British Archaeology banner

Cover of British Archaeology 118

Issue 118

May / June 2011



All the latest archaeology news from around the country

on the web

Fieldwork opportunities and the South Downs


Your views and responses


THE BIG DIG: Gough's Cave, Somerset

6 Threatened Sites

Overthrowing Egypt's Past


ISSN 1357-4442

Editor Mike Pitts


THE BIG DIG: Gough's Cave, Somerset

Gough's Cave has long been known for its huge collection of ice age remains retrieved in many excavations since 1892. Over the past 30 years, archaeologists have been applying modern analysis and new excavation to the site, with dramatic results. Silvia Bello, Simon Parfitt and Chris Stringer here describe the latest and most sensational discovery: the world's oldest directly dated cups made from human skulls.

People had been digging in Gough's Cave, and debating the meaning of what had been found there, for over a century. But the cave acquired special interest in the 1980s. New research and excavation confirmed the great age of the remains, and threw important light on a well-aired controversy that went back at least 50 years: for the first time, in a report published in 1991, some marks on the human bones were shown beyond doubt to be the result of people removing flesh. In other words, the spectre, if not proof, of ancient cannibalism – a word avoided by the modern archaeologists – was firmly on the agenda.

But what had inspired scientists to go back to Gough's Cave? The natural tunnel, about 50m long, was discovered in 1890. It rapidly became a feature of the popular tourist attraction of Cheddar Gorge, Somerset's famous craggy limestone valley. Most excavation in the cave occurred when its owners sought to improve facilities for visitors. Investigators realised that, millennia before, people had sheltered there, leaving rubbish and human remains that became entombed in the gravel and silt accumulating around them – curiously, most of the human remains did not seem to have been deliberately buried.

Freshly uncovered, these remains became part of the tourist draw, especially a complete skeleton found in 1903 and dubbed "Cheddar Man". Ancient DNA was recently extracted from one of this man's teeth, and further studies claimed to show that his modern descendants lived in the area. We now know this body to be mesolithic (dated to 10,250 years ago), an unusual burial for this era. But the main interest in the cave is older still.

Particular flint tools, and carved reindeer antler and mammoth ivory, linked these older remains to finds on the continent. Similar material was associated with the famous ice age painted caves in France, where it was known as Magdalenian. In Britain finds of this age were very rare; the even rarer quantities of human remains gave Gough's Cave international interest.

Roger Jacobi initiated new excavations in 1986. In the course of his intensive study of hunter-gatherer remains across Britain, he had noticed what seemed to be a small undisturbed ancient deposit near the cave mouth that was in danger of being accidentally destroyed. He decided to record it before that happened, and what is now the Natural History Museum (NHM) joined him the following year. The small dig produced a rich assemblage of flint tools, and animal and human bones. Radiocarbon dates showed these to be at least 12,000 years old.

Meanwhile, the previously excavated remains had been re-examined. Andy Currant had studied the many animal bones, and Chris Stringer had compiled the first comprehensive report on the human remains, after the Cheddar Caves Museum loaned most of them to the NHM. Stringer had confirmed that these bones represented late ice age people – that is, modern humans. He had also tentatively noted several "possible" cutmarks and scratches, code for butchery with flint tools.

Jill Cook, at the British Museum, had looked at these marks with a scanning electron microscope. She saw numerous grooves and scratches on the fragments of skull and jaw, but concluded these were most likely to have been inflicted on the bones by trampling after they had been buried. There was just one bone she was less sure about: marks on a jaw, she said, might have been made when the tongue had been cut out.

However it was the new excavation that really galvanised prehistorians. This demonstrated that most of the human remains could, as suspected, be properly associated with the ice age tools and animals. This meant that all the material could be safely studied as evidence for ice age use of the cave. And the new bone finds offered greater promise than the by now much handled and preservative-coated old ones.

The animal remains threw light on how the world looked outside the cave. The commonest mammals represented were wild horse and red deer (both apparently hunted), but there were also bones of wolf, brown bear, lynx, saiga antelope, arctic fox and arctic hare. This was an open landscape, cold in winter and hot in summer, during a warm interlude near the end of the ice age's final big freeze.

The 1986–87 excavations added some 120 human remains to the collection. When Cook looked at these, she realised how much her earlier study had been hindered by the bones' condition.

This time she was in no doubt. About a quarter of the human bones, the great majority of which came from the head, showed "discrete incisions". In every case she had been able to examine, she wrote in 1991, "these incisions can be confidently ascribed to human activity rather than to natural causes". People had removed heads and jaws shortly after death, and cut the skin and flesh away from skulls. "[C]annibalism", she added, "is difficult to confirm or refute from the evidence available" – but if it had happened, it could not have been because people were starving; the animal remains showed there was plenty to eat out there.

People for food

The Natural History Museum investigations in the cave continued into 1992, with a second small deposit being rescued (see plan, page 13). Intensive study of all finds, new and old, has added substantially to our understanding of the scene 14,700 years ago.

That date is itself one of the more surprising outcomes. It was obtained by Tom Higham at the Oxford Radiocarbon Accelerator Unit, applying increasingly sophisticated analytical techniques. He has shown that people arrived at Gough's immediately the climate warmed, with dramatic speed, after the extreme cold of the Last Glacial Maximum (see Science, Jul/Aug 2009/107). The short time in which the cave became filled with debris – possibly less than a century, perhaps three or four generations at most – supports a disputed case that the unusual quantities of material reflect the cave's use by groups coming together from a wide area, and not just by a few people over a very long time.

Jacobi published a major study of the cave's contents in 2004, following another loan from the Cheddar Caves Museum, this time of the flint tools to the British Museum. He argued that hunters had used the cave in both winter and summer, for butchery (especially of horse, whose bones were smashed to extract brains and marrow) and working hides and bones (for example to make awls and needles for sewing skins).

