The development of monumental landscapes in many parts of Western Europe in the third millennium BC, characterised by the construction of structures in earth, stone and timber, is argued to represent the physical manifestation of significant social change within prehistory (Bradley 1998; Thomas 1999; Tilley 2004). The construction of monuments represents either a change, or formalisation, in the relationships between place, space and society, and establishes locations for key social practices and performances (Cummings 2008; Gillings et al. 2008). Stone rows and settings are amongst the least understood monument types to be found in north-west Europe. Their form and scale can be visually impressive, and a number of examples appear to be part of larger ceremonial complexes; however, despite their ubiquity in parts of western Britain, Ireland and Brittany (Burl 1993), some of the basic understandings and premises concerning these monuments (e.g. function and chronology) are currently unclear.
Dense concentrations of stone rows and settings can be found on the uplands of south-west England, and the various rows and settings on Dartmoor, Bodmin Moor and Exmoor have been the subject of surveys and gazetteers which have detailed the shape and topographic positioning of monuments (e.g. Emmett 1979; Johnson & Rose 1994; Butler 1997; Riley & Wilson-North 2001). Although various functional hypotheses have been proposed, there is little consensus as to their chronology or the role they played within prehistoric society. It has been argued that rows may have provided guides to astronomical alignment (Lockyer 1906; Ruggles 1985), played a mnemonic function to confirm or establish the location of an individual within their known world (Tilley 1995), marked different land use zones (Barnatt 1982; Johnston 2005), or acted as revelatory monuments, leading the individual to key visual locations that reference or reveal other parts of the landscape (Herring 2008). It is not the purpose of this paper to review or critique these suggestions; rather, it is to consider one of the more fundamental uncertainties pertaining to stone rows; the question of chronology. The age of these monuments is currently unclear. None have been directly dated, and chronological assumptions typically rely on association with other monument types. In 2004 a new stone row was discovered on Cut Hill on central northern Dartmoor that has potential to shed crucial light on the dating of stone rows. This paper presents the results of survey and trial excavation at Cut Hill designed expressly to recover dating samples, and considers the significance of the results for the wider understanding of stone rows on Dartmoor and the wider region.
Stone rows: form and assumed chronologies
In his comprehensive survey of stone rows from Western Europe, Burl provides a model for the chronology of this type of archaeological site (1993), dividing them into three main groups. Long single rows, with more than seven stones, are broadly assigned to the period 2100-1600 cal BC. Multiple rows, with parallel or fan-shaped arrangements, are thought to date to the period 3000-1500 cal BC. Short rows, with six or fewer stones, are broadly assigned to the period 1800-1000 cal BC. The basis for this proposed chronology of stone rows is, however, poor, with little or no secure, independent, scientific dating control. Chronologies are dependent upon one or more of: (a) astronomical dating, tied to interpretative frameworks of monuments' functions (e.g. Ruggles 1985), where rows are assumed to relate to particular stellar alignments whose position in prehistory can be calculated; (b) stratigraphic or (more usually) spatial associations with other datable archaeological features, including artefacts and structures; or (c) general typological analogy across regions, which inevitably becomes circular in the absence of secure, independent dating.
The majority of stone rows in south-west England fall into the first of Burl's groupings-long stone rows--although there are also a number of multiple rows. There are three such rows on the Isles of Scilly (English Heritage Archaeological Item Database nos. 147091, 139488 and 130489), at least nine in mainland Cornwall (Johnson & Rose 1994; Herring 2008), eight on Exmoor (Riley & Wilson-North 2001) and, although the precise number is contested, about 80 on Dartmoor (Figure 1). None of these stone rows has had, until now, reliable dates on any of the component elements. Probably the most securely dated row in Cornwall is that described by Herring (2008) on Searle's Down, Bodmin Moor. The row itself is not dated directly; in 1977-78 three cairns were excavated in advance of the construction of a reservoir (Griffith 1984). One of the elements of the stone row was sealed by the outer ring of one of the excavated cairns. The relationship between the inner ring of the cairn and the orthostat of the row was such that the excavator believed there was no break in activity at the site. A radiocarbon date from under the inner ring of the cairn made on oak charcoal calibrates to 2040-1620 cal BC (Griffith 1984; Jones 2005), providing a best-estimate date for the row at around the early second millennium BC.
