Back to Book Reviews

Building Memories: The Neolithic Cotswold Long Barrow at Ascott-Under-Wychwood, Oxfordshire, Edited By Don Benson and Alasdair Whittle
Oxbow Books (Cardiff Studies in Archaeology) 2007, 379 pages, 269 figs., 24 plates, 57 tables, hb, ISBN 1-84217-236-0 (£55)

This very handsome Oxbow publication represents the definitive report on the excavation directed by Don Benson of the Ascott-under-Wychwood long barrow (hereafter abbreviated to AuW), which was fully investigated in 1965–69. For a variety of reasons familiar to those of us who are archaeologists of Don Benson’s generation the report has been a long time coming, but its appearance now, in extremely comprehensive form, is all the more welcome. At last we can avoid having to search for that early interim in Current Archaeology (Selkirk 1971 – interestingly a reference not mentioned in this report) or for John Evans’s initial accounts of the pre-barrow environment (Evans 1971; 1972), or in particular having to struggle with interpreting that strange debate in Man about the AuW burials (Benson & Clegg 1978; Chesterman 1977); these are superseded by the present volume (though perhaps not entirely – see below). For the appearance of this report now we are largely and very gratefully indebted to Alasdair Whittle, who, with funding from English Heritage, provided a base for the publication project at Cardiff University, assembled a team of specialist collaborators to work on various aspects of the AuW archive, and drove the project to its conclusion.

Thankfully the report (despite its redundant ‘clever’ prefix to the title – do these really ‘sell’ books?) follows the format of a traditional excavation monograph. After an entertaining and evocatively illustrated introductory chapter by Benson giving the background and explaining how the excavation progressed, inevitably now with a large dollop of historical perspective, 14 chapters follow on: The Pre-Barrow Contexts (Lesley McFadyen, Benson & Whittle); The Environmental Setting (John Evans, Susan Limbrey & Richard Macphail); The Long Barrow (McFadyen, Benson & Whittle); The Layout, Composition and Sequence of the Human Bone Deposits (Whittle, Dawn Galer & Benson); The Human Remains (Galer); Interpreting Chronology: The Radiocarbon Dating Programme (Alex Bayliss et al.); The Animal Bones (Jacqui Mulville & Caroline Grigson); Carbon and Nitrogen Stable Isotope Compositions of Animal and Human Bone (Robert Hedges, Rhiannon Stevens & Jessica Pearson); The Early Neolithic Pottery and Fired Clay (Alistair Barclay & Humphrey Case); Organic Residue Analysis (Mark Copley & Richard Evershed); The Flint (Kate Cramp); The Worked Stone Objects(Fiona Roe); Post-Neolithic Finds (Edward Biddulph, Peter Guest & William Manning); and a final lengthy discussion section entitled ‘Place and Time: Building and Remembrance’ (Whittle, Barclay, McFadyen, Benson & Galer).

Illustration plays a key role in publications of this type, and the present volume is well served by both numerous contemporary excavation photographs (including a colour section at the end of the book) and by excellently conceived and reproduced plans and sections. Particularly noteworthy are the pull-out section drawings which incorporate colour photographs of the matching sections (eg, fig.4.29) and the successive plans of the burial deposits with numbering of each bone and the use of colour to distinguish bones belonging to different individuals or to different parts of the skeleton (eg, figs.5.39-40). These reflect well on the quality of the original excavation and recording process and on the skill and care applied during post-excavation towards publication; thus it is entirely appropriate that the illustrator Ian Dennis is acknowledged on the title page.

The basic ‘story’ of AuW is of course familiar from various other publications since the excavation (eg, Case 1986 – also curiously omitted from the references in this report). In a Cotswold location which saw earlier Mesolithic activity, some kind of Neolithic ‘settlement’ was established involving structures and the deposition of ‘domestic’ debris. Over this formerly inhabited spot a laterally chambered tomb of trapezoidal outline was erected, oriented east-west, with a blind façade between projecting horns at the east end. The main body of the barrow was constructed in a series of rectangular bays on either side of an axial alignment. The bays, some with timber demarcation as well as stone revetting, were built up of mixed deposits of soil, rubble, and stone, and finally externally edged with stone revetments, the outermost of which was finely finished. Burial deposits occurred in and around the chambered area, which unusually in this case is seen as comprised of two pairs of closed ‘cists’ on either side of the axis rather than having the normal entrance, passage, and chamber configuration, though the chambered area did have external passage and entrance arrangements which were only fully preserved on the north side.

