these post-built platforms, the rite of excarnation does not leave intensely visible traces in the archaeological record, other than scattered bones. However, we can use osteological analyses of the bones themselves, as well as the type and degree of disarticulation, which follows a predictable sequence: beginning with the skull and (whole) limbs, proceeding to the ribs, individual limb bones (such as carpals separating from the radius/ulna), and lastly the vertebrae.17 Carr and Knüsel write: “If (articulated but incomplete skeletons) are found in the ‘correct’ units for the natural
disarticulation sequence of humans (i.e. a whole arm found in articulation with the hand still attached), these may be read as evidence for the ‘intermediary period’ of the excarnation rite before secondary burial. If an
incomplete body is found in an ‘incorrect’ unit of disarticulation, then damage by animals is the more likely cause of the separation of body parts.”18 In the case of human remains found in disarticulated states in riverine
contexts, this scattering is probably a result, not of scavengers, but river taphonomic processes. Another aspect to scattered or disarticulated, mixed remains could be a result of dismemberment. In this case, it is likely cut marks would appear on the bones.19 In summary, the evidence for exposed bodies includes: (1) traces of gnawing by animals, (2) scattered, isolated, fragmentary, weathered or split bones, (3) disarticulated skeletons, and (4) incomplete skeletons lacking small bones (phalanges, carpals, vertebrae) or missing limbs (no evidence of cut marks).20
Another approach in discussing burial patterns, evidence of excarnation or dismemberment, is anthropologie de terrain, which is a field method developed in France and focuses on burial
excavation. The ultimate goal is to reconstruct the burial from the time of deposition through the
14 Carr/Knüsel, 1997: 167.
15 Carr/Knüsel, 1997: 168. For percentage authors referred to Wait, 1985: Ritual and Religion in Iron Age Britain. 16 Carr/Knüsel, 1997: 168. Exposure scaffolds, authors referred to first reference by Ellison, Drewett, 1971: “Pits and Post-holes in the British Early Iron Age.”
17 Carr/Knüsel, 1997: 168-169. 18 Carr/Knüsel, 1997: 169. 19 Carr/Knüsel, 1997: 169. 20 Carr/Knüsel, 1997: 170.
11 | P a g e
taphonomic processes that occurred until excavation. This approach is concerned with human anatomy, the decomposition process and natural influences, such as flooding, root activity, insects and burrowing animals.21 Anthropologie de terrain operates under the ideology that “every deviation from anatomical articulation has an explanation worthy of discovery.” If applied, this field method can help
distinguish the causes of hyper-flexed burials, whether disarticulation was caused by post-interment human disturbance, and if deviations in anatomical position could suggest dismemberment or natural decomposition processes.22
Now that the process of excarnation has been laid out, how do we interpret the use of this rite? Humphreys proposed in her 1981 research that the concept of death had three key timescales: when the right time to die is, what the structure of the transition from living, dying and death is, and how the remembrance of the dead and the world of the dead is organized through time.23 She also provided examples from ancient Greece of Achilles’ choice to either die young and gloriously, or old and anonymously, and Victorian England’s societal view that the death of the young and innocent was particularly poignant. Looked at in this way, deaths that counter the typical demographic of a community are often singled out for a special treatment.24 This paradigm could be an explanation for exposure rites used in British and Dutch prehistoric sites, and may offer insight into the extremely rare inhumation burials noted by Antje Nieuwhof in the Netherlands, where cremation is considered to be the dominant rite through the Roman period.25 Practices of reburial of exposed (and sometimes curated) remains on settlements can have several ideological purposes. On one hand, excarnation and secondary burial is a specific mortuary treatment for the release of the soul of the deceased, and on the other, bringing human remains back to the settlement and interring them (and sometimes animal bones) in middens, ditches, house foundations and other domestic contexts was interpreted as “an attempt at bringing about the regeneration of life and fertility in the wider domain of cultivated land and animals.”26 This idea can also be connected with the ritual destruction of metal and ceramic objects, and also of people (cremation, bone modification, cannibalism, excarnation). Brück
suggested that literal and symbolic destruction occurred in both metalworking and pottery
production in the early stages of smelting and casting, but also in later actions of recycling (use of
21 Knudsen/Stojanowski, 2008: 407. 22 Knudson/Stojanowski, 2008: 408. 23 Pierce, 1997: 175. 24 Pierce, 1997: 175. 25 See Nieuwhof, 2015: 248-251. 26 Carr/Knüsel, 1997: 171.
12 | P a g e
grog in temper), and she concluded that “the destruction of both people and artefacts acted not only as a
means of transportation, but was seen as essential to the regeneration of life in the face of death.”27
1.4 Reading Rivers: Palaeogeography and Archaeological Interpretation
Thus far, this paper has outlined the conceptions of water usage, especially relating to death, in Bronze and Iron Age cosmologies. I have also discussed the funerary rite of excarnation and how this could be viewed in the archaeological record. Ideas of death, water and mortuary treatments can provide a backdrop with which we can interpret depositions of human remains and other objects in rivers. In addition to these interpretations, it is also important to understand how rivers functioned in prehistoric times, not just as a symbol, but as a physical part of the landscape. Alluvial environments played a huge role in settlement patterns, as well as hunter-gatherer and agrarian activities. Tracing the contexts of archaeological remains recovered from rivers requires understanding the
palaeogeographical and -geological aspects of these bodies of water, and how prehistoric populations interacted with these vital resources.
