UDK 903.7(4i0)"634"
Documenta Praehistorica XXXV (2008)
The Stonehenge Riverside Project:
exploring the Neolithic landscape of Stonehenge
Mike Parker Pearson1, Joshua Pollard2, Colin Richards3, Julian Thomas3,
Chris Tilley4 and Kate Welham5
1 Department of Archaeology, University of Sheffield; 2 Department of Archaeology and Anthropology,
University of Bristol, Julian.Thomas@manchester.ac.uk; 3 School of Arts, Histories and Cultures,
University of Manchester; 4 Department of Anthropology, University College London;
5 School of Conservation Sciences, University of Bournemouth
ABSTRACT - The Stonehenge Riverside Project is a collaborative enterprise directed by six academics
from five UK universities, investigating the place of Stonehenge within its contemporary land-
scape. In this contribution, a series of novel approaches being employed on the project are outlined,
before the results of investigations at the Greater Stonehenge Cursus, Woodhenge, the Cuckoo Stone
and Durrington Walls are discussed.
IZVLEČEK - Stonehenge Riverside Project je skupen projekt, ki ga vodi šest profesorjev s petih uni-
verz Združenega kraljestva Velike Britanije. Ukvarjamo se s položajem Stonehenga v takratni pokra-
jini. V prispevku predstavljamo vrsto novih pristopov, ki smo jih uporabili v projektu, kot tudi rezul-
tate raziskav Greater Duringhton Cursus, Woodhenge in Durrington Walls.
KEY WORDS - Stonehenge; Durrington Walls; Southern Britain; monumentality; landscape
Introduction: the landscape of Stonehenge
Stonehenge is a national symbol, recognised through-
out the world, and interpreted in different ways by
a wide variety of constituencies, from Druids to New
Age enthusiasts (Chippindale 1990) (Fig. 1). It has
served as a focus for contemporary cultural and polit-
ical struggles, and has a special place in popular cul-
ture and the public imagination (Bender 1998;
Worthington 2004; 2005). Yet the attention that
Stonehenge attracts sometimes occludes its place
within a broader landscape, a World Heritage Site
composed of a great many structures and deposits
that built up over dozens of generations (Darvill
2005) (Fig. 2). This process arguably began with the
construction of an arrangement of huge post-holes
dating to the eighth millennium BC (in the earlier
Mesolithic), discovered when the car-park for Stone-
henge itself was constructed in 1966 (Cleal, Walker
and Montague 1995.43). This clearly refutes the
argument that only agriculturalists build monuments
(e.g. Rowley-Conwy 2004.85), but it also potentially
demonstrates the longevity of special places within
this particular landscape. During the early 1980s, a
very important programme of investigation was con-
ducted by Julian Richards, under the rubric of the
Stonehenge Environs Project. This combined target-
ed excavations with extensive field walking to iden-
tify the surface concentration of lithics (Richards
1990). The intention of this project was to place the
known field monuments into a clearer chronological
framework, and to identify complementary domestic
and industrial activity in their immediate surround-
ings.
The publication of the Stonehenge Environs Project,
and that of the various excavations by Gowland,
Hawley, Atkinson, Piggott, Stone, and the Vatchers
at Stonehenge itself during the course of the twen-
tieth century (Cleal, Walker and Montague 1995)
represent an unparalleled contribution to knowl-
edge. However, for more than twenty years the pur-
suit of archaeological investigations on any scale
within the Stonehenge landscape has been curtailed
by the continuing deliberations over the future of
the A303 trunk road (which runs immediately to the
south of the monument). Potential options have
included the possibility of running the major road
through a bored tunnel, closing the stretch of the
A344 road that severs Stonehenge from the Avenue
which connects it to the River Avon, and establish-
ing a visitor centre to replace the present subter-
ranean structure beside the car-park. At present, it
appears that only the improvements to visitor facil-
ities are likely to proceed in the foreseeable future
(Harris 2007). Irrespective of the view that one
might take of this outcome, the effective hiatus
affecting archaeological research in the Stonehenge
landscape has coincided with a period of heightened
debate over the character of social archaeology,
particularly as it relates to the British Neolithic (e.g.
Barrett 1994; Bradley 1998; Whittle 2003). As a re-
sult, until now the opportunity has not arisen to
'field trial' a variety of new ideas and approaches in
the immediate context of Stonehenge. While hypo-
theses and arguments concerning Stonehenge have
continued to be constructed, they have had to rely
on existing evidence, often collected according to
the research agendas of past generations.
New approaches to the landscape
The Stonehenge Riverside Project is a collaborative
research initiative directed by six academic archaeo-
logists from five different British universities. It
brings a series of novel approaches to bear on the
development of the Stonehenge landscape, and we
can begin this contribution by outlining each. The
first is a concern with what we might call the 'mate-
riality of monuments': that is, an interest in the
physicality and constituent substances involved in
monument building. These issues animated a pio-
neering study by Parker Pearson and Ramilsonina
(1998), who drew on a parallel with contemporary
Madagascar to suggest that monuments constructed
of timber and stone respectively may have been un-
derstood in different ways by Neolithic people. For
many Malagasy communities, the human body is
considered to be soft and wet at birth, maturing to
greater hardness and dryness, and culminating in
the exceptionally hard and dry character of the
bones of the ancestral dead. Consequentially, while
the living inhabit houses made of wood, whose or-
ganic character has much in common with human
flesh, stone tombs and standing stones are the exclu-
sive prerogative of the dead. In a similar way, Par-
ker Pearson and Ramilsonina noted that while Stone-
henge is distinguished by its multiple stone settings,
the presence of numerous cremation burials, and a
general paucity of ceramics and human remains, the
much larger henge monument at Durrington Walls,
Fig. 1. Stonehenge, seen from the north-east (photo: Aerial-Cam).