Relationships between stone rows and Bronze Age enclosure on Dartmoor are enigmatic, although Butler (1997) has argued that relative dating between rows and the extensive field systems may be possible. The chronology of enclosure on Dartmoor has been widely explored by Andrew Fleming (see for example 2008). He argues that the construction of Bronze Age field systems and territorial boundaries (reaves) dates to around 1700-1600 cal BC. Although there is still debate around the process and chronology for the construction of the boundaries (see Johnston 2005), this broad date seems likely and is supported by palaeoecological work (Fyfe et al. 2008). Whilst some rows undoubtedly do predate territorial boundaries, where rows and boundaries do run parallel to each other it is practically impossible to determine relative age relationships. If stone rows were markers to subdivide different land use zones in the later Neolithic (Barnatt 1982; Johnston 2005), then it should come as no surprise that later boundaries are in broadly the same place, assuming continuity of land use.
To date, attempts to resolve the question of chronology have assumed a single event horizon for stone rows, where it has been tacitly accepted, though not proven, that rows were erected at broadly the same time and that all the orthostats in any given row were positioned as part of a single episode of construction, rather than monuments developing by accretion. This may have significant implications for understanding the age of the monuments when their chronology is reliant on the association of a particular stone with other structures. Their survival implies that they were rarely robbed or slighted as a result of the construction of field systems in the middle Bronze Age, suggesting that they may have had considerable longevity as monuments, albeit experienced and used differently by subsequent generations and groups (Bradley 2002).
The Cut Hill stone row
In April 2004 an alignment of stones was recognised on Cut Hill, northern Dartmoor (Figure 1), at National Grid Reference SX 59928275 at a height of approximately 600m OD (Greeves 2004a & b). It was adjacent to a previously unrecognised barrow. The Cut Hill row consists of nine granite slabs so far recorded, with a length of about 215m (Figures 2 and 3). The axis of the row is approximately ENE/WSW, which is an orientation followed by the majority of Dartmoor rows (Butler 1997: Figure 171). It is located in an area of spatially extensive blanket peat, up to 1.8m thick in places, which overlies the granite that forms the whole of the upland area. There is considerable erosion of the peat around Cut Hill. Of particular interest is that all of the nine stones so far recorded at Cut Hill are lying flat, with seven of them lying on a similar orientation, approximately ESE/WNW. The northernmost stone lies NE/SW and the southernmost approximately N/S. This raises the question as to whether the stones were purposely laid out, and indeed whether they were ever erected. Stone 1 appears to have a cluster of packing stones around its WNW end (Figure 4), though these are slowly being dislodged and dispersed through peat erosion. If interpretation of the packing stones is correct, this stone at least was once upright and set in peat, and we can infer that perhaps the others were too. Four of the stones lie within a 100m-wide strip of ground where the peat has been totally removed, probably largely by human agency since medieval times when peat was cut for charcoal production (Fox 1994), exposing an old, presumably prehistoric, land surface. The most south-westerly visible stone lies on eroding peat and this was the subject of the first sampling for radiocarbon dates (see below). In 2007 a previously unrecorded stone was located some 25m further to the south-west through probing within the extant peat, and was the subject of the second sampling programme (see below). It is not unreasonable to expect the discovery of further stones of the row buried in the peat, or indeed other stone features associated with it. However, the orientation of the stones at the south-west and north-east ends of the row is different from that of the other stones discovered so far and this may suggest that the two ends of the row have possibly now been determined. The barrow sits on top of one of the peat hags on Cut Hill rather than the exposed, prehistoric land surface, suggesting considerable temporal discontinuity with the stone row.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
Dartmoor rows are diverse in terms of design, length, size of stones, spacing and location; however, there are several unusual aspects of the Cut Hill row. It is located on one of the highest hills of Dartmoor, in the heart of the northern moor, and lies 100m above the previously recorded highest row (at Conies Down), in an area otherwise devoid of known prehistoric structures (within 2.6km of the site), with the exception of the newly-recognised barrow. This dearth in prehistoric activity most likely reflects the generally extensive covering of blanket peat across this part of the upland and is thus ah artefact of visibility rather than a genuine absence of activity. The spacing of the stones on Cut Hill is much greater than elsewhere on Dartmoor, being between 19m and 34.5m apart. The relatively large size of the stones, too, is unusual, being between 1.53m and 2.6m in length, and between 0.5m and 1.2m in width (Table 1), with an approximate thickness of 0.20m, giving a thin slab-like appearance. Only two other rows on Dartmoor have a significant number of stones of comparable length/height. There are six large stones at the northern end of the Stalldown row on southern Dartmoor which stand 1.9-2.6m above the ground surface, and the Piles Hill row, also on southern Dartmoor, has fallen stones averaging 1.6m in length (Butler 1997).