Prehistoric Fluvial Environments
Fluvial environments such as the Dutch Delta area provide insight into human habitation and activities because some taphonomic processes such as river meandering, flooding and gradual shrinking of habitable areas due to ‘drowning’ are visible in human timescales, and can show potential uses of the landscape.28 The early 20th century saw the beginning of palaeogeographic studies of the Dutch Rhine-Maas (Meuse) Delta and detailed soil mapping, particularly studies by Modderman (1949, 1955), traced fluvial development with (pre)historic habitation.29 Focus on Holocene development in the delta increasingly appeared in combined palaeoenvironmental- and archaeological studies, which allowed for better, larger and more detailed palaeogeographical reconstructions of different spatial and temporal scales to be used in archaeological investigations of Holocene environments.30 Understanding of active and passive fluvial systems is integral to
deciphering human habitation patterns and activities in the Rhine-Maas Delta.31
Active fluvial systems do not typically provide suitable environments for permanent
habitation due to frequent flooding, though levees and crevasse deposits may provide some areas for settlement activities, such as cattle grazing on the lower-lying areas. Extensive and thick crevasse 27 Brück, 2011: 398. 28 Arnoldussen, 2008: 29. 29 Arnoldussen, 2008: 29-30. 30 Arnoldussen, 2008: 30. 31 Arnoldussen, 2008: 55.
13 | P a g e
deposits may have been suitable for occupation, such as the Middle Bronze Age area near
Dodewaard.32 As I will discuss later, activity in many wetland areas stretches back to the Mesolithic and Neolithic. In addition to the high biodiversity in riverine and marine areas, rivers also provided sources of granite, flint, and quartz for various lithic tools, which additionally made wetlands attractive areas for prehistoric habitation.33 Passive fluvial systems provided similar ecological opportunities as active systems, however, due to silting, changes in vegetation, fish and waterfowl occurred. Softwood trees and shrubs were gradually replaced by hardwood strands such as ash, elm and oak, which could have provided construction wood and acorns for human consumption.34 The combination of freshwater lakes, and the development of sandy dunes and ridges provided an environment for developing habitation and agrarian activities.35 The creation of crevasse splays in active and passive fluvial systems is important in discovering these patterns, because developing vegetation was easier to clear for agriculture than more wooded areas on older, passive fluvial landscapes, although risk of flooding may have led to seasonal abandonment of many of these areas. These crevasse deposits were located between two major agrarian landscapes for prehistoric
farmers—the upper lands could be used for crop cultivation, while the lower flood basins provided grasslands and drinking pools for cattle.36
Another type of wetland environment is the estuary. Estuaries provide an environment that is conducive to the preservation of archaeological remains which are rarer on dry land. Their ecological changes can be traced through dendrochronology, and illustrate the connection between
environment and past human activities.37 Similarly to the study of fluvial systems in the Dutch River area, researchers at the Severn Estuary in Wales also found patterns in archaeological remains of structures (roundhouses and rectangular buildings) which suggested two main periods of human activity: the Middle to Late Bronze Age and the Early Iron Age (400-300 BC).38 Early stages of marine transgression in the estuary system were conducive to human activity, particularly cattle grazing in lower wetlands, as evidenced by preserved hoof prints, many of which have been found at Goldcliff39; this pattern is also evidenced at the levees in the Rhine-Maas Delta. Bell and Neuman concluded that many of the activities in the Severn Estuary may have been seasonal, which means
32 Arnoldussen, 2008: 55-56.
33 Van den Biggelaar/Kluiving, 2015, 545. 34 Arnoldussen, 2008: 57.
35 Van den Biggelaar/Kluiving, 2015, 538-539. 36 Arnoldussen, 2008: 57.
37 Bell/Neuman, 1997: 95-96. 38 Bell/Neuman, 1997: 107-108. 39 Bell/Neuman, 1997: 108.
14 | P a g e
permanent sites of these time periods are likely to be on dry land.40 In this case, tracing areas of seasonal activity can lead to discovery of other permanent prehistoric land sites.
The effects of wetland environments are illustrated well in the case of the Balaton Lake community in Hungary (Copper Age/Chalcolithic, ca. 3500-3000 BC). Like local groups in northwestern Europe, the agrarian activities consisted of animal husbandry supplemented by crop cultivation. The population was non-nomadic, but practiced transhumance, moving with their domesticated animals. During the winter months, when grazing herds slowed, people gathered in larger numbers. Consequently, local and perhaps supralocal rituals occurred in greater frequency during this time.41 Therefore, the marsh environment affected the movements and activities of the people, and this is reflected in the archaeological record in certain concentrations of artefacts (i.e. burials on the settlement rather than in cemeteries), which may differ from dry land sites. To understand how prehistoric people interacted with the landscape, we need to acknowledge to dynamic nature of fluvial systems. Braiding, meandering and anastomosing river systems affect development or inundation of ridges, terraces, and levees. The types of soils present in the area determine the vegetation and fauna of a given locale, which in turn affect the types of activities which may occur at a site. Agriculture occurs typically on higher ground, while cattle grazing and perhaps, fishing and hunting occur on lower grounds. Habitation may be restricted to seasonal or semi-permanent timescales, which affects the disbursement of archaeological artefacts, including burial or deposition of human remains. Changing river courses can scatter settlement remains or erode foundations, but can also make these sites more visible and easier to explore and protect, and wetlands can preserve organic materials better than dry soil. Both wetland and dryland sites complete the prehistoric picture; opportunities for greater understanding of prehistory would be lost if only one were studied without the other.
