1

The Pebblebed landscape

The landscape of the East Devon Pebblebed heathlands, whose bedrock is composed entirely of water-rounded pebbles, is geologically unique in the UK. Today they form a low ridge forming the watershed between the river Otter to the east and the Exe estuary to the west. The ridge is approximately 13 km long N–S, extending inland from the coast; it is 2–3 km wide and almost continuous. It is broken up today by pockets of improved agricultural land. The heathland soils are thin, poor and acidic and most of the area has never been cultivated. The western side of the Pebblebeds are defined by a distinct scarp slope about 20 m or so high rising up quite steeply from the undulating lowlands that extend to the Exe. The highest point on the western scarp is marked by the Iron Age hillfort of Woodbury Castle (183 m). The heathlands dip away gently to the south and the east toward the Otter valley and the sea. To the north they run down into the Clyst valley. The overall dip of the land across the greater part of the heathland area is from the northwest (high) to the southeast (low). At the base of the western scarp there is a spring line. To the east the sloping heathlands are broken up by small valleys. These sometimes originate in broader and boggy irregular basins. The heathland area is highly porous and drains quickly. Small east to west or northwest to southeast fast-flowing perennial streams now occur in the valleys where the water has cut down to underlying clays and marls. Further up beyond the surface streams, there are dry valleys formed in permafrost conditions during glacial periods. Wherever the surface mantle of vegetation is absent or disturbed, along the stream beds, on exposures on the often steep sides of valleys, and on paths and trackways crossing the heathlands, pebbles are exposed at the surface (Figure 1.1).

The Pebblebed heathlands are surrounded by rich pasturelands on clays and marls and arable fields on the surrounding Otter sandstones. The boundary between these lowland areas and the heath itself, on which the vegetation is gorse and heather, pine and birch, is abrupt and distinct. Today it is mostly marked by hedged banks and fencing. Contemporary farms and villages are sited along the stream courses where they emerge from the heathland habitat. Parish boundaries typically extend from the rich pastureland and arable fields up on to the heathlands, formerly common land both to the west up the scarp slope and to the east up the dip slope, ensuring that each had its share of fertile agricultural land as well as uncultivated grazing land. Although few animals graze the heaths today, in the past they provided important and substantial areas of rough grazing during the summer months. The heathland habitat is a mosaic of dry heath where the principal vegetation is gorse and heather and the lowland heath characterized by bog valley plant communities, birch and alder. Substantial areas of the dry heath have been afforested with conifers since the 1920s. The natural woodland cover of the heath is deciduous. Pine trees were introduced in the late eighteenth century as landscaping features (see Chapter 10).

This area of Devon has long been noted for its long, warm summers and mild winters. Today during the summer months temperatures reach on average 20 degrees C or more during July and August. Given the proximity of the sea, winters are mild with little snow and few frosts. Temperatures range between 3 and 8 degrees C. Situated in the rain shadow of Dartmoor, with prevailing winds being southwesterly, the area is relatively dry. The rainfall average is about 800 mm, peaking during the winter months.

Looking out to a world beyond

Standing on the western high scarp of the Pebblebeds there is a magnificent and panoramic view of the surrounding landscape. To the west one looks across the course of the Exe estuary to the unbroken ridge of the Haldon Hills running along its eastern edge (Figure 1.1; Figure 1.2). Further west still, over the line of the Haldon Hills, there are glimpses of the tor-crowned high peaks of Dartmoor. High Willhays and Ugborough Beacon are just visible some 48 km distant. To the northwest, the Raddon Hills, capped by a Neolithic causewayed enclosure and later Iron Age hillfort, frame the near landscape. To the north, the line of the Blackdown Hills is prominent, with another Neolithic causewayed enclosure and Iron Age hillfort at Hembury occupying a prominent southern spur. Way beyond the highest point on Exmoor, Dunkery Beacon, some 58 km distant, and the Quantock Hills can be seen on a clear day.

To the northeast, the hill island of Dumpdon (Figure 1.3) crowned by a hillfort and possibly another Neolithic causewayed enclosure is prominent in the Honiton gap created through the Blackdown Hills by the river Otter. To the east the landscape is framed by the broad Otter valley and the almost unbroken line of the East Hill and Peak Hill ridges, which block any view further in this direction (Figure 1.15a). Between these two ridges there is a prominent gap through which the sun rises at the equinoxes. This is a prominent landscape feature visible for long distances from the west and north, as far away as the southern edge of Exmoor to the northwest (Figure 1.15b). To the southeast, High Peak, with its distinctive triangular-shaped profile, is a dominant coastal landmark (Figure 1.4). Beyond it there are more distant views across LymeBay to the Isle of Portland 70 km distant – glimpses into other worlds and different landscapes.

But for the most part, views out from the remaining and lower areas of the Pebblebed ridge are strikingly restricted by the higher hills that surround it: the unbroken line of the Haldon Hills to the west, the more irregular line of the Blackdown and Raddon Hills to the north, and the East Hill and Peak Hill ridges to the east. These all rise up fairly abruptly above river valleys and are flat-topped. The eastern scarp slopes of the Haldon Hills and the western scarp of the Peak and East Hill ridges appear remarkably uniform from the Pebblebeds. This contrasts markedly with their appearance from the other side, where all these ridges are deeply indented with coombes and valley systems. Their most uniform and regular scarp slopes face towards the Pebblebeds and are framed by the surrounding hills, creating a sense of interiority and difference: a landscape that is peculiarly distinctive, framed and bounded, a world apart. When the Otter and Exe valleys fill with mists, the ridge and the hilltops are dramatically transformed, appearing to be islands enveloped in a grey sea.