What do we learn from this report which is new? Undoubtedly what would have been the volume’s main excitement, the extensive series of radiocarbon dates and the establishing of a detailed absolute chronology for AuW, has been somewhat deflated by the simultaneous publication of the ‘Histories of the dead’ radiocarbon-dating supplement to the Cambridge Archaeological Journal (Bayliss & Whittle 2007). The paper on AuW therein (Bayliss et al. 2007), and the associated discussion (Whittle et al. 2007), are likely to become the first port of call for most students (and perhaps most prehistorians) when working on this aspect of the British Neolithic, rather than the AuW volume itself, and this will be a pity, because the report has so much more to offer. Suffice it to say that the dating programme, after the application of Bayesian statistical analysis, indicates pre-barrow Neolithic activity in the 40th–39th centuries cal BC, ending some 50 years or so before the initial barrow was constructed in c.3760–3700 cal BC. Human burials start in c.3755-3690 cal BC and end in c.3645–3595 cal BC, with the preferred dating model suggesting between 60 and 155 years of burial activity here around the 37th century BC. (The Late Neolithic date from an adult femur in the southern passage conflicts so markedly with the other results from the area that it seems more probable that this old British Museum determination is anomalous rather than, as suggested here (p.165), that it indicates a later insertion. It is not possible to identify this bone on the relevant plans (figs.5.33–35).)

We should celebrate the achievement of the advances in the application and interpretation of radiocarbon dating in British prehistoric studies which are reflected here, so strongly underpinned by the work of the Radiocarbon Dating Laboratory at Oxford University, which remains at the forefront internationally of the science involved. Nevertheless, we must remember that the interpretation of the radiocarbon data given here rests on modelling, and that subtle reinterpretations are possible (as indeed in Bayliss et al. 2007) and will no doubt be further developed, and also that the main revelation, that monuments like AuW were actually very short-lived in terms of burial usage, had already been predicted (Saville 1990, 265–266).

At the heart of the volume, and the part of the AuW story which will probably receive most attention other than the dating, are the chapters on the human remains. Dawn Galer’s elegant presentation of the data on human bones and teeth, taking advantage of various advances in physical anthropological analyses since these remains were excavated, reveals an assemblage with a minimum number of only 21 individuals (i.e. close to the MNI of 23 in the original interim (Selkirk 1971, 7) and far fewer than Chesterman (1977) concluded). From amongst the usual superficially jumbled disposition of bones which awaits the investigator of any Cotswold tomb, Galer has isolated the burials of a mature adult male, a young unsexed adult and an unsexed juvenile from the southern inner cist; a young adult ?male and an unsexed juvenile from the southern outer cist; two unsexed adults, a juvenile/young adult ?male, and an unsexed newborn infant from the northern inner cist; and a mature female from the northern passage (no mention here of the original reference to a mismatched atlas vertebra amongst the bones of this individual – Selkirk 1971, 8). Of particular interest is the fragment of a leaf-shaped flint arrowhead embedded in the third lumbar vertebra of the young adult ?male from the southern outer cist, and the lethal injury this represents is considered in detail by Christopher Knüsel (pp.218-220). Also noteworthy are the cremated remains of at least one adult ?male from the northern inner cist.

As for the process of burial deposition the authors wrestle commendably with the various clues available, and concede, somewhat grudgingly perhaps (a reflection of divergent opinions?), that at least some of the remains reflect the decay of successive in situ inhumations. Other remains are considered possibly to represent alternative rites of burial practice involving disarticulated skeletal parts, though there is no enthusiasm expressed for excarnation, which has continued to loom large in previous discussions of the AuW burial rites (cf. Darvill 2004, 147).

In considering the skeletal evidence the question of post-depositional alteration of the burial deposits, as a result of natural decay of both the human remains and the enclosing monument, and as a consequence of human, animal and vegetable intrusion during and subsequent to the burial phase(s), is perhaps underplayed. The empty northern outer cist, with its single tooth, is surprisingly given very little attention in this report, leaving readers to speculate for themselves as to reason or implication. I have no doubt that the human bone assemblage from AuW will, now that it is published, receive further attention from researchers, and deservedly so.