154
Fig. 2. Map of prehistoric monuments in the Stonehenge area (drawing:
Anne Leaver).
3 kilometres to the north-east, showed a very diffe-
rent pattern. Here, there were multiple settings of
upright timbers inside a massive earthwork enclo-
sure, and colossal quantities of animal bones and
Grooved Ware pottery (Wainwright and Longworth
1971). On this basis, Parker Pearson and Ramilso-
nina hypothesised that the two monuments might
have been focal to two distinct areas of the land-
scape, reserved for the living and the dead, and
linked by the River Avon. The transformation of the
newly dead into ancestors might then be physically
expressed through the passage downriver, from Dur-
rington Walls to Stonehenge. Such an argument at
once draws our attention to the complementarity of
the two monuments, to the way that what are often
Fig. 3. The Greater Stonehenge Cursus under excavation, sum-
mer 2007 east (photo: Aerial-Cam).
understood as separate struc-
tures may form parts of a sin-
gle complex, and to the axial
role of the River Avon within
the Stonehenge landscape.
This emphasis on the materi-
al substance of monumental
constructions is complement-
ed by a concern with the con-
struction of monuments as a
collective social practice (Ri-
chards 2004). Rather than a
simple exercise in ergonom-
ics, the creation of elaborate
works like Stonehenge and
Durrington Walls involved
the mobilization of large
numbers of people, materials,
animals and food, in an en-
terprise that could generate prestige and fame for
the builders, but which also risked shame and fail-
ure if the desired outcome was not achieved. More-
over, addressing the physical composition of mon-
uments encourages us to think about their varying
temporal qualities: the way that their decay, de-
struction or endurance conditions and contributes
to quite different histories or biographies of place
(Thomas 2004).
A second theme is provided by a new attentiveness
to the disposition of materials in the archaeological
record, informed by the concept of 'structured depo-
sition'. This originated in work undertaken by two
of the authors in the 1980s, re-assessing the evi-
dence from Geoffrey Wainwright's exca-
vations at the Durrington Walls henge,
and suggesting that many of the depo-
sits at the site had been deliberately pla-
ced, as one aspect of ritual activity (Ri-
chards and Thomas 1984). More recent-
ly, increasingly sophisticated analyses
have drawn attention to the important
role of depositional practice in transfor-
ming the meaning of place, and in en-
gendering memory (Pollard 1995;
2001). Both within monumental struc-
tures and in isolated pits dispersed
across the landscape, the placement of
artefacts and other materials appears
to have been one of the key means by
which people expressed their connec-
tion with specific locations during the
British Neolithic (Garrow 2006).
Fig. 4. Section of the Greater Stonehenge Cursus at its western terminal (drawing: Julia Roberts).
Our third preoccupation is what we might call the
'phenomenology of landscape', or an approach to
field survey that stresses the experiential qualities of
places and monuments (Tilley 1994). Although the
area surrounding Stonehenge has been subject to
exhaustive survey and mapping, from the work of
Sir Richard Colt Hoare and Philip Crocker (Hoare
1810) down to the high quality investigations of
English Heritage (e.g. McOmish, Field and Brown
2002), it is arguable that a concern with the way
that the landscape might be engaged with from a hu-
man perspective is capable of generating fresh in-
sights. Both in terms of the architectural organisa-
tion of specific monuments, and in relation to the
wider landscape, a number of novel observations
have been generated (Tilley et al. 2007). In harmo-
ny with some of the arguments already outlined, it
is clear that the River Avon and the system of dry
valleys with which it articulates had a fundamental
role in influencing the location of a variety of struc-
tures throughout the Neolithic and Early Bronze
Age. As well as delimiting areas of higher ground,
the valley systems define a series of potential routes
through the landscape, so that significant structures
may have been positioned in such a way as to be en-
countered by people and their animals in the course
of their daily or seasonal movements. Equally, Bea-
con Hill, a distinctive natural eminence formed by
the intersection of the chalk with the pebble deposits
of the Reading Beds, seems to be visible from or
aligned upon by almost all of the Neolithic and Early
Bronze Age constructions in the whole landscape, in-
cluding Stonehenge itself (Tilley et al. 2007.189).
While Beacon Hill possesses no upstanding prehis-
toric features of its own, its evident influence on the
development of the monumental landscape demon-
strates that 'natural' topographic features often hold
great significance.
Finally, the Stonehenge Riverside Project has sought
to employ a series of new field technologies, many
of which have not previously been used in the Stone-
henge area. As well as very large areas of GPS, mag-
netometer and resistivity survey, the project has
made use of ground-penetrating radar, laser scan-
ning of archaeological features, and unmanned pho-
tographic aircraft. At the Durrington Walls henge,
for instance, this work has revealed two formerly
unknown blocked entrances through the henge
bank, and the causewayed character of the ditch, in-
dicating that this was probably dug in sections by a
series of work-gangs.