What makes this landscape so special is not only the local presence of the Pebblebed heathland but also the hills that physically and visually hem it in with the significant gap on the eastern side. None of the surrounding ridges and hills have any Pebblebed outcrops or exposures. To the west, the Haldon Hills are covered by grey and white flinty gravels. The Blackdown Hills to the north and the East Hill and Peak Hill ridges to the east are capped with substantial layers of clay with flints and chert derived from the underlying greensand (paradoxically grey to grey-brown to yellow in colour), as is High Peak to the southeast (Woodward and Ussher 1911: 62ff.). All these surrounding hills thus contrast greatly with the much lower, rolling Pebblebed heathlands in terms of their far greater height, their much more pronounced scarp slopes and the sharp, angular and jagged stones that cover them. Sensorially encountering the bones of this landscape, we move from the smooth and rolling heathlands covered with smooth, rounded and multicoloured pebbles, to higher flat-topped hills with steep scarps covered with brittle, irregular and jagged material of fairly uniform and dull colour, an important series of visual, tactile and colour contrasts (Figure 1.5).

Another contrast occurs between the stones that may be observed along the ridges and the hills and those exposed along the rivers. The numerous river cliffs that occur along the lower course of the Otter are all exposures of the red Otter sandstones, as are those found along the Exe estuary at Lympstone. Along the Exe there are very limited exposures compared with those along the lower course of the Otter. Immediately to the north and south of Ottery St Mary these are grey-green in colour. Beyond here all the way to the south to the sea, where the river passes the Pebblebed heathlands to the west, these river cliffs are all bright red in colour. They occur along the Otter’s eastern side except in a short stretch between Newton Poppleford and Colaton Raleigh, where they are on the western side. By contrast, nowhere along the course of the Otter can the exposed stratum of the Pebblebeds be seen.

Walking north, east or west off the heathlands, one notes that the pebbles rapidly disappear under the surrounding marls. None are exposed along the sands and muds of the Exe. Redeposited material down-washed from the Pebblebeds occurs locally in the river bed along the Otter river valley, particularly in its lower stretches from Ottery St Mary southward to the sea at Budleigh Salterton. In the upper reaches of the river the pebbles are few and the river bed is largely made up of angular gravels derived from the clay with flint capping of East Hill. Lower down the river in some places between Tipton St John and Colaton Raleigh, Pebblebed material locally dominates. Newton Poppleford is named after the ford crossing the pebbles, or ‘popples’, derived from the Pebblebeds, that are numerous here along the river course. Everywhere in the surroundings of the heathlands, pebbles are numerous in arable fields and gardens, down-washed into the soils formed on the surrounding red Otter sandstones.

The geology of the Pebblebeds and its significance

There are very few natural inland rock exposures in this area of East Devon, and none are very large. The only places where the underlying rocks can be seen occur in the river cliffs along the Otter and the Exe valleys and along a few valleys with streams flowing east into the river Otter. The Steamer Steps cliffs to the west of Budleigh Salterton rise up sharply from the beach in a series of staggered ledges to West Down Beacon (129 m), the highest point before the land drops to the west and the Exe estuary. The cliff exposure here provides a dramatic and huge cross-section through the landscape – elsewhere almost always mantled in soil and covered in vegetation. An inspection of the sea cliffs thus permits a unique glimpse of another concealed world, the hard structure, or the ‘bones’ of the land beneath the constantly changing soft, damp and fleshy surface (Figure 1.6).

From Exmouth eastward, the cliffs and headlands are composed of the relatively soft and warm red-coloured Devonian New Red Sandstone formation. West Down Beacon marks the point where the geologically famous Triassic Budleigh Salterton (or Bunter) Pebblebeds first outcrop and attain their maximum thickness in the cliff face of up to 26 m (Selwood et al. 1984: 96). This strip of pebbles dips diagonally down through the otherwise red sandstone cliffs in a distinctive band, dipping and narrowing to the east. It finally disappears in the cliff face near to the edge of a small valley cut down to the beach by the stream at Budleigh Salterton. It marks the eastern limit of the Pebblebed outcrop.

The contrast between this band of pebbles and the New Red Sandstone appearing both above and below it could not be greater. The fine-grained red sandstone is smooth and uniform in colour. The only variation in its surface appearance is caused by localized honeycomb wind weathering creating numerous rounded hollows eating into the cliff’s face. The Pebblebed formation in the middle is dense and infinitely varied in terms of texture and the forms and colours of the stones. They are composed of well-rounded spherical or oval, clearly water-worn pebbles bedded in a coarse and gritty or finer and sandier matrix. Within the sand and grit lenses, pebble-filled channels can be observed, proof of the riverine origins of these formations.

The Pebblebeds were laid down by a huge river that flowed north through a hot red sandy desert about 240 million years ago (Figure 1.7). The direction of river flow is based on the discovery of fossils in the pebbles: brachiopods, bivalves and trilobites, whose probable origin was in the mountains of Brittany, northern France (Vickery 1864; Audley-Charles 1970: 52; Audley-Charles 1992). This river (or rivers) was huge, stretching from East Devon as far east as the Isle of Wight. Side streams probably flowed into it from Dartmoor, south and north Wales, bringing down material. Mountains lay to the west and north of what is now Devon and Somerset, forming part of the supercontinent Pangaea. It flowed from Brittany and Normandy and northward extensions of these areas in the English Channel, across the middle of England. The delta of the river, where it entered what may have been an open sea, was in the Cheshire basin/north Midlands. Almost all its course is only known today through geological bore holes and quarries. The only area where part of the river channel, now upraised, is exposed at the surface is the East Devon Pebblebed landscape.

The river consisted of braided channels and alluvial fans, was subject to flash floods and fast flowing. The large size of some of the pebbles is evidence for this and the fact that they are well rounded suggests a long transport history. The climate of the British Isles throughout the Triassic was hot, with alternating dry and wet seasons, and was tropical to sub-tropical in character.

