THE ARCHAEOLOGICAL RECORD FROM 500,000 years ago stands in stark contrast to the earlier periods. There are more sites containing large numbers of artefacts, new stone tools in the form of handaxes and for the first time direct evidence of humans with the fossil bones of Homo heidelbergensis. The richness of the record, though, has only been evident for the last 30 years and has been realised through better understanding of the Pleistocene sequence, better dating and the discovery of new sites. Prior to 1985 any thought of a pre-Anglian occupation of Britain was limited to a small assemblage of chemically weathered and disputed flakes from Westbury-sub-Mendip in Somerset and to a handful of derived handaxes from the Breccia at Kent’s Cavern in Devon (Cook & Jacobi, 1998; Andrews et al., 1999). Crude handaxes on some of the higher river terraces were added to the mix, but none provided incontrovertible evidence of pre-Anglian humans.
Two areas of research changed the scene and the dynamic of Lower Palaeolithic studies. The first was new geological work in the Midlands and East Anglia on what came to be known as the Bytham River (Rose, 1992). As part of this, new excavations in 1988 at the well-known site of High Lodge clearly placed the rich stone tool assemblages in a pre-Anglian context (Ashton et al., 1992). The second area lay on the coastal plains of Sussex, where old raised beach deposits were beginning to produce a startling array of stone tool industries associated with a rich fauna. The site was Boxgrove. In combination these two regions have transformed our understanding of humans and their habitat in Britain from over 500,000 years ago.
WAVERLEY WOOD AND THE BYTHAM RIVER
The Bytham was one of the great rivers of central and eastern England, named after the small village of Castle Bytham in Lincolnshire (Figure 53). Its catchment encompassed Derbyshire in the north, Worcestershire in the west and Oxfordshire in the south, flowing eastwards across central East Anglia into the North Sea. The river was destroyed by the Anglian Glaciation, 450,000 years ago, but its former course can be traced by small pockets of quartz and quartzite-rich gravel that occasionally survive on hilltops and valley slopes, carried by the river from the Midlands (Rose, 1992).
FIG 53. Map of southern Britain showing palaeogeography about 500,000 years ago with sites mentioned in the chapter. (Craig Williams)
One of the more westerly upstream sites to survive is Waverley Wood, a gravel pit that lies 5 km south of Coventry (Shotton et al., 1993; Keen et al., 2006). The gravel underlies glacial sediments that have been dated to the Anglian glaciation, so the underlying river sediments appear to date to the preceding interglacial, about 500,000 years ago. This is supported by the recovery of water vole teeth without roots (Arvicola cantiana) and amino acid ratios (Penkman et al., 2011). Although there has been no large-scale excavation, regular fieldwork and monitoring of the pit has produced a small archaeological assemblage together with rare and highly important environmental information about the Bytham River at this time. Extensive quarrying in the late 1970s and early 1980s revealed quartz and quartzite-rich gravels, typical of the Bytham River. It was not until 1984 that organic sediments were discovered in a succession of four overlapping channels at the base of the sequence that were filled with sands and sometimes muds. The channel fills were extensively sampled and sieved to produce rich assemblages of molluscs, ostracods, insect remains and mammals, together with pollen and other plant remains. In combination they paint a detailed picture of the changing environment during an interglacial.
The sands infilling Channels 1 and 4 suggest flowing water that sandwich the gentler or still-water conditions of the muds of Channels 2 and 3. Molluscs were recovered from the lower three channels, with those in Channel 1 dominated by Pisidium that live in flowing water. By contrast the assemblage from Channel 2 indicates very gently flowing or still water, perhaps an oxbow cut-off with seasonal flooding and aquatic vegetation shown by species such as Bathyomphalus contortus and Armiger crista. The Channel 3 molluscs indicate marsh conditions dominated by species such as Lymnaea truncatula and Caychium minimum, although drier grassland was also nearby, as shown by various species of Vallonia.
A more detailed picture of Channel 2 is provided by the ostracods, insect and plant remains. The ostracods show a development from quiet water to more muddy conditions with a hint of cooling towards the top provided by the species Ilyocypris decipiens and Candona levanderi. The much larger beetle assemblage of over 200 taxa shows a similar picture. Elmis aenea and Oulimnius tuberculatus show flowing water at the base, but most of the sequence is dominated by species that prefer still, poorly oxygenated water such as Helophorus aquaticus. Other species such as Carabus granulatus and Leistus rufescens suggest densely vegetated wet habitats, or meadows shown by Trechus secalis. Although there are very few species indicating trees, two exceptions show that both coniferous and deciduous woodland was nearby. Blastophagus piniperda lays eggs under the thick bark of both pine and spruce, causing severe damage. The bark beetle, Ptelobius vittatus, attacks elm, sycamore and willow, creating ‘bark roses’. The rarity of dung beetles shows that larger mammals were only occasional visitors. One intriguing species is Agonum quadripunctatum, which often occurs in burnt woodland, particularly under the bark of burnt branches and stumps of trees. The beetle may simply reflect natural fires in the area, although humans could perhaps have been responsible. Generally the beetles show temperate conditions through the sequence of Channel 2 similar to northern England today with mean July temperature estimates of 15°C. These estimates drop towards the top with July temperatures of only 10°C before returning to a more temperate climate. The markedly colder conditions are described as near-Arctic and one notable species is Helophorus obscurellus, which is found today on the Kanin Peninsula in Arctic Russia and further east across northern Siberia.