Monuments as places of enduring significance
The earliest site investigated under the aegis of the
project is also the largest. The Greater Stonehenge
Cursus is a linear enclosure over a mile long, which
runs between the King Barrow Ridge and Fargo
Ridge, immediately north of Stonehenge (Stone
1947; Christie 1963) (see Fig. 2). The Cursus is in-
timately associated with a series of Early Neolithic
long barrows, including Amesbury 42, which runs
parallel with its eastern terminal (Richards 1990.
96). However, the only radiocarbon date that has
previously been derived from it falls in the mid-third
millennium cal BC, in our Later Neolithic (2890-
2460 cal BC; OxA-1403). In the summer of 2007,
excavations were able to demonstrate that this date
had come from one of a series of intrusive features,
which formed the first of two phases of re-cutting in
the cursus ditch (Fig. 3) (Thomas et al. 2008). Clearly,
the cursus was a very long-lived structure, which was
repeatedly re-established over a long period of time.
This was underlined by radiocarbon determinations
from a piece of antler located on the bottom of the
ditch at its western terminal, which calibrated to
3632-3375 BC and 3630-3370 BC at the 95.4% con-
fidence level (OxA-17953 and OxA-17954) (Fig.4).
This is roughly half a millennium earlier than the
earliest phase of construction at Stonehenge itself,
so that the cursus can be said to have had an impor-
Fig. 5. The Cuckoo Stone: excavations 2007 (photo:
Aerial-Cam).
tant role in structuring the landscape into which
Stonehenge was placed. In the course of excavation
it was also recognised that the initial laying-out of
the structure involved an alignment on Beacon Hill,
tying the enclosure into the local topography. Nei-
ther geophysics nor excavation could locate any in-
ternal features, and the ditches contain so little ma-
terial culture that we were very
lucky to recover the one piece
of antler noted above. So, un-
like other Neolithic structures
in the area, the cursus gives lit-
tle indication of having been
used for ceremonial, consump-
tion or deposition, and this sup-
ports the idea that it enclosed
a venerated, sanctified or cur-
sed area, which was set apart
from the rest of the landscape.
Immediately to the east of the
Cursus, and in line with its axis,
excavation was conducted du-
ring 2007 at the Cuckoo Stone,
a formerly upstanding sarsen
monolith (Fig. 5). This is one of
two isolated sarsen stones that have been investi-
gated by the project, and both here and at the Tor
Stone on Bulford Hill, the stone socket and the hole
from which the stone was quarried were discovered.
This is of particular importance as there has been
a continuing debate in the literature over the ques-
tion of whether some of the sarsens at Stonehenge
could have been acquired locally, or whether they
must all have been dragged from the Marlborough
Downs, nearly 20 miles to the north (Stone 1924.69;
Atkinson 1956.110). Evidently, we have two exam-
ples of sarsen stones recovered from the chalk of Sa-
lisbury Plain. Both the Cuckoo Stone and the Tor
Stone seem to have been set up in the Neolithic,
and to have formed a focus for Neolithic pits, Early
Bronze Age urned cremations, and in the case of the
Cuckoo Stone, a Roman shrine. So as at the Cursus
we have a sense of a particular site maintaining its
importance over an exceedingly long period.
The same is true of Woodhenge, the small late Neo-
lithic enclosure just south of Durrington Walls orig-
inally excavated by Maud Cunnington in the 1920s
(Cunnington 1929). Here, excavation in 2006 re-
vealed that the bank overlay a tree-hole which had
been filled with large quantities of Early Neolithic
carinated bowl pottery, before being covered over
by a chalk capping. Moreover, the concentric timber
circles that Cunnington had excavated proved to
have been succeeded by two separate phases of
stone settings, again indicating a very long-lived
structural sequence (Pollard and Robinson 2007.
162) (Fig. 6). In 2007, immediately to the south of
Woodhenge, in an area of Bronze Age ring-ditches
which had also been investigated by Cunnington,
Fig. 6. Excavations at Woodhenge, 2006 (photo: Aerial-Cam).
three separate Late Neolithic
timber structures were encoun-
tered, each composed of four
main uprights, with two en-
trance posts and, in some ca-
ses, an enclosing palisade (Fig.
7). Although these were not
roofed structures, their archi-
tecture seems to relate to the
small houses of the Late Neoli-
thic Grooved Ware tradition
(see below). In this respect,
they are very relevant to the
evidence that was recovered
immediately to the north, at
Durrington Walls.
Durrington Walls
Fig. 7. Exacavation of Late Neolithic timber structures, amongst Early
Bronze Age ring ditches, south of Woodhenge 2007 (photo: Aerial-Cam).
Durrington Walls is the largest
henge monument in Britain, with an overall diam-
eter of nearly half a kilometre (Fig. 8). Henges are
a type of enclosure dating to the later Neolithic,
from 3000 BC onwards, distinguished by having an
external bank surrounding an internal ditch. For the
most part, they are considered to have been cere-
monial, and non-domestic in character, the enclo-
sure keeping something in - or at least secluding it,
rather than keeping enemies out (Wainwright 1989.
