BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND: CONSTRUCTION, FORMS AND INTERPRETATIONS

Transcription

1 ojoa_ RICHARD BRADLEY AND ELISE FRASER BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND: CONSTRUCTION, FORMS AND INTERPRETATIONS Summary. The Bronze Age barrows on the downs of southern England have been investigated and discussed for nearly 200 years, but much less attention has been paid to similar structures in the areas of heathland beyond the chalk and river gravels. They were built in a phase of expansion towards the end of the Early Bronze Age, and more were constructed during the Middle Bronze Age. They have a number of distinctive characteristics. This paper considers the interpretation of these monuments and their wider significance in relation to the pattern of settlement. It also discusses the origins of field systems in lowland England. crichel down, launceston down and beaulieu heath A large number of prehistoric burial mounds were excavated during the Second World War as their sites were taken over for use by the armed forces. Among them were the barrows on Crichel Down and Launceston Down on the Wessex chalk, and those on Beaulieu Heath in the New Forest. Both excavations were conducted by the same people. Stuart Piggott and Margaret Piggott worked together at Crichel Down, where the sites of a series of barrows were to be used as a bombing range (Piggott and Piggott 1944). Margaret Piggott also worked at Beaulieu Heath where the mounds were removed to make way for a military airfield (Piggott 1943). Again Stuart Piggott contributed to the excavation report. These projects were important as it was rare for so many barrows to be excavated in a single location, or for them to be explored by the same techniques. Of course the circumstances of the excavations imposed some constraints. At Crichel and Launceston Downs the work focused on 16 of the smaller Bronze Age barrows so that more monuments could be explored in the time available. When the site reverted to farmland in 1958, two larger mounds (Long Crichel Barrows 5 and 7) were completely excavated (Green, Lynch and White 1982). At Beaulieu Heath there were fewer limitations, but parts of the ten barrows were only sampled (Fig. 1). Work at Long Crichel Barrows 5 and 7 was more extensive than the earlier research, and when it was published the authors of the report were able to draw on the results of a number of projects that had taken place during the intervening years. As a result they paid more attention to the stages by which the structures were built. They also discussed the reuse of older graves. In doing so, they rejected the traditional view that the barrows were intended to house individual burials. Instead each could be considered as a cemetery in its own right. 29(1) Journal compilation, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street Malden, MA 02148, USA. 15

2 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND Figure 1 Plans and profiles of two chalkland round barrows (Long Crichel 5 and 7) and two heathland barrows (Beaulieu 5 and 6). The mounds at Long Crichel were rebuilt and were associated with a sequence of intercutting graves. Those at Beaulieu illustrate a widespread structural sequence with a turf core and a gravel capping. Information from Piggott (1943) and Green, Lynch and White (1982). 16 Journal compilation

3 RICHARD BRADLEY AND ELISE FRASER There have not been many projects of similar character in recent years as most excavations have investigated single monuments or have been concerned with barrows that had already been levelled by the plough. For that reason these particular sites retain a special importance. They are often mentioned in accounts of prehistoric burials, but, considering that the projects were conducted by the same people, the striking differences between their results have hardly been discussed. At a time when the character of Bronze Age burials is being reconsidered (Brück 2004; Last 2007) the problem needs to be investigated. The graves and their contents The more recent report on sites at Crichel Down draws attention to a feature that was mentioned only briefly in the Piggotts paper (Green, Lynch and White 1982). Many of the Beaker and Early Bronze Age graves had been reopened after an interval and often contained more than one body. Some of the burials seem to have been disturbed and not all the skeletons were complete; one of the primary graves included a small number of disarticulated bones belonging to another person. It seems possible that relics were removed from the original deposits and that others were introduced from outside. Apart from the containers that held cremated bones, there were not many grave goods at Crichel and Launceston Downs. Long Crichel Barrow 5 was associated with one anomalous deposit. Large fragments of a single decorated vessel were distributed between two different graves. The remainder of the pot was missing. Either it had been brought to the barrow in pieces, or parts were removed during the Bronze Age. One of the cremation burials from Barrow 7 posed a similar problem. It was inside two urns, one inverted over the other. The outer vessel was broken and may have been damaged subsequently, yet it was clear that its companion was incomplete when it came to the site. It is tempting to suggest that such items were heirlooms (Woodward 2002; Woodward et al. 2005). It seems likely that the missing fragments continued to circulate among the living. Both artefacts and human remains may have been treated in the same ways, and even those buried in graves could later be removed. There was little evidence of the same process at Beaulieu Heath where most of the barrows contained only one burial and almost all the associated artefacts were pots. There was nothing to suggest that any of the graves had been reopened. An amber necklace was found beneath one of the monuments. Since it was not complete, Margaret Piggott made the prescient suggestion that it was probably an heirloom (1943, 14). The mounds and their materials For the reasons mentioned earlier, the Piggotts excavations were limited to the smaller barrows on Crichel and Launceston Downs. That introduced a certain bias, but it was also revealing, for nearly all the structures they investigated belonged to just two phases: the Beaker period and the Middle Bronze Age. The Beaker burials could be associated with shaft graves, yet the covering mounds were inconspicuous. The Middle Bronze Age round barrows were built on a similar scale, but in this case they were associated with small pits containing cremations (Piggott and Piggott 1944). On the other hand, two of the Early Bronze Age monuments, Long Crichel Barrows 5 and 7, were quite distinctive (Green, Lynch and White 1982). Each was a prominent feature of the landscape and provided evidence of several phases of enlargement and modification. Barrow Journal compilation 17

