Archaeology Group

• Archaeology Group
• About the Group
• Research Themes
• Landscape and Townscape Archaeology
• Durotriges Research Page and Dig Diary
• Stonehenge Riverside Project
• Medieval Novgorod
• The Palaeo-environment of medieval Novgorod and its Hinterland
• A Chronological Framework for early medieval Towns in NW Russia
• Craft Production, Environment and Landscape
• Billown Neolithic Landscape Project
• Knowlton Prehistoric Landscapes Project
• Wiggold Project Gloucestershire
• The Wessex Barrow Project
• Early Bronze Age Studies
• General Analysis of Barrows
• A Specific Analysis of Barrows
• Impact of Results
• Neolithic Flint mines of Sussex
• Piltdown Man
• Landscape Perception Project
• New Light on Stonehenge
• Environmental Archaeology
• Forensic Archaeology and Anthropology
• Cultural Resource Management
• Science-based Archaeology
• Technology, Production and Ancient Materials
• Maritime Archaeology
• Field Schools
• Bournemouth Archaeology
• Undergraduate programmes
• Postgraduate programmes
• PhD Studentships
• Staff

General Analysis of Barrows

In the following case study an alternative typological method was developed that takes into account both similarities and differences in the archaeological record to classify variation in Early Bronze Age burial mounds.  In order to examine the range of variation in burial practices of the Early Bronze Age, three studies were conducted on a collection of barrows from Wiltshire and Dorset. 

The first study aims to establish the predominant or general burial practices of Early Bronze Age Wessex.  Since opposing groups often use similar material culture, a typological method that classifies artefacts, practices and structural traits according to their form will inevitably fail to notice any differences.  However, since material culture is used in different ways by opposing worldviews, this distinction in practice could be a way to identify differences between groups.  By sorting the mounds by context – the manifestations of use – it was believed a much more reliable taxonomy could be produced.

The second study examines the anomalous contexts that do not fit any of the general types.  By investigating the general contexts of these anomalies, a possible interpretation for these is suggested.   The third study then examines the particular sequence of construction for one of these anomalous contexts to test this hypothesis.
                                                                                 
As part of the ongoing Wessex Barrow Project at Bournemouth University, excavation reports of all barrows excavated in Wiltshire and part of Dorset up to 1960 have been compiled and records created for them.  The aim of this project was to make these records available on an online database with links to over a thousand site plans and artefact images.  The database is now available (Martin 2007) and contains records for over eight hundred mounds.   As more excavations are entered (from the 1960 onwards) the site will be updated and a complete study carried out.

Relevant Dimensions of Variation

Before any analysis could be conducted, a better system than arbitrary selection had to be found to select the relevant elements for catagorisation of the mounds (Hodder 1991:139).  These elements needed to be relevant to each predominant burial practice, but also be the dimensions over which each differed.  In order to prevent these dimensions from being arbitrarily selected, the relevant dimensions were chosen by identifying the particular context types that varied the least in usage.   This was based on the assumption that the most significant elements of worldviews would be less changeable than elements that were less significant (Latour 1987:208).

In order to find the context types that varied the least, it was necessary to calculate the frequency of correlations between every trait and every other trait, regardless of its apparent significance. Five years ago such a comparison would have been fairly difficult, but recent information technology, now available, allows us to process enormous amounts of information in contextual ways.   Modes 1.99, a cataloguing system used by museums, is perfect for this because it preserves detailed artefact and structural context data, and enables that data to be indexed.  By comparing the indexes of indexed traits (one at a time) a list of the context types that varied the least was compiled.  Due to the lack of space, only four of the least variable types are included here:

1. The type of burial within primary graves (Inhumation or Cremation)
2. The type of burial within secondary graves (Inhumation or Cremation)
3. The mound structure (Bowl, Bell, Disc or Saucer Barrow)
4. The pottery types (Beaker, Incense, Collared Urn and Food Vessel)

‘Cenotaph’ mounds without burials were excluded from this analysis, as were bowl barrows with cremations that had no Early Bronze Age artefacts (since these could have been from the Middle or Late Bronze Age).  While there have been several classifications of barrow types - eight by Colt Hoare (1810) and fourteen by Grinsell (1957) – I have taken Ashbee’s (1960:24) lead and classified Bell and Disc barrows together as ‘Wessex’ barrows, also including the Saucer barrows due to their similarity in structure and deposits (Ozanne 1972:53).

The First Study

The first study attempted to identify the predominant burial practices from the collection of mounds studied.  This was achieved by identifying the common combinations of the above context types.  Modes 1.99 was also invaluable for this.  Modes uses a form of Natural Language Processing software similar to a Google browser that allows subsets of similar context combinations to be formed using an advanced search feature.

From these correlations, three groups appeared to stand out.  The first group related to Inhumation burials and Inhumations with Beaker burials, which correlated predominantly with Bowl barrows.  These ‘Beaker’ barrows occurred roughly from 1800 to 1450 bc (Burgess 1986).

The second group related to Cremation burials in Bell, Disc or Saucer barrows and other unusual barrows.  These ‘Wessex Culture’ barrows occurred roughly from 1600 to 1450 bc (ibid).