Larger and smaller pebbles embedded in a coarse to fine gravel and sand matrix comprise the Budleigh Salterton Pebblebed formation that in the area of Black Hill quarry on the western scarp attains a maximum thickness of 31 m (Selwood et al. 1984: 96). Metaquartzite pebbles and cobbles up to 0.45 m in diameter make up the greater part of the material (up to 90 per cent). Other pebbles and cobbles of schorl, vein quartz (up to 7 per cent), porphyries, tourmaline, feldspathic conglomerate and sandstone are present. There is little variation in the relative proportion of these types of pebbles throughout the exposures. Pebbled strata are commonly interleaved with horizontally bedded red-brown gravel and silty sand layers. These are more frequent towards the top of the formation. The frequency of pebbles to matrix varies between 80 per cent and 20 per cent (Henson 1970: 175).

The quartzites are the product of low-level regional metamorphism. In other words, the sand grains of which they are composed have been comparatively little altered by subsequent heat and pressure. The pebbles are poorly sorted, so very large ones and smaller ones may be found side by side. The mean size of the pebbles declines northwards along the Pebblebed heathlands from about 16 cm at the coast to 10 cm inland to the north (Henson 1970: 97; Edwards and Scrivener 1999: 88) and from the lower to the upper strata of the exposure (Ussher 1913: 89).

At the top of the Pebblebed formation, visible in the cliffs at Budleigh Salterton, there is a striking bright-yellow band of sandstone appearing immediately beneath the Otter (new red sandstones) (Figure 1.8). Below this there is a thin black deflation layer including ventifacts up to 7 cm thick. These are wind-faceted and polished pebbles with two or more smooth faces with a distinctive ridge between them, and one rough face. Some, termed dreikanter, have a very distinctive triangular appearance with three facets at the top; others have four or more. The dark colour of these pebbles when newly exposed is only a surface varnish caused by desert weathering. When split open they are ordinary quartzites like the others (Perkins 1971: 130; Leonard et al. 1982). They lay exposed on the floor of the Triassic desert and were wind-polished. The side of the pebbles that lay on the desert floor is typically rough and unpolished with shatter pitting of the surface, while the faces exposed to the polishing actions of the desert sands are beautifully smooth.

The pebbles seen on the beach at Budleigh Salterton are all derived from the cliffs above. Long shore drift and violent storms have swept them eastwards into a substantial ridge backing up to Otter Point and Otter Ledge to the east of the mouth of the river Otter and almost blocking its passage to the sea. Formerly a port, mud flats and marshes have formed behind the great pebble ridge. The beach is approximately 40 m wide and up to 3.5 m above high-water mark.

Despite the derivation of the pebbles on the beach from those in the cliffs there are substantial differences. First, many of the cliff pebbles have numerous shatter marks originating in the transport of the pebbles along the river course. These are rapidly removed and scoured by the waves. Second, while the cliff pebbles are ungraded, with smaller and large pebbles occurring next to each other, those on the beach are sorted by wave action, with smaller pebbles occurring next to the sea and the largest ones higher up the beach nearer to the cliffs with mean pebble size decreasing to the east. Third, the surfaces of the pebbles in the cliffs are stained brown by the gravel and sandy matrix in which they occur. Those on the beach are scoured clean by wave action and the salt and are much more brightly coloured. Many of the pebbles in the cliffs are fractured or broken. Those on the beach are smoother and well rounded, with wave action removing any rough, fractured edges. Fourth, the beach pebbles tend to be flatter and more rounded in form, again a product of abrasive wave action. Statistical sampling has shown that mean roundness values are lowest for cliff samples, with little variation along the beach (Carr and Blackley 1975: 306). By contrast, length variation was considerable, varying from sample to sample, reflecting the composition of the Pebblebeds themselves.

Bunter, the term for this geological formation, derives from the German for ‘brightly coloured’. The variation in colours of the pebbles is quite extraordinary, although curiously somewhat little remarked in contemporary geological accounts. All the colours of the rainbow and more are here. The pebbles range in colour from pure black to pure white. Brown, red, green, yellow, blue and grey pebbles all occur. Some have mottled surfaces with many different colours. Others have striking and intricate quartz veins and inclusions. This, above all, is what makes the pebbles of the Pebblebed formation on the beach and throughout their inland distribution so distinctive compared with beach pebbles found elsewhere throughout Britain.

The modern rational geological account of the Pebblebeds, referred to above, is of recent date, little more than a hundred years old. Like the modern geologist, the prehistoric cosmologist might have attempted to understand what was under his or her feet by asking some ‘geological’ questions and making similar observations. Observing fallen pebbles from the cliffs, the origin of those on the beach might have been easily deduced. But why the great uplifted ridge of pebbles running inland and also seen in watercourses and exposed patches so distant from the sea? What were their origins, when smooth, rounded pebbles are normally only found by the sea? The logical premises for interpretation would, of course, have been radically different, but the answer might have been the same: these were the remains of an ancient river or beach thrust up towards the sky.

The Mesolithic and Neolithic (9500–2400 BC)

Mesolithic and Neolithic finds from the Pebblebed heathlands consist of a few finds of Neolithic axes and mixed surface flint scatters near to Blackhill on the western scarp, containing both Mesolithic and Neolithic material and two Mesolithic flint scatters to the south and southwest of Woodbury Castle (Smith 1956; Wymer and Bonsall 1977). These are all on the highest areas of the heathlands. Beyond the heathlands Mesolithic material was recovered from the excavations at Hembury of the Neolithic causewayed enclosure and Iron Age hillfort (Berridge 1986).