The plant remains from Channel 2 build on this picture. Plant remains from five species of pondweed (Potamogeton) were found, while fruit stones from several species of Ranunculus including Creeping Buttercup (Ranunculus repens) and Cursed Buttercup (Ranunculus sceleratus), reflect a mix of open ground by rivers and muddy, shallow pools. The pollen has been derived from further afield and shows a mix of pine, spruce and birch forest with grassland. The open areas were dominated by grasses (Gramineae) and sedges (Cyperaceae) with aquatic and water-edge plants such as Bulrush, Watermilfoil (Myriophyllum), Bur Reed (Sparganium) and Sweet Grass (Glyceria).
Vertebrate remains are very rare, with only a few mammals represented and many of these were found as isolated bones on the quarry floor. The small mammals include the extinct species of shrew (Sorex araneus), water vole (Arvicola cantiana) and pine vole (Pitymys arvaloides) together with the extant European Mole (Talpa europaea). Most of the larger mammal bones were from the straight-tusked elephant but none of the remains were found in situ. Horse (Equus ferus) was found in the base of Channel 1. We can only guess at the range of other mammals that would have populated the river valley.
In combination, the evidence from Waverley Wood shows a sluggish river with ox-bow lakes and marshland supporting a variety of vegetation from pondweeds, reeds, sedge and grassland meadows flowered with varieties of buttercup and with woodland of pine, spruce and birch beyond. The climate was temperate, but slightly cooler than today other than the marked cooling towards the end of the sequence, before a return to temperate conditions.
Onto this canvas we can almost certainly add humans, although frustratingly very few of their stone tools have been found within the channels. The exception is one small quartzite flake from the bottom of Channel 1. Other artefacts were found on gravel discard heaps or more recently from other nearby quarries. The majority are flakes, cores and rather crude handaxes made from quartzite, but the most intriguing finds are a small series of beautifully made handaxes of andesitic tuff (Figure 54). Their fresh condition means that they almost certainly derive from the channels. The puzzling aspect is the andesite, with the nearest outcrops being in Ordovician rocks in the Lake District. It was initially suggested that humans travelled with these handaxes from the north, but more recently andesite has been found as occasional erratics within the local gravels. It must have been quickly recognised as an excellent workable material and valued more highly by early humans above the rather intractable quartzites. Perhaps better evidence of human travel is provided by two flint handaxes. The only flint available in the area occurs as small, degraded pebbles, totally unsuitable for handaxe manufacture. The nearest sources of good quality flint are the Chilterns, 100 km to the south, or East Anglia, 125 km to the east. The Bytham was a natural routeway from the east coast to the Midlands, so the flint handaxes could be indications of seasonal movements of people between the two regions and they certainly emphasise the value placed on well-made handaxes if they are being carried those sorts of distances.
FIG 54. Andesite handaxe from Waverley Wood. (Terry Hardacre; courtesy of Warwick Museum)
THE BRECKLAND SITES: MAIDSCROSS HILL, HIGH LODGE AND WARREN HILL
Moving downstream, the Bytham flowed northeast towards Leicester via the sites of Brooksby and Castle Bytham, crossed what is now the Fen Basin, then veered south towards Bury St Edmunds before heading east across East Anglia to Lowestoft. On this course there is one particularly rich area where Bytham River deposits survive. The Breckland, a gently undulating landscape of heathland and coniferous forest, lies on the eastern margins of the Fens. The area contains a remarkable number of geological and archaeological sites, not just former courses of the Bytham River. Several different river terraces of the Bytham have been identified, the highest and oldest of which might date back to perhaps a million years. Unfortunately biological remains are rare, so that the dating is reliant on the assumption that new terraces were formed through down-cutting of the river during each new glacial cycle (Chapter 2). It is in the lower two terraces that the first evidence of humans appears in the form of handaxes.
Maidscross Hill
One such site is Maidscross Hill that now overlooks the runway of RAF Lakenheath. In 1869, decades prior to any thought of air-travel, John W. Flower, a Fellow of the Geological Society of London, read a paper to the society entitled ‘On some recent discoveries of flint implements of the Drift in Norfolk and Suffolk, with observations on the theories accounting for their distribution.’ In the paper he described the discovery of handaxes in quarries with quartzite gravel at a height of 90 feet (27.4 m). The old workings can still be seen on the eastward brow of the hill and many of the handaxes are in the collection of the British Museum. These important and detailed observations were confirmed by recent fieldwork which indeed showed quartzite-rich gravel on the hilltop at a height of 27 m. The elevation of the gravel is well above the lowest and youngest Bytham terrace and suggests that the handaxes come from the 2nd terrace of the river (Ashton & Lewis, 2005). The lowest terrace is pre-Anglian in age and dated to about 500,000 years ago (Voinchet et al., 2015). If terrace formation relates to 100,000 year glacial-interglacial cycles as discussed in Chapter 2, then the 2nd terrace would date to c. 600,000 years ago, making Maidscross Hill the earliest handaxe site in northern Europe.
The handaxes seem to divide into two groups, with some being rather thick and crude in manufacture, bearing evidence of battering and rolling from transport in the river bed. The second group are more refined ovates and cordiforms with less abrasion to the edges. The simplest explanation is that the handaxes represent earlier and later groups of humans using the floodplains of the river. The more rolled handaxes could be derived from a higher terrace, which would imply even earlier handaxe-users in Britain.