4 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND 7 may have begun as a small mound in the Beaker phase, but it ended as a considerable earthwork, 3 m high and nearly 30 m in diameter. Although it contained the remains of several people, it is clear that not every funeral resulted in the enlargement of the mound (Barrett 1991; Barrett 1994, chapter 5). Again there is a significant contrast between the sites excavated by the Piggotts. Individual barrows on Beaulieu Heath could be as large as those on Crichel Down, but fieldwork suggested a much simpler sequence of development. There was little to suggest that the more conspicuous monuments were created by augmenting older structures, and only one of the ten mounds had been enlarged. The structural sequence was equally simple, for in most cases an inner core of turf was covered by a layer of gravel (Piggott 1943). The barrows on Beaulieu Heath are distinctive because they provide such limited evidence for cremation burials: the dominant rite on Crichel Down at the time when they were built. On the other hand, two of the mounds covered timber structures which could have housed a corpse (Piggott 1943, 6 9, and 24 5). Since these sites are located on acid soils it is possible that certain of them were associated with inhumations or with deposits of unburnt bones. Chronology There are no radiocarbon dates from either group of barrows but the associated artefacts are of types with an established chronology. The mounds can also be dated by comparison with a scheme recently published by Garwood (2007). Both methods provide similar results. Garwood distinguishes between three different phases: In the first, between about 2400 and 2100 BC, inhumation burials were often placed in deep graves, some of which were later reopened. Not all were marked by mounds, and where barrows were built they were usually small and inconspicuous. (Later Neolithic/Chalcolithic) The second phase took place between approximately 2100 and 1850 BC. It was characterized by larger, more complex monuments, some of which were built over the positions of flat cemeteries or enclosures. These barrows were considerably larger than their predecessors and were often reconstructed. Many of the graves were reopened to include inhumation and cremation burials. Any one mound might contain the remains of a large number of people. (earlier Early Bronze Age) That contrasts with developments in Garwood s final phase, which extends from 1850 to 1500 BC. Now newly built barrows were constructed in a single operation and were frequently associated with just one primary burial, usually a cremation. Sometimes they were erected over the sites of pyres. Later deposits might be inserted into the tops of the mounds, but the monuments were rarely enlarged. Some of the mounds took the specialized forms known as bell barrows and disc barrows. (later Early Bronze Age) These three phases span the period from the introduction of Beakers to the end of the Early Bronze Age. Middle Bronze Age barrows were not included in Garwood s analysis, but their characteristics are easy to summarize. They consist of smaller circular mounds, a few of which were on sites with evidence of burning. They were apparently built in a single phase and might cover one or more cremations. More can be set in the edge and top of the mound, and others occur in shallow pits beyond the barrow altogether (Ellison 1980; Petersen 1981, chapter 11). After about 1200 BC round barrows of any kind are rare in southern England. 18 Journal compilation

5 RICHARD BRADLEY AND ELISE FRASER The periods of use of these different structures may have overlapped, but it is easy to locate the cemeteries excavated by the Piggotts within this general sequence. The mounds at Crichel Down extend through all four phases, beginning with Beaker shaft graves covered by low mounds and ending with a series of small Middle Bronze Age barrows. Those on Beaulieu Heath were confined to the last two stages in this sequence, meaning that the cemetery should have started later than its counterpart on the chalk. It lasted for a shorter time, although both groups of monuments may have gone out of use together. the wider context: the early bronze age The changing significance of the past It is a sequence that spans an important development in the archaeology of central southern England. Round barrows were built continuously between the first introduction of metals and the middle of the Bronze Age, but in other respects the landscape underwent a change as the influence of older ceremonial centres decreased. Many of the first round barrows were built near older structures, especially long barrows, enclosures, cursuses and henges dating from the Neolithic period. They are a special feature of the chalk and the river gravels and do not occur in areas that are currently occupied by heathland. The positions of the Bronze Age barrows not only acknowledged the existence of important structures dating from the past, new research has shown that those on the chalk of south Dorset, Cranborne Chase and Salisbury Plain were located in areas of fertile land that had never been densely wooded. These regions attracted an unusual concentration of earlier prehistoric monuments, most of which would have been built in places that had already been settled (French et al. 2007, chapter 4). The character of these monuments changed over time. The process is epitomized by the most famous example, Stonehenge. For a long time it was thought that the major stone settings on the site were contemporary with the Bronze Age barrows that cluster around it. Now that seems unlikely. Instead the burial mounds appear to have been built close to what was certainly an important place, but one which was already becoming an ancient monument (Parker Pearson et al. 2006). The same applies to other henges in Wessex, which were not rebuilt after the Beaker period. The monuments on Crichel Down and Launceston Down illustrate the complexity of the evidence. They commanded a view over the Dorset Cursus once the largest monument of its kind in Britain (Barrett, Bradley and Green 1990, 35 58). There had been a Neolithic burial on Crichel Down itself (Piggott and Piggott 1944, 51 2), but excavation suggests that the area occupied by the mounds contained a settlement associated with Beaker pottery (Piggott and Piggott 1944, 56; Green, Lynch and White 1982, 48 50). Perhaps the recent history of the site became more important in the Early Bronze Age. There is yet another way in which the political geography of southern England was changing. For too long the discussion has been based on the main concentrations of round barrows on the downland, in particular those on Salisbury Plain. New research suggests that greater attention should be paid to the English Channel coast where there was a series of exceptionally rich graves and other deposits reaching from Kent to Cornwall. Their contents provide evidence of long-distance contacts extending around the shoreline and into Continental Europe (Needham 2006). Needham and Woodward suggest that communities near to the sea may have maintained a certain independence from those who buried their dead in inland areas (2008, Journal compilation 19