Three sections were opened within the peat on Cut Hill with the primary aim of recovering peat samples for dating and palaeoenvironmental analysis (Figures 2 and 4). Section 1 (0.52m of well-humified peat) was excavated from underneath stone 1 in 2005 to recover material for dating the stone and to provide a palaeoenvironmental context for the monument. Section 2 (1.84m deep) was excavated from an exposed section of eroding peat hag 5m west of section 1 in 2005, to provide a long, undisturbed record of vegetation and climatic change in the region (analysis in progress). Section 3 was taken from the face of a 1.0 x 0.7m slot trench that exposed the end of the buried stone, which was located by probing 25m to the SW. The intention of this excavation was to recover dating material for the stone without exposing any more of the sealed part of the stone than was necessary, and hence packing stones were neither sought nor identified. Sections 1 and 3 were undercut below the stones (and in the case of section 3, overcut above) to ensure that samples had a direct stratigraphic relationships with the stones; all sampling was undertaken using overlapping monolith tins from cleaned faces.
[FIGURE 4 OMITTED]
Radiocarbon dating results
Two samples for AMS radiocarbon dating were taken from section 1 and both the humin and humic acid fractions of these samples were dated at the SUERC radiocarbon facility. These were positioned between a point 2cm below the stone at the top of the section, and the base of the section. Two samples for AMS radiocarbon dating were taken from section 3 positioned immediately above and below the stone within the section, with both the humin and humic acid fractions dated at the [sup.14]C Chrono laboratory at Queen's University Belfast. Ali AMS samples comprised 1cm slices of peat. The results of the analyses are presented in Table 2, with dates presented as calibrated ages BC/AD using the CALIB5.0.2 program (Stuiver & Reimer 1993).
The dates from section 1 indicate that peat development on Cut Hill began during the early to mid-sixth millennium BC, although there is a disparity between the humic and humin acid fractions. Shore et al. (1995) present various scenarios under which both the humic and the humin fractions of a sample can be affected by post-depositional processes resulting in a poor age estimate, including intrusion of fine rootlets from younger plants (leading to a 'young' humin date), decay of fine rootlets (leading to a 'young' humic date), net transport of water-soluble organics down profile (leading to a 'young' humic date), and incorporation of older, reworked, carbon (leading to an 'old' humin date). The position of the section on a plateau high on top of Dartmoor makes inwash of older carbon unlikely, and as a result, the humin fraction in this instance is considered a more reliable estimate of the true age of the peat. It remains possible, though, that the younger dates are correct.
The date from under stone 1 implies that it was in its final position no earlier than 3700-3540 cal BC. There is a 2cm gap between the underside of the stone and the radiocarbon sample from under the stone. The accumulation rate between the radiocarbon dates is between 34 and 50 years [cm.sup.-1] (the wide range given by the difference between the humic and humin acid fractions from the basal date). This means that the stone may have reached its final position between the thirty-seventh and thirty-fifth centuries BC, assuming that no peat was removed or disturbed during this process. The dates from above and below the stone in section 3 show a similar disparity to the basal date from section 1. The date from the humin fraction from the sample immediately below the stone demonstrates that the stone was in its final resting position no earlier than the 3347-3100 cal BC, that is, between the thirty-fourth and thirty-second centuries BC. The date for the sample from above the stone indicates that this element of the row was sealed within the peat unit by 2476-2245 cal BC, that is, the twenty-fifth to twenty-third centuries BC.
Palaeoenvironmental context of the Cut Hill row
Eight 1.0[cm.sup.3] sub-samples were taken from section 1 for pollen analytical work to provide a landscape context for the monument (Figure 5) and in addition spot macrofossil analysis was undertaken. Pollen samples were prepared using standard procedures (see Moore et al. 1991). A minimum of 300 land pollen grains (including Cyperaceae) were identified flora each level, and charcoal fragments were counted from each level in two size classes (10-50 microns, 50-100 microns). Four pollen zones were differentiated visually, and are described in Table 3.