High Lodge
High Lodge is a rather different type of site and has been surrounded in controversy since its discovery through gravel and clay extraction in the late nineteenth century with various interpretations put forward for the complex sequence of glacial tills, clays and gravels (Ashton et al., 1992). The old pit is now surrounded by pine forest. The largest-scale excavations were in the 1960s by Gale Sieveking of the British Museum with Richard West and Charles Turner from Cambridge University, but due to differences of opinion over the age of the site the work remained unpublished for 30 years (Figure 55). The main problem stemmed from the unusual stone tool assemblage consisting of elegant scrapers, likened by the archaeologists to Mousterian industries in France of less than 100,000 years old. By contrast there were geological grounds for suggesting that the site was much older. Although not all are agreed, fieldwork in 1988 led to a consensus among most researchers that it forms part of the Bytham River story.
FIG 55. High Lodge during British Museum excavation in the 1960s.
The oldest sediments consist of 2 m of clays and silts, which not only contain the scraper industry, but also pollen, insect remains and the occasional bone. The clays and silts were formed as floodplain sediments of the Bytham River as shown by analysis of minerals deriving from the Midlands and by the occasional gravel component rich in quartz and quartzite. Large-scale shearing and thrust structures within the clays and interleaving with glacial till indicate their disturbance by glacial processes, which created an inverted stratigraphy with the older clays overlying younger till (Figure 56). The clays seem to have been transported as frozen rafts by Anglian ice. Up to 6 m of sands and gravels, formed by glacial melt waters, overlie the sequence. A rather different stone tool assemblage consisting of ovate and cordiform handaxes has been recovered from the base of the sands and gravels, possibly derived from a different part of the underlying floodplain clays.
FIG 56. Schematic sections of High Lodge showing the formation of river silts, movement by glaciation and deposition of meltwater gravels. (Phil Dean)
A pre-Anglian age for the site is supported by part of a right upper premolar from the extinct Etruscan rhinoceros now attributed to Stephannorhinus hundsheimensis (Figure 57; Stuart, 1992). Elsewhere in Europe this species became extinct about 450,000 years ago. Other mammalian remains are rare and fragmentary, consisting of straight-tusked elephant, horse, a medium-sized deer and a large bovid, none of the latter attributable to species.
The environment in the immediate vicinity of the site is shown by the insect remains. From the lower part of the clays 47 taxa of insects have been identified with most reflecting marshy conditions (Coope, 2006). The weevil, Notaris acridulus, occurs in virtually all the samples and feeds on marshland grasses, particularly Sweet Grass. Various species of Cercyon feed on rotting vegetation in muddy, stagnant ponds, while areas of sedge and swamp grasses are shown by several species of Bembidion. Other taxa such as Otiorhynchus feed on the leaves of various low-growing herbs and their larvae eat the roots. Dry habitat species are rare, as are those showing more open aquatic conditions. As with Waverley Wood, dung beetles are virtually absent, other than Aphodius, reflecting the rarity of larger mammals in the immediate area. Climate estimates can also be calculated from the beetle remains, with mean July temperatures between 15 and 16°C, while January temperatures were probably between −4 and 1°C. A modern-day analogue would be northern Britain or southern Scandinavia.
FIG 57. Fragment of Etruscan rhinoceros tooth from High Lodge showing how the fragment (a) forms part of a complete example (b). (Phil Dean)
The pollen adds to the environmental picture from the wider landscape showing that pine and spruce were the dominant trees, but with significant herbaceous, grassland and aquatic elements (Hunt, 1992). Wetland habitats from slow-flowing water to marsh are shown by algal spores, particularly Spirogyra and pollen from Eurasian Watermilfoil (Myriophyllum spicatum) and Broad-leaved Dock (Rumex obtusifolius). Drier ground supported juniper (Juniperus), ferns (Filicales), bracken (Pteridium), heathers (Ericaceae), Crowberry (Empetrum) and buttercups. The vegetation is in accord with the beetle evidence suggesting a slightly cooler climate than East Anglia today, probably similar to central Scandinavia.
The 1960s’ excavations produced two main stone tool assemblages, with that from the clays totalling over 1,200 artefacts. The majority were flakes and cores, some of which formed distinct knapping scatters, which could be refitted back together (Figure 58). It is remarkable that the scatters survived the transport of the clays, by ice. A small, but important element of the assemblage are the highly unusual scrapers. Large, thick flakes have been retouched to form elegant scraping edges (Figure 59). Although use-wear does not survive on the edges, they would have made ideal tools for removing the fatty tissues from the insides of hides. Their presence implies the processing of animal skins either for clothing or perhaps shelters and the first hint of humans developing ways of countering the cold of northern Europe. The other unusual aspect of this assemblage is the complete absence of any handaxes or flakes from their manufacture. By contrast handaxes are an important element in the assemblage from the glacial sands and gravels immediately overlying the clays. They are relatively fresh in condition and similar in form to the ovate and cordiform-shaped handaxes from Maidscross Hill and Warren Hill (see here). They are unlikely to be contemporary with the formation of the glacial gravel and probably derive from a different part of the floodplain. Whether the scraper industry and the handaxe industry were made by the same human group with different activity areas, or indeed whether they are contemporary, is not clear. Alternatively they could represent different human groups with completely different traditions of tool manufacture.
FIG 58. Excavation of refitting flint scatter from the 1960s’ excavations at High Lodge.
