Placing Boxgrove in its prehistoric landscape

Placing Boxgrove in its prehistoric landscape Matthew Pope The Lower Palaeolithic site at the Boxgrove quarry in Sussex was featured in the first (1997/98) issue of Archaeology International, since when it has become internationally celebrated for the exceptionally detailed evidence it has provided of human occu­ pation of a coastal habitat 500,000 years ago, towards the end of an interglacial period. New investigations of Boxgrove's place in the wider landscape are now under way, and there are plans for further excavations and the establishment of afield school at the site.

T he desolate terrain of British gravel quarries contrasts starkly with the rural beauty of the countryside that commonly surrounds them. The open, sunken expanses of bare shattered rock and sand in the quarries are perhaps not unlike the barren ice-age landscapes that gave rise to the stony deposits they contain -landscapes in which ground-penetrating freezing and thawing created thick, mobile blankets of sediments stripped from the surrounding hillsides by melting ice. Over much of Britain these flows of broken bedrock ripped away the ancient land surfaces, lake beds and stream channels occupied by our earliest ancestors. Yet in some localities these sediments had the opposite effect, by covering fine-grain deposits in deep protective mantles of periglacial and fluvial gravels. For this rea son, the study of Palaeolithic archaeology in Britain has had a long and close associ ation with the aggregates industry, and for the past two centuries most significant col lections ofPalaeolithic artefacts have been made in these bleak environments.
In most cases, the conditions under which such discoveries are made are far fr om ideal for a discipline relying heavily N eChichester CO. \ST, \L l'L lN on scientific rigour in the interpretation of often meagre and fragmentary finds. From the earliest collections made by workmen at small quarry pits on country estates, to archaeological watching briefs alongside the imposing machinery of modern quar ries, discoveries have often been made fortuitously. Undertaking thorough inves tigations amid the potential dangers of machinery and unstable sediments, or the obscuring effects of flooding, is a challenge to the modern discipline of Palaeolithic archaeology. However, as the past 25 years of research at the Boxgrove quarry has demonstrated, persistent endeavour in these superficially unattractive environ ments can yield unique insights into the life of archaic Hom o sapiens. The new phase of work now under way at Boxgrove and its surroundings in West Sussex ( Fig. 1) promises to provide a regional focus for Palaeolithic studies over at least the next 10-1 5 years .1

Previous research at Boxgrove
My personal involvement in the Boxgrove project began in the mid-1990s under rel atively comfortable conditions and clem ent weather at the site, as a member of the team that was following up the discovery  of a human tibia (shin bone) in 1993. It had taken many seasons of dedicated field work in rough conditions at the quarry before the importance of the site was recognized by the wider community of Palaeolithic specialists. It was while living for much of the year in a derelict, rat infested and unheated farm in the early 1 980s that Mark Roberts and his small team from the Institute of Archaeology had started to investigate fine-grain deposits at the base of the Boxgrove quarry (Fig. 2). Their dedication to the task stemmed from the recognition, at an early stage, that they had something unprecedented in Euro pean prehistory on their hands: a series of extensive and perfectly preserved ancient land surfaces containing untouched scat ters of flint tools and fo ssilized animal bones of remote antiquity. During the 1 980s, the initial exploratory excavations were consolidated into what became an established multidisciplinary research project, coordinated by the Institute under the direction of Mark Roberts.
As the project developed, over 90 areas of the buried prehistoric landscape were sampled, producing evidence of excep tionally well preserved areas of human activity associated with the butchered Figure 3 Reconstruction by John Sibbick of humans butchering a rh inoceros carcass at the Boxgrove waterhole site. Stone biface tools were manufactured here with wh ich large mammal carcasses were systematically processed, including the extraction of bone marrow (righ t fo regro und). Th e humans wo uld have defended such valuable carcasses against other carnivores such as scavenging hyaenas (left background).