Only one possible instance of an accompanying ceramic gravegood was recorded – 24 sherds from the southern passage, representing the decay of a half-complete ‘Abingdon’ plain bowl. Whilst in the pottery report (p.279) this is accepted as a deliberate deposition, perhaps freshly broken for the purpose, the account in other places is more equivocal: ‘[i]ts deposition should not necessarily be regarded as an act linked to the deposition of human bone deposits at the end of the passage’ (p.131). Pottery fragments from elsewhere within the cists and passages are considered to be derived from the pre-barrow occupation (p.278), which leaves a fossil echinoid and an antler comb, both from the area of the southern passage, as the only other mooted gravegoods, though in neither case is this possibility affirmed; rather the opposite (pp.131–132). However, the lithic analyst regards the flint knife and leaf-shaped arrowhead from the southern inner cist as associated with mortuary ritual, not derived from the pre-barrow occupation (p.306). From the original interim (Selkirk 1971, 8) we learn that this arrowhead was located beneath a rib, but unfortunately none of the above-mentioned items, not even the bowl, is shown in position on the drawings of the burial deposits. It might also be queried why no consideration is given to the possibility that some of the ‘pre-barrow’ Neolithic flints may in fact have been part of the burial deposits, in view of the very suggestive distribution below the chambered area of some of the diagnostic pieces (fig.2.9). And finally in relation to the finds, the antler comb surely warrants more attention than it gets (p.248). These implements are usually associated with causewayed enclosures in southern England (Piggott 1954, 83; Smith 1965, 125-127) from a slightly later phase within the Neolithic and it would be useful to have this example radiocarbon dated.

Centrally beneath the barrow is the enigmatic feature F11. In the original accounts of this feature it was described as a sub-soil hollow (Evans 1971; 1972), but in the present report it has become an ‘incontrovertible’ tree-throw pit (p.75). The reasoning for this (other than that every excavation of a Neolithic site these days has to have at least one tree-throw pit!) is not spelled out, and there is some special pleading required to explain the bipartite fill as the result of more than one fallen tree in the same spot (pp.57–59). This is an explanation claimed to be supported by the wide range of flintwork from these fills, though this is not apparent from the flint report itself (pp.297–301). What is also not so clear from the present account, and which is much easier to appreciate from Evans’s original work (1971, figs.6-7), is that this feature is in the centre of ‘timber structure 1’. Whilst it clearly pre-dates post-hole F10, which is cut into it, the locational coincidence should be mentioned, if only to allow any actual association to be dismissed.

The pre-Neolithic inhabitation at AuW is characterized from the flint typology as including both Early and Later Mesolithic activity. The Early Mesolithic aspect, principally isolated on the basis of the obliquely blunted microliths, is compared, for not entirely obvious reasons, with the somewhat indistinct ‘Horsham’ assemblage from St Catherine’s Hill near Guildford, against which the smaller size of the AuW microliths is explained by greater distance from raw material sources. Comparisons might better have been drawn from further north than further south, for example with the ‘Honey Hill’ Early Mesolithic assemblages from Warwickshire and Northamptonshire (Saville 1981a; 1981b), but, even so, separating the Early from the Late within the Mesolithic assemblage at AuW will always be difficult in the absence of any obvious spatial separation within the buried soil, for which subsequent Neolithic disturbance can perhaps be blamed.

There is significant evidence for Neolithic activity in and on the buried soil below the barrow, both artefactual and by way of features, including two possible structures and a hearth. The report perhaps fails to recognize sufficiently that this preserved evidence is only preserved by virtue of the protection afforded by the overlying barrow. Strictly speaking, there is no necessity to imagine the barrow is where it is because of what is beneath it – such traces from Neolithic activity (if indeed they were still visible in any sense when the barrow was built) would no doubt be widespread in the landscape and sub-barrow surfaces merely provide a snap-shot of this. And a health warning should be attached to the ‘midden’ and the importance accorded to it by Whittle, who regards it as somehow ‘highly charged’ (p.357) and would like to see it having a continuing relevance during barrow construction (pp.81, 96, 102, & 135). This reviewer would prefer to interpret the midden as the remnant of the inevitable concentrated mix of cultural and organic residues left behind by Neolithic occupational activity but rarely glimpsed in this fashion simply because of the lack of preservation in other contexts. This reviewer would also reject the interpretation of the Mesolithic artefacts as deliberately gathered and redeposited by Neolithic people (p.351) as a way of actively engaging with the past (p.35). I would of course admit that my disagreements are on the level of subjective interpretation and reflect my recognition of chance rather than intentionality as often a more parsimonious explanation for patterning in archaeological residues, as opposed to the authors’ reluctance to countenance the mundane (p.35).