6 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND 41 3). This distinction can be recognized as early as 1950 BC but continued throughout the period studied in this paper. Needham argues that these long-distance contacts provide evidence of a distinct Channel Bronze Age (Needham 2006). The case is certainly persuasive as there are important concentrations of surface finds at the mouths of the main rivers in the Solent Basin (Field 2008a). Barrows and the occupation of new land The nature and distribution of Bronze Age barrows changed while such developments were taking place. The excavated monuments at Crichel Down and Beaulieu Heath typify these wider developments. From the Beaker period onwards round barrows were established on the downs and river gravels across large parts of southern England. Their sizes varied considerably and so did their distributions on the ground. Some of the mounds occur singly, but others were built in groups or laid out in more elaborate configurations. The barrows in both these regions were constructed for more than 1000 years. Similar monuments occur in limited numbers in areas of sand or plateau gravel which are now occupied by heathland (Dimbleby 1962). They include parts of the Solent Basin, the Weald, the Surrey Greensand and the hinterland of the Thames and Kennet Valleys (Fig. 2). Although the density of monuments varies, the regions with round barrows extend across approximately 2000 sq km between the Solent and the River Thames. They are adjacent to areas in which round barrows were built in larger numbers. They were also close to places with a longer history of settlement. Most fieldwork has taken place in the south of Wessex, and it is that region which is studied in most detail here. The heathlands of southern England assumed their present form at different times, but the oldest monuments there were constructed in the Bronze Age (Dimbleby 1962; Branch and Green 2004, 13 15; Grant and Barber 2008). Apart from Mesolithic artefacts, they produce few surface finds dating from earlier periods, and those that have been recorded rarely extend back before the adoption of metals (Gardiner 1984; Needham 1987; Cotton 2004, 24 7 and 33 5; Field 2008a). None of the barrows contained a Beaker burial. The first diagnostic artefacts associated with them belong to the later part of the Early Bronze Age. Some of the smaller monuments were used again during the Middle Bronze Age when other examples were built, but few of these areas were occupied after that time. Pollen analysis shows that they had been used for cultivation and for pasturing animals (Bradley and Keith-Lucas 1975). It seems as if the heathland barrows developed some time after those in other areas. The monuments on Beaulieu Heath are typical of a much wider pattern which has been traced by piecemeal excavation and field survey in Dorset, Hampshire, Sussex, Surrey and in Berkshire south of the Thames. The mounds often form small groups but larger cemeteries are rare or absent. In certain respects this evidence is unusual. The excavated mounds may be contemporary with those in other regions, but the burials associated with them contained few, if any artefacts. The expansion of settlement seems to have been relatively short lived, and the surface beneath the round barrows is normally podsolized. People may have cleared fresh areas of marginal land as the quality of the soil deteriorated. An alternative was retrenchment, and it may have been in this context that the first co-axial field systems were created, for in central southern England they have a restricted distribution on the lower ground of the Solent Basin where they are associated with deposits of gravel (Yates 2007, chapter 7). The clearest evidence comes from 20 Journal compilation

7 RICHARD BRADLEY AND ELISE FRASER Figure 2 The study area, showing the main sites and regions considered in this paper in relation to the distribution of surviving round barrows. Heathland mounds are indicated in black. Information from Grinsell (1941). fieldwork at Bestwall Quarry, Wareham where the field ditches were associated with sherds of biconical urn (information from Lilian Ladle and Ann Woodward). That would make them contemporary with the barrows constructed on the chalk and the heathland towards the end of the Early Bronze Age. It is possible that early fields existed on the chalk at this time, but none of the evidence is entirely satisfactory. the character of early bronze age round barrows Structural sequences at the heathland barrows Most of the heathland barrows were built in a single phase (Fig. 3). The process was usually the same. An extensive area was stripped of turf, although the position of the monument Journal compilation 21

8 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND Figure 3 (Upper) The surviving remains of Beaulieu Heath Barrow 6, parts of which escaped destruction in the 1940s. The ditch is filled with water. Photograph: Richard Bradley. (Lower) A heathland round barrow on Mortimer Common. Photograph: Elise Fraser. For a plan of the cemetery, see Figure 4. was not affected, and a mound of sods was built. This would have been separated from its surroundings by an area of exposed subsoil. Then the barrow was enlarged by enclosing it with a ditch and capping it with gravel or sand. Occasionally this process was followed more than once. It would have altered the whole appearance of the monument, and the barrows would have changed their colour from black to yellow or white (Fraser 2005). That sequence has been recognized on most excavated sites, but one question is never asked. Why was the turf deposited separately from the bedrock? It rarely happened in regions where older mounds were constructed. A typical heathland mound began as a structure that appeared dark and potentially fertile, enclosed by an open area that would have looked like a cultivated field. It was transformed into something bleached and inert. At the same time its composition was altered. The initial turf stack was composed of an organic deposit which could have continued living for some time. When the ditch was dug, the original mound was covered over and the turf of which it was made would rapidly die. On the evidence of pollen analysis at Ascot in Berkshire it seems possible that 22 Journal compilation