The pollen assemblage from the base of the section 1 (CH1-lpaz1 : Figure 5) shows that the local environment was dominated by heather-dominated heath (Calluna). This is supported by macrofossil analysis which recovered a heathland community assemblage dominated by hare's-tail cottongrass (Eriophorum vaginatum), at least seven different sedges (Carex sp.), bell heather (Erica cinerea), heather (Calluna vulgaris), bugle (Ajuga reptans) and tormentil (Potentilla erecta). Although the levels of arboreal pollen within the sequence are low, other pollen sites on Dartmoor at lower elevations show a higher level of pollen of trees and shrubs. Pinswell, at 460m OD, lay within the tree line at around 5000 cal BC (Caseldine & Hatton 1993). The arboreal taxa in the pollen record at Cut Hill will represent pollen sourced from within the tree line on lower slopes, and although it is not possible to determine the precise location of this tree line, it is possible to imagine 'islands' of open heath above a mostly wooded landscape.
[FIGURE 5 OMITTED]
The changes between zones CH1-lpaz-1 and CH1-lpaz2 (Table 3, Figure 5) suggest a local shift from heather-dominated heath to a mosaic of grass, sedge and heather heath. Although arboreal types increase, this is likely to be an artefact of the percentage method used to present the data rather than any real expansion of woodland taxa. Calluna produces a greater amount of pollen than either grasses or sedges (Brostrom et al. 2008), thus if Calluna declines, there will be a decline in the absolute levels of pollen being deposited at the site. In relative terms, this will increase the proportions of arboreal pollen (Fyfe 2006). The highest sample in the sequence (CH1-lpaz4) is broadly similar to preceding levels (Figure 5), although it contains elevated levels of arboreal taxa. It is possible that this level, on an eroding peat surface, includes elements of contamination from either modern pollen, or pollen eroded from within the peat system in recent times. As a consequence, interpretation of this single sample is neither simple, nor reliable.
The chronology of the monument
The dates obtained from underneath and above the two stones within the Cut Hill row are open to two interpretations which are not resolvable from the evidence presented here. These interpretations hinge on resolving the original form of the site itself, and crucially centre on: (i) whether the stones were laid out in a recumbent position, or were originally upright and have toppled (or been placed) into their current positions; and (ii) whether the stones themselves are contemporary elements placed as a single constructional event. It is assumed that the peat is intact around the stones, i.e. there was no removal of peat, or preparation of the surface which would have damaged the integrity of the peat stratigraphy; there was no evidence for disturbance in either of the sections which were cleaned and recorded meticulously.
If the stones were originally upright, then the dates from underneath the stones represent the date at which they fell into their final positions, i.e. they represent a terminus ante quem and the stones were erected at a time before the thirty-seventh and thirty-fifth centuries BC (stone 1) and thirty-fourth and thirty-second centuries BC (stone 9). If the stones were laid out in a recumbent form then these dates become a terminus post quem, although given the supposed integrity of the peat, these dates should be very close to the 'true' date of the placement of the stones. The difference in the dates between the two stones does little to help to resolve the two possible models. The hypothesis that the stones were placed upright could be supported by different 'toppling' dates, several hundred years apart; the hypothesis that the stones were placed recumbent could be supported by a model of incremental or piecemeal positioning of stones over centuries, rather than construction as a single event. Only one of the exposed stones has anything which may resemble packing stones around it (stone 1). The answer to this conundrum may lie within the extant peat around the submerged stone. The date from above the buried stone is simpler to interpret than those from below. This stone, at least, became submerged in the peat by the twenty-fifth to twenty-third centuries BC, and it is probable that the others were sealed within peat at the same time.
This latter date is around the period of time to which most authors attribute the start of construction of linear monuments on the uplands of the south-west and elsewhere (Burl 1993; Butler 1997; Jones 2005). But it here seems irrefutable that this particular row is of fourth-millennium BC date (even if the humic dates from under stone 9 are preferred) and hence Neolithic rather than early Bronze Age, and whichever form model is preferred (upright or recumbent), most likely to belong to the earlier, rather than later, Neolithic. Further dating control from future excavation may allow for formal modelling of these informal date estimates to refine the chronology of the site further.
