The first hint of a ring ditch emerged in 1972, when Martin Green noticed a faint, curving line in freshly ploughed soil on his Cranborne Chase farm in Dorset. Lying a mere 20m from the boundary of the great Dorset Cursus, the ring ditch was just another very small element of a vast ‘sacred landscape’ associated with the longest Cursus monument in England. This landscape comprised several hundred monuments dating to the Later Neolithic and Bronze Ages, which embellished the environs of the great spinal earthwork.
Thirty-five years later, the ring ditch appeared once more, this time on a 2007 geophysical survey undertaken by Bournemouth University students supervised by Paul Cheetham. This picked up faint traces of a ditch and a substantial central feature. As plough erosion was continuing, Martin decided to excavate the site in 2009. The results were completely unexpected.
Digging the ring ditch
Longstanding readers may well be familiar with Martin Green. He has spent several decades investigating archaeological features, many prehistoric in date, on his farm (CA 67 and 138). While many of the published explorations have focused on monuments, artefacts, and environmental evidence, with the help of staff and students at Bournemouth University Martin is now re-examining the human remains. The catalyst for this work was provided when Martin’s latest excavation revealed the remains of seven further individuals at the Canada Farm ring ditch, ranging in date from the Beaker period (c.2,500-1,700 BC) through to the Middle Bronze Age (c.1,500-1,150 BC).
Detailed osteological studies of the primary burial by Francine O’Malley and of secondary interments by Lauren Bailey were undertaken as Master’s Degree dissertation projects at Bournemouth University, supervised by anthropologist Martin Smith. This work produced evidence for remarkable post-mortem treatment of the bodies, along with a surprisingly early radiocarbon date. Together these discoveries are challenging preconceptions of the treatment of the dead in these periods within Britain.
The 2009 excavations revealed two phases of ditch. Both were very shallow, with the outer one about 0.4m deep and the later, inner ditch that cut it only surviving to 0.1m. Such shallowness underscored the need for excavation before the feature was lost forever. These rings are often the only surviving traces of quarry ditches cut to win material for a prominent burial mound, long since obliterated by the plough. The recent reconstruction of a burial monument on Down Farm emphasises the dramatic nature of such a mound: the brilliant, gleaming white sepulchre clearly demonstrates the powerful visual impact of earthworks cut from fresh chalk.
Nine postholes lie directly north-west of the rings, representing a structure around 3m square. While the presence of a nondescript set of postholes might seem ambiguous at best, careful study of the human remains has revealed that the bodies were not simply buried and left to rest in peace immediately after death. The proximity of this structure to the burial monument suggests it may have been where cadavers were temporarily laid after death, either in the form of a shielded morgue-style building or an excarnation platform where the deceased would be exposed to the elements, and any passing carrion birds.
A Beaker burial
The earliest burial within the ring ditch occupied a shallow recess, which may originally have held a rectangular, wooden coffin. This recess had been dug into the bedrock floor of a larger burial pit that was later in-filled with rubble.
The skeleton belonged to a 25- to 30-year-old male. Interred in a foetal position (known technically as semi-flexed), he had been placed on his left side, facing east towards the path of the rising sun. While the body was mostly articulated when buried, the jaw had been removed and placed in the north-west corner of the ‘coffin’. The left and right arms were also slightly out of position, possibly after being disturbed when the jaw was removed. Alternatively, their movement may simply have been a consequence of the fragile nature of the sinews binding the bones together at the time of burial. Telltale traces of gnawing by scavenging carnivores suggest that the body had experienced short-term exposure prior to burial, although defleshing was incomplete when the remains were interred. The cause of death was not certain, but the deceased had received an only partly healed head injury.
Accompanying grave goods included two slivers of a boar’s tusk and an antler pendant or ‘toggle’ left touching the man’s jaw. A Wessex/Middle-Rhine-style pottery Beaker and single flint flake were positioned by his feet. This particular form of pot is associated with the Beaker culture that developed in the Rhine region. It is the appearance of such pottery in Britain that triggers the development of the Wessex ‘Beaker culture’ – a term that was initially adopted to refer to what is now seen as a specific social stratum during the Early to Middle Bronze Age, characterised by ‘wealthy’ and elaborate grave goods. Although such cultural transmission implies a movement of people between the Rhine and Britain, strontium-isotope analysis on the deceased’s teeth indicates that he was native to the Wessex chalk lands.