Turning to the barrow itself, AuW has always seemed somewhat anomalous within the architectural canon of Cotswold tombs. There are numerous interesting aspects to its construction and plenty of room for debate over the interpretations offered. AuW seems to be the only clear-cut instance of a chambered Cotswold long barrow with underlying stake-alignments. The report havers somewhat over the interpretation of these, in some cases stating that the stake-holes represent the position of hurdling or shuttering which continued through the height of the barrow, forming part of the constructional elements of some bays, at other times being much less positive about this. The excavation evidence appears equivocal. A minimalist interpretation could be proposed whereby some of these features relate to the pre-barrow Neolithic occupation and that some are merely to do with preliminary marking out of aspects of the barrow plan and do not have direct structural significance or in all cases prefigure the actual architecture – as is the case in bays 3, 4, 5 and 19. In terms of the earliest stages of barrow construction there is a certain lack of clarity about the ‘axial divide’, a structural entity which is fundamental to the whole barrow-building enterprise, and the presence and character of the ‘turf stacks’ (p.93), which are supposedly an important part of this, are difficult to appreciate from the sectional and photographic record presented.

At this point, and before making any further observations, one must acknowledge the difficulties under which the authors of this report were working. AuW was by no means an entirely preserved barrow – it had suffered considerable depredation by ploughing and other disturbances which compromised the available record and although the excavation was virtually total, it was conducted over seasons by the cuttings technique (fig.1.7) rather than by complete open-area exposure. Given the complexity of these ancient decayed monuments, cuttings can create interpretative problems (as I know from my own work at Nympsfield: Saville 1979). In view of these drawbacks, the level of information we are presented with about the barrow construction at AuW is impressive.

One does wonder, however, if the authors have quite got to grips with the kind of constructional technique involved in Cotswold barrow building. This comes across when dealing with the internal and external ‘walling’, where it is not always apparent that the nature of this as revetment to the material to which it is bound has been appreciated. (McFadyen, p.353) does recognise the structural significance of the dumped material, but elsewhere in the report the bays are ‘infilled’ and slabs are ‘propped up’.) Moreover, the interpretation of a two-phase construction, with an initial eastern termination in a flat façade at the east end of bays 5 and 19 is frankly unsupportable. The interpretation rests on there being a corner return of the inner revetment on the SE of bay 19. This is shown as a definite corner on various plans (eg, fig.1.6 & fig.4.38), but these are themselves schematic and interpretative. The photograph (fig.4.48) and the most detailed stone plan (fig.4.37) make it clear that there is no actual return here, other than the impression of one created by the medieval disturbance immediately to the east. There is no return to the revetment to the NE of bay 5 and there is no façade worthy of the name. Subsequent support for the authors’ interpretation is claimed from a change in the character of the external revetments east of bays 5 and 19 (p.117), but this has little validity because of the disturbed nature of the entire east end of the monument.

Similarly, when the ‘transverse corridor’ in the area of the burial ‘cists’ is referred to as being ‘blocked with stone packing’ (p.329) this, in my view, misunderstands the nature and demands of building around orthostatic elements within barrows. The stonework here is not ‘packing’, but the essential and integral supporting structure for the orthostats and the dry-walling and corbelled and/or capstone roofing which undoubtedly formed the lost superstructure, which was decayed, collapsed, and/or robbed well before the time of excavation. Offset 15/16, with its slabs pitched to the east (fig.4.44), would seem to be part of this construction in forming the outer face of the chambered area stonework, raising the possibility that in this area at least the bay 16 (‘transverse corridor’) construction preceded, not followed, the building of the bay to the west.