14). When Geoffrey Wainwright excavated at Dur-
rington in 1966-7, in advance of road-building, the
strip that he cut across the enclosure was the largest
prehistoric excavation that had ever taken place in
Britain (Wainwright and Longworth 1971; Pitts
2000.55). Wainwright's excavation revealed colos-
sal banks and ditches and massive timber circles,
and produced prodigious quantities of Grooved
Ware, animals bones, and stone tools. His work
transformed our understanding of the Neolithic in
southern Britain but, because it was a rescue project
conducted under formidable time constraints, it left
a series of questions unanswered. Forty years later,
new excavations conducted as part of the Stone-
henge Riverside Project sought to complement the
extensive work of the 1960s with a more targeted,
intensive approach.
The initial decision to excavate outside of the east-
ern entrance of the Durrington Walls henge was
based on the hope of discovering an avenue con-
necting the henge to the River. This would confirm
the link between Durrington and Stonehenge. Just
such an avenue was found in 2005, actually a huge
metalled roadway, 30 metres wide, with a bank and
gully on either side, leading for 170 metres down to
the river. The avenue had traces of extensive tram-
pling down the middle, and large quantities of highly
fragmented animal bones and Grooved Ware pottery
scattered on either side. It was re-surfaced on two
occasions, and had a line of upright sarsen stones
running down one side (Parker Pearson 2007.130).
Fig. 8. Durrington Walls: areas excavated 2004-7
(drawing: Mark Dover).
Fig. 9. Plan of excavations at the eastern entrance, Durrington
Walls, 2004-7, showing the Avenue and Neolithic house floors
(drawing: Mark Dover).
This last point invites comparison with the West
Kennet Avenue, connected to the Avebury henge in
north Wiltshire, but in the local context, it is clear
that this roadway is the equivalent of the Stone-
henge Avenue, linking the henge to the river, and in
a way that has an astronomical alignment comple-
mentary to that at Stonehenge. The stone settings at
Stonehenge face the midsummer sunrise, while the
Stonehenge Avenue is aligned on the midwinter sun-
set; the southern timber circle inside Durrington
Walls faces the midwinter sunrise, and the Dur-
rington Avenue aligns on the midsummer sunset.
The implication is that one might process from Dur-
rington to Stonehenge at midwinter, and in the op-
posite direction at midsummer, passing over or
through the purifying or transforming waters of the
river in the process. Recent reconsideration of the
radiocarbon dates from Stonehenge has demonstra-
ted that the sarsen stones and the
avenue can be placed in the mid-third
millennium cal BC, contemporary with
Durrington Walls (Parker Pearson et
al. 2007.627), so we are entitled to
see the two monuments as parts of a
single, integrated structure.
Although the potential presence of
the avenue was the initial reason for
excavating at the eastern entrance, the
presence of seven small houses of Late
Neolithic date, clustered around the
roadway (Fig. 9) was a complete sur-
prise. Two of these were located on
opposite banks of the avenue, and ap-
pear to have been open on their east-
ern sides, facing toward the river.
Only a very few such houses have
been found on the British mainland,
and never as a substantial settlement,
so the only real parallels are the vil-
lages of stone cellular buildings in the
Orkney Isles of northern Scotland,
such as Skara Brae, Rinyo and Barn-
house (Childe 1931; Childe and Grant
1947; Richards 2003). Like the Orca-
dian houses, the Durrington buildings
have clay floors and central hearths,
but their walls were of wattle and
daub rather than stone (Fig. 10). Se-
veral of the houses had floor levels
that had been terraced back into the
hillside, and some were separated by
fence-lines, against which waste mate-
rials in the form of burnt flint, flint
cores and animal bones had been flung (Fig. 11).
Associated with the houses were borrow pits, from
which the daub had been acquired, and other pits
containing dense deposits of animal bone and pot-
tery, as well as large numbers of flint arrowheads.
The buildings appear to have been abandoned with
some formality: three had a single human bone de-
posited close to them, and two had deposits of cat-
tle vertebrae placed into their hearths.
Several of the structures were stratified beneath the
henge bank, which indicates that both the houses
and at least the first phase of the avenue pre-date
the construction of the bank and ditch. It is likely
that the enclosure of the great natural amphitheatre
of Durrington Walls may have been made at the very
end of the Neolithic (at around 2500-2400 BC), and
that there was a complex sequence of structures of
various kinds that culminated in
this event, and which paralleled
the sequence at Stonehenge. At va-
rious points around its circumfer-
ence, excavations (whether for the
purposes of research or for pipe
trenches) have cut through the
bank, and in every case a dense
spread of cultural material has been
encountered, similar to that spread
over the settlement area (e.g. Stone,
Piggott and Booth 1954). The im-
plication is that a Late Neolithic set-
tlement covered the entire area co-
vered by the henge bank, and that
it was very large indeed. None the
less, this settlement was clustered
around a roadway leading to an
enormous timber circle (see below)
and was close to the group of non-
domestic timber structures identi-
fied south of Woodhenge (see
above), indicating that the imme-
diate location was rather special
even before the henge bank and ditch were con-
structed. It is open to question whether the huge
accumulation of houses at Durrington represents a
typical Late Neolithic habitation, fortuitously pre-
served under the combination of bank and hill-wash,
and whether we should expect to find numerous
such settlements in future.