40 Bell/Neuman, 1997: 108. 41 Horváth/Kӧhler, 2012: 468.
15 | P a g e
2.
Wetland Contexts
Now that I have discussed the symbolic significance of water in relation to Bronze and Iron Age ritual, conceptions of death with regards to the excarnation rite, and the palaeogeographical and -geological roles of fluvial and lacustrine environments, the next chapter will encompass discussion on finds of human remains from wetlands. These archaeological remains originate from rivers/river sediments and fenlands, and generally span the Early Bronze to the mid-to Late Iron Ages and a variety of contexts. The last case discussed in this section is of the threatened preservation and heritage management of prehistoric Alpine lake settlements, which I believe also supports to need for study and preservation of the remains recovered from the Thames, Tollense, Maas, Ouse and other rivers.
2.1 England: The Thames River Skulls
There are several hypotheses regarding the deposition of skulls in rivers—whether skulls entered the water as select depositions or as part of complete bodies, and if the preponderance of crania
recovered reflect specific interest in the skull on the part of antiquarians or amateur archaeologists. Some researchers suggested that a large proportion of skulls originated from ritual and funerary activities, while others believed in more pragmatic reasons, such as murder, suicide, and accidental drowning. The first is supported by excavations revealing much deposition of metalwork and human remains, a practice that may have persisted from the Bronze Age; the second idea is reinforced by
16 | P a g e
studies of post-mortem forensic taphonomy on human remains in rivers.42 Ritual activity occuring at the Thames in the Bronze and Iron Ages is supported by evidence from excavations at an area in the Middle Thames Valley. Palaeoenvironmental and -geological investigations started at the Eton College rowing lake, which is situated on a 0.75km wide, 2km long section of floodplain in the Thames Valley.43 A settlement was found at Dorney, which dated back to almost the start of the Neolithic (3950-3640 cal BC), based on radiocarbon dating of a charred hazelnut shell and grain of emmer wheat. A barrow cemetery and a discovery of a substantial bridge across the Thames channel, evidenced by a double row of oak piles, proved that activity continued into the Early and Middle Bronze Age; the radiocarbon dating of several of the oak piles placed the date of the bridge between 1530-1310 cal BC and 1420-1210 cal BC.44 An Iron Age Bridge, dated between 770-400 cal BC and 770-390 cal BC, eventually replaced the previous structure. Excavations around the area discovered four other Iron Age bridges, as well as various hurdle structures, and votive offerings of pottery and human bones, which were dated to 940-550 cal BC.45
The ideas of ritual deposition are supported by two studies of skulls from the Thames and its Walbrook tributary. Bradley and Gordon analyzed around 300 skulls from the Thames, recovered from dredging activities in the 19th century, which were stored in museums. They suggested that based on large quantities of Bronze Age metalwork recovered from silty areas along the river, that the skulls from this period may have been deposited as part of a ‘rite of passage’ for the dead.46 Marsh and West studied 48 skulls from the Walbrook, recovered by workmen in the 19th and 20th centuries. The material was determined to be of Iron Age and Romano-British origins, supported by
radiocarbon dating of three skulls (completed by Bradley and Gordon) and preserved timber in close proximity to the others (completed by Marsh and West). The authors concluded that these skulls may also have been selectively deposited.47 Edwards, Weisskopf and Hamilton’s 2009 study included 18 Thames skulls (see Fig. A) and 33 Walbrook skulls from the Museum of London (MOL). Of the 33 Walbrook skulls, 24 were originally part of Marsh and West’s 1981 study, though reassessment was decided, as parts of the osteological analysis were not published.48
42 Edwards/Weisskopf/Hamilton, 2009: 35-36. 43 Parker/Robinson, 2003: 43. 44 Parker/Robinson, 2003: 54. 45 Parker/Robinson, 2003: 55. 46 Edwards/Weisskopf/Hamilton, 2009: 36. 47 Edwards/Weisskopf/Hamilton, 2009: 36. 48 Edwards/Weisskopf/Hamilton, 2009: 36-37.