Establishing a chronology for an unusual stone row on the top of Dartmoor does not demand the wholesale rejection of a body of literature and ideas concerning the dating and function of the more 'typical' linear stone monuments (if there is such a thing as a typical stone row) within the wider region, and these may still date to the late Neolithic/earlier Bronze Age as argued elsewhere (Burl 1993; Fleming 2008; Herring 2008). It is clear, however, that there is now an earlier Neolithic precedent to the linear stone monument. This may fit well within a tradition of linear monuments in the early Neolithic, with monuments such as the Avenue at Raunds (dated to 3860-3620 cal BC: Harding & Healy 2008), early fourth-millennium BC split timber settings in Scotland (e.g. Lochhill and Pitnacree: Noble 2006) and late fifth-millennium BC linear monuments in northern France (Kinnes 1999), followed by the construction of cursus monuments from about 3500 cal BC (Barclay & Harding 1999). Refinement of the chronology of earlier Neolithic monuments is ongoing, but a general model has been produced by Whittle et al. (2008). They describe an initial Neolithic with no or little monument construction (with deposition in pits, e.g. Jones 2005); increasing emphasis on marking the dead in local social groups (reflected in long barrow construction) around 3800 cal BC; and larger congregation and construction of enclosures within southern Britain around 3700 cal BC (Oswald et al. 2001). The dating presented here suggests that the stone row on Cut Hill is most likely to fall into the latter period described by Whittle et al. (2008). The dating of this monument may also in turn lend support to evidence for an earlier Neolithic standing stone and row tradition in Brittany, where some stones carry early Neolithic carvings (e.g. Kermarquer at Moustoirac, Morbihan) and decorated standing stones have been incorporated into chambered tombs, as at Mane Rutual, La Table des Marchand and Mane-er-Hroek (L'Helgouach 1983; Patton 1993).
Local landscape contexts of the row
The location of the monument on Cut Hill is remarkable in a modern sense, as it lies in the most remote and inaccessible location within the region. It should be remembered, though, that this sense of remoteness and inaccessibility is a modern cultural construct and almost certainly does not reflect the views or approach to the landscape held by earlier Neolithic groups. It has become clear that later Mesolithic groups had the ability to manipulate their environment through the use of fire to support resource exploitation strategies, on Dartmoor and elsewhere (Caseldine & Hatton 1993; Bell 2007). Locally, this will have resulted in the mosaic landscape described within the Cut Hill pollen sequence (Figure 5), with earlier Neolithic clearance of lowland river valleys (Fyfe et al. 2003); whilst fern-rich woodland persisted on the fringes of the upland (Fyfe et al. 2008).
The picture that emerges from palaeoecological sequences across the region is an earlier Neolithic during which there is unlikely to have been any clearly definable upland moorland; rather, there would have been patches of open heath or bog within a generally wooded landscape. The Cut Hill row was constructed on one such open heath, and the people or groups who came to Cut Hill are likely to have travelled on paths within broadly wooded country. Emerging into open areas free from trees may have contributed in part to a feeling, or recognition, of place, different from other locales in the region, as has been argued for earlier periods (e.g. Davies et al. 2005; Fyfe 2007). This may in turn have contributed to the selection of this particular location for the construction of this monument. Although the form of the monument is unparalleled on Dartmoor, the marking of these types of locations, and in the wider landscape, is not. Bender et al. (1997) have argued for the importance of tors and distinctive hill tops within the routines of Neolithic life and this is most clearly reflected in the earlier Neolithic by the certain tor enclosures at Carn Brea (Mercer 1981) and Helman Tor (Mercer 1997), and the location of other possible, unconfirmed, enclosures in the south-west (Oswald et al. 2001). Cut Hill itself is a very striking, massive and gently rounded hill, without any significant rock outcrops to break its skyline silhouette. It lies at the heart of upland, at the head of the watersheds of the East Dart and the River Tavy. The choice of Cut Hill to locate this row seems significant, although it is not apparent why, without a clear understanding of the original purpose or function of the stones.
[FIGURE 6 OMITTED]
An isolated monument?