There are several strands of evidence that suggest
that the situation was not straightforward. Some of
the anticipated signs of year-round domestic activity
are missing. Despite having subjected enormous
numbers of soil samples to flotation, no cereal grains
or glume fragments have
been recovered from the set-
tlement area: the only cereal
remains came from the sur-
face of the avenue. This is
complemented by a complete
absence of grinding stones.
Amongst the assemblage of
80 000 pig and cattle bones,
there are no neonates at all,
suggesting that animals were
brought to the site and not
raised in the immediate area.
The lithic assemblage from
the settlement is dominated
by transverse arrowtead.^ Fig. 11. Laser scan of the surface of a Neolithic house-floor, eastern en-
with few scrapers and knives, trance, Durrington Walls (image: Kate Welham/Mark Dover).
Fig. 10. Two of the house floors at the eastern entrance, Durrington
Walls (photo: Aerial-Cam).
and no flint or stone axes. All of this suggests tempo-
rary (perhaps seasonal) rather than permanent ha-
bitation. There are very large numbers of domesti-
cated pig bones, but some of these animals seem to
have been shot with arrows and then barbequed.
This does not suggest conventional culinary behav-
iour, and there are strong indications that periodic
feasting took place (Albarella and Serjeantson
2002). An unusual proportion of the animal bones
are complete and hundreds were discarded still in
articulation. Tooth eruption evidence suggests that
most of the pigs were killed at about nine months
old, most likely in the midwinter period. Given the
Fig. 12. The Southern Circle, Durrington Walls: plan of excavations
1967 and 2005-6 (drawing: Mark Dover).
midwinter solstice alignments at Stonehenge and at
four of the Durrington timber monuments, it is like-
ly that this was a major calendrical event. The first
few radiocarbon determinations for the settlement
show that it was occupied in the 26th century BC.
Some of the house floors were re-plastered up to
seven times, and the inter-cutting of the borrow pits
associated with each house suggests up to a dozen
episodes of repair for the walls and floors. However,
this might still mean that the overall inhabitation of
the settlement was comparatively brief. Further ra-
diocarbon dates may tell us whether the site served
as a centre for seasonal gatherings or pilgrimage for
many decades, or whether it represents a single
significant episode, such as an accumulation of pop-
ulation to build the stone settings at Stonehenge.
The Southern and Western Circles at Durring-
ton Walls
Some indication of the character of the activity at
Durrington Walls is provided by a series of struc-
tures enclosed inside the henge bank and ditch fur-
ther to the west. The Southern Circle was a two-
phase timber structure composed of six concentric
rings of upright posts, 60% of
which was excavated by Wain-
wright in 1967 (Wainwright and
Longworth 1971.23) (Fig. 12). A
further timber structure, the
Northern Circle, was also identi-
fied in the same excavation. On
its eastern side the Southern Cir-
cle intersected with a chalk and
gravel platform, which we can
now identify as the western ex-
tremity of the Durrington Ave-
nue, leading down to the river.
This provides a stratigraphic link
that places the circle earlier than
the henge bank and ditch: the
timbers of the Southern Circle pe-
netrated the surface of the ave-
nue, but their post-holes were in
some cases masked by it. Another
element of the Southern Circle
complex, originally identified as
an enclosed midden, can now be
reconsidered in the light of the
discoveries at the eastern entran-
ce. It is very likely that this repre-
sented a large, hall-sized build-
ing with a terraced floor area sur-
rounded by stakehole-defined
walls. Wainwright considered that the timber circle
had itself been a massive roofed building, compa-
rable with the 'council houses' of certain Native Ame-
rican communities (Wainwright and Longworth
1971.232). However, subsequent investigation of
some even larger timber circles has demonstrated
that they were simply too big to be roofed, and the
same was probably true of the Southern Circle (Da-
vid et al. 2004) (see Fig. 13 for the probable ap-
pearance of the Southern Circle in its second phase).
While there were indeed two phases of construction,
our re-excavation in 2005-6 demonstrated that some
of what had been identified as postholes of the first
phase were actually integral to the second, standing
on either side of individual larger posts, and perhaps
supporting sections of fencing or shuttering. The
structure thereby defined establishes a secluded in-
ner space within the circle, comparable with the in-
nermost area of Stonehenge. Tellingly, this structure
respects the spiral entrance passage to the second-
phase circle. The implication of this is that the first
phase circle was comparatively modest, composed
of four main posts, surrounded by a single post-ring,
attached to an avenue and facade. This would make
it very similar to both the Northern Circle, and the
structures excavated south of
Woodhenge (see above).
Another issue addressed through
re-investigating the Southern
Circle was that of deposition.
Wainwright's original interpre-
tation for the concentration of
finds in the upper parts of the
post-hole fills was that, within
the roofed building, sherds of
pottery, animal bones and other
objects had been placed as of-
ferings at the bases of the tim-
ber uprights. When the latter
had rotted out, the objects fell
into what he referred to as 'wea-
thering cones' (Wainwright and
Longworth 1971.24-5). These
he argued to have been formed by the erosion of
the post-packing, following the decay of the posts.