17 | P a g e
Osteological Analysis
Seventeen of the skulls were determined to be adult or young adult, while the 18th was that of a child about 5 years old, and none of the skulls displayed fatal ante-mortem injuries, although there was evidence of healed trauma on several. The damage done to the facial and lateral structures, as well as single-rooted teeth was probably the result of rolling and erosion in river-bed deposits, although very few of the skulls displayed severe pitting and cracking from continuous movement through a fast-flowing body of water—this suggests that the remains were likely submerged in a silty, slower-moving section of the river.49 The Thames group consisted of 8 males (including possible males) and 4 females (including possible females), while sex assignation was indeterminable for 2 adults, 3 young adults, and a child. The Walbrook skulls were comprised of 10 males (including possible males), 12 females (including possible females), while 1 juvenile, 3 young adults, and 7 adults were indeterminable.50 Radiocarbon dating was completed on 6 Thames skulls (see Fig. B). One was dated to the Late Neolithic, 2 to the Bronze Age, 1 to the Late Iron Age, and 2 to the Medieval period. Another skull, recovered from the Chelsea foreshore (MOL L344) was dated (twice) in 2001 (by Haughey and Hamilton) to the Bronze Age.51 The only other dates available from the 18 Thames skulls were obtained by Bradley and Gordon. Of these 6, 4 date to the Middle and Late Bronze Age, and the last two dated to the Neolithic and Anglo-Saxon periods. In summary, of 13 out of 18 Thames skulls, 2 dated to the Neolithic, 7 to the Bronze Age, 1 to the Iron Age, 1 to the Anglo-Saxon period, and 2 to the later Medieval period.52
(See Fig. 14, Appendices) Violence, Ritual or Fluvial Effects? Forensic Taphonomy Provides Answers
There are three contentious topics surrounding the discussion of Thames skulls: how the skulls came to be in the water, selective deposition of cranial elements versus depositions of complete bodies, and
49 Edwards/Weisskopf/Hamilton, 2009: 39. 50 Edwards/Weisskopf/Hamilton, 2009: 39. 51 Edwards/Weisskopf/Hamilton, 2009: 43. 52 Edwards/Weisskopf/Hamilton, 2009: 43-44.
Fig. B Radiocarbon dates for 6 Thames Skulls; Note: The Chelsea foreshore dates are from the same sample; Edwards/Weisskopf/Hamilton, 2009: 43.
18 | P a g e
whether these depositions were related to ritual.53 This section will look at how forensic taphonomy evaluates both natural forces and human agency. There has been discussion that rivers and other ‘watery places’ “represent a boundary with a spiritual world after death and are therefore fitting places to lay
ancestral bones to rest.”54 The idea of ritual deposition developed in response to patterns that were
difficult to explain by more mundane means, such as an abundance of cranial bones or a high proportion of Bronze Age human and animal remains associated with contemporaneous metalwork.55 In his dissertation on sacrificial landscapes, Fontijn discussed the idea of ‘cultural biographies’ of objects as an aspect of selective deposition, though his focus was mainly metalwork. In his view, cultural biographies are created through socially constructed meanings attributed to objects, which then give them a certain ‘life-path.’56 While the Thames skulls are not objects like swords or coins, humans too have a ‘life-path’ and in death, in the minds of Bronze and Iron Age people, that path led to the Otherworld, which was represented by water. Therefore, depositing bodies or parts of bodies of the dead, and associated objects, would help the departed into the
afterlife. Evidence to support some possible instances of selective deposition are finds of skulls dating to the Bronze Age on the Middle Thames at Wallingford, Runnymede, the river Lea at Enfield, and at Goldcliff in the Severn Estuary.57
While ritual is one explanation for the depositions of skulls in the Thames, taphonomic processes are another. The disparity between cranial and post-cranial elements recovered from dredging, antiquarian and osteoarchaeological excavations can be partially explained by forensic taphonomy, which shows that when complete bodies are deposited in moving water, heads are more easily separated than in land burials and depositions.58 Erosion of riverbanks where bodies may have been buried is another taphonomic aspect which may have resulted in a significant quantity of human remains in the river. These skeletons would have been affected by fluvial sorting, and the disbursement would depend on the weight and shape of various parts. It is precisely this scenario which makes the understanding of alluvial sedimentation, erosion levels and river courses in the Holocene important. Edwards, Weisskopf and Hamilton suggest that “river action and erosion are very pertinent to this discussion” because the tidal range and levels of sediment ablation have shown much variation over the millennia, and the sea-level changes in the last 3,000 years have had a profound effect upon these processes. Excavations at Eton in the Middle Thames revealed burials of
53 Edwards/Weisskopf/Hamilton, 2009: 44.
54 Edwards/Weisskopf/Hamilton, 2009: 46. See also Brück, 2011 and Oestigaard, 2011. 55 Edwards/Weisskopf/Hamilton, 2009: 46.
56 Fontijn, 2002: 26.
57Edwards/Weisskopf/Hamilton, 2009: 46. 58 Edwards/Weisskopf/Hamilton, 2009: 44.
19 | P a g e
about 15 individuals (1300-200 BC) which had been placed on sandbank islands in the middle of the river. The skeletons were from in-situ graves which seemed to have been “weighted down” in case of flood, suggesting that prehistoric populations were also aware of changing river patterns.59
Another factor in the recovery of numerous cranial elements is that skulls and craniofacial bones are more recognizable as human, especially to dredging crews and untrained osteologists, than mandibles, torso and long bones. Still, the question remains, if this can explain areas where research excavations have produced large quantities of skull material, rather than dredging.60 If a number of intact bodies were deposited in the river, it would be expected that some mandibles would also be recovered, even if disbursed downstream. In addition to the 10% of skulls that were accompanied by mandibles in the Thames and Walbrook collections, 3 mandibles (MOL collection) were also found downstream, 4 are listed in the BMHN Thames collection, and 14 loose mandibles were reported by Bradley and Gordon. It is this recovery of a significant number of mandibles in addition to some post-cranial material, which led the researchers to conclude that many of the skulls probably entered the water as an intact or partially intact body, not through selective depositions of crania alone.61 While the subject of selective deposition of skulls versus deposition of mostly complete bodies may be unresolved, based on excavation evidence revealing prehistoric bridges, piers and other structures, a large quantity of Bronze Age metalwork and of course, the crania, ritual deposition must have been occurring at the Thames. Edwards, Weisskopf and Hamilton conclude, “…the situation is likely to be
more complex with other mechanisms shaping these data” and “[there is a] need for more complex investigations taking into account such factors as geographical location, patterns of sedimentation/alluviation in or near to rivers, and associated human settlement patterns.”62
59 Edwards/Weisskopf/Hamilton, 2009: 47-48. 60 Edwards/Weisskopf/Hamilton, 2009: 45. 61 Edwards/Weisskopf/Hamilton, 2009: 45. 62 Edwards/Weisskopf/Hamilton, 2009: 47-48.