FIG 59. Scraper from High Lodge. (Phil Dean)
Warren Hill
Just over 1 km to the south of High Lodge lies the site of Warren Hill, an area of sloping heathland that bears the scars from old gravel workings, overlooking the valley of the Lark (Figure 60). It is famous for producing over 2,000 handaxes, mainly during the late nineteenth and early twentieth centuries. This was during the heyday of collecting, when handaxes often fetched £1 a piece. It has been suggested that digging for handaxes was more profitable than digging for gravel and is the reason for such quantities of stone tools being recovered from Warren Hill (Figure 61). As with Maidscross Hill there are two groups of handaxes, although the majority are the more refined ovates and cordiforms. Geological work has established that the gravels and underlying sands and silts are part of the youngest terrace of the Bytham River and date to about 500,000 years ago (Voinchet et al., 2015). One curiosity is the similarity of the handaxes to some of those from Maidscross Hill on the 2nd terrace. This might imply that the Warren Hill artefacts are in fact derived from much older deposits.
FIG 60. Recent view of Warren Hill. The bumps and hollows are from the spoil heaps and small pits from earlier quarrying. (Simon Lewis)
FIG 61. Quarry workmen possibly sieving for handaxes at Warren Hill in the 1930s. (Henry Dewey)
The gravels, sands and silts at Warren Hill tell a further story. They form a series of dipping units called ‘foresets’, which are characteristic of rivers flowing into shallow lakes and meeting still water. It seems that the deposits at Warren Hill were flowing into a lake delta, which formed as a result of damming of the river by ice downstream and marking the onset of the Anglian Glaciation. It was only a matter of time before more extensive ice covered the region and destroyed forever the river courses, valleys and tributaries of the great Bytham River.
HAPPISBURGH SITE 1
Happisburgh Site 1 also lies on an eastward-flowing river, although probably not the Bytham. The site was discovered in 2000 by local resident, Mike Chambers, about 1 km to the south of Happisburgh village (Figure 62). The handaxe poking out of rich organic muds on the foreshore eventually led to a full excavation in 2004. Any fieldwork was a daunting task, the site being only visible at occasional low tides. This might explain why any major excavation was delayed for four years, when I was persuaded by my friend and colleague Simon Parfitt that not only was the site of great importance, but that the local hostelry served excellent beer and had cheap accommodation. The pub, the Hill House, later gave its name to the geological formation at Site 3. The work was conducted in the narrow window between the tides with a 16-tonne excavator at hand to remove any in-washed sand from the latest inundation (Figure 63). Controlled excavation was difficult, but all the sediment was sieved at 1 mm mesh in the sea to ensure maximum recovery. Excavations were continued from 2009 to 2012 by a team from Leiden University, led by Mike Field, in collaboration with AHOB.
FIG 62. View of Happisburgh looking southeast towards Site 1 from the church tower.
FIG 63. Excavation at Happisburgh Site 1. (Nigel Larkin)
The work established the age, the contemporary environment and a better idea about the humans at the site (Ashton et al., 2008b). The sediments consist of peat and mud above organic sands that infill a channel some 100 m across. All these sediments form part of the Cromer Forest-bed Formation. The channel is probably the youngest of the sites at Happisburgh, dating to about 500,000 years ago, which has mainly been deduced from the recovery of the water vole with unrooted teeth of Arvicola cantiana.
The botanical remains are still being studied but show wetland habitats with forest of birch and pine. The beetle remains also reflect aquatic and dry-land habitats (Coope, 2006). Limnius volckmari requires running water, but most taxa such as Cercyon reflect marshland and shallow pools with rotting vegetation. Sweet Grass and sedge provided food for Donacia semicuprea and Notaris aethiops. Large numbers of Micrelus ericae indicate heather on the drier ground, while coniferous woodland is shown by Polygraphus polygraphus, which burrows under the bark, particularly of spruce. Some deciduous woodland is also indicated by the presence of Hylastes attenuatus, which attacks willow. The dung beetle, Aphodius, points to the presence of medium to large mammals. Most of these species of beetle live in Britain today, but a few more northerly types indicate cooler conditions, such as Pelophila borealis and Notaris aethiops. In combination the beetles give estimates of average summer temperatures between 12 and 15°C, while average winters dropped to between −11 and −3°C. It is worth noting that winter temperatures are always more difficult to estimate and that the values given do not reflect the range of temperatures, but show that the average temperature is somewhere within the range. But however the beetle data is interpreted, they still reflect cool summers and distinctly chilly winters.
The vertebrate remains are from fish and mammals and a single bone of a frog. The fish indicate slow-flowing fresh water with dense aquatic vegetation and include Pike, Tench, Three-spined Stickleback (Gasterosteus aculeatus) and carp (Cyprinidae). Some of the mammals also reflect aquatic environments with extinct beaver and water vole. Other mammals include the Bank Vole (Clethrionomys glareolus), Red Deer, Roe Deer (Capreolus capreolus) and indeterminate species of bison and rhinoceros, which collectively reflect open grassland environments. The most unusual bone is from the extinct pliobatrachid frog (Pliobatrachus cf. langhae), which is the last member of the Palaeobatrachidae, the only family of frogs to become totally extinct (Holman, 1999).
FIG 64. The handaxe from Happisburgh Site 1. (Courtesy of Norfolk Museums Service)
The small flint assemblage consists of the one handaxe discovered in 2000 and about 300 other artefacts (Figure 64). They include a few notched flakes, possibly for stripping bark or working wood, but mainly simple flakes, some of which could have been used for cutting meat or hide. Butchery was undoubtedly taking place at the site as several of the bones from Roe Deer and bison have cut marks made with flint tools. The small size of the assemblage and thin distribution of artefacts probably reflects the rarity of flint in the channel area with most knapping taking place closer to a raw material source, perhaps a gravel bar nearby.