The third group related to Cremation burials and Cremation burials with E/MBA Urns in bowl barrows.  This group occurred roughly from 1450 to 1250 bc (ibid) and can be related to a later development of the first two groups.

Table 1: Analysis of the range of Early Bronze Age burial practices

So far, these correlations are commensurate with Grinsell’s and Ashbee’s divisions of the Early Bronze Age and earlier interpretations (Piggott 1938; Childe 1956).  Their understanding of this period was that a dynamically different burial practice emerged during the Late Beaker period.  The practice of cremation, a new repertoire of artefacts skilfully crafted in gold, bronze and amber, and very different types of barrow appeared together roughly at the same time.  This new burial practice also corresponded with the abandonment of Stonehenge. 

Yet the exact relationship between the ‘Wessex culture’ and the ‘Beaker culture’ is still poorly understood.  This is because archaeologists have had great difficulty in distinguishing the characterising differences between the Wessex Culture and the Beaker Culture.  Although radically different in many ways, the particular ways they differ sometimes change from cemetery to cemetery.  This means that any general analysis that has tried to pin down their characteristics has always run into a number of exceptions. 
           
Largely on the basis of these anomalies, most researchers since the 1960s have rejected the classification of barrows into groups in favour of studying general elements that all barrows have in common, such as topographical location.

The Second Study: The Context of Anomalies

The first study highlighted two sets of anomalies, specifically 13 Wessex barrows that had primary inhumations and 13 Wessex barrows with secondary inhumations.  According to the above classification of Wessex barrows these barrows should only have had cremation burials.  The other exception – the presence of primary and secondary cremations in Bowl Barrows – is, of course, to be expected if the practice of cremation in bowl barrows followed the practice of cremation in Wessex barrows. 

Before dismissing the predominant groupings completely on the basis of these anomalies, it is worth investigating whether there are any local explanations for the anomalies.  Thus the second study examines the context or background behind each anomaly in order to find any general correlations between them.

An examination of the contexts of the first set of anomalies revealed that 11 of the 13 anomalies occur in cemeteries that have both Wessex and Beaker Barrows (see table 2).  This combination of Wessex and Beaker barrows only occurs in 17 of the 55 cemeteries studied (cemeteries that contain over four mounds) indicating fairly exceptional circumstances for this first set of anomalous mounds.

Table 2: Analysis of the first set of anomalies
                           
The second set of anomalies (13 secondary inhumations above primary cremations in Wessex barrows) has suggested to many prehistorians that cremation and inhumation was practiced simultaneously by communities (Bradley 1984:84; Woodward 2000:23).  However, these anomalies also appear to have occurred in exceptional circumstances.  An examination of the contexts of these anomalies revealed that out of those 13 secondary inhumations, 10 occur in cemeteries that have both Wessex and Beaker barrows (see table 3).  Four of these also occur in the same cemeteries as the first set of anomalies.  This brings the total number of cemeteries with anomalous mounds that have both Wessex and Beaker barrows to 13 out of the 17 cemeteries (or 76%). 

Table 3: Analysis of the second set of anomalies

On the basis of these calculations it is clear that something strange was going on at the interface between the two burial practices.  While Wessex culture mounds and Beaker Culture mounds normally subscribed to stringently different burial practices, where they came together in the same cemeteries there was a distinct fusion of character traits. 

A similar correlation between anomalous forms and interactions is noticed by Jones (1997:95) and Chamberlin (2006:44).  They argue that when fundamentally different ideas come together, there is a rapproachment between them producing new hybrid traditions (see Barth 1971:184). 

This phenomenon is also noticed by Bruno Latour (1987:208) who adds that interactions also result in some of the most explicit and graphic expressions of knowledge.  Latour states that when confronted with a threatening difference in opinion, both sides “end up mobilizing the most heterogeneous and distant elements, thus mapping for themselves, for their opponents, and for the observers, what they value most, what they are most dearly attached to” (Latour, 1987: 205).  This may also explain why the anomalies occur in the richest barrow cemeteries in Wiltshire and Dorset, including the famous Normanton Down, Winterbourne Stoke and Oakley Down cemeteries.

These coincidences offer a good reason to re-evaluate the position taken to disregard any differences between burial mounds, and to look once again at the possibility that different groups existed within the Wessex area, albeit sub-cultural ones.  In order to examine the above explanations of anomalies – whether they represent transformations due to interactions and produced any graphic expressions – the third study examines the cemetery context of one of the anomalies.  This was the Net Down barrow cemetery in Shrewton, Wiltshire.

Excerpt from Martin, A. 2008 ‘The Alien Within: the forgotten subcultures of Early Bronze Age Wessex’ to be published in Jones, A. and G. Kirkham (ed.s) Beyond the Core: reflections on regionality in prehistory, Oxford: Oxbow Books.

<< Previous       Next >> An individual cemetery study

• Search for a Course
• Register for Open Days, Events & Exhibitions
• Apply to BU
• Fees & Funding
• Virtual Tour