remains oflarge mammals. A picture began to emerge of the life and environment ofthe people who had occupied a landscape in which, towards the end of an interglacial period, a retreating sea had created an intertidal environment of saltmarsh and grassland overlooked by an imposing line of high chalk cliffs fringed by a beach (traceable today as the Boxgrove raised beach, Fig. 1). There was a ready supply of flint -the raw material for all the stone tools found at the site -in the tal us (scree) slopes at the base of the cliff, and large herds of herbivores grazed amid the open grassland and freshwater springs of the lowlands below. The animal-bone evidence showed that the people were gaining access to carcasses of horse, red deer, rhi noceros and bovids such as bison. They were using the tools to extract meat and bone marrow ( Fig. 3) but they had to com pete with other carnivores such as wolf, hyaena and lion, whose gnaw marks were found to overlie the fine tracery of flint tool marks on some of the bones. Some extensive activity areas, such as a horse butchery site (GTP17), perfectly preserved the evidence of butchery in the form of scatters of flint fragments left over from making (knapping) stone tools on site (Fig.  4). The knapping scatters surrounded the remains of a large horse and provided a snapshot of a few hours of activity left vir tually untouched since the human group moved on, sated on horse flesh, some 500,000 years ago.
By the beginning of the 1990s the archaeological significance of Boxgrove had become widely recognized, but it was not until 1993, with the discovery of the human tibia, that the work of the research team gained international recognition. 2 The project was transformed by this dis covery, unprecedented in Britain, and, with funding from English Heritage, a major excavation was undertaken that involved an international team of special ists and students. The new investigation focused on a suite of fr eshwater deposits the remains of a small spring-fed pool -in the area (which we now refer to as the waterhole site) surrounding the spot where the tibia was found. In 1994 the team recovered two human teeth (lower front incisors) and the excavations revealed dense concentrations of stone tools and butchered animal bones. The concentra tions of stone tools were dominated by sev eral hundred bifaces (hand axes) in mint condition (Fig. 5). They had been worked finely on both sides and finished by expertly removing a tranchet sharpening flake (a slice struck fr om the tip of a biface to make the cutting edge razor sharp).
Numerous bifaces had previously been recovered throughout the areas excavated at Boxgrove, together with the remains of knapping debitage (flint debris) produced during their manufacture. By refitting this debitage to re-constitute sequences of knapping, we have been able to show that tools were routinely carried around by human groups and were subjected to minor modifications and re-sharpening as their edges became blunt through use. However, at the waterhole site, people appear to have ab andoned this conserva tive approach to tool use and routinely dis carded bifaces and blocks of raw material in large numbers, despite their obvious future usefulness. This behaviour has been noted at many other Lower Palaeolithic (Acheulean) sites and is usually, as at Box grove, associated with areas of springs, streams and other areas of fresh water. It is not the only type of enigmatic behaviour observed at the water hole site: a small clus ter of incised chalk blocks of unknown use were unearthed from the deposits there. Linear striations, apparently cut with flint tools, cross the upper surfaces of the blocks and may possibly relate to the processing of plant or animal resources, although it is equally possible that they derive from some activity not related to subsistence.
Results of the analysis of finds from the waterhole and horse-butchery sites and from other excavation areas at Boxgrove have been published in a series of papers and a major monograph.3 The multi disciplinary approach, incorporating the work of experts in fi elds as diverse as micropalaeontology, avifauna, palynol ogy, soil micromorphology and human anatomy, has allowed a detailed picture to be built up of environmental changes and human activities at Boxgrove. The depos its have been dated to around 480,000 years ago by correlating stratigraphical sequences of mammalian remains.4 This date is consistent with the tentative attri bution ofthe human remains to the species Homo heidelbergensis, an archaic form of Homo sapiens that is currently thought to have be the common ancestor of popula tions of both neanderthals in Europe and anatomically modern humans in Africa.