Radiocarbon analysis of the primary burial returned a date of 2620-2470 cal BC, just predating the start of the Beaker phase as currently defined in Britain. This seemed remarkably early, so a second date was obtained from the same bone thanks to Mike Parker Pearson and the Beaker People Isotope Project. This produced a range of 2470-2290 cal BC. Failing a T-test for significance, it is frustratingly clear that there is a problem somewhere. A further test is required to determine which of these results is more accurate. Nevertheless, it is worth noting that even if the later date for the primary burial is closer to the truth, it would still be an exceptionally early date for a burial that included a Wessex/Middle-Rhine-style Beaker.
Strip flesh from the bones
The burial mound’s second phase of use belongs to the Middle Bronze Age. Then, burials were placed either within the ditches or, in one case, just outside them. The individuals were all relatively young, including two adolescents (F3 and F4), one possible child or adolescent (F5), one infant of around 3 or 4 years old (F6), and, lying south-west of that, a cremation scatter associated with Middle Bronze Age pottery. On the basis of the teeth, the final burial to be uncovered (F8), which cut the phase 1 ditch terminal, was an adolescent or young adult. Aside from this last individual, whose skeleton had been severely damaged by ploughing, most of the Canada Farm skeletons are relatively complete.
The remains of F3 and F4 have been radiocarbon dated to 1620-1390 cal BC. While not all the bodies have been dated, similarities in burial style suggest a similar time-frame. These five skeletons were all interred in a foetal position, with the recovered bones lying in their correct anatomical position. While F3 had been covered with flint nodules, none of the burials was accompanied by any grave goods, conforming to the normal tradition at that time. There was plenty of evidence that considerable attention had been paid to the bodies themselves.
Further study of the Middle Bronze Age remains in the lab revealed some particularly unusual examples of skeletal modification. The first discovery was of multiple chop-marks on the right shoulder blade of F4. Cut-marks were also detected on the socket of the upper arm bone. The repetition of such scoring in small, localised areas does not match the kind of wounds inflicted during fighting. Instead it has far more in common with chop-marks found on butchered animal bones. This parallel may suggest that similar methods were being used to strip the flesh from human corpses at Canada Farm.
Unfortunately, the poor preservation of the remaining shoulder blades prevents any appreciation of how widespread this practice was. While three other potential cut-marks were observed within the sample (including the left and right shinbone of F3), it is not clear whether they were genuinely ancient in origin or the result of excavation damage. This is also true of F4’s right kneecap, which has a distinct slice visible on its outer surface.
More intriguing than the cut-marks, however, was the presence of several circular holes positioned at the ends of the leg bones. These were identified on the thigh and lower leg bones of F5 and F3. The holes measure approximately 9mm in diameter, and penetrate into the ends of the bones to varying depths. They are not consistent with any disease process or any previously detected form of natural post-mortem change. In fact, the only explanation is that these holes had been manually – and deliberately – drilled into the bone.
As these individuals were relatively young, their leg bones were still growing. The curving end plates at the long bone joints do not permanently fuse to the shaft of the bone until it stops growing. With this in mind, it is particularly interesting that while all the drill cavities in the long bones were associated with the ends of the bones, not all of the holes located in the shafts had holes in the corresponding end plates. In the case of F3, for example, the drill holes penetrate both ends of the right femur but only into the unfused shafts – the end plates had not been similarly modified. In contrast, the bottom end of the left femur exhibits a similar drill hole through the end plate, with only a slight continuation of this hole into the shaft. This differential distribution can only mean one thing: the holes must have been made after death when the bodies were in a sufficiently advanced state of decomposition that the ends of these bones were accessible.
In addition to the holes noted on the leg bones was a smaller drill hole (approximately 3mm) through the centre of the first metacarpal – the bone that articulates with the thumb – on the juvenile remains of F5. This hole required skilled workmanship, and it is unlikely to have been made while the bone was still surrounded by soft tissue. It is also interesting that while the bones of the hands and feet were extremely well preserved in all cases, this was the only example of a drill hole in such an area.
The modifications to the long bones must have occurred after they were no longer fixed in their anatomically correct locations. Despite this, the individuals were still buried as though they remained articulated. While trepanation, that is removing a disc of bone from the skull, has been widely observed throughout pre-history, the drilling of other human skeletal parts has little, if any, precedence in published literature. One other British example currently under investigation is visible on an elbow bone from a skeleton excavated by the Cambridge Archaeological Unit in Over, Cambridgeshire. This skeleton has also been dated to the Bronze Age, raising the tantalising prospect of a shared motivation for the ritual.
For further parallels one must travel far from Bronze Age Wessex. Drilled human forearms were removed as macabre war trophies in prehistoric California. Human crania were similarly adapted for suspension and display in prehispanic Peru, although again this relates to trophy-taking rather than an aspect of funerary practice.
So what was going on?