The stone row at Cut Hill is at present unusual in both form and geographic location. There may be no other comparable stone rows on Dartmoor, bur there are rows on Bodmin, to the west, that are more like Cut Hill in form (Figure 6). These Cornish rows are divided into two main types: short rows, with closely-spaced small stones; and long rows with larger stones with longer spacing (Johnson & Rose 1994). For example, the Nine Maidens row at St. Columb has stones averaging 1.6m in height above ground level, at Cardinham Moor 1.3m, and at East Moor 1.0m. These three mainland Cornish rows have relatively wide spacing between the stones, averaging between 10m and 14m (Johnson & Rose 1994), although these distances are rarely uniform (Figure 6). If the Cut Hill row conforms to this Cornish long row typology, it is possible to suggest that these Cornish rows may also be significantly earlier in date than current thinking suggests.
Radiocarbon dating of peat underneath two elements of the stone row on Cut Hill, and a date from above one of the elements, has provided the first secure, independent, date for a stone row within the wider region. The dates imply the laying out of the row during the second half of the fourth millennium BC. This age for a stone row on Dartmoor is significantly older than the assumed age of such monuments, which are usually argued to date to a period from the late third to the middle of the second millennium BC. The row on Cut Hill is a typological outlier within the stone row tradition identified on Dartmoor, and as such, the age of this row may not be a reflection of that of the more typical rows on Dartmoor. There are, however, morphological parallels with (undated) rows found to the west in Cornwall. An earlier Neolithic date for some stone rows in the south-west adds an additional monument type to the earlier Neolithic 'package' of long barrows, cursus, causewayed and for enclosures that are known from across southern England, and lends support to early dates for a number of standing stones within Brittany.
This research was funded by Dartmoor National Park Authority and English Heritage (RF). The authors are grateful to the Exeter English Heritage office for permission to use their plan of the row, and to Chris Chapman for the use of Figure 3. The authors would like to thank Derek Collins, Katie Head, Richard Hartley, Bill Radcliffe and Vanessa Straker for assistance during fieldwork. The authors are grateful to Andrew Fleming, Peter Herring and Jane Marchand for valuable discussion, to Mark Gillings and Heather Adams for comments which improved an earlier draft on this paper, and the referees (Emmanuel Mens and Alasdair Whittle) for their constructive comments.
Received: 13 March 2009; Accepted: 5 May 2009; Revised: 12 June 2009
BARCLAY, A. & J. HARDING. 1999. Pathways and ceremonies. The cursus monuments of Britain and Ireland. Oxford: Oxbow.
BARNATT, J. 1982. Prehistoric Cornwall: the ceremonial monuments. Wellinborough: Turnstone Press.
BELL, M. 2007. Prehistoric coastal communities: the Mesolithic in western Britain (Council for British Archaeology Research Report 149). York: Council for British Archaeology.
BENDER, B., S. HAMILTON & C. TILLEY. 1997. Leskernick: stone worlds; alternative narratives; nested landscapes. Proceedings of the Prehistoric Society 63: 147-78.
BRADLEY, R. 1998. An archaeology of natural places. London: Routledge.
--2002. The past in prehistoric societies. London: Routledge.
BROSTROM, A., A.B. NIELSEN, M.-J. GAILLARD, K.
HJELLE, F. MAZIER, H. BINNEY, M.-J. BUNTING, R.M. FYFE, V. MELTSOV, A. POSKA, S. RASANEN, W.
SOEPBOER, H. STEDINGK, H. SUUTARI & S. SUGITA. 2008. Pollen productivity estimates--the key to landscape reconstructions. Vegetation History and Archaeabotany 17: 461-78.
BURL, A. 1993. From Carnac to Callanish: the prehistoric stone rows and avenues of Britain, Ireland and Brittany. New Haven (CT): Yale University Press.
BUTLER, J. 1997. Dartmoor atlas of antiquities. Volume 5, the second millennium BC. Tiverton: Devon Books.
CASELDINE, C.J. & J. HATTON. 1993. The development of high moorland on Dartmoor: fire and the influence of Mesolithic activity on vegetation change, in F.M. Chamber (ed.) Climate change and human impact on the landscape: 119-31. London: Chapman & Hall.
CUMMINGS, V. 2008. The architecture of monuments, in J. Pollard (ed.) Prehistoric Britain: 135-59. Oxford: Blackwell.