A variety of evidence suggests the hunters' territory included much of southern Britain. Good flint for the thousands of tools was obtained 70km to the east in Wiltshire, while a piece of amber was probably picked up still further east on the shore of the North Sea. As we saw, the major butchered animal at Gough's is the wild horse. Yet stable-isotope analysis of human bone indicates that the people represented had eaten mainly red deer and wild cattle (incidentally showing they really were hunters) – so they had presumably travelled from an area where these animals were being killed. Finally the mammoth and reindeer remains from the cave are not meal leftovers, but carved artefacts – so these might have been brought from some distance by exchange or expedition from a place where such animals roamed.

By 2004 substantial study of the human remains had been completed, involving scientists from the NHM, Arizona State University, the National Museum of Natural Sciences, Madrid, and elsewhere. The evidence for cannibalism had been examined by the NHM's Peter Andrews and Yolanda Fernández-Jalvo from Madrid, and widely accepted. Human and animal bones had been cut and smashed in exactly the same way, and found mixed together in the ground. The practice, then, was "nutritional cannibalism" – people for food. The one possible exception related to the skulls: despite being fragile and covered in cutmarks, human skulls were relatively complete, suggesting an element of ritual.

Jacobi proposed an alternative. Disarticulation, defleshing and smashing of bone, he said, might be ways of reducing a deceased person to a package that could be carried by a hunter on the move. You might die on a steppe in Essex: but your remains would be laid to rest in a cave in Somerset. But an extraordinary discovery has shown beyond reasonable doubt that nearly 15,000 years ago, people in Somerset chewed parts of human bodies.

Fernández-Jalvo and Andrews asked students to chew cooked pig and sheep bones, thus obtaining a collection of marks they knew had been made by human teeth. They then compared these to marks on some of the bones from Gough's Cave. Their early suggestion that these had been made by human teeth, was proved right. People had chewed human toes and at least one rib (see end note).

This was the state of play when we proposed a new study of the human crania. There are 41 pieces altogether, 37 from skulls and the rest from mandibles (lower jaws). According to Chris Stringer and his NHM colleague Louise Humphrey, these represent at least five individuals: a young child (a little over three years old), two adolescents, a young adult and an older adult. By carefully refitting pieces, some of them from different excavations, Silvia Bello established that among the bones there were three almost complete mandibles and three skull-cups.

More generally, all the indications are that weathering and trampling had not significantly altered the human remains after their burial. Many have cutmarks typical of those known to have been made by stone tools, confirmed by Bello in a new three-dimensional imaging study using an Alicona Infinite Focus optical surface measurement system. Most of these marks were made by slicing; some chopping had occurred, but signs of scraping were rare. Many bones had also been broken up, in a way consistent with the striking of fresh or "green" bone.

The skulls, however, were different. Less than half (46%) showed the percussion marks found more commonly on other parts of the body. But cutmarks were extremely common (on 95% of the pieces), and seen on all individuals regardless of age. The skulls had been cut only on their outer surfaces, and no fragments showed obvious fire damage. Such details are important, for the implications of what follows are controversial.

Alas! our brains are gone

The pattern of cutmarks and other damage on the Gough's Cave bones indicates skilled post-mortem processing of the head, which included careful removal of soft tissues followed by controlled percussion. The complete procedure can be clearly seen.

First, the head was detached from the body: the evidence for this is in cutmarks at the base of the skull, and on skullneck vertebrae. It is likely that this took place shortly after death, before the soft tissues dried out or decomposed leading to natural disarticulation.

Next the mandible was removed. In the case of two upper jaws, the front teeth showed post-mortem scratches and percussion fractures that may have been caused when a lever was inserted in the mouth to disjoint and separate upper and lower jaws.

Then the major skull muscles (masseter and temporalis) were removed, as were the tongue, lips, ears and nose. Cutmarks around and inside the eye sockets and on the face suggest extraction of the eyes and cheeks. The high incidence of cutmarks around the vault, far from muscle attachments, points to scalp removal.

Finally the face and the base of the skull were struck off with minimum damage to the vault, and the broken edges were chipped away to make them more regular.

The modification of these skulls is unique in Britain, but they fit into a wider Magdalenian context. Broken and cutmarked human bones are known, for example, from the Dordogne area in France and the Rhine valley in Germany. There, however, such evidence is more often interpreted as ritual behaviour, and cannibalism is refuted. However, at Gough's Cave human and animal remains had been similarly treated, and were found discarded in the same context. Both human and animal mandibles, for example, were severed from the head, and carefully defleshed and broken, suggesting a common butchery practice aimed at extracting edible tissues.

Despite the evidence indicating cannibalism, the high frequency of cutmarks on the Gough's Cave human bones is unusual. In north America where such frequencies have been observed, post-mortem face damage has been interpreted as a mutilation practice. At Gough's Cave, however, the lack of obvious signs of trauma makes this hypothesis unlikely, and may instead suggest that bodies were stiff from rigor mortis or freezing when they were cut up. Another unusual characteristic of the Gough's Cave remains is the completeness of the cranial vault. At sites where nutritional cannibalism has been documented, the skulls are invariably much broken, often from being struck hard on the top or base.

A more likely explanation for the cutmarks and state of the vaults is that the Gough's skulls were scrupulously cleaned with flint tools, in the first stage in making skull-cups. Scalping and defleshing were followed by removal of extraneous bone and shaping of the vault with a hammerstone and anvil. The large number of small pieces from the cranial base, none of which can be refitted to the more complete vaults, suggests this skull processing took place in the cave.