Overall the evidence points to humans inhabiting the muddy fringes of a slow-flowing river with still ponds and marsh alongside occasional stands of willow (Salix) and alder, extending to drier grassland and heathland beyond. The grasslands would have supported a range of deer, bison and rhinoceros, while spruce and birch woodland probably dominated the valley slopes. The cool climate might have been a portend for the ensuing Anglian Glaciation, when ice covered the region, depositing the sediments that form the cliffs at Happisburgh today.
THE PROTO-THAMES AND SOLENT RIVERS
The two other major rivers of southern and central England were the proto-Thames and the Solent. The Thames at this time was flowing through the Vale of St Albans and reached the sea near Colchester. It deposited large spreads of gravel across Essex which is mapped as a series of terraces known collectively as the Kesgrave Sands and Gravels. Despite extensive quarrying there have been no traces of artefacts and very little surviving faunal evidence. This might be due to the lack of diligent collectors or large-scale fieldwork, rather than an absence of humans. Further upstream it was rather different. The Reading area was fortunate to have one of the great collectors of the early twentieth century, Llewellyn Treacher (Hosfield, 2009). It is in large part due to his efforts, and those of his wife Mabel, that we have so much material from the Middle Thames. Treacher was a market gardener by trade, but also a much respected amateur geologist, receiving an award from the Lyell Fund of the Geological Society of London in 1913. One of his many collecting areas was a series of quarries on the north side of Reading that cut into the course of the Caversham Ancient Channel. The Channel is a remnant of the proto-Thames immediately prior to the forced diversion of the river into its current course by the arrival of Anglian ice. Although there is little fauna or other environmental evidence, the pits produced large quantities of handaxes, giving us an insight into human occupation of the Thames during this period.
The Solent was once the largest river of central southern England, but unlike the Bytham and Thames was largely unaffected by glaciation (Figure 53). It rose in western Dorset and flowed parallel to the present coast along the course of the Frome, north of the Isle of Wight, before veering south to join the Channel River. The Isle of Wight was linked to the mainland at this time by a chalk ridge that ran from the Needles to Portland Bill. The river had major tributaries with the Dorset Stour, the Hampshire Avon and the Test flowing in from the north, while the Medina flowed from the Isle of Wight in the south.
Terraces of the Solent and its tributaries have been mapped across the region and the gravels have produced a huge number of Palaeolithic sites, second only to the Thames. Where the Anglian Glaciation on the Thames and Bytham left clear upheavals to those rivers which help with the dating, no such signs can be recognised on the Solent. Although the various archaeological assemblages can be put into order according to the terrace staircase, there is still only a poor understanding of their absolute ages, except in the lower terraces. The near absence of faunal remains or other environmental information has also hampered research into their dating through biostratigraphy. Despite these problems we can guess that the higher terraces probably date back to beyond a million years and are devoid of artefacts, but in the middle terraces of the Stour and the Test handaxes first occur. If the introduction of handaxe technology across England is synchronous, then we can guess that these earliest sites might date to about 500,000 years ago, or perhaps a little older. One such site is Corfe Mullen, to the northwest of Bournemouth (McNabb et al., 2012; Davis, 2013). Here, collections of handaxes were recovered during gravel extraction in the late nineteenth and early twentieth century. One intriguing feature of some of the fresher handaxes is their similarity to the Boxgrove artefacts (see here), being generally ovate in shape and with a tranchet resharpening removal across the tip. Boxgrove was 90 km down the Solent and along the coast to the east. Does the similarity in technology show a connection between the sites, with perhaps the same group of humans exploiting this stretch of the Solent River and nearby coast?
BURIED CLIFFS, LAGOONS AND COASTLINES – BOXGROVE
The coastline of Britain has been constantly changing through rises and falls in sea-level and through the build-up of sediments in some areas and erosion in others. Occasionally relict coastlines have been preserved, particularly where there has been tectonic uplift of the land and raised beaches and ancient cliffs have been covered and survive. One of the best examples is the Slindon Raised Beach in Sussex which extends from almost Havant in the west to Arundel in the east. Quarrying across this area has revealed chalk cliffs against which rest ancient beach deposits, themselves gently covered by lagoonal silts and sands of the coastal plain. The best-known site is Boxgrove, which preserves an astonishing landscape that has been buried and undisturbed for 500,000 years. It is probably the most remarkable Palaeolithic site from a choice of many world-wide and could be described as the Pompeii of the Stone Age.
Boxgrove
In 1982 Mark Roberts was a student at the Institute of Archaeology in London, but spent part of his summer working on a Roman villa site on land above the Eartham Quarry at Boxgrove. His primary interests were prehistory and geology, so many hours after work were spent examining the sand and gravel deposits in the quarry below. Handaxes had been previously recovered from there on several occasions and reported on by the East Sussex County Archaeologist, Andrew Woodcock (1981). The following winter he decided to have a more detailed look at the geology and with four fellow students dug a large test pit in sediments at the base of the quarry. The section revealed gravel overlying silts and sands with an ancient soil horizon, but importantly associated with the land surface was a handaxe in very fresh condition, which had lain undisturbed since it was discarded in the Palaeolithic. The rarity of in situ sites led to further fieldwork the following summer and, through English Heritage funding, large-scale excavations continued until 1996 (Figures 65 and 66; Roberts et al., 1994; Roberts & Parfitt, 1999).