In fact the whole concept of the ‘transverse corridor’, as what the authors appear to feel was, until a relatively late stage in the constructional sequence, an open space which allowed access externally to the ‘cists’ for burials, bears revisiting since this reviewer sees it as entirely possible that the area of the orthostats would have been roofed at a sufficient height for there to have been access for burial in the conventional way along the entrances and passages, at least to the ‘outer cists’. The photographic evidence appears to permit the belief that there could have been access from the south to the southern outer cist, either over the top of orthostat 1 or by the movement of this thin slab (which resembles a blocking slab, despite the authors’ explicit denials p.86). The fact that this orthostat and orthostat 17 are so deeply embedded (fig.4.7) actually supports this idea, since frequent movement of the slabs would encourage scooping and lowering, whereas the other orthostats, as is common in Cotswold tombs, are fixed elements which are only slightly embedded, if at all. The angle of the lateral orthostats 3, 6, 13, and 16 as found is almost certainly due to post-construction deterioration of the architecture of the chambered area and they would originally, from architectural necessity as load-bearing elements, have all been upright (contra the authors p.181). Once erect, and with dry-walling /corbelling above them (see fig.4.43), there would easily be sufficient room for movement of the outer transverse orthostats and access to the outer cists, and my best guess is that there would have been some kind of access to the inner cists as well (the photographs in figs.4.55 and 4.57-58 are certainly suggestive of the possibility of ingress into the northern chambered area, assuming roofing at sufficient height). Perhaps there cannot now be certainty on these points, but the uncertainties demand the keeping of an open mind.

Another interpretative problem concerns the ‘Roman’ quarry, which is given insufficient critical attention in the report. There is an inherent improbability in a quarry of any kind being dug in this location next to the barrow when the barrow itself would have been a far easier and more productive option for quarrying. Indeed, what would such a shallow quarry like this be for in Roman times, either in terms of the material it would produce or in terms of what that material would be used for? Far more likely, surely, is that the ‘Roman’ quarry is merely the late upper infill of a major Neolithic quarry which provided much of the material for the barrow construction. The location adjacent to the north side of the barrow (fig.1.6) is so reminiscent of the setting of the southern quarry at Hazleton North (Saville 1990, fig.10), that the failure of the authors to consider other explanations seems blinkered.

The final discussion is extensive and a little meandering in its desire to cover as many interpretative angles as possible. It is rather Evenlode/Oxford-centric in its geographic focus. AuW is after all a Cotswold-Severn tomb and the Severn valley deserves some consideration in thinking about the Neolithic context as well as the Thames valley; indeed a pan-Cotswold SW/NE axis may have been as relevant to the tomb builders as a Cotswold/Thames axis. Some specific references to Hazleton North which contradict the original excavation report are only referenced to McFadyen’s unpublished thesis (p.352–353). In fact, apart from the useful overview (pp.327–330) the discussion section is for this reviewer the least interesting, least convincing part of the book, and it is certainly the most redolent of its time. Had the report been published in the 1970s or 1980s we would have been spared agency, temporality, and materialities and such tendentious pontifications as ‘places are powerful’ (p.344) and ‘bodies, like places, are powerful’ (p.358), and we would surely have avoided cloying purple prose such as:

[t]o dwell, in any form, is to remember, and there is commemoration at work in the formation of middens … just as later … the midden itself is incorporated, like an egg under a nesting bird, into the nascent mound (p.347).

Any possible import of this is dissipated by attempting to resolve the conundrum of how an egg is incorporated into a nesting bird. These annoyances are a small price to pay, however, given the overall excellence of the report.

As far as usability is concerned, in general the layout of the volume is very good, with the caveat that the distribution plots, which are all grouped together on pages 36–50, would have better been split between the relevant categories concerned and placed in the chapters to which they directly relate; so that, for example, the pottery plots (figs.2.19–24) would have been more useful in the chapter on ceramics. And a consolidated list of the contexts would have helped greatly in deciphering the sections. Putting copies of the basic barrow plan with the bay numbers (Fig.1.6) and the cist plan with orthostat and burial deposit numbering (fig.5.1) on to the end-papers would have made using the report much easier.