Figure 1.9 shows the distribution of all known flint finds from the heathlands and along the Otter valley to the east. These range from the Mesolithic to the Iron Age and very few are broadly dateable diagnostic finds. There are concentrations along the lower Otter valley but they also occur all over the landscape from the highest points of the heathlands to the surrounding lowlands. Given the character of the heathland vegetation, making the collection of flints impossible except in disturbed areas, it is interesting to note that the frequency of find spots is nevertheless as common as in arable land to the east, illustrating their extensive prehistoric visitation and use.

It is worth mentioning here in a book concerned with pebbles and their significance one class of ceremonial artefact long held as being diagnostic of the Mesolithic: pebble maceheads with hour-glass perforations. These are recorded across all of southern England from Cornwall to Kent. In Wymer and Bonsall’s (1977) catalogue 10 are recorded from Cornwall, three from Devon, 16 from Dorset, 22 from Wiltshire and between 4 and 12 in other southern English counties. Sometimes, as at Portland in Dorset, they form part of excavated Mesolithic assemblages (Palmer 1999). In other cases they are isolated finds.

They represent the first direct evidence that we have of an interest in and ceremonial use of pebbles. Although some are recorded from coastal sites, others are found far inland and at a considerable distance from the sea. They must represent long-distance movement or exchange between hunter-gatherer groups, something that is also known from the widespread distribution of Portland chert across southern England during the Mesolithic (Palmer 1970). None are so far recorded from East Devon. The sources of the pebbles used, their colours and characteristics (see Chapter 5) would clearly repay further study.

Neolithic flint scatters are recorded along the coast to the west of High Peak and to the south and north of Otterton, on Mutter’s Moor, part of the Peak Hill ridge, at Patterson’s Cross just to the north of Ottery St Mary, and a series of others much further north along the Exe valley around Nether Exe (Griffith and Quinnell 1999b; Miles 1976; Pearce 1979). In addition to these surface flint scatters, Neolithic settlement and ritual deposition in pits is documented from the A30 excavations at Castle Hill and Long Range (Fitzpatrick et al. 1999), at Hayes Farm, Clyst Honiton and at Pixie’s Parlour, Ottery St Mary (Mudd and Joyce 2014). A pit at Hayes Farm contained 16 clay loom weights, sherds from a carinated bowl, burnt sheep/goat bones, parts of a bird and charred plant remains (Hart et al. 2014: 7–9). At Pixie’s Parlour a 5-m-long and 2-m-wide pit interpreted as a possible tree-throw hole contained 57 sherds of earlier Neolithic pottery with a wide variety of different fabrics, some as far away as the vicinity of Dartmoor, and 56 pieces of worked flint (Mudd and Joyce 2014: 17).

A house structure together with possible enclosures (for animals? The land was never ploughed) on top of the Haldon Hills (Gent and Quinnell 1999b; Willock 1933, 1937) has long been known at Haldon Belvedere. Recent excavations on the outskirts of Ottery St Mary carried out ahead of a housing development have provided additional rare evidence of Early Neolithic occupation. Around 40 pits and post pits, some of which were arranged in arcs, were associated with pot sherds, many of which were from carinated bowls (DAS Newsletter 2013: 6–7).

There are three known Neolithic causewayed enclosures and/or hilltop settlements on High Peak, and at Hembury and Raddon, a much greater distance away to the northwest (Gent and Quinnell 1999b). There is the possibility that another may exist under the Iron Age hillfort of Dumpdon in the Honiton gap to the northeast of the heathlands.

High Peak (Figure 1.4) is the highest and most distinctive point along this stretch of the East Devon coastline. Although it is considerably lower (157 m) than either the Peak Hill or East Hill ridges (highest point 246 m) to its north it appears both higher and more prominent because of its relative isolation, distinctive triangular shape and coastal situation. Excavations on High Peak revealed traces of a possible Neolithic causewayed enclosure on top of the hill, virtually all of which has been subsequently destroyed along with the ramparts of an Iron Age/Dark Age hillfort by coastal erosion. The Neolithic remains included a short ditch segment rock-cut in its lower part through the greensand and underlying chert beds with a primary fill that included charcoal, bone fragments and flint flakes with pottery in the upper fill (Pollard 1966: 41). Pollard also identified ‘cooking areas’ with flint and pottery scatters and three pits. The pottery recovered included sherds of gabbroic ware originating in the Lizard peninsula, Cornwall but this seems to have been rare, constituting only 3 per cent of the total assemblage (Quinnell and Taylor in Rainbird et al. 2013: 37) compared with 10 per cent of this material at Hembury (see below) (Quinnell and Taylor in Rainbird et al. 2013: 37). A few sherds contained temper with a granite (Dartmoor) source and the bulk was manufactured locally: some of the clay was of Lias origin, the nearest source being the Devon/Dorset border to the east (Quinnell and Taylor in Rainbird et al. 2013: 37). Most flints were of local material but they included two pieces of Portland chert and black flint derived from Beer (Pollard 1966: 47–8; Tingle 1998). Among the groundstone axe fragments there is more exotic material: a jadeite piece with an Alpine origin and a picrate piece from Callington, Cornwall. Other groundstone axes were made from the local greensand. A number of pebbles, some showing signs of usage, from the Budleigh Salterton Pebblebeds, were found among the Neolithic material (Pollard 1966: 52). Recent excavations in 2012 recovered 22 pebbles derived from the Pebblebeds and local beach deposits, one of which with faint polishing may have been used as a rubber (Taylor in Rainbird et al. 2013: 42).