FIG 65. Excavation of Q1B at Boxgrove. (Courtesy Boxgrove Project)
FIG 66. Excavation of handaxes in Q1B at Boxgrove. (Courtesy Boxgrove Project)
The site was situated within a semi-enclosed marine embayment beneath a marine-cut chalk cliff. A raised beach with gravel and marine sands was overlain by intertidal and lagoonal silts. With slight retreat of the sea a soil developed, marked by an iron-rich horizon, and overlain by organic silts. The sequence was then covered by a series of hillwash and solifluction deposits that slid downslope under cold conditions during the following Anglian Glaciation. The gentle sedimentation has led to the preservation of scatters of animal remains associated with flint artefacts across the land surface extending for over 800 m in a wide band in front of the chalk cliff.
The environment at Boxgrove has been reconstructed from the ostracods, molluscs and a remarkable diversity of vertebrates. The lagoonal silts are interspersed with poorly developed soil horizons suggesting occasional drying out and the spread of grassland. Over 30 non-marine molluscan taxa were recovered, which generally reflect moist, well-shaded habitats with some pools. Several marine species were also identified, including the Dog Whelk (Nucella lapillus), Blue Mussel (Mytilus edulis) and the Northern Horse Mussel (Modiolus modiolus), which could have provided extra nourishment for the humans (Preece & Bates, 1999).
The fish remains show the change from marine to freshwater environments as the sea receded (Parfitt & Irving, 1999). Blue-fin Tunny (Thunnus thynnus), Atlantic Cod (Gadus morhua) and Conger Eel (Conger conger) are replaced by fish also tolerant of fresh water such as Three-spined Stickleback, European Eel and Brown Trout (Salmo trutta). The large number of Stickleback recovered from a pool may be a sudden ‘death assemblage’ where the fish were introduced to the pool through flooding and then became isolated and died.
A rich diversity of herpetofauna were recovered with nine species, a diversity of amphibians and reptiles that would be unusual to find in one location in Britain today (Holman, 1998). They reflect both pools and dryland habitats. Spadefoot Toads (Pelobates fuscus) dig burrows in sandy soil, but breed in pools, while the Natterjack Toad (Bufo calamita) is today found in coastal dunes, breeding in still, sometimes brackish ponds. Both the Moor Frog (Rana arvalis) and the Common Frog (Rana temporaria) prefer damp meadows and bog and Grass Snake (Natrix natrix) are usually found on damp ground, but often hunt in water.
The wide range of birds adds to the coastal landscape (Harrison & Stewart, 1999). Wetland habitats are shown by Whooper Swan (Cygnus Cygnus), Greylag Goose (Anser anser), Widgeon (Anas penelope) and Tufted Duck (Aythya fuligula). Grassland is indicated by the Grey Partridge (Perdix perdix) and woodland by the Tawny Owl (Strix aluco) and Goldeneye (Bucephala clangula). The latter nests in hollow, often coniferous, trees but close to water. The great auk (Pinguinus impennis) is the most unusual species and is the earliest record of this flightless seabird. The bird is now extinct, with the last individual killed in Iceland in 1844.
The largest group of vertebrates are the mammals, with over 50 taxa reflecting a wide range of habitats. Open grassland species include the Field Vole (Microtus agrestis) and grazers such as Horse and extinct rhinoceros. Both grassland and open woodland are shown by the large number of Pine Vole (Microtus cf subterraneus). Other woodland mammals include Wood Mouse (Apodemus sylvaticus), Hazel Dormouse (Muscardinus avellanarius) and two types of bat – the Whiskered Bat (Myotis mystacinus) and Bechstein’s Bat (Myotis bechsteini). The full range of mammals probably indicates a slightly more continental climate with habits similar to the mix of coniferous and deciduous woodland found in southern Scandinavia today.
The ostracods have been used to estimate the climate at Boxgrove, which suggest average July temperatures between 14 and 20°C, but Januarys of −4 to 4°C (Holmes et al., 2010). These estimates contrast with present-day values for nearby Eastbourne of 17°C and 6°C. It would seem that Boxgrove had similar summers, but cooler, more continental winters than southern England today.
Handaxes, hunting and butchery
For humans, the coastal lagoons and grasslands of Boxgrove would have been a Garden of Eden, teeming with wildlife. Not only did it offer rich hunting grounds, but the chalk cliffs provided a ready supply of good quality raw material as flint eroded out of seams within the chalk. The excavations recovered huge numbers of artefacts and fragments of teeth, bone and antler, which through their distribution document how humans used the different parts of this rich landscape.
There was no shortage of game, but how did they acquire it? The first clue comes from butchery cut marks on the bones. Time and again they are over-printed by hyaena gnawing showing that humans were the first to the feast and that humans were the top carnivore whether as scavengers or hunters. A single puncture wound on a horse scapula provides the other clue, as evidence of hunting with a wooden spear (Figure 67). It is more likely to have been a thrusting rather than a throwing spear, to give sufficient power to pierce both the hide and the bone. Close-encounter hunting also requires skill, cooperation and communication among the hunting group to track, corner, ambush and kill. Other butchered animals include deer, bear and even rhinoceros. Whether all the butchered carcasses had been hunted is not clear, but it is likely that scavenging opportunities were also taken. Perhaps this explains the butchered rhinoceros, which would have been a highly dangerous and difficult animal to hunt or even kill.