9 RICHARD BRADLEY AND ELISE FRASER these stages were separated by a year or more (Bradley and Keith-Lucas 1975, 104). The progression is surely revealing. When the dead were buried, they were covered by a deposit of living matter. After an interval which was possibly quite brief, that process came to an end and any burial was cut off from the world outside. Where an unburnt body was buried, the stages of construction may have happened as the corpse decayed. Now the turf beneath the mound could no longer grow, and as this happened the living were separated from the dead by a ditch. This sequence is found throughout the heathlands. Heathland barrows were built on newly opened ground and seem to have been located some distance away from areas of long-established settlement, and even from the monuments built by earlier generations. Few of these earthworks were on sites with many artefacts, and just as they were established in areas that had not been inhabited before, the mounds themselves were sealed after they had been built. The dead were kept apart from the living to an extent that had rarely happened during previous phases, and the newly built structure created an impermeable barrier between the corpse and the mourners. Few of these structures were associated with grave goods. One possibility is that they were associated with the poorer members of the community, but that is improbable since the building of the covering mounds would have involved the same investment of labour as in many of those on the chalk. It is more likely that in places which lacked a strong connection to the past, special objects were less often buried with the dead (cf. Woodward 2002 and Woodward et al. 2005). Instead they might have been used as offerings in other locations. It may be no coincidence that it was at just this time that bronze daggers were first deposited in rivers (Bradley 1998a, ). Boundaries between the living and the dead A few heathland barrows took specialized forms. Among them were bell barrows where a circular mound was separated from a ditch by a strip of unexcavated ground (Grinsell 1934). It seems likely that many examples were built in a single operation. Disc barrows are also found on the heathland of southern England, but less often. They were circular ditched enclosures with one or two low mounds at the centre (Grinsell 1974). It is hard to know how to interpret them. Grinsell suggested that the differences reflected the burials found there bell barrows were for men and disc barrows for women but the field evidence was not convincing, and a significant proportion of the sites combined the attributes of both kinds of monument (Grinsell 1974, 86 7; Petersen 1981, ). Perhaps it is time to abandon a typology of barrows which does not accommodate the evidence from the heathlands. It is more informative to emphasize the common links between the mounds that were built towards the end of the Early Bronze Age. Several characteristics call for comment. These mounds illustrate a quite specific sequence in which an organic core of turves was sealed beneath a deposit of inert material. Once those barrows were built they were rarely modified. Their distinctive character was emphasized by the very features that make them so difficult to classify: the strips of open ground between a mound and its ditch; the presence of circular enclosures associated with some of these structures; and even the multiplication of earthwork barriers defining the limits of these sites. For example, a number of bell barrows also possess the external bank characteristic of quite different forms of monument (Fig. 4). As a result the dead were no longer accessible. A series of physical barriers was erected to keep them at a distance. Journal compilation 23

10 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND Figure 4 Outline plans of the heathland barrow groups on Mortimer Common, Berkshire, and Elstead, Surrey, showing the unusual configuration of the mounds. Ditches are indicated in light tone. Information from Grinsell (1932 and 1939). 24 Journal compilation

11 RICHARD BRADLEY AND ELISE FRASER Structural sequences in the chalkland barrows Paul Garwood s scheme defines three phases in the use of barrows on the chalk and, indeed, on the river gravels (Garwood 2007). The early stages date from the period before the heathland barrows were built, but those constructed between about 1850 and 1500 BC are particularly relevant to this study. They do not need to be described in detail, for in most accounts of this period they are treated as the norm. It is important to distinguish between barrows of the proportions of those in heathland areas and significantly larger mounds which have a more restricted distribution on the chalk. Similarly, it is necessary to consider the mounds in major cemeteries separately from the rest. Not only were those monuments laid out in formal patterns that are less apparent in other areas, they often include more specialized kinds of earthwork. The principal cemeteries contain most of the more conspicuous mounds. They also include a higher proportion of richly furnished burials, although there is no obvious correlation between the character of the individual barrows and the artefacts that were deposited in them. It seems as if the deployment of certain objects was especially appropriate in these places. That could be because the cemeteries had a lengthy history, in some cases extending back to the Beaker phase and beyond. Particular artefacts had a history, too. There is evidence that they had been worn, broken and even repaired long before they entered the grave (Woodward 2002; Woodward et al. 2005). It had happened during Garwood s first two phases, but the practice continued after the widespread adoption of cremation. Other offerings may have been destroyed on the pyre. It was in Garwood s third phase (the later Early Bronze Age) that monuments assumed new forms. In some respects they ran in parallel with developments on the heathlands. More of the mounds were constructed in a single operation, and they were generally associated with only one burial. They followed a comparable sequence to those in other areas, with a core of turves covered by a mantle of chalk quarried from a ditch. Again this could happen more than once. The most conspicuous mounds were occasionally organized in cemeteries aligned on the position of an older monument or even on the setting sun (Garwood 2003, 60 1). Burial mounds had been built in groups before, but it was only at this stage that there are indications of a more formal layout. Even in their final configuration, barrow cemeteries were usually divided between clusters of different earthworks (Exon et al. 2000, chapter 8). Boundaries between the living and the dead On one level the barrows just described acted in a similar fashion to those built on the heathlands, but there were important variations among the monuments established on the chalk. Mounds like Long Crichel 5 and 7 are of roughly the same proportions as barrows south of the chalk, but that would not apply to those in a major cemetery like Oakley Down, which was only 8 km away (Royal Commission on the Historical Monuments of England 1975, 102 4). The distinction is apparent in other ways. In complexes of this kind more artefacts were associated with the dead, and these groups also included a significant proportion of bell barrows and disc barrows where the limits of individual monuments were marked in special ways. If access to these cemeteries was restricted to certain sections of society, the forms taken by these mounds could have reinforced that exclusiveness. The distinction is apparent from the distribution of a number of elaborate artefacts, but it also depended on enjoying privileged access to the past. Journal compilation 25