Forensic analysis of the horse remains Fi gure 4 Excavating in situ knapping scatters at the horse butchery site GTP1 7. found at Boxgrove produced evidence of wounds fr om both bone breakage and pro jectile impact. The latter evidence sup ports finds from Lower Palaeolithic sites at Clacton in Essex and Schi:iningen in Lower Saxony (Germany), which suggests that humans were hunting large mammals with wooden spears. This evidence, together with the observation that butchery marks underlie carnivore gnaw marks, suggests that the Boxgrove people were competent hunters at the top of the fo od chain, who used tools and exploited the land in com plex ways.5 Finds of resharpened tools, and of hammers made of antler and used in the production of the fine bifaces, suggest that the hunters routinely curated valued tools, frequented areas such as waterholes where game was intercepted, and system atically defended and exhaustively butch ered animal carcasses (Fig. 3). The local availability of fertile grassland, fresh water and abundant raw material for making tools would undoubtedly have attracted human groups. These factors, combined with the excellent conditions of preserva tion at the site and the use of modern exca vation techniques, have combined to make the record of human behaviour at Boxgrove the most complete for any Lower Palaeo lithic (and Middle Pleistocene) site in Europe.

Mapping beyond the Boxgrove quarry
During the past 150 years Acheulean arte facts have been found at several locations near Boxgrove, such as Slindon and East Lavant (Fig. 1), and some of them are asso ciated with suites of marine and terrestrial deposits identical to those at Boxgrove, at a similar altitude of about 40 m above present sea level. In the late 1990s, Box grove was shortlisted for inclusion in the UNESCO list of World Heritage Sites, with the proviso that the full extent of the Box grove deposits should first be established so that the entire palaeo-landscape that they represent could be included in the designated area ofthe World Heritage Site. Therefore, in 2001, a geological survey referred to as the Raised Beach Mapping project was initiated with the aim of map ping the limits of preserved land surfaces, characterizing differences in local sedi mentary sequences and prospecting for deposits with archaeological potential. The Raised Beach Mapping project com bined re-assessment of historical records with archaeological trial trenching and borehole surveys (Fig. 6) across the north ern part of the coastal plain of Sussex and eastern Hampshire. As a result of 42 fi eld investigations, the extent of the Boxgrove palaeo-landscape became apparent for the first time. The marine deposits and asso ciated cliff line have now been traced for some 26km, between Westbourne and Arundel (Fig. 1),6 and in this zone some 13 km of intact fossiliferous former land surfaces have been identified. The depos its are very similar to those discovered at Boxgrove and may well preserve extensive evidence of human activity in situ. One stratigraphical horizon, designated the Unit 4c palaeosol, is a single land surface that has been shown to have developed over a period of 10-5 0 years. It represents a developing grassland habitat and pre serves in situ evidence of human activity over the timespan of a single human gen eration. It has now been traced east-west for over 10 km and offers a unique oppor tunity to investigate patterns of variation in human activity within such a short time period across a very extensive area.
The survey has also revealed some local environmental variation and has allowed us to reconstruct the topographical setting of the palaeo-landscape in finer detail. The saltmarsh and grassland environments that developed as the interglacial sea retreated appear to have been partially enclosed by two pronounced chalk ridges (the Portsdown and Littlehampton anti clines, Fig. 1), since eroded away by marine action. Half a million years ago they ended in two headlands that defined a large embayment backed by imposing 80 m-high chalk cliffs , the line of which can be traced today along much ofthe course of the Boxgrove raised beach between West bourne and Arundel. A series of streams flowing across the chalk surface behind the cliffs emptied into the embayment, form ing gravel fans at their estuarine mouths. By means of the survey we have been able to explore the interface of these gravels with the grassland palaeo-landsurfaces. We have also identified beds of preserved organic remains, which may enable local vegetation to be reconstructed. At the west ern and eastern margins of the embayment, where unfortunately preservation is not so good, we have identified areas where the chalk cliffs were replaced by lower bluffs cut into soft clays of Tertiary age rather than the Cretaceous chalk. These varying local environments would have produced different local ecologies and may have pro voked different behavioural responses among the human groups occupying them -differences that we hope to investigate by excavation in the future. One testable hypothesis is that human groups would have curated and re-used flint tools more extensively in these marginal areas farther from local sources of raw material.
The results of the Raised Beach Map ping project show that Boxgrove should no longer be considered a single, isolated site. Instead, we now know that it is at the heart Figure 6 Sampling geological deposits with a borehole rig at West Stoke, 8 km west of Boxgrove. Th e sedimentary sequence in the core obtained here was fo und to be identical to that recorded at the Boxgrove quarry.
of a vast, preserved prehistoric landscape buried between 2 m and 20 m below the modern surface of the Sussex coastal plain. Although quarries such as Boxgrove still provide the best chance of investigating extensive tracts of this landscape. the dis covery of several valley-side locations, where the deposits rest close to the surface, offers the possibility of discovering new locales rich in traces of human activity.