However, this interpretation was open to question.
If we accept that the structure was unroofed, it is
hard to see how pottery sherds and animal bones
could have survived on the surface for some de-
cades in an unabraded state, before falling into the
weathering cones. In all of the post-holes excavated
in 2005-6, it was clear that the so-called 'weathering
cones' were actually conical re-cuts, dug after the
posts had rotted out (see Fig. 14).
Inside these re-cut features, deposits of flints and
animal bones had clearly been placed, or at least
dumped, rather than having fallen haphazardly into
eroding post-pipes. Animal bone predominat-
ed, but flint occurred as clusters of waste,
often higher in the fill (Thomas 2007.149).
In all cases, pottery sherds were found
almost exclusively in the upper part of the
re-cut fill. This suggests a pattern in which
sherds were being carefully placed into the
tops of the re-cuts following the more sum-
mary deposition of flint and bone. It is clear,
though, that our excavated area, located op-
posite the entrance to the circle, produced
far smaller quantities of cultural material
than the postholes facing toward the avenue
and the river, dug in 1967. In other words,
the densities of objects placed in each post-
hole reflected the individual importance of
each feature. If this material had been depo-
sited in features that were cut after the posts
had rotted out, it must have post-dated the
construction and initial use of the circle by
Fig. 13. Full size reconstruction of the Southern Circle, constructed
for the Time Team TV programme (photo: Julian Thomas).
one or two generations, if not more. It is possible
that this re-cutting took place at much the same time
as the enclosure of the Durrington landscape by the
henge bank and ditch. It follows that the Southern
Circle was ancient and ruinous by the time the re-
cutting took place, indicating that the depositional
activity was essentially commemorative in charac-
ter. That is to say, digging a hole and placing cul-
tural materials in it was a means of venerating the
Circle, its component elements, and the past activ-
ities that had taken place within it. The richness of
the material deposited in the re-cuts reflected the
relative significance of the different parts of the tim-
ber circle, even though the structure was by then
decrepit. The physical manifestation of the circle
Fig. 14. Section of post-hole 071 in the Southern Circle, sho-
wing re-cut pit in the upper fill (drawing: Julia Roberts).
Fig. 15. Comparison of the plans of the Stonehenge stone settings and
the Southern Circle (drawing: Julian Thomas).
was then one of collapsed timbers, slumped post-
holes, and memories that were brought to mind
through acts of deposition. So this was now an 'ar-
chitecture of memory', commemorated in its absence.
We have noted that the discovery of the roadway de-
monstrates that Durrington Walls and Stonehenge
are effectively parts of a single structure linked by
the two avenues and the river. This challenges some
of our implicit expectations of monumentality, for
rather than being a cultural imposition onto a natu-
ral landscape, the Stonehenge-Durrington complex
threads together built elements and topography.
We might argue that both the Stonehenge and the
Durrington avenues were conceptually indistinguish-
able from the river, and that the two henge enclo-
sures were linked by flows and movements of var-
ious kinds. This encourages us to consider the rela-
tionship between the Southern
Circle and the Stonehenge stone
settings at each end of this pas-
sage. We might see them as com-
plementary structures: while
composed of stone and timber
respectively, they are remark-
ably similar in plan (Fig. 15).
The principal sarsen and blue-
stone settings at Stonehenge
have much the same diameters
as the four inner rings at Dur-
rington, and similar units of
measurement appear to have
been used in laying them out
(Chamberlain and Parker Pearson 2007). Signifi-
cantly, the interiors of both structures are much the
same size, and would have admitted the same num-
ber of people. Moreover, we have seen that the South-
ern Circle may have had a secluded inner space
comparable with the Stonehenge sarsen 'horseshoe',
while a geophysical survey of the unexcavated por-
tion of the Southern Circle undertaken in 2006 sug-
gests some elements of the Southern Circle may be
oval rather than truly circular. It is highly likely that
some relationship of mirroring or mimicry existed
between the Durrington Southern Circle and the
stone settings at Stonehenge.
Further to the west again, on a terrace overlooking
the Southern Circle and the eastern entrance, a
group of at least six penannular enclosures have
been revealed by geophysical survey. While these
DURRINGTON WALLS 2006	_
Western Enclosures Trench 14	metres
Fig. 16. Plan of enclosed building (Trench 14), Western Enclosures, Durrington Walls 2006 (drawing:
Julian Thomas).
were reasonably expected
to have enclosed burials or
more timber circles, the
two that were excavated ac-
tually contained small buil-
dings not dissimilar to those
at the eastern entrance
(Thomas 2007.152). How-
ever, in each case the buil-
ding was enclosed inside a
timber palisade, while each
'house' had four post-holes,
presumably roof-supports,
surrounding the central
hearth (Fig. 16). These were
not present in the houses
at the eastern entrance. The
building in the larger of the
two ring-ditches investigated also had a fagade of
huge posts, so closely-set that they may have repre-
sented the equivalent in timber of the Stonehenge
trilithons (see Fig. 17). Both buildings appear to have
been kept clean in comparison with the filthy hous-
es at the eastern entrance, and the larger one had
a pit immediately outside the entrance to the pal-
isade containing distarticulated animal bones and
abraded pottery that may have been accumulated in
the course of cleaning up.