As elsewhere in southern England, causewayed enclosures began to be built in the thirty-seventh century cal. BC (Whittle 2007: 137–8; Whittle et al. 2007). Radiocarbon dates have suggested that the Neolithic settlements on High Peak and Hembury were roughly contemporary but they were from bulk samples and not very reliable. The enclosure at Raddon is somewhat later (Gent and Quinnell 1999a: 64). The causewayed enclosure at Hembury occupies the southern tip of a prominent spur of the Blackdown Hills with extensive views to the south across the Pebblebed heathlands to the sea. Liddell’s excavations revealed eight ditch and low bank sections with intervening causeways cutting across the spur, and house structures and substantial occupation debris inside indicating permanent settlement (Liddell 1929–1932a, 1929–32b, 1929–32c, 1936). A second ditch line was found to the north, as well as additional ditches, by Todd’s re-examination of the northern part of the spur, indicating the presence of multiple enclosures (Todd 1984).

Artefact finds included pottery tempered with local quartzites derived from crushed Bunter pebbles, imported gabbroic pottery from the Lizard peninsula, Cornwall, implements made from Beer flint and a few of Portland chert. Others were from closer flint sources only a few kilometres away; greenstone axes of Cornish origin; and from North Devon, querns and rubbing stones from the local Pebblebeds, beads of steatite, and jet, possibly from Spain and Brittany (Liddell 1929–32a, 1929–32b, 1929–32c).

The excavated materials from High Peak and Hembury indicate a systematic gathering of raw materials and artefacts from (1) the immediate locality; (2) the Pebblebed heathland that had to be crossed to move between these two places, and (3) more distant sources at a variable distance away – Beer Head, Portland, Exmoor, Dartmoor, Cornwall, and those from very distant origins as far away as the Alps and Spain. Materials and artefacts used in these two Neolithic enclosures thus brought together and incorporated elements drawn from the immediate and more distant landscapes at various scales of movement. Some of these, such as the Pebblebed heathlands and Beer Head (for high-quality black flint), could be visited in a day. Other more distant places (Portland, Dartmoor, Exmoor) could be seen on the far horizon from the vantage point of the heathlands or High Peak. Finally there were artefacts and materials brought from places that could never be experienced by people remaining in place or travelling only through the local landscape.

This pattern of raw material utilization seems to contrast with the Neolithic and Bronze Age domestic assemblages found during the A30 and gas pipeline excavations to the north of the heathlands in which stone material other than flaked flint and chert is rare and of local origin (Mepham 1999: 210–21; McSloy 2014: 55–62.). It appears that the curation and use of pebbles was confined to meeting places and settlements of especial significance and ceremonial importance. During the Neolithic the pebbles were associated with the living, whereas in the Early Bronze Age they became associated with ritual monuments in the landscape: pebble cairns.

Bronze Age cairns and settlement (2200–500 BC)

There are 32 recorded Bronze Age pebble cairns on the East Devon Pebblebed heathlands (Figure 1.10). These were the first structures to be built from pebbles. There is no known evidence of monument construction during the Neolithic or of other mortuary practices. The excavations at High Peak, Raddon and Hembury revealed no human remains from the enclosure ditches or interiors. One rectilinear structure at Castle Hill just to the north of the heathlands excavated in advance of the A30 road construction has been suggested to be a ‘long mortuary enclosure’, but there is a lack of evidence to suggest such a funerary use (Fitzpatrick et al. 1999: 213). Another rectilinear enclosure has been suggested to be part of a possible cursus monument, but again the evidence is equivocal.

It remains the case that the first funerary structures and monuments to be constructed in this area of East Devon are cairns and barrows of Early Bronze Age date. The distribution of the cairns is entirely confined to the heathlands. There are also a number of ring ditches just beyond the limits of the present-day heathlands, revealed as cropmarks, through an important campaign of aerial photography undertaken by Griffith since 1983 (Griffith 1999: 8). These ring ditches may be earthen barrows or, alternatively, traces of round houses. Although the A30 excavations revealed the presence of round houses with circular timber post settings, Bronze Age barrows or other evidence of funerary activity was almost absent from this lowland area.

The partial excavation of a ring ditch at land southeast of Broad Oak, Ottery St Mary, provided no positive evidence of the presence of a barrow (Mudd and Joyce 2014: 32–4) and it is probably also a round house. Recent excavations and geophysical survey have revealed the presence of a small sub-rectangular enclosure on the top of a low knoll just to the west of the river Otter at Colaton Raleigh dated to the Middle Bronze Age from residues on a biconical urn. This is an exceptionally earlydate for enclosures of this type in Devon (Farnell and Quinnell 2015). Further north along the Otter river valley other lowland Middle Bronze Age enclosures have been recorded at Patteson’s Cross and Castle Hill. The former contained a single round house supported by wooden posts. The latter was a much larger rectangular feature set within a system of rectilinear fields (Fitzpatrick et al. 1999). Thus although there is a complete absence of Bronze Age settlements on the heathlands contrasting with the surrounding lowlands to the east and the north, we know that Middle to Late Bronze Age settlements existed on the heathland fringe, suggesting seasonal use rather than settlement of the heathland area by Bronze Age inhabitants.

The Pebblebed heathlands appear to have been a reserved area for the construction of ritual structures and the burial of the dead from surrounding areas. There are, however, three exceptions to this general picture. One is the discovery of a pit at Salston, Ottery St Mary, on flat ground at the valley edge to the west of the river Otter. The pit was about 0.5 m in diameter and 0.2 m deep. It contained cremated human bone and mostly oak charcoal. The remains were of an adult woman with a radiocarbon date of 1948–1772 cal. BC. There were no associated grave goods or evidence of a mound or barrow above the pit (Mudd and Joyce 2014: 24). Excavations in advance of a new town at Cranbrook situated 8 km to the northwest of the heathlands revealed an Early Bronze Age ring ditch on top of a prominent knoll that may have surrounded a mound. It surrounded a central pit containing a beaker and a stone bracer. A ring ditch at Hayes Farm, Clyst Honiton, to the northwest of the heathlands, may represent another lowland Early Bronze Age earthen barrow. It was superseded by a rectangular enclosure with associated fields (Hart et al. 2014).