DAVIES, P., J.G. ROBB & D. LADBROOK. 2005. Woodland clearance in the Mesolithic: the social aspects. Antiquity 79: 280-8.
EMMETT, D.D. 1979. Stone rows: the traditional view reconsidered. Proceedings of the Devon Archaeological Society 37:94-114.
FLEMING, A. 2008. The Dartmoor Reaves: investigating prehistoric land division. Oxford: Windgather.
Fox, H.S.A. 1994. Medieval Dartmoor as seen through its account rolls. Proceedings of the Devon Archaeological Society 52:149-71.
FYFE, R.M. 2006. GIS and the application of a model of pollen deposition and dispersal: a new approach to testing landscape hypotheses using the POLLANDCAL models. Journal of Archaeological Science 33: 483-93.
--2007. The importance of local-scale openness within regions dominated by closed woodland. Journal of Quaternary Science 22: 571-8.
FYFE, R.M., A.G. BROWN & B.J. COLES. 2003. Mesolithic to Bronze Age vegetation change and human activity in the Exe Valley, Devon, UK. Proceedings of the Prehistoric Society 69: 161-81.
FYFE, R.M., J. BROCK, R. JOHNSTON, H. LEWIS, T. ROLAND & H. WICKSTEAD. 2008. Historical context and chronology of Bronze Age enclosure on Dartmoor, UK. Journal of Archaeological Science 35: 2250-61.
GILLINGS, M., J. POLLARD, D. WHEATLEY & R. PETERSON. 2008. Landscape of the Megaliths: excavation and fieldwork on the Avebury monuments, 1997-2003. Oxford: Oxbow.
GREEVES, T. 2004a. Major prehistoric find on Cut Hill. Dartmoor Magazine 75: 4.
--2004b. Megalithic stone row discovered on Dartmoor's remotest hill. PAST 47: 10-2.
GRIFFITH, F.M. 1984. Archaeological investigations at Colliford Reservoir, Bodmin Moor, 1977-1978. Cornish Archaeology 23: 49-140.
HARDING, J. & F. HEALY. 2008. The Raunds Area Project: a Neolithic and Bronze Age landscape in Northamptonshire. Swindon: English Heritage.
HERRING, P. 2008. Stepping onto the commons: south-western stone rows, in P. Rainbird (ed.) Monuments in the landscape: 79-88. Stroud: Tempus.
JOHNSON, N. & P. ROSE. 1994. Bodmin Moor, an archaeological survey. Volume 1, the human landscape to c. 1800. London: RCHME.
JOHNSTON, R. 2005. Pattern without a plan: rethinking the Bronze Age coaxial field systems on Dartmoor, south-west England. Oxford Journal of Archaeology 24: 1-21.
JONES, A.M. 2005. Cornish Bronze Age ceremonial landscapes c. 2500-1500 BC (British Archaeological Reports British Series 394). Oxford: Archaeopress.
KINNES, I. 1999. Longtemps ignores: Passy-Rots, linear monuments in northern France, in A. Barclay & J. Harding (ed.) Pathways and ceremonies. The cursus monuments of Britain and Ireland: 148-54. Oxford: Oxbow.
L'HELGOUACH, J. 1983. Les Idoles qu'on abat. Bulletin de la Societe Polymathique du Morbihan 1983: 57-68.
LOCKYER, N. 1906. Stonehenge and other British stone monuments astronomically considered. London: Macmillan.
MERCER, R. 1981. Excavations at Carn Brea, Illogan, Cornwall, 1970-1973; a Neolithic fortified complex of the third millennium BC. Cornish Archaeology 20: 1-204.
--1997. The excavation of a Neolithic enclosure complex at Helman Tor, Lostwithiel, Cornwall. Cornish Archaeology 36: 5-61.
MOORE, P.D., J.A. WEBB & M.E. COLLINSON. 1991. Pollen analysis. Oxford: Blackwell.
NOBLE, G. 2006. Neolithic Scotland: timber, stone, earth and fire. Edinburgh: Edinburgh University Press.
OSWALD A., C. DYER & M. BARBER. 2001. The creation of monuments. Neolithic causewayed enclosures in the British Isles. Swindon: English Heritage.
PATTON, M. 1993. Statements in stone: monuments and society in Neolithic Brittany. London: Routledge. RILEY, H. & R. WILSON-NORTH. 2001. The field archaeology of Exmoor. Swindon: English Heritage.