FIG 67. Part of horse scapula with puncture wound from spear from Boxgrove with reconstruction of remainder of bone in black (courtesy of Boxgrove Project and Smithsonian Museum of Natural History)
FIG 68. Handaxe from Boxgrove. (Matt Pope)
With a larder full of carcasses an efficient tool for butchery was required – and this was the handaxe (Figure 68). Microscopic use-wear and polish on the handaxe edges confirms that this was their primary purpose (Mitchell, 1997) and experiments using professional butchers show them to be ideal tools for a range of tasks from slitting the hide, filleting the meat, disjointing and cutting through ligaments and tendons (Schick & Toth, 1993; Machin et al., 2007; Bello et al., 2009). So how were they made? Here we can use the evidence of the distribution of artefacts across the landscape.
Nodules were selected from the flint rubble at the base of the eroding cliff, with initial shaping of the blocks into handaxe roughouts nearby. Further shaping and finishing seems to have taken place up to 100 m away on the coastal flats, presumably located closer to the hunted game, ready for butchery. There are numerous examples of distinct knapping episodes, but one in particular stands out (Austin & Roberts, 1999). In an area of ‘Quarry 1A’, 157 flakes made an isolated scatter, the waste products of handaxe manufacture. The scatter was in the shape of a fan with sharp lateral edges and larger flakes concentrated towards the base. The fan formed the area between the flint knapper’s thighs, sitting or kneeling on the ground, with the flakes falling from their lap. A few of the larger flakes had been selected and placed on the ground to the left. This remarkable ‘imprint’ provides a very real image of these early handaxe makers (Figure 69).
FIG 69. Knapping scatter from Q1A at Boxgrove. (Courtesy of Boxgrove Project)
The ‘GTP17’ site is an instance of both knapping and butchery, representing perhaps a single afternoon (Roberts, 1999). It is clear that at least six handaxes were made in this single location and in one example a large number of the flakes can be refitted to reform most of the original nodule. In the middle is a void of a partially finished handaxe. Initial knapping was probably with a hard stone pebble as a hammer, but was quickly replaced with a ‘soft’ hammer of antler or bone. Examples of these have been found elsewhere on the site, where the shafts of the hammers have tiny flint chips embedded in their battered sides or ends. Soft hammers produce thinner flakes and allow the knapper more control over the shaping and finishing of the handaxe.
Strangely none of the handaxes were left at GTP17, but there is clear evidence of their use. The butchered remains of a single horse were found among the flint scatters (Roberts & Parfitt, 1999). From the distribution of cut marks and modification to the bones and teeth, the Boxgrove humans were clearly skilled butchers. The marks and impact damage indicate skinning, disarticulation of the vertebral column and other joints, filleting, extraction of fat from the mandible and the breaking of bones for marrow extraction. When the butchery was complete the handaxes and presumably the meat were carried away, the meat for eating and the handaxes for use elsewhere (Figure 70).
FIG 70. Experimental butchery of deer with handaxe.
Whereas the scatters in Q1A and GTP17 are single events, the Q1B pond site shows persistent activity in one place over perhaps several generations. The small pond, over 40 m across, formed from a spring at the base of the cliff. As the pond silted up flint artefacts were incorporated into the sediment, but this time dominated by over 400 handaxes and flakes from handaxe modification and resharpening. Associated with the handaxes are the butchered remains of deer, horse and other mammals. The pond was clearly a focus for processing carcasses with repeated visits over several decades and shows how the landscape was used in an intelligent and logical way.
One further question arises from the butchery around the Q1B pond and elsewhere in the landscape – were the humans processing the hides? Being first to the kill to recover intact hides, prior to shredding by the big cats or hyaenas, was critical and the evidence does suggest that people had prime access. It seems inconceivable that the hides were not valued and used. Processing for long-term use would have been important to prevent any adhering tissues from turning rancid. Although there are few flake tools, such as scrapers, suggestive of hide-working, the pond at Q1B would have been a good location for processing with a ready supply of fresh water. Even better would have been stagnant ponds, rich in tannins to give longer-term preservation.
We can only guess at other activities, and one of the big questions is where were the humans living? Were they sheltering in primitive tents beneath the cliff, or perhaps in the more forested areas on the clifftop? One thing is certain, the coastal flats littered with carcasses would have been a dangerous place to live at night with nocturnal carnivores such as hyaena and lion scavenging, hunting and feasting on the game. Scaring off these predators during the day was achievable, but having protection at night was quite a different problem. Fire would have helped, but there is a complete absence of charcoal, which would have survived in the sediments if present. Nor are there other sites in Europe that have evidence of fire at this time.
The Boxgrove humans
Boxgrove is the earliest site where we can answer the question of who these humans were. In December 1993 a human tibia was discovered close to the pond of Q1B, but in a higher context (Roberts et al., 1994). As the bone was reworked from other sediments, it is not certain whether it was contemporary with the environmental information described above. The likely original length of the bone was about 400 mm and together with measurements of the mid-shaft circumference and diameter, they indicate a robust male with an estimated height of between 1.77 and 1.82 m. The discovery directly led to the excavation of the Q1B pond in the hope of recovering further human remains. In 1995 the work paid dividends when a lower incisor was discovered, joined by a second lower incisor from a metre away the following year (Figure 71). The teeth seem to be from a single individual and from their wear are thought to belong to a mature adult. Slightly oblique grooves on the teeth indicate some form of cutting and it is suggested that, in common with more recent hunter-gatherers, the jaw was used as a vice to hold meat while cutting, leaving damage to the teeth. The slant of the cuts down to the right of the individual indicates slicing with the right hand (Figure 72; Hillson et al., 2010). As to the species of human, both the tibia and the teeth are thought to belong to Homo heidelbergensis, a recent arrival in Europe, probably from Africa (Stringer, 2012b).
FIG 71. Human tooth from Boxgrove. (Simon Parfitt; courtesy of Natural History Museum)
FIG 72. Detail of human teeth from Boxgrove showing cut marks from cutting meat. (Silvia Bello; courtesy of Natural History Museum)
The picture that emerges from Boxgrove is of coastal mudflats with the periodic development of grassland beneath the chalk cliff. As the sea retreated a more permanent mosaic of grassland, freshwater pools, scrub and woodland became established, which supported a diverse mix of mammals, amphibians, reptiles, birds and fish. Humans were attracted to the area by the promise of rich game and the ready supply of flint to make handaxes for butchering their prey. As the top carnivore they hunted horse and probably deer, while also using the meat from bear and rhinoceros. At night they may have sought the shelter of the cliffs or the more wooded slopes above. Several generations exploited the rich pickings of this landscape before the downturn in climate and the onset of the Anglian Glaciation.
ENVIRONMENTS, HUMANS AND SUCCESSFUL COLONISATION
By 500,000 years ago humans seemed at home in the landscapes of southern Britain. Whereas earlier periods reflect fleeting visits by pioneering populations, now we have a larger range of sites with an abundance of evidence in the form of stone tools. So what had changed – the environments, the human species or technological improvements? Each of these aspects probably has some bearing on their successful occupation.
There is a remarkable consistency in the landscapes painted by the environmental evidence from the few sites where it survives. Waverley Wood, High Lodge and Happisburgh Site 1 lie within river valleys, the latter close to its estuary. They all contain a mosaic of river-edge habitats, with slow-moving water environments, still-water marshes and bog, grasslands, heathlands and coniferous forest on the higher ground. Boxgrove is similar, except in a coastal setting with mudflats, lagoons and freshwater ponds and with coniferous forest on the chalk downs above. All the sites suggest summers similar to Britain today, but a more continental climate with cooler winters. These landscapes supported a wide range of fish, amphibians, reptiles and birds, most of which are found in Britain today. The mammals included more exotic animals, such as straight-tusked elephant, rhinoceros, giant deer, bison and bear, together with Red, Fallow and Roe Deer and horse.
The environments, it would seem, were not so different from those at the earlier sites of Happisburgh Site 3 or West Runton. There was though one difference – the demise of the big cats (Turner, 1992). By 500,000 some of the bigger carnivores disappear from the record, either due to low populations or extinction. The giant hyaena and the sabre-toothed cat (Megantereon cultridens) had both disappeared by 500,000 years ago, while Homotherium latidens had become more rare. Ironically it may have been due to the increase in size of the larger herbivores, such as giant deer and bison, which made them more difficult to hunt. Whatever the cause, this change in the carnivore guilds of Europe must have had an important impact on the ability of humans to compete. There were still of course lions, hyaenas and wolves, but the gap in the ecological niche may have been filled by humans, who were certainly the top carnivore. This seems to have been achieved at least in some cases through close-encounter hunting with the use of spears, followed by quick, efficient butchery with handaxes. Scavenging, particularly of larger carcasses such as rhinoceros, also contributed to their diet.
Being the top carnivore may have had a side benefit. Caves in the limestone areas of Britain would have been an obvious haven for warmth and protection, if it was not for the other occupants seeking shelter in the form of bears and hyaenas. Competition for this resource could have been fierce. Unfortunately we only get glimpses of human use of these landscapes from the sites of Kent’s Cavern in Devon and Westbury-sub-Mendip in Somerset. At both sites the stone tools could have been washed into the cave systems, and so actual occupation by humans is not clear. Sharing game was one thing, but they were perhaps not yet ready to give up their dens.
The other benefit of being first to the kill was access to fresh, complete hides. The elaborate scrapers from High Lodge suggest processing of animal pelts, presumably for clothing and perhaps simple shelters. Their effectiveness would depend on how well they trapped the air and body heat. Sinews or tendons might have helped to bind the pelts together, but more elaborate stitching and tailoring is unlikely at this time. In whatever way they were used, hides must have helped in the quest to fend off winter cold, particularly with the apparent absence of fire.
The development of hunting, use of hides and the introduction of new tools such as handaxes and scrapers seem to be associated with a new species of human, Homo heidelbergensis. Although the evidence is limited to the tibia and two teeth from Boxgrove, from fossil evidence elsewhere in Europe and Africa we can say a little more about these humans (Stringer, 2011). Their brain size was on average about 1,200 cc, only a little smaller than modern humans at 1,350 cc, but larger than the 1,000 cc of Homo antecessor. Their stature and gait was similar to ours with estimates of average heights of 1.75 m for males and 1.55 m for females. Most researchers argue that they emerged from Africa into Europe about 600,000 years ago.
Drawing this evidence together, we can now suggest that Homo heidelbergensis did not arrive empty-handed, but brought with them a whole suite of skills and equipment that gave them an advantage in the cooler climates of northern Europe. Part of this package may have included different social structures, such as larger group size and more developed language. Both would have helped in tasks such as hunting and fending off competing carnivores while butchering the meat.
One aspect of these changes was the increase in material culture or possessions, whether these were spears, handaxes, rudimentary clothing or shelters. Shelters would have given a greater sense of place within the landscape and together with other possessions an appreciation of ownership. This is perhaps one of the reasons for the elaboration in tool form, beyond the purely functional, where personal possessions might have the additional role of expressing individual or group identity. Ownership brings potential conflict, so society would have had to cope with the pressures that this brought. By now humans are actors within their adopted landscape, which they are beginning to modify to their own ends. This a far cry from the early pioneers who were part-time extras on a vast and foreign stage.