The copy-editing is of a high standard and I noticed few faults in grammar other than a slight tendency for putting plural verbs with singular nouns. The use of the word ‘plaque’ throughout this report to indicate thin laminated pieces of limestone as used in the revetments is not explained (p.79); I know of no precedent for this usage and it seems very odd. Actual errors spotted include: p.26 the lower fill of F11 is shown on fig.4.32, not figs 2.5 or 4.29; p.33 Chancerel and Kinnes 1991 is cited here but not given in the bibliography; p.51 reference to the grid on fig.2.1 should read fig.2.7; fig.4.20 incorrectly shows the axial division continuing in bay 11; p.93 the axial divide 10/12 (context 62) is referenced to the section on fig.4.34 which does not impinge on the axis and on which context 62 does not appear; p.260 for sample 869 read sample 103/34; fig.4.34 orthostat 6 is labelled 66; context 66 overlies the buried soil in fig.4.31 lower, but disappears from the continuation of this section in fig.4.32 upper. There is no index. And at the risk of appearing very po-faced, I cannot see that the colour photograph of the excavator up a photographic tower (pl.1.4) adds anything to the volume, nor do parts of the late John Evans’s chapter do him or his field of study any favours – the editors should have heeded the advice of their referee (p.77).

But I do not wish to end on a sour note. This volume is a fantastic achievement and it contains a wealth of information which will be pored over for decades to come. Academic libraries and scholars of the Neolithic period should purchase this book now. Once out-of-print it will be highly sought after; I see no way in which a digitally scanned on-line version of an excavation report like this, where one needs constantly to refer back and forth between text, plans, sections and photographs, could in practice be used by archaeologists as a substitute for hard copy (English Heritage please note!).

Alan Saville
Edinburgh, National Museums Scotland

References Bayliss, A. & Whittle, A. (eds), 2007. Histories of the dead: building chronologies for five southern British long barrows. Cambridge Archaeological Journal 17.1 (supplement)
Bayliss, A., Benson, D., Galer, D., Humphrey, L., McFadyen, L. & Whittle, A., 2007. One thing after another: the date of the Ascott-under-Wychwood long barrow. Cambridge Archaeological Journal 17:1 (supplement), 29–44
Benson, D.G. & Clegg, I.N.I., 1978. Cotswold burial rites? Man 13, 134–137
Case, H., 1986. The Mesolithic and Neolithic in the Oxford Region. In G. Briggs, J. Cook & T. Rowley (eds), The Archaeology of the Oxford Region, 18–37. Oxford: Oxford University Department for External Studies
Chesterman, J.T., 1977. Burial rites in a Cotswold long barrow. Man 12, 22–32
Darvill, T., 2004. Long Barrows of the Cotswolds and Surrounding Areas. Stroud: Tempus
Evans, J.G., 1971. Habitat change on the calcareous soils of Britain: the impact of Neolithic man. In D.D.A. Simpson (ed.), Economy and Settlement in Neolithic and Early Bronze Age Britain and Europe, 27–73. Leicester: Leicester University Press
Evans, J.G., 1972. Land Snails in Archaeology. London: Seminar Press
Piggott, S., 1954. The Neolithic Cultures of the British Isles. Cambridge: Cambridge University Press
Saville, A., 1979. Further excavations at Nympsfield chambered tomb, Gloucestershire, 1974. Proceedings of the Prehistoric Society 45, 53–91
Saville, A., 1981a. Honey Hill, Elkington, a Northamptonshire Mesolithic site. Northamptonshire Archaeology 16, 1–13
Saville, A., 1981b. Mesolithic industries in central England: an exploratory investigation using microlith typology. Archaeological Journal 138, 40–71
Saville, A., 1990. Hazleton North: the Excavation of a Neolithic Long Cairn of the Cotswold-Severn Group. London: English Heritage
Selkirk, A., 1971. Ascott-under-Wychwood. Current Archaeology 24, 7–10
Smith, I.F., 1965. Windmill Hill and Avebury: Excavations by Alexander Keiller, 1925–1939. Oxford: Clarendon Press
Whittle, A., Barclay, A., Bayliss, B., McFadyen, L., Schulting, R. & Wysocki, M., 2007. Building for the dead: events, processes and changing worldviews from the thirty-eighth to the thirty-fourth centuries cal BC in southern Britain. Cambridge Archaeological Journal 17.1 (supplement), 123–147

Review Submitted: December 2007

The views expressed in this review are not necessarily those of the Society or the Reviews Editor.

Home Page The Prehistoric Society Home Page