RUGGLES, C.L.N. 1985. The linear settings of Argyll and Mull. Archaeoastronomy 9:105-32.
SHORE, J.D., D.D. BARTLEY & D.D. HARKNESS. 1995. Problems encountered with the [sup.14]C dating of peat. Quaternary Science Reviews 14: 373-83.
STUIVER, M. & P.J. REIMER. 1993. Extended [sup.14]C database and revised CALIB radiocarbon calibration program. Radiocarbon 35:215-30.
THOMAS, J. 1999. Understanding the Neolithic. London: Routledge.
TILLEY, C. 1995. Rocks as resources: landscapes and power. Cornish Archaeology 34: 5-57.
--2004. The materiality of stone: explorations in landscape phenomenology Oxford: Berg.
WHITTLE A., A. BAYLISS & F. HEALY. 2008. The timing and tempo of change: examples from the fourth millennium cal BC in southern England. Cambridge Archaeological Journal 18: 65-70.
Ralph M. Fyfe (1) & Tom Greeves (2)
(1) School of Geography, University of Plymouth, Drake Circus, Plymouth, PL4 8AA, UK (Email: firstname.lastname@example.org)
(2) 39 Bannawell Street, Tavistock, Devon, PL19 ODN, UK
Table 1. Size and spacing of elements within the Cut Hill stone row (numbering of stones is shown on Figure 2). Sizes of stones 7 and 9 have been established through probing. Stone number Long axis (m) Short axis (m) 1 2.3 0.9 2 2.4 0.8 3 1.85 0.7 4 1.9 1.2 5 2.6 0.7 G 2.1 1.05 7 (partially buried) 1.8 1.1 8 1.53 0.5 9 (buried) 2.0 0.7 Table 2. Results of radiocarbon dating. Depth (cm) [sup.14]C age lab code fraction Section 1 2-3 4835 [+ or-] 35 SUERC-10211 humic 2-3 4850 [+ or-] 35 SUERC-10212 humin 2-3 (comb.) 4858 [+ or-] 25 50-51 6475 [+ or-] 35 SUERC-10209 humic 50-51 6960 [+ or-] 35 SUERC-10210 humin Section 3 104-105 3814 [+ or-] 21 UBA-8856 humic 104-105 3903 [+ or-] 35 UBA-8855 humin 120-121 4402 [+ or-] 26 UBA-8854 humic 120-121 4505 [+ or-] 24 UBA-8853 humin [[delta].sup.13] Depth (cm) C cal range BC Section 1 2-3 -28.4 3710-3530 2-3 -28.2 3710-3530 2-3 (comb.) 3700-3540 50-51 -27.8 5510-5360 50-51 -27.3 5980-5740 Section 3 104-105 -30.3 2338-2150 104-105 -31.3 2476-2245 120-121 -39.7 3095-2922 120-121 -33.6 3347-3100 Table 3. Description of pollen zones, Cut Hill section 1. CH1-lpaz1 52-48cm c. 5900-5700 cal BC Calluna The lowest pollen level is dominated by Calluna (80%). Low levels of arboreal taxa are also present at around 5% TLP, notably Quercus, Corylus and Pinus. Charcoal is abundant within the sample. CH1-lpaz2 48-23cm c. 5700-4550 cal BC Calluna-Quercus-Corylus Calluna continues to be the dominant pollen type with the zone, but is reduced to between 40% and 50% TLP Poaceae and Cyperaceae rise to levels of around 10% TLP Within the arboreal taxa Quercus (15% TLP) and Corylus (12% TLP) increase. Charcoal levels are reduced, but increase again in the upper most level in the zone. CH1-lpaz3 23-2cm c. 4550-3550 cal BC Calluna Calluna remains dominant in the diagram into zone CH1-lpaz3, with levels similar to zone CH1-lpaz1. Arboreal taxa, and those of Poaceae and Cyperaceae, decline to similar levels as those recoded in the lowest pollen zone. CH1-lpaz4 2-0cm c. 3550-3450 cal BC Calluna-Corylus-Quercus The upper-most level in the diagram is different from preceding levels owing to an increase in Corylus (to 25% TLP) and Quercus (to 12% TLP), and a decline in Calluna (to 51% TLP).…