20 | P a g e
2.2 The Netherlands: Kessel/Lith at the Maas
The discussion in this case revolves around the 1999 study of recovered material by Muuk ter Schegget. The ‘new’ unpublished material, which I analyzed, will be discussed in Chapter 4. This section will summarize a brief history of depositional patterns of human remains in the Netherlands, and an overview of activities at the site, including work by amateur archaeologists and dredging. I will then discuss Ter Schegget’s analysis of the bones and proceed to interpretations of the site.
The study of prehistoric human remains recovered from rivers has not been at the forefront of Dutch archaeology. There is a series of articles, entitled “River Valley People”, which describes human bone material recovered from rivers, and contains the analysis of around 150 finds, but at this time, there is no systemic inventory of human bones deposited in Dutch rivers.63 Main find spots of human bones seem to be clustered around the bank zones of the Maas and IJssel, which have yielded assemblages of predominately skulls and femura, such as sites at Kessel and Roermond. In addition to the bone material, there has also been an accumulation of metal objects. The attitude that these finds were eroded and lacked context began to change by the end of the 1980s, but the problem of dating still remains, at least in regards to the bone material, which is a result of dredging activities disturbing the stratigraphy.64
63 Ter Schegget, 1999: 201-202. 64 Ter Schegget, 1999: 202.
Fig. C Human remains recovered from Kessel, mostly Late Iron Age; Roymans, 2004: 128 and Ter Schegget, 1999: 210.
21 | P a g e
Material which can be radiocarbon dated is limited in the Netherlands, and Kessel is the only site where a large portion of human remains can be analyzed in conjunction with other finds, such as pottery, weapons, brooches, bronze cauldrons, harvest implements, and unburnt animal bones—C14 dates range from 360 cal BC-cal AD 1260, and show a peak in the Iron Age, and a decrease in the Early Middle Ages.65 (See Fig. 16, Appendices) Updated dating for the new material will be discussed in Chapter 5. Britain has been the main region, in which human remains found in rivers were seriously studied, but this is changing with the discovery of sites such as Kessel/Lith in the Netherlands and the Tollense Valley in north-eastern Germany.
The Archaeology is in the Pits! The History of Dredging at Kessel/Lith and the Discovery of an Iron Age Settlement
Kessel and Lith are situated in North Brabant, at the area where the Maas and Waal almost meet.
(See Fig.11, Appendices) Dredging activities in the 1930s unearthed a small number of finds, but with large scale sand and gravel extraction in the 1950s, serious archaeological collection and study of artefacts began and continued through the 1990s. The southern bank of the Maas at Kessel/Lith is where two dredge pits, which became a single pond, were created.66 Amateur archaeologists Olaf and Leo Stolzenbach discovered a significant quantity of unburnt human remains in a dredge pit at Kessel. Between 1991 and 1993, they collected more than 650 human bones (see Fig. C), along with large quantities of Late Iron Age pottery, unburnt animal bones and metal objects.67 As of Ter Schegget’s study, an enormous quantity of animal bones and fragments have been recovered, represented first by cattle, then by pig at one-fifth the total, one-tenth represented by horse,
sheep/goat at 5 percent, and dog at less than 1 percent.68 The best preserved (human) bone material includes the teeth and distal part of the humerus, as well as complete bones. Less compact bones, such as the proximal end of the humerus and vertebrae are less well preserved. The types of bones found in an environment such as at Kessel/Lith depend on the characteristics of the bones
themselves, the nature and intensity of the dredging activities, and the degree of interest of the crew.69 The following section includes a short summary of dating, determination of sex, and minimum number of individuals (MNI).
65 Ter Schegget, 1999: 202-203. 66 Roymans, 2004: 105. 67 Roymans, 2004: 127-128. 68 Ter Schegget, 1999: 208. 69 Ter Schegget, 1999: 208.