So were these buildings inside the Western Enclo-
sures elite residences, or were they shrines or cult-
houses? The key to this question may lie in their si-
milarity with the first phase of the Southern Circle,
with the Northern Circle, and with the structures in-
vestigated south of Woodhenge. Here, the familiar
architecture of the Grooved Ware house was elabo-
rated upon, and elements that were more usually
found in monumental contexts were added to draw
attention to the separation between the small enclo-
sed space of the building and the outside world, in-
cluding the more obviously domestic dwellings. Sig-
nificantly, Richards (1993) has described a similar
process in Neolithic Orkney, in which the modular
form of the house found echoes in the layout of
chambered tombs and henge monuments. At an
early stage, the natural amphitheatre of Durrington
Walls was occupied by a series of structures that de-
veloped a single basic plan in different ways, pro-
Fig. 17. Reconstruction of enclosed building (Trench 14), Western Enclo-
sures, Durrington Walls (image: Aaron Watson).
viding spaces for dwelling and for ritual. Only later
did the more complex concentric architecture of the
Southern Circle and Woodhenge develop, at a time
when the former came to be physically linked to
Stonehenge in a new and grand design that drew
the entire landscape together. Recognising this, how-
ever, depends on acknowledging the long and com-
plex histories of place that the Stonehenge Riverside
Project has revealed.
-ACKNOWLEDGEMENTS-
We thank our colleagues on the Stonehenge Riverside
Project, particularly Umberto Albarella, Mike Allen,
Mark Dover, Charly French, Karen Godden, Dave Ro-
binson, the site supervisors (Dave Aspden, Ian Heath,
Neil Morris, Bob Nunn, Becca Pullen, Jim Rylett, Dave
Shaw, Anne Teather), the outreach team and the
many students and volunteers who have made this
work such a success so far. We also thank Amanda
Chadburn, Richard Osgood, Mike Pitts and Julian Ri-
chards for their advice and support. Funding for the
SRP was provided by the Arts and Humanities Re-
search Council, the British Academy, English Heri-
tage, the McDonald Institute, the National Geographic
Society, the Prehistoric Society (who awarded it the
Bob Smith Prize), the Royal Archaeological Institute
and the Society of Antiquaries. We thank the Rawlins
family, the National Trust, Wiltshire County Council
and the Ministry of Defence for permission to exca-
vate on their land.
REFERENCES
ALBARELLA U. and SERJEANTSON D. 2002. A passion for
pork: meat consumption at the British Late Neolithic site
of Durrington Walls. In P. Miracle and N. Milner (eds.),
Consuming Passions and Patterns of Consumption. Mc-
Donald Institute. Cambridge: 33-49.
ATKINSON R. J. C. 1956. Stonehenge. Hamish Hamilton.
London.
BARRETT J. C. 1994. Fragments from Antiquity. Black-
well. Oxford.
BENDER B. 1998. Stonehenge: Making Space. Berg. Lon-
don.
BRADLEY R. J. 1998. The Significance of Monuments.
Routledge. London.
CHAMBERLAIN A. and PARKER PEARSON M. 2007. Units
of measurement in Late Neolithic southern Britain. In M.
Larsson and M. Parker Pearson (eds.), From Stonehenge
to the Baltic. British Archaeological Reports. Internatio-
nal Series 1692. Archaeopress, Oxford: 169-74.
CHILDE V. G. 1931. Skara Brae: A Pictish Village in Ork-
ney. Kegan Paul. London.
CHILDE V. G. and GRANT W. G. 1947. A Stone-Age settle-
ment at the Braes of Rinyo, Rousay, Orkney (Second Re-
port). Proceedings of the Society of Antiquaries of Scot-
land 81:16-42.
CHIPPINDALE C. 1990 (ed.). Who Owns Stonehenge?
Batsford. London.
CHRISTIE P. 1963. The Stonehenge Cursus. Wiltshire Ar-
chaeological Magazine 58:370-382.
CLEAL R., WALKER K. and MONTAGUE R. 1995. Stone-
henge in its Landscape: Twentieth Century Excavations.
English Heritage. London.
CUNNINGTON M. E. 1929. Woodhenge. Simpson. Devizes.
DARVILL T. C. 2005. Stonehenge World Heritage Site: An
Archaeological Research Framework. English Heritage.
London.
DAVID A., COLE M., HORSLEY T., LINFORD N., LINFORD P.
and MARTIN L. 2004. A rival to Stonehenge? Geophysical
survey at Stanton Drew, England. Antiquity 78:341-358.
GARROW D. 2006. Pits, Settlement and Deposition dur-
ing the Neolithic and Early Bronze Age in East Anglia.
British Archaeological Reports. British Series 414, Archa-
eopress. Oxford.
HARRIS T. 2007. A303 Stonehenge improvement scheme:
statement by the Parliamentary Under-Secretary for Tran-
sport. http://www.dft.gov.uk/press/speechesstatements/
statements/a303. Consulted 15th March 2008.
HOARE R. C. 1810. The Ancient History of Wiltshire. Wil-
liam Miller. London.
MCOMISH D., FIELD D. and BROWN G. 2002. The Field
Archaeology of the Salisbury Plain Training Area. En-
glish Heritage. London.