In addition to the prehistoric heathland cairns there are 14 late eighteenth- or early nineteenth-century landscaping mounds that resemble prehistoric barrows or cairns. At least two of the prehistoric pebble cairns have landscaping additions (see Chapter 10, Figure 10.1). Furthermore, there are numerous modern mounds of military origin, or dumps of farm or building materials, that under the current dense vegetation cover of gorse and heather cover, up to 2 m high, sometimes also resemble cairns (Figure 14.4). The area has been used for military training, and occasionally military occupation, for the last 200 years (see Chapter 11). This makes the process of recording pebble cairns of genuine prehistoric date based on field survey alone often difficult and frustrating. There may be many more than those recorded on the map: only a few of the ‘thirty spots’ (Carter 1936: 1) that Carter mentions that he excavated in the 1930s ‘of dimensions barely perceptible except on waste cleared by fire’ (Carter 1936: 1) have been possible to locate today despite intensive field walking of the area over a three-year period. He left no record or map of their precise location (see Chapter 2).

At least nine of the cairns have a surrounding ditch and in three cases this appears to have been a modern addition. The cairns vary in diameter from small, discreet structures 4–8 m in diameter to much more substantial cairns, three of which are over 20 m in diameter. Two of the largest cairns, including the very largest (32 m in diameter), appear today as flat-topped rather than rounded in profile but this is almost certainly the result of later landscaping activities, following which Scots pines were planted on top of them. The smaller cairns are rarely more than 1 m high, whereas the larger cairns vary in height between 1.5 and 3.5 m (Table 1.1; Figure 1.10).

Table 1.1The dimensions (height and diameter, in metres) and height above sea level (HASL) of the Pebblebed cairns and location and other notes.

No.

Height

Diameter

HASL

Location and other notes

 1

0.6

13.0

159

On summit of ridge with panoramic views. Ploughed out

 2

150

As above. Ploughed out

 3

1.0

3.8

110

Highest of a staggered row of seven cairns running down slope from NW–SE. on Venn Ottery Hill. Towards top of W (high)–E (low) slope. Above boggy area and spring line to E. Land drops steeply to E of cairn row that overlooks Otter valley. 18.5 metres NNE of 4

 4

1.2

7.5

110

7.5 m NW of 5

 5

1.0

7.0

105

5 m N of 6

 6

1.0

5.0

105

1 m N of 7

 7

0.7

4.0

105

57 m NW of 8

 8

1.0

4.0

100

24 m NW of 9

 9

0.6

3.8

100

24 m SE of 8

10

1.8

20.0

160

Edge of flat summit area of Aylesbeare Common. Surrounded by ditch. Land drops to south from which cairn skylined

11

1.5

25.0

160

Centre of Aylesbeare Common hill summit. Surrounding ditch map be a landscaping addition. Formerly planted with Scots pines. One dead trunk known as the Lone Pine, a local landmark, stood until 2009, when it fell in a storm

12

0.8

3.5

110

Twin Cairn A. Near top of gentle NW–SE spur with valleys to E and W. Excavated cairn (2011 Pebblebeds project)

13

1.0

4.8

110

Twin Cairn B. 7 m below Twin Cairn A

14

1.5

12.0

95

Great Tor Cairn. On top and in centre of spur with valleys to E and W. Overlooks enclosure (cropmark) site to SW. Excavated by Pebblebeds project in 2011

15

0.6

6.0

90

Tor Cairn. On western side of NW–SE sloping spur with valleys to E and W. Overlooks excavated enclosure (cropmark site) to SW. Excavated cairn (2008–9 Pebblebeds project)

16

0.4

2.6

90

Little Tor Cairn. 4.6 m south of 15. Excavated cairn (2009–10 Pebblebeds project)

17

0.5

7.0

120

Carter’s Woodbury ε. Centre of spur on SE-facing slope, valleys to E and W. Cairn excavated by Carter in 1930 and 1937–8

18

2.3

20.0

175

On high point on western edge of Pebblebed escarpment. Has added eighteenth-century landscaping ditch and planted with Scots pines

19

3.7

32.0

175

The Beacon. Situated on high point on western edge of Pebblebed escarpment. Modified as a fire beacon and has surrounding landscaping ditch. Planted with Scots pines

20

1.0

2.8

95

Carter’s ‘holy mound’. Approx. 20 m south of bog and stream source. Land dips to E and rises to W. On edge of Pebblebed heath beside arable fields. Excavated by Carter. Trench still visible in centre of cairn

21

3.0

16.0

110

On flat area with steep valley and stream to south. Cut on N side by Woodbury to Yettington road, near to a milestone on road edge. Excavated by Carter in 1960. Trench still visible

22

1.5

5.0

90

In valley bottom in boggy ground

23

1.5

5.0

90

In valley bottom in boggy ground. About 50 m to E of 22

24

1.8

23.0

140

Jacob’s Well. Burnt mound. Oval-shaped over spring at foot of western edge of Pebblebed escarpment in bog. Land rises steeply to E. Excavated by Carter in 1938, whose central trench survives, and in 2010 by Pebblebeds project

25

1.3

8.0

150

Western end of a row of three adjacent cairns. Near end and in centre of gently sloping W–E ridge. Surrounding ditch

26

1.7

8.0

150

In middle of row. Surrounding ditch

27

2.0

8.0

150

At E end of row. Surrounding ditch

28

0.4

5.0

110

On flat low area with stream in valley to north

29

0.4

5.0

110

As 28

30

1.5

9.0

120

In centre of sloping spur above Bystock stream which runs in deep valley to N. Dry valley to W. Rise to S. Drop to E. Dip to N and W. This is at N end of a row of three cairns that align NE–SW

31

0.8

2.0

120

As above

32

0.8

5.8

120

S end of row

Note: Cairn numbers refer to Figure 1.10.