22 | P a g e
The researchers selected 7 right femura, 6 skull fragments (with evidence of trauma) and a humerus (with evidence of malnutrition) for dating. The dates span more than a millennium, with two peaks, the Iron Age (8 bones), and the Merovingian Period (3 bones).70 For sex and age determination, the study focused on the skull and pelvis, using sexing guidelines from the
Arbeitsgruppe Europӓischer Anthropologen (AEA), and the results (see Fig. D) showed 15.8% definite
male, 5.4% definite female, 5.7% probable male, 4.6% probable female, 58.7% indeterminate adult, and ca. 10% indeterminate juvenile. The age demographic consisted of 90% adult, and 10% juvenile, with an MNI of 55 established by the number of right parietal bones, or an MNI of 45 based on right femura.71 (See Fig. 26, Appendices) Two Late Iron Age skulls depict cut marks, a third displays a round hole above the left orbit, possibly caused by a spear, and the large portion of the frontal on a fourth was hewn off, possibly caused by a sword blow.72
The bones do not display gnawing marks, are remarkably well-preserved, and have not suffered river erosion or serious damage from dredging. The main hypothesis for the origin of the bones in the river (put forth by Ter Schegget) was that the bodies of the individuals must have been deposited in the water shortly after death as part of a ritual activity at a cult place.73
Historical sources and current cartography have brought to light some disagreement as to the location of the Maas/Waal confluence. According to Caesar, the Rhine was split into two branches, the northern part bore the name Rhine, and the southern branch was called the Waal, which was joined by the Maas at or around Kessel/Lith. However, according to Berendsen and Stouthamer’s palaeogeographic study, the Maas and Waal did not form a confluence at Kessel/Lith, rather the Waal’s course was located further to north, flowing from Tiel via the Linge system toward the coast, and the more southern course did not emerge until Late Roman times.74 Roymans maintains that this modern reconstruction may not be accurate because the river course between Tiel and Kessel/Lith has been subject to erosion, making it difficult to assess whether the course existed there in the Late
70 Ter Schegget, 1999: 209-210.
71 Ter Schegget, 1999: 212-213. Table 3, 212. 72 Roymans, 2004: 129.
73 Roymans, 2004: 129. 74 Roymans, 2004: 131-132.
Fig. D Sex Distribution, Kessel; Ter Schegget, 1999: 213
23 | P a g e
Iron Age/Early Roman Period. Therefore, he chose to utilize the historical sources in conjunction with archaeological evidence in his reconstructions of the area.75
Results and Interpretations: Conflict and Cult Activity
The researchers determined that there was no indication that skulls were selectively deposited, and while the assemblage displayed more right bones than left, this was most likely due to chance recovery, rather than purposeful selection of right limbs associated with males.76 Distinguishing between ante-mortem injury and post-mortem or modern damage is difficult, so the term peri-mortem is applied to possible ante-mortem damage, in which bone reaction (healing, inflammation) is absent.77 Fifteen skeletal elements were found to show clear signs of injury, and the majority were caused by pointed/sharp objects, rather than blunt-force trauma. Most wounds were found on adult males, and only males or adults of indeterminate sex displayed metal-bladed weapon injuries. Six fragments with weapon injuries were radiocarbon dated, and 4 originated from the Late Iron Age, and 2 from the Early Middle Ages.78 Interpretation of the site at Kessel/Lith is difficult due to limited systematic excavations. The Late Iron Age complex covers an area of about 2km x 0.5km and
contains two main contexts of material: (1) the Late Iron Age river channel, and (2) the nearby southern riverbank. Late Iron Age handmade pottery and animal bones were distributed over the entire site, and stone quern fragments, bone tools, loom weights, spindle whorls and La Tène glass bracelet fragments were also uncovered. The amounts of these artefacts do not necessarily suggest the density was uniform along the entire bank, but the settlement was unusually large for the time, and concentrations of finds seem to point to Kessel as the center.79
Several possible hypotheses regarding the use of the site have been put forward. The first was that the complex could be an eroded inhumation cemetery. But this has several problems because the complex was used to deposit all kinds of objects over centuries, and the large quantity and vast timeframe would require several cemeteries to be involved and this was unlikely. Also, most of the finds date to the Late Iron Age, and the dominate funerary practice was cremation, not inhumation. None of the bones displayed gnawing traces, which would be expected, if the remains had been deposited on an eroded settlement.80 The third interpretation was that Kessel was a community engaged in warfare. Although the majority of the bones seemed to be from adult males, the presence 75 Roymans, 2004: 132. 76 Ter Schegget, 1999: 215. 77 Ter Schegget, 1999: 219. 78 Ter Schegget, 1999: 219. 79 Roymans, 2004: 133. 80 Ter Schegget, 1999: 223.
24 | P a g e
of at least 10 women, several children and a number of individuals aged between 50 and 80, suggests the site was not purely associated with military activity. This is reinforced by the presence of pottery, animal bones, and many bent or broken weapons, which would not occur in combat.81 The last hypothesis proposed is that the site at Kessel/Lith was part of a regional cult place, based on the presence of coins, fibulae, weapons, pottery, and animal bones, which makes it similar to other pre-Roman cult places that have been excavated.82
Also noteworthy, is that the remains of a massive, monumental temple (see Fig. E) were discovered by amateur archaeologists in 1976, of which parts of the wall foundations (eroded by the Maas) lay under 4 meters of sediments in a dredge pond. In 1977, the State
Archaeological Service completed a small rescue
excavation, examining a 20m x 30m area under difficult conditions, and no further work has been carried out since. It is unfortunate that the remains now lie under half a hectare of rubble, which was dredged and dumped in the pond.83 Kessel is an important case for preserving river deposits because the bones and other objects provide two contexts: one of conflict, and one of local communal, and possibly supralocal cult use. Without recovery of these remains, the demographic and historical significance of the region, as depicted by Roymans and others, would be
less known.
2.3 Germany: Tollensetal (Tollense Valley)
The valley along the Tollense River near Mecklenburg, Western-Pomerania has been yielding skeletal material, especially cranial fragments, along a 2km stretch of the riverbank since the end of the 1970s. Yet, the importance of these finds was not highlighted until 1996, with the
discovery of a human humerus in which a flint arrowhead was deeply embedded.84 Weltzin 20 (see Fig. F) is the most heavily excavated find site, producing a large concentration
81 Ter Schegget, 1999: 223-224. Roymans, 2004: 133. 82 Ter Schegget, 1999: 224. Roymans, 2004: 134. 83 Roymans, 2004: 135.