PARKER PEARSON M. 2007. The Stonehenge Riverside
Project: excavations at the east entrance of Durrington
Walls. In M. Larsson and M. Parker Pearson (eds.), From
Stonehenge to the Baltic, British Archaeological Reports.
International Series 1692. Archaeopress, Oxford: 125-
44.
PARKER PEARSON M. and RAMILSONINA 1998. Stone-
henge for the ancestors: the stones pass on the message.
Antiquity 72: 308-26.
PARKER PEARSON M., CLEAL R., MARSHALL P., NEEDHAM
S., POLLARD J., RICHARDS C., RUGGLES C., SHERIDAN A.,
THOMAS J., TILLEY C., WELHAM K., CHAMBERLAIN A.,
CHENERY C., EVANS J., KNÜSEL C., LINFORD N., MARTIN
L., MONTGOMERY J., PAYNE A. and RICHARDS M. 2007.
The age of Stonehenge. Antiquity 81: 617-39.
PITTS M. 2001. Hengeworld. Arrow Books. London.
POLLARD J. 1995. Inscribing space: formal deposition at
the later Neolithic monument of Woodhenge, Wiltshire.
Proceedings of the Prehistoric Society 61:137-56.
2001. The aesthetics of depositional practice. World
Archaeology 33:315-33.
POLLARD J. and ROBINSON D. 2007. A return to Wood-
henge: the results and implications of the 2006 excava-
tions. In M. Larsson and M. Parker Pearson (eds.), From
Stonehenge to the Baltic. British Archaeological Reports.
International Series 1692. Archaeopress, Oxford: 159-
68.
RICHARDS C. C. 1993. Monumental choreography: archi-
tecture and spatial representation in late Neolithic Ork-
ney. In C. Tilley (ed.), Interpretative Archaeology. Berg.
London: 143-78.
2004. A choreography of construction: monuments,
mobilization and social organization in Neolithic Ork-
ney. In J. Cherry, C. Scarre and S. Shennan (eds.), Ex-
plaining Social Change: Studies in Honour of Colin
Renfrew. McDonald Institute. Cambridge: 103-30.
RICHARDS C. (ed.) 2003. Dwelling among the Monu-
ments: The Neolithic Village of Barnhouse, Maeshowe
Passage Grave and Surrounding Monuments at Sten-
ness, Orkney. McDonald Institute for Archaeological Re-
search. Cambridge.
RICHARDS C. C. and THOMAS J. S. 1984. Ritual activity
and structured deposition in later Neolithic Wessex. In R.
Bradley and J. Gardiner (eds.), Neolithic Studies. British
Archaeological Reports. British Series 133. Oxford: 189-
218.
RICHARDS J. 1990. The Stonehenge Environs Project.
London: English Heritage.
ROWLEY-CONWY P. 2004. How the west was lost: a re-
consideration of agricultural origins in Britain, Ireland,
and southern Scandinavia. Current Anthropology 45:83-
113. Supplement.
STONE E. H. 1924. The Stones of Stonehenge. Robert
Scott. London.
STONE J. F. S. 1947. The Stonehenge Cursus and its affi-
nities. Archaeological Journal 104: 7-19.
STONE J. F. S., PIGGOTT S. and BOOTH A. 1954. Durring-
ton Walls, Wiltshire: recent excavations at a ceremonial
site of the early second millennium BC. Antiquaries Jour-
nal 34:155-77.
THOMAS J. S. 2004. Materiality and traditions of practice
in Neolithic south-west Scotland. In V. Cummings and C.
Fowler (eds.), The Neolithic of the Irish Sea: Materiality
and Traditions of Practice. Oxbow. Oxford: 174-84.
2007. The internal features at Durrington Walls: inves-
tigations in the Southern Circle and western Enclo-
sures, 2005-6. In M. Larsson and M. Parker Pearson
(eds.), From Stonehenge to the Baltic. Oxford: British
Archaeological Reports. International Series 1692. Ar-
chaeopress. Oxford: 145-58.
THOMAS J., PARKER PEARSON M., POLLARD J., RICHARDS
C., TILLEY C. and WELHAM K. 2008. The date of the
Greater Stonehenge Cursus. Antiquity 82:1-14.
TILLEY C. 1994. A Phenomenology of Landscape. Lon-
don. Berg.
TILLEY C., RICHARDS C., BENNETT W. and FIELD D. 2007.
Stonehenge - its landscape and architecture: a re-analysis.
In M. Larsson and M. Parker Pearson (eds.), From Stone-
henge to the Baltic. British Archaeological Reports. Inter-
national Series 1692. Archaeopress, Oxford: 183-204.
WAINWRIGHT G. J. 1989. The Henge Monuments. Tha-
mes and Hudson. London.
WAINWRIGHT G. J. and LONGWORTH I. H. 1971. Dur-
rington Walls: excavations 1966-1968. Society of Anti-
quaries. London.
WHITTLE A. 2003 The Archaeology of People: Dimen-
sions of Neolithic Life. Routledge. London.
WORTHINGTON A. 2004. Stonehenge: Celebration and
Subversion. Alternative Albion. Loughborough.
(ed.) 2005. The Battle of the Beanfield. Enabler Pub-
lications. Teignmouth.
back to contents