The small cairns occur in the middle of low sloping spurs bounded by valleys. They typically occur on sloping ground and do not command panoramic views. The larger ones were clearly intended as monumental constructions punctuating and marking the landscape and visible for long distances. They occur on ridge tops and localized high points, so there is an important association between cairn size and height. The small cairns (see Chapter 3) are of later Beaker/earlier Bronze Age date c. 2000–1700 BC). In terms of Needham’s (1996) revision of British Chalcolithic and Bronze Age chronology used as a general chronological framework in this book, they all fall into the early part of his period 3. The very largest cairns are in all probability (none have been excavated) of Early Bronze Age date, period 4 according to Needham’s scheme (c. 1700–1500 BC) (see Chapters 3 and 8). Thus through time the cairns move up in the landscape and assume a monumental form.

The large ridge-top cairns all occur in the western and northern areas of the overall distribution. Cairns 18 and 19 (Figures 1.10 and 1.11; 1.12) are both situated on the edge of the steep western scarp slope, are unusual in that they can be seen skylined on the horizon far away to the west, from both the Exe valley and from the top of the Haldon and Raddon Hills. They punctuate the skyline and must have been located so as to be highly visible landmarks when seen from the west or the northwest. These cairns are also visible from long distances away to the east and can be seen from the East Hill and Peak Hill ridges. They also have the highest degree of intervisibility with others on the Pebblebed heathlands (Figure 1.13). Other large cairns are sited in the landscape so as to be most visually impressive when seen from long distances away only from the east. Few can be seen from more than a short distance away to either the north or south. Some groups consisting entirely of small cairns such as those on the slopes of Venn Ottery Hill and others on Bicton Common and Withycombe Raleigh Common are not intervisible with any others, whereas those in the southeast in Colaton Raleigh Common are only locally intervisible.

Six of the cairns occur singly (Figure 1.10: 14, 17, 18, 19, 20, 21). There are five pairs and three groups of three or more cairns. As a whole these cairns occupy every major topographic situation in the landscape:

1.highest points on the western escarpment (18, 19);

2.flat ridge summits (1, 2, 10, 11);

3.in the middle of and towards ends of sloping ridge tops (25–7);

4.upper sloping sides of ridges (28–9; 30–2);

5.on low sloping spurs between valleys (12–17);

6.on upper slopes of valley sides (3–9);

7.in valley bottoms and bogs (22–4).

The earliest and small dated cairns are all found on spurs (location 5 above). The very largest and presumed later cairns are on the highest points of the scarp edge and hill summits. Small cairns are located on sides of ridges, upper slopes of valley sides and valley bottoms. The close associations of these cairns with valleys and/or water sources is strong. Cairn 18, although situated on the western escarpment edge, is also set just to the north of a shallow valley that gives birth to a stream. Cairns 22–3 are set almost at the bottom of the head of another stream valley above a substantial boggy area. Cairns 12–17 are all on southeast sloping spurs between valleys and near to the source of streams. The cairns, as a whole, then, occupy both high and ‘dry’ locations in the landscape and are associated with water and valleys that give birth to streams running in beds of pebbles. The cairns are all associated with streams draining the Pebblebed heathlands that flow east or southeast to join the river Otter in its passage to the sea. There are only a few barrows/ring ditches known from the marls due west of the heathlands between them and the river Exe. Others cluster in the vicinity of Exeter along the Exe valley itself to the northwest (Griffith and Quinnell 1999c: map 6.5). The cairns on the Pebblebed heathlands are linked with each other and the Otter by valleys and streams that have their sources near to or beside them.

The Exe estuary to the west of the Pebblebed ridge is a wide and shallow valley of muds and shifting sands (Figure 1.14). The Otter valley, by contrast, is a valley of pebbles and gravels. Along its course it mixes and combines pebble material washed down from the heathlands and more jagged flints and cherts from the East Hill and Peak Hill ridges. It flows beneath Dumpdon Hill, and its northern tributary, the Tale, is born, or has its source, on the eastern side of the spur occupied by the Early Neolithic Hembury causewayed enclosure. The Otter flows to the east of High Peak with another Neolithic enclosure, before entering the sea near to the east of the cliffs at Budleigh Salterton, where the Pebblebeds are most dramatically exposed.

The sea, to the south, is visible from most of the cairn locations. The Peak Hill and East Hill ridges flanking the Otter valley to the east are visible from all but a few (Table 1.2). A series of larger and smaller flint cairns once crowned the tops of these ridges, but because of afforestation only a couple now survive on East Hill (see Chapter 3). Grinsell (1983) records the former presence of at least 6 from the Peak Hill ridge and 14 running along the spine of East Hill. There are extensive views from these ridge tops across the Otter valley. Some of these cairns on these ridge spines would certainly have been visible from almost all the pebble cairns on the heathlands below them. By contrast the Haldon Hills and Hembury are visible from only those cairns situated on ridgetop locations or along the western scarp of the heathlands. None occur on the Hembury spur. On the Haldon Hills there are at least twenty-six small cairns (Grinsell 1983: 13; Finneran and Turner 2003: 242–3). Because of their small size, distance and their specific locations (mostly on the upper and western slopes of Little and Great Haldon), none of these are visible from the cairns on the Pebblebed heathlands. All these cairns running along the East Hill, Peak Hill and Haldon ridges, constructed of angular and dull materials, would have made a striking visual and tactile contrast with the brightly coloured Pebblebed cairns, perhaps objectifying in their material form different social identities and relationships to the east of the Otter and to the west of the Exe: differing landscapes and social worlds (see Chapters 3 and 8).