12 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND the wider context: the middle bronze age The distribution of field systems The expansion onto heathland soils seems to have lasted to the end of the Middle Bronze Age, although the distribution of dated barrows suggests that some of the first areas to be cleared were abandoned in favour of others. It was now that larger areas were enclosed. Field systems on the gravels and the chalk were associated with circular houses, pits and water-holes. A few sites were bounded by a bank and ditch (Barrett 1994, ). Burial mounds of the same date can often be found in the vicinity. By the end of this phase field systems had extended into many of the areas that had been occupied by downland barrow cemeteries, but now those monuments were little used. Elsewhere existing mounds were incorporated in land divisions. Their importance was respected, but most had gone out of use (Field 2008b). the character of middle bronze age round barrows (Fig. 5) Structural sequences at the heathland barrows Some of the heathland barrows were reused in the Middle Bronze Age, but they were rarely reconstructed. Instead cremations were deposited in pits in the tops of existing mounds, around their edges and even in their ditches. At the same time the distribution of the burials extended well beyond the perimeter of the monument itself (Ellison 1980; Petersen 1981, chapter 11). More often new barrows were established. Most were smaller than their predecessors and some of them were built on new sites. Although they often illustrate the same structural sequence, with a turf mound overlaid by a deposit of sand or gravel, there were important differences between them and their predecessors. There were fewer layers of capping, and normally only one (Petersen 1981, 198). Where Early Bronze Age barrows had been demarcated by a continuous boundary, the new construction was provided with an entrance where the ditch was interrupted by a causeway; this is a particular characteristic of sites in the Solent Basin. The later monuments were no longer separated from the surrounding area by a band of unexcavated ground or by the construction of an outer bank. Instead they were simpler and less conspicuous. These mounds occur singly or in groups, but there is only limited evidence of a formal layout among the barrows themselves or in the organization of the burials at a single monument. There might be one or more cremations below the centre of the mound, but they might also be found in its upper levels. Others occur around its limits or outside it altogether. The number of cremations varies from a single deposit to a series of different burials. Within the larger cemeteries they can sometimes be divided into clusters (Ellison 1980). Some were in pottery vessels whilst others lacked any associations and were in small pits. At most sites there were no grave goods, perhaps because the offerings provided by the mourners had been deposited somewhere else (Bradley 2007, 200 4). The relationship between the living and the dead Of course there was some continuity between the heathland barrows of the Early and Middle Bronze Ages, and it is possible that for a while their currencies overlapped. Even 26 Journal compilation

13 RICHARD BRADLEY AND ELISE FRASER Figure 5 Outline plans of the Middle Bronze Age barrows at Crichel Down, Knighton Heath and Simons Ground, Dorset, and Barrow Pleck, Wiltshire, indicating the distributions of cremation burials and related features. Information from Piggott and Piggott (1944), Petersen (1981), White (1982) and Barrett, Bradley and Green (1990). so, several contrasts are important. The newer monuments were smaller, the associated burials were more plentiful, and many were not covered by the mound. Most important of all, the forms of these structures interposed less of a barrier between the living and the dead. Journal compilation 27

14 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND Some of the later barrows were associated with many more burials than those of the previous phase. Moreover the mound itself might provide the focus for cremation deposits, but only rarely did it seal them off entirely. In that sense the monuments were permeable. Fewer people were buried beneath these barrows than in its surface or outside them altogether. The provision of an entrance is particularly revealing, for it has very different connotations from the elaborate barriers constructed around Early Bronze Age mounds. It suggests that these places were intended to appear open and accessible. The forms of these monuments imply a change in the relationship between the living and the dead. So do the details of the burials themselves. In contrast to Early Bronze Age practice, most of the vessels that contained cremations were upright in the ground and could have been reopened. In the previous phase more had been inverted. Structural sequences at the chalkland barrows In many respects developments in the Middle Bronze Age took a similar course to those on the heathlands. Older round barrows were reused and smaller ones were built. They usually took the form already described. They appear singly or in groups, and in the southern part of Wessex they were often defined by a causewayed ditch which was often open towards the south. Again the burials were in pits beneath the mound; others might be cut into its surface; and some were around its outer edge or beyond the structure altogether. There is little evidence that these monuments were constructed in the same way as their immediate predecessors, with an inner core of turves and an outer capping of chalk. A feature of the chalkland barrows was the deposition of large quantities of worked flint in their ditches (Barrett, Bradley and Green 1990, ). There was a change in the siting of Middle Bronze Age burials. Where existing mounds were reused, they were usually the smaller barrows of the previous period rather than the more elaborate bell barrows. When they did occur in close proximity to Early Bronze Age cemeteries the later barrows were sometimes located towards their limits. A number of these barrows were organized into groups in their own right. They show important contrasts in the sizes of different mounds, in the character of the deposits, and even in their positions in relation to the monuments (Barrett, Bradley and Green 1990, ). The relationship between the living and the dead Again the chalkland barrows share many features with their counterparts on the heathlands. They seem to have been open to the surrounding area and many of the cremation deposits were outside the monuments altogether. Those associated with pottery might have remained accessible as they were in upright vessels. Another development was of more significance, for on the Wessex chalk these monuments were closely associated with settlements and field systems: the first to leave clear traces on the ground. It is possible that the same was true in the heathlands where the earthworks of fields have also been identified (Smith 1999; English 2007). The downland barrows were normally located within 200 m of the settlements of the same period (Bradley 1981). Other connections were less direct. It seems possible that barrows with entrances opening towards the south were constructed in the image of the round houses found nearby, for they were about the same size as these buildings. Where the mounds were associated with a group of burials, they may have signified the importance of the household (Bradley 1998b, 28 Journal compilation

15 RICHARD BRADLEY AND ELISE FRASER 150 8). The same link is emphasized by the use of domestic pottery in the cemeteries and also by the presence of large amounts of worked flint like that associated with settlements. The link is so widespread that it has been possible to discover previously unrecorded cemeteries by investigating the surroundings of already known occupation sites (Bradley 1981). summary round barrows and the prehistoric landscape Table 1 summarizes a series of similarities and contrasts. This paper ends by suggesting why they were significant. The first question to consider is chronology. The earliest round barrows on the chalk were built during the later third millennium BC, but those on the heathlands of southern England were constructed many years afterwards. Their distribution extends from the English Channel to the Kennet and the Thames. Most of these monuments were strikingly similar to one another, and few were associated with many artefacts. The heathland barrows included smaller versions of the specialized monuments found on the chalk. It seems likely that most of them were built in areas which lacked a history of monumental architecture. More barrows were constructed in both areas during the Middle Bronze Age, and only then is there any indication of regional variation. It seems table 1 The main characteristics of chalkland and heathland barrows between 2100 and 1200 BC BC (Earlier EBA) BC (Later EBA) BC (MBA) Barrows were generally permeable and were often rebuilt. There were multiple burials, and graves were sometimes reopened. Most barrows are found on the chalk, but in lowland areas others may have been levelled by the plough. Some groups of round barrows were close to older monuments, the first of which date from the Neolithic period. Grave goods were usually represented. There is evidence for the recycling of heirlooms and perhaps of human remains. Barrows were generally impermeable and were usually built in one operation. There were single burials, and the graves were not reopened. The bodies were covered by a turf core. They were protected by a capping of inorganic material and by the construction of earthwork boundaries. THE FIRST HEATHLAND BARROWS WERE BUILT. Other barrows were constructed on the chalk and in lowland areas. Some groups were close to older monuments, and specialized types were especially common in the major barrow cemeteries on the chalk. Other less conspicuous mounds were more widely distributed. Heathland barrows were of similar proportions to the majority on the chalk and were often isolated or in small groups. Specialized types did occur there, but they were not as large as their counterparts on the chalk. Grave goods were present at a number of chalkland barrows, where some objects may have been heirlooms. Grave goods were rarely deposited in heathland barrows. Barrows were generally permeable and were built in a single phase. Many had an entrance and were associated with multiple burials, some of which extended beyond the limits of the monument. Barrows are found on the chalk and on the heathlands. On the chalk, small groups of barrows were built on the edges of major cemeteries. Others were nearby. Some mounds were directly related to settlements and field systems. The same types of barrows are found on the heathlands. Grave goods were generally absent on both the chalk and the heathlands. Journal compilation 29

16 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND as if mounds with entrance causeways are a feature of southern Wessex, but it is impossible to discuss more general patterns as so much of the excavated evidence comes from a restricted area near the south coast. The construction of these monuments had ended by 1200 BC. A second question concerns the character and distribution of round barrows. The areas with heathland mounds had not been occupied during the Neolithic period and some of them seem to have been colonized for the first time during the Early Bronze Age. It happened as part of a more general expansion of settlement in England (Bradley 2007, 170 2). Heathland barrows had distinctive features of their own. In contrast to earlier developments on the chalk, there is little evidence for the prolonged use of the mounds or for the reopening of graves and the circulation of heirlooms. Each monument was employed only once before it was closed. Rather than emphasizing the links between the generations, the significance of certain individuals was accepted without the need to emphasize their relationships to other people. If that information was provided at all, it was expressed by the positioning of different barrows within the same cemetery (Barrett 1994, 123 9). That is also relevant to sites on the Wessex chalk where there were unusually large concentrations of monuments in places with a long-established significance. Here distinctive artefacts are associated with the graves. In other areas, and particularly on the heathlands, it rarely happens, perhaps because funeral offerings were deposited in different locations. How was this development related to other changes in the pattern of settlement? There are two factors to consider here: social organization and ecology. It is notoriously difficult to infer prehistoric social structure using the evidence of burials, but in this case the attempt is worth making. The earliest barrows considered here were essentially permeable, for they contained a sequence of successive deposits. There is an emphasis on continuity between the generations. In the later Early Bronze Age practices began to change. Nowhere is this more obvious than on the heathlands. The structure of round barrows was modified. Now they were impermeable and most of them were sealed after only one person had been buried. There was a greater emphasis on the individual and less concern with the past. Descent may have played a smaller part than achievements in life, and it may be no accident that the people commemorated in this phase had settled in new areas, some of them well away from long-established ceremonial centres. Did they free themselves from the constraints of tradition as they colonized new land? In the Middle Bronze Age round barrows changed their character again. Now they were closely associated with the domestic world and individual monuments could even have been built in the image of the house. They were entirely permeable monuments, and the clusters of cremations associated with them may have emphasized the importance of the household. That would certainly be consistent with the character of the associated settlements. A second factor is ecology. The colonization of poorer soils continued in the Middle Bronze Age, but the dead were not always buried in the same areas as before. Perhaps the land could not sustain a protracted period of exploitation. It seems likely that food production was reorganized as the expansion approached its limits. The earliest evidence of ditched field systems comes from four sites on the southern margin of Wessex, where they probably date from the end of the Early Bronze Age (Fig. 6; Yates 2007, chapter 7). Their creation could have been a response to the poor returns obtained from marginal areas, but another factor could have been equally important. It is possible that the development of field systems in the south of Dorset began at such an early date because the inhabitants of that area had different outside contacts from the people who buried their dead on Cranborne Chase or Salisbury Plain. The axis described as the Channel Bronze Age involved connections with 30 Journal compilation

17 RICHARD BRADLEY AND ELISE FRASER Figure 6 (Upper) The distribution of Bronze Age co-axial field systems in the study area, showing the four examples attributed to the Early Bronze Age. Information from Yates (2007). (Lower) The location of the study area in relation to the position of Dartmoor, and the geographical extent of the Channel Bronze Age. Information from Needham (2006). south-west England where parts of Dartmoor were enclosed at approximately the same time (Needham 2006; Fleming 2008). The development of the first field systems in the Solent Basin may have been suggested by developments outside Wessex altogether. There is too little evidence to take this argument further, but two features are clearly documented. Field systems became much more important during the Middle Bronze Age. If the first examples were in south Dorset, they soon extended across a wider area which included large tracts of chalk downland and the river valleys. At the same time the significance of ancient ceremonial centres lapsed as the new land divisions developed around them. The last round barrows on the chalk were constructed during this phase, yet by the Late Bronze Age their period of use was over. This paper has discussed one category of field monument, the round barrow, and has emphasized how its character changed over space and time. It began by contrasting the results of two excavations undertaken at the onset of the Second World War. Some of the differences may Journal compilation 31

18 BRONZE AGE BARROWS ON THE HEATHLANDS OF SOUTHERN ENGLAND shed light on more general developments. It is unfortunate that perceptions of this evidence have been biased by the well known evidence from the chalk, for it should have been apparent from the excavation at Beaulieu that heathland barrows had a distinctive character of their own. It is by comparing the monuments in both these regions that their wider significance becomes apparent. The opportunity was missed when so many barrows were investigated 60 years ago. It is important to take advantage of it now. Acknowledgements We must thank Ann Woodward and Lilian Ladle for permission to refer to their current research on Bestwall Quarry. We would also like to thank Ann for her comments on an earlier draft of this article and Margaret Mathews for the figure drawings. Department of Archaeology School of Human and Environmental Sciences University of Reading Whiteknights Reading RG6 6AB references barrett, j. 1991: The monumentality of death: the character of Early Bronze Age mortuary mounds in southern Britain. World Archaeology 162, barrett, j. 1994: Fragments from Antiquity (Oxford). barrett, j., bradley, r. and green, m. 1990: Landscape, Monuments and Society. The Prehistory of Cranborne Chase (Cambridge). bradley, r. 1981: Various styles of urn : cemeteries and settlement in southern England, BC. In Chapman, R., Kinnes, I. and Randsborg, K. (eds.), The Archaeology of Death (Cambridge), bradley, r. 1998a: The Passage of Arms (2nd ed.) (Oxford). bradley, r. 1998b: The Significance of Monuments (London). bradley, r. 2007: The Prehistory of Britain and Ireland (Cambridge). bradley, r. and keith-lucas, m. 1975: Excavation and pollen analysis on a bell barrow at Ascot, Berkshire. Journal of Archaeological Science 2, branch, n. and green, c. 2004: The environmental history of Surrey. In Cotton, J., Crocker, G. and Graham, A. (eds.), Aspects of Archaeology and History in Surrey (Guildford, Surrey Archaeological Society), brück, j. 2004: Material metaphors: the relational construction of identity in Early Bronze Age barrows in Ireland and Britain. Journal of Social Archaeology 4, cotton, j. 2004: Surrey s early past: a survey of recent work. In Cotton, J., Crocker, G. and Graham, A. (eds.), Aspects of the Archaeology and History in Surrey (Guildford, Surrey Archaeological Society), dimbleby, g. 1962: The Development of British Heathlands and their Soils (Oxford). ellison, a. 1980: Deverel-Rimbury urn cemeteries: the evidence for social organisation. In Barrett, J. and Bradley, R. (eds.), Settlement and Society in the British Later Bronze Age (Oxford, BAR Brit. Ser. 83), english, j. 2007: Fieldwork on Whitmoor Common. Bulletin of the Surrey Archaeological Society 398, 5 6. exon, s., gaffney, v., woodward, a. and yorston, r. 2000: Stonehenge Landscapes (Oxford). field, d. 2008a: The Use of Land in Central Southern England during the Neolithic and Early Bronze Age (Oxford). field, d. 2008b: The development of an agricultural countryside. In Pollard, J. (ed.), Prehistoric Britain (Oxford), Journal compilation

19 RICHARD BRADLEY AND ELISE FRASER fleming, a. 2008: The Dartmoor Reaves: Interpreting Prehistoric Land Divisions (2nd ed.) (Oxford). fraser, e. 2005: Colour, Structure and Stratigraphy in Bronze Age Barrows (Unpublished M.A. thesis, University of Reading). french, c., lewis, h., allen, m., green, m., scaife, r. and gardiner, j. 2007: Prehistoric Landscape Development and Human Impact in the Upper Allen Valley, Cranborne Chase, Dorset (Cambridge, McDonald Institute Monographs). gardiner, j. 1984: Lithic distributions and Neolithic settlement patterns in central southern England. In Bradley, R. and Gardiner, J. (eds.), Neolithic Studies (Oxford, BAR Brit. Ser. 133), garwood, p. 2003: Round barrows and funerary traditions in Late Neolithic and Bronze Age Sussex. In Rudling, D. (ed.), The Archaeology of Sussex to AD 2000 (Brighton, University of Sussex), garwood, p. 2007: Before the hills in order stood: time and history in the interpretation of Bronze Age barrows. In Last, J. (ed.), Beyond the Grave. New perspectives on Barrows (Oxford), grant, m. and barber, k. 2008: Palaeoecology of human impact in the New Forest. ceest/palaeo/nfnew_forest.htm green, c., lynch, f. and white, h. 1982: The excavation of two round barrows on Launceston Down, Dorset (Long Crichel 5 and 7). Proceedings of the Dorset Natural History and Archaeological Society 104, grinsell, l. 1932: Some Surrey bell barrows. Surrey Archaeological Collections 40, grinsell, l. 1934: Bell barrows. Proceedings of the Prehistoric Society of East Anglia 7, grinsell, l. 1939: Berkshire barrows: Part IV. Berkshire Archaeological Journal 43, grinsell, l. 1941: The Bronze Age barrows of Wessex. Proceedings of the Prehistoric Society 7, grinsell, l. 1974: Disc barrows. Proceedings of the Prehistoric Society 40, last, j. 2007: Beyond the grave. New perspectives on barrows. In Last, J. (ed.), Beyond the Grave. New perspectives on Barrows (Oxford), needham, s. 1987: The Bronze Age. In Bird, J. and Bird, D. (eds.), The Archaeology of Surrey to 1540 (Guildford, Surrey Archaeological Society), needham, s. 2006: Networks of contact, exchange and meaning: the beginning of the Channel Bronze Age. In Needham, S., Parfitt, K. and Varndell, G. (eds.), The Ringlemere Cup. Precious Cups and the Beginning of the Channel Bronze Age (London, British Museum), needham, s. and woodward, a. 2008: The Clandon Barrow finery: a synopsis of success in an Early Bronze Age world. Proceedings of the Prehistoric Society 74, parker pearson, m., pollard, j., richards, c., thomas, j., tilley, c., welham, k. and albarella, u. 2006: Materialising Stonehenge: the Stonehenge Riverside Project and new discoveries. Journal of Material Culture 11, petersen, f. 1981: The Excavation of a Bronze Age cemetery on Knighton Heath, Dorset (Oxford, BAR Brit. Ser. 98). piggott, c.m. 1943: Excavation of fifteen barrows in the New Forest, Proceedings of the Prehistoric Society 9, piggott, s. and piggott, c.m. 1944: Excavation of barrows on Crichel and Launceston Downs, Dorset. Archaeologia 90, royal commission on the historical monuments of england 1975: County of Dorset. Vol. 5 East Dorset (London). smith, n. 1999: The earthwork remains of enclosure in the New Forest. Proceeding of the Hampshire Field Club and Archaeological Society 54, white, d. 1982: The Bronze Age Cremation Cemetery at Simons Ground, Dorset (Dorchester, Dorset Natural History and Archaeological Society Monograph Series No. 3). woodward, a. 2002: Beads and Beakers: heirlooms and relics in the British Early Bronze Age. Antiquity 76, woodward, a., hunter, j., ixer, r., maltby, m., potts, p., webb, p., watson, j. and jones, m. 2005: Ritual in some Early Bronze Age grave goods. Archaeological Journal 162, yates, d. 2007: Land, Power and Prestige. Bronze Age Field Systems in Southern England (Oxford). Journal compilation 33