84 Brinker et al., 2013: 131-132. Fig. E Fluted column, Kessel; Roymans, 2004: 136.
Fig. F Map of Tollense River sites; Brinker et al., 2013: 132.
25 | P a g e
of disarticulated horse bones and human remains, including a skull displaying peri-mortem trauma, and several wooden clubs, which were discovered by the Landesamt für Bodendenkmalpflege.85 The first radiocarbon dating revealed a date of 1250 cal BC for the site, and the research team plans to
continue assessments of pathology and trauma, as well as stabile isotope analysis to try to determine the origins of the human remains, as well as diet and other environmental factors.86 In addition to the land excavations, there was also systematic (see Fig. G) diving research in the known find spots along the river. This allowed the riverbed and its stratigraphy, where numerous assemblages of human remains were unearthed, to be further explored. The goal of these explorations, like the land excavations, was to clarify the origins of the remains, as well as study river taphonomic processes which contributed to the disbursement of the material.87 With assistance from various research and diving groups from Mecklenburg-Pomerania and Greifswald University, the first expedition took place in June 2008. The goal was to explore Weltzin 20, however due to livestock in the fields, the teams were forced to relocate to Weltzin 21 and 13, where they searched both banks and the riverbed for potential finds.88 Between these points, numerous human and animal bones were recovered, as well as a wooden post or stake made of oak, which was dated to 1300 BC. The second diving expedition took place in May 2009 between Weltzin 13 and 32, but unlike the first section, very few human or animal bones were recovered.89 The combination of land and diving excavations has been
fruitful; through diving, prospecting the entire find area is possible, and study of riverbed stratigraphy can contribute to the overall understanding of the levels of archaeological remains at the site. Future diving projects are planned to continue to search for evidence.90
As of March 2011, 2,900 human and about 900 animal bones, mostly
disarticulated, have been recovered from the Tollense river area, many from the Weltzin 20 site, and from the diving expeditions. The team used skulls and pelves for sex and age 85 Brinker et al., 2013: 132. 86 Brinker et al., 2013: 133. 87 Brinker/Krüger/Lübke, 2010: 42. 88 Brinker/Krüger/Lübke, 2010: 42. 89 Brinker/Krüger/Lübke, 2010: 43. 90 Brinker/Krüger/Lübke, 2010: 46-47.
Fig. G Weltzin 32, Eastern riverbank, disarticulated human remains; Brinker/Krüger/Lübke, 2010: 44.
26 | P a g e
determination, but they also analyzed bones for traces of pathology and trauma, as well as evidence for taphonomic processes of weathering, fluvial factors, or disbursement by scavengers.91 Like the Kessel and Thames/Walbrook sites, Brinker et al. also wanted to investigate whether the enormous amount of remains could have come from flooded burial grounds, whether there were visible traces of violence on the bones, if so, what types, and whether the bodies were deposited in articulated or disarticulated states.92
Results: Numbers Tell Part of a Story
Weltzin 20 yielded 2,642 human bones, including 25 crania, 28 mandibles, 36 skull fragments, 74 isolated teeth, 397 long bones, and 2,082 other post-cranial elements. The minimum number of individuals (MNI) was determined based on femura, 40 left and 37 right, of which 21 pair were identified, and in conjunction with the remaining 19 left and 16 right, the MNI calculated was 56.93 The composition of the assemblages consisted mainly of long bones, of which femura were most predominant, followed by crania, mandibles, and parts of the shoulder girdle, while smaller bones, such as hand and foot bones, and less dense bones (ribs and vertebrae), made up the smallest group. This is similar to other river and land sites, though the fact that these smaller bones still make up a significant percentage of the assemblage, and the overall preservation quality is high, the researchers must consider other explanations besides erosion for the amount of material in the river.94 Also similar to Kessel/Lith deposits, 24 individuals were determined to be males, with one possible male, and 5 were determined to be female, with one possible female, while juveniles made up less than 25 percent of the population. The male to female ratio and number of young individuals is much different than would be expected if this was a representation of the ‘normal’ demographic composition of a community; other sites along the Tollense also displayed similar trends.95
The types of injury discovered on many of the bones from Weltzin 20 were associated with interpersonal conflict and Bronze Age weapons (see Fig. H), such as arrowheads and wooden clubs. Many of these peri-mortem injuries are depicted on skulls and skull fragments, manifesting in circular depressed fractures caused by blunt-force trauma, and triangular or pointed defects, caused by sharp-force or projectile trauma. An adult right humerus displayed sharp-force trauma, and an
91 Brinker et al., 2013: 133. 92 Brinker et al., 2013: 134.
93 Brinker et al., 2013: 134-135. These data were calculated as of March 2011; work at the site is ongoing, therefore bone and MNI counts, as well as trauma and isotope analyses may differ from what is discussed in this paper.
94 Brinker et al., 2013: 135. 95 Brinker et al., 2013: 136, 138.
27 | P a g e
ray showed a flint arrowhead which was lodged in the bone. The ages of the victims were young adults, and included a 12-18-year-old probable female.96
Conclusions
Based on these initial analyses, the researchers concluded that as of yet, there was no evidence of disbursal of the bones by scavengers, nor were there traces of human manipulation, such as
dismemberment, as seen in other land sites, and the human remains must have entered the water as complete bodies.97 Like Kessel/Lith, the idea that many of the bones arrived in the river due to eroded cemeteries by high flood waters was discounted, and this was based on the demographic representation (mostly male, many fewer women, juveniles and older individuals), which is atypical
96 Brinker et al., 2013: 138-139. 97 Brinker et al., 2013: 143.
Fig. H Frontal injury by blunt-force trauma; Flint arrowhead embedded in humerus; Brinker et al., 2013: 136, 139.
Bronze implements from Tollensetal-Spearheads, arrowheads, pins, adze, sickle, Spindlersfeld type fibula; Jantzen et al., 2011: 420.
Wooden weapons from Tollensetal; Jantzen et al., 2011: 423.
28 | P a g e
for a local communal burial place.98 The idea of a massacre was discredited because the numbers and types of injuries reflect the use of Nahkampfwaffen (short-distance weapons), and there was no
standardized treatment of the victims; the likelier hypothesis based on the number of young adults and the presence of both healed and lethal trauma, is that the supralocal communities represented at the Tollense were engaged in armed conflict here and previously at other Bronze Age sites.99
2.4 England: Godwin Ridge, Great Ouse River, Cambridgeshire Fenlands
Landscapes in the Great Ouse Valley and Cambridgeshire fens have been subjects of archaeological and palaeoenvironmental investigations (carried out by the Cambridge Archaeological Unit and English Heritage) before extensive commercial sand and gravel extraction work was set to begin. This uncovered many prehistoric settlement, ceremonial and field system sites, some of which were located on the river or fen-edge and are still waterlogged, and thus may still be well preserved.100 The Great Ouse Valley consists of an alluviated floodplain, up to 4km wide which flows into fenland peat and marine deposits.101 It is this environment in which Godwin Ridge is located. (See Fig. 1, Appendices) Godwin Ridge is a Late Bronze Age settlement, which was reused as a
later Iron Age ritual complex. Activity areas of both periods are situated on a sand ridge, which at one time, lay mid-stream in the Great Ouse River. The Pleistocene ridge is 575m long, and nestled on the floodplain of the Ouse Valley between palaeochannels of the river.102 More than 70,000 artefacts, depicting a Mesolithic to Iron Age sequence, have been unearthed from an estimated 700 hand-dug meter-square test pits, suggesting that finds may be in excess of one million. The western edge, which reaches an elevation of 3m, is the site of the intense Mesolithic and Late Bronze Age activity, and it is this area at which the expansive Iron Age complex yielded amazing numbers of ‘votive offerings’ and evidence of riverside human-mortuary and bird-related ritual activities.103
98 Brinker et al., 2013: 143. 99 Brinker et al., 2013: 143-144. 100 French/Heathcote, 2003: 81. 101 French/Heathcote, 2003: 82. 102 Evans, 2013: 55. 103 Evans, 2013: 55.
29 | P a g e
Godwin Ridge is also located near the O’Connell Ridge, a larger, ridged terrace-island to the south. This site yielded four urn-barrows, one which preceded a beaker cemetery, and two pond barrows. Both sites contained Neolithic and Early Bronze Age artefacts, as well as Middle Bronze Age enclosures, however, O’Connell contained an extensive field system and a channel-side settlement. Habitation at O’Connell appears to have ceased due to rising marsh levels and was eventually abandoned, while Godwin saw continued occupation due to being situated on relatively higher sediment deposits.104 Late Bronze Age usage of Godwin Ridge was discovered in the sandy loam soils which covered the site.105 The LBA3 level contained 6,150 sherds of plainware pottery, disbursed over the site in densities of around 14 sherds per square meter, leading excavators to estimate that the site’s surface deposits contained more than 85,500 LBA sherds. In addition to the ceramic assemblage, the west-central area of the ridge contained traces of 3-5 contemporary roundhouses, plus post-holes and scattered pits.106
(See Fig. 2, Appendices) Around 2,800 sherds of
Middle to Late Iron Age pottery were also recovered, including 500 wheel-made specimens, of which 200 dated to the LIA, and the remaining seemed to correspond with the “Romanization”, including 13 pieces of first century AD Roman ware.107 The fact that the Iron Age features of the site contain only modest buildings, suggests the area did not function as a shrine, but rather as a point for mortuary and ritual activities. Permanent habitation is questionable, because in addition to limited development, palaeoenvironmental data showed that only around one hectare of the ridge top would have been accessible at the time, restricting sustenance.108
Ritual Activity
A platform (7m x 10m and 0.15m-0.30m thick) was found on the northern riverside area of the site. Here, the remains of four dismembered horses, disarticulated/partially disarticulated remains of a dog, two cows, a pig, and 12 sheep (in addition to large quantities of fish bone) were recovered. (See Fig. 3, Appendices) Bones of at least 15 different species of birds, some of which were butchered and
consumed, were also unearthed in the platform area. Also, 51 percent of the total Iron Age pottery assemblage (1,460 sherds) were also found near the platform, and more strikingly, this area also
104 Evans, 2013: 58.
105 Evans, 2013: 58. See French, Heathcote, 2003: 84-85 for further palaeoenvironmental data. 106 Evans, 2013: 58.
107 Evans, 2013: 61. 108 Evans, 2013: 61.