Table 1.2The visibility of principal hills and ridges from the Pebblebed cairns.

No.

High Peak

Peak Hill Ridge

East Hill Ridge

Dumpdon Hill

Hembury

Haldon Hills

Raddon Hills

Sea

1–2

+

+

+

+

+

+

+

+

3–9

+

+

+

+

+

+

10–11

+

+

+

+

+

+

+

+

12–13

+

+

+

+

+

14–16

+

+

+

+

+

17

+

+

+

+

+

18

+

+

+

+

+

+

19

+

+

+

+

+

+

+

+

20

+

+

21

?

+

22–3

+

24

+

+

25–7

+

+

+

+

29–32

?

+

Notes:Cairn numbers refer to Figure 1.10.

? = Probable but trees now block view.

Ancestral hills and the birth and death of the sun

It is striking that High Peak is visible from almost all the pebble cairns on the heathlands, whatever their position in the landscape. Given the presence of the Neolithic occupation and probable causewayed enclosure on its summit, this peak was likely a hill of paramount significance to the Bronze Age populations living in the vicinity of the Pebblebeds. It is situated to the east, southeast or east-southeast of all the cairns and may be suggested to have been associated with the rising sun.

Celestial events and in particular the birth or rising of the sun in the east and its setting or death in the west were, and still, are, an important part of the experience of the Pebblebed heathlands. The sun at the midsummer solstice rises to the northeast towards the northern end of the East Hill ridge. It then slowly slips to the south along the ridge. Sunrise at the spring and autumn equinoxes would first be visible through the Sidmouth gap between the Peak Hill and East Hill ridges and visible from some of the cairns (Table 1.3). The ridges on either side frame and intensify the effect. The sunrise can be seen most dramatically at this time of year from the large summit cairns on Aylesbeare Common (Figure 1.15; Figure 1.10: 10–11). Thereafter the sun slips down the Peak Hill ridge and shines through the gap between the Peak Hill and High Peak in the late spring and autumn. By midwinter it rises to the southeast out of the sea just to the west of High Peak.

Table 1.3The visibility of the ridge and hill gaps from the cairns and the main directions from which they look most impressive.

No.

Sidmouth Gap

Peak Hill Gap

Honiton Gap

Most impressive from

1–2

+

+

+

n/a; destroyed

3–9

+

+

+

East but small

10–11

+

+

+

East and 10 to south

12–13

+

+

+

East but small

14–16

+

+

+

14: east or west; 15–16 very small

17

+

+

+

East or west, but small

18

+

+

+

East or west

19

+

+

+

East or west

20

+

Very small

21

+*

South

22–3

+

n/a; in valley bottom

24

n/a; in bog at base of scarp

25–7

+*

+*

+*

South or north*

29–32

+*

West but small

Notes: Cairn numbers refer to Figure 1.10.

* = extrapolated because of presence of modern plantations.

At the midwinter solstice the sun sets in the southwest behind the Haldon Hills in the dip between Great Haldon to the north and Little Haldon to the south. At the spring and autumn equinoxes it sets due west and can be seen dipping down behind Great Haldon and the high peaks of Dartmoor beyond, an event visible from the large pebble cairns situated on the western scarp edge. At midsummer the sun, having passed along the line of the Haldon Hills, sets to the northwest in the gap between the Raddon and Blackdown Hills (Figure 1.15).

The only prehistoric cairns from which all these celestial events could be seen are the two summit cairns on Aylesbeare Common (the view is now blocked in some directions by pine plantations) and the largest cairn on the heathlands known as the Beacon (Figure 1.10: 19). From other cairns such as Tor Cairn and Little Tor Cairn in the southeast of the heathlands or those on the side of Venn Ottery Hill, while the rising of the sun over the hills surrounding the heathlands is visible, its setting is not because of the manner in which the heathlands rise to the north and the west. The fact that the rising of the sun is the most dramatic event in relation to the surrounding landscape and that this can be seen from all the cairns, whereas the setting of the sun can be seen only from a few suggests that a view of the rising sun was of particular importance in relation to the locations of cairns in the landscape.

During the course of the year the sun effectively moves back and forth (north and south), rising over the East Hill and Peak Hill ridges and between them at the equinox. It moves back and forth (north and south) along the line of the Haldon Hills to the west, setting over them at different points during the course of the year. The only time when it is seen to be born from the sea to the south is at midwinter sunrise. At this time of the year it has a ‘wet’ birth. Seen from the visual perspective of the heathlands it always has a ‘dry’ death. This may be of great significance in relation to fire rituals taking place at the prehistoric cairns (see Chapter 3).

The presence of these three gaps to the east of the barrow distribution thus points to the significance of the rising sun as seen from the cairns at significant points during the year. The gaps through the hills effectively served to frame and thus dramatize and animate these important celestial events and the brilliant changes in the colour of the sky from red to yellow. By contrast, the setting sun in the west over the Haldon Hills, visible from relatively few of the barrows, is not framed by any dramatic gaps. The Raddon Hills, with their Neolithic causewayed enclosure, may have represented another, more distant place of ancestral significance. Situated to the northwest of the barrows they might have been associated with the setting of the sun on the summer solstice. However, the effect would not have been dramatic and was visible only from a few of the barrows (Table 1.2).

Dumpdon Hill, despite its quite considerable distance from the cairns, about 20 km away, is visible from a surprising number of them (Table 1.2). This, like High Peak, is a hill island situated in the middle of the Honiton Gap. As is the case with High Peak, the river Otter runs beneath it, but to the west rather than the east. Dumpdon Hill is, like Hembury and High Peak, crowned by a hillfort. This is a very likely location for another Neolithic hilltop enclosure. Like High Peak, this hill may have had an especial ancestral significance for the Bronze Age pebble cairn builders.

Share