It has been suggested by Martin Smith and Michael Allen that the drill holes at Canada Farm may represent an attempt to keep the skeletons held together following death. The skeletons certainly appear to have been arranged anatomically correctly at the time of burial, despite the drilling proving that at least parts of two individuals had disintegrated into loose bones prior to this.
That comparable drill holes are so rare need not mean the customs they are associated with were unusual. If the holes were made in an attempt to keep skeletons connected, it is entirely plausible that other, less destructive methods may also have been used, both at Canada Farm and elsewhere. The tightly flexed, foetal positions of the Bronze Age skeletons have prompted Martin Green to speculate that the bodies may have been wrapped or bound in position. Such binding could have slowed the decay of some of the body’s more fibrous soft tissues, in particular the ligaments and perhaps at least some of the skin. This would effectively produce a partially ‘mummified’ corpse. If such a cadaver was not immediately buried but retained in a structure such as a house, this treatment would help the deceased retain their form as an individual rather than a loose assortment of bones.
Inevitably, the greater the length of time that elapsed prior to burial, the greater the degree of deterioration to a body. Eventually the corpse would fall apart. Given that the drill holes could only have been produced some time after death, it seems plausible that they represent an attempt to keep the body intact by pegging it back together, presumably using wooden pegs or dowels that have not survived over time.
The Cladh Hallan mummies
The characteristics displayed by the two phases of burial conform to some general archaeological norms for these periods. The grave goods associated with the earliest burial are typical of Beaker interments, while the later phases re- defined an ancient monument before insertion of further burials that were devoid of grave goods. But what was really happening at Canada Farm?
The answer may lie in an alternative explanation for the early date of the primary skeleton in comparison to the Wessex/Middle Rhine Beaker. We now know that a variety of complex burial practices took place in Neolithic and Bronze Age Britain, frequently involving manipulation of the skeleton, including examples where considerable periods of time elapsed between death and eventual interment. One well-known example of this involves the mummified human remains discovered at Cladh Hallan on South Uist in the Hebrides (see CA 265).
The tightly flexed Cladh Hallan skeletons had been temporarily interred in a peat bog in what appears to have been a deliberate act of mummification, as opposed to the burial of bodies in peat bogs with no intention of retrieving the remains. The mummies caused further excitement when it was discovered that they were composite works containing body parts from at least six different individuals, whose remains have been dated to between 1,600 and 1,300 BC, but which were not finally buried until around 1,000 BC. That the remains came from individuals whose lives spanned three centuries of South Uist history has led researchers to suggest that selection may have been linked by kinship.
No DNA analysis has been performed on the Canada Farm skeletons, but a shared genetic defect between the primary skeleton (F1) and F3 may be an indication of a genetic lineage. Both of these skeletons exhibit a mutation that results in the presence of an extra tooth in an abnormal location. In this case, both additional teeth were located in the upper jaw. The presence of supernumerary teeth is thought to affect between 0.1% and 3.8% of a population, making it very rare. To find this anomaly in two out of the seven Canada Farm individuals strongly suggests that the feature has been inherited.
While the Cladh Hallan remains have been dated considerably later than the earliest activity at Canada Farm, they are contemporary with the second phase of site use, and share broad similarities such as the tightly flexed arrangement of the skeletons. It is possible that shared customs were being performed, which raises the question of how deep these similarities may run. It is also possible that the position of the Canada Farm bodies was changed at the point when they were finally laid to rest in the ground.
The Canada Farm discoveries must raise the possibility that a widespread practice of curating the dead, through practices such as mummification or desiccation – literally drying – of the corpse, was being performed in different parts of the British Isles rather than being peculiar to the Hebrides.
Returning to the primary skeleton, if this body had also been retained and ‘curated’ after death for an extended period, it could explain the apparent discordance between the date of the burial and its associated Beaker. Such a suggestion presents a note of caution to anyone dating human burials purely on the presence of associated objects. But if the curation of ‘mummified’ bodies was a tradition that had a long pedigree, exactly how far back do such practices go? At the nearby burial site of Wor Barrow, also on Cranborne Chase, investigations are currently under way by Mike Allen and Martin Smith to test this question further, inspired again by anomalous dating results and by the intriguing arrangement of skeletons found there. Watch this space for further details in the near future.
We would like to extend our thanks to all the people who have facilitated the excavation and analysis of the Canada Farm ring ditch, and this subsequent article. In particular, we would like to acknowledge Francine O’Malley, for her osteoarchaeological analysis of the primary burial; Dave Cousins, for his invaluable help presenting our images; Mike Allen and Mike Parker Pearson, for their help dating the remains and reviewing this article; and Chris Evans from the Cambridge Archaeological Unit.
All photos: Martin Green, unless otherwise stated.