The next p hase: future excavations and a fie l d school In 2001 the close relationship that had long existed between the gravel-extraction industry and Palaeolithic archaeology in the UK was cemented by the establishment of the Aggregates Levy Sustainability Fund. The fund was created by the govern ment by taxing quarrying and landfills to provide fi nancial resources for environ mental improvements to old sites and also for archaeological investigations. One of the first benefits of the fund was the pro vision of a grant that enabled English Heri tage to purchase part of the Boxgrove quarry in 2003. The resulting 9 ha property incorporates the waterhole site as well as several other areas of rich archaeological potential and it is now owned by English Heritage and managed by the Institute of Archaeology. which has assumed respon sibility for its partial restoration, ecologi cal management and development.
Throughout the winter of 2003-2004, plans drawn up by the Boxgrove project for the restoration of the site were imple mented. This involved the movement of tens ofthousands of tonnes of gravel. amel iorating the quarry edges and covering the site in a layer of topsoil sufficient to allow the establishment of coppice and pasture. In addition, the area around the water hole site has been levelled to within 2 m of the underlying archaeological deposits to facilitate future excavation, increasing eightfold the area available for study. This enlarged area should be enough to include the interface between the edge of the water hole and the nearby grassland environ ment, as well as being sufficient to trace the deposits that produced the dense concen trations of flint tools and animal remains. Working northwest from the area previ ously excavated, we hope to follow the margins of the stream channel to where the originating spring emerged from the beach at the base of the cliff. Artefacts. together with the remains of butchered animals, were densely concentrated around this channel and the human tibia was recov ered from its upper fills. The new area is likely to merit years of excavation, aimed at gaining a more detailed understanding of human behaviour at this locality, including the possibility of recovering further bones or teeth of Homo heidelberg ensis.
Excavation will be renewed in 2005, when we will try to determine the overall topographical setting of the waterhole, and also establish basic on-site amenities essential to the running of a fu ll season of excavation. It is hoped that in the future the project will attract sufficient external funding to allow the Institute to establish a permanent fi eld school at the site. The school will provide training for undergrad uate and postgraduate students in the run ning of modern scientific excavations as well as facilities for on-site post-excava tion work. It will also function as an impor tant resource for the Institute's teaching programmes in human origins, prehistoric archaeology, ancient technology and pal aeoecology. While research will continue to focus primarily on the Boxgrove quarry, other sites, including those located by the Raised Beach Mapping project, will begin to be investigated through the field school, thus broadening the project into a wider investigation ofthe archaeology of the Sus sex coastal plain.
In contrast to its humble beginnings, the opportunity now exists for the Boxgrove project to move into a new phase of poten tially very rewarding research. The provi sion of permanent accommodation and facilities for laboratory work, teaching and site interpretation for visitors lie some dis tance in the future. Establishing a field school will consolidate the past 25 years of research at this unique site and provide a much-needed regional focus for Palaeo lithic studies in a valuable new facility managed by the Institute of Archaeology. Through the restoration work already under way, the Boxgrove quarry is begin ning to be transformed from a bleak post industrial landscape into an appropriate base for modern Palaeolithic archaeology in Britain.
Notes form ation in the Lower Palaeolithic (PhD dissertation. University of Southampton, 2002). 6. The Sussex coastal plain is crossed east west by a series of approximately parallel fo rmer (raised) beaches, the oldest and most northerly of which is the Boxgrove raised beach. The results of the Raised Beach Mapping project will be published by English Heritage as a monograph within the next few years. New geological maps will be deposited with local plan ning authorities to assist in the manage ment and protection of the preserved prehistoric landscapes in Sussex. 7. I would like to thank Mark Roberts.
Jemima Mieville and Kate Russell for their comments on an earlier draft of this arti cle. As a multidisciplinary project, the work at Boxgrove could not have been undertaken without the dedicated work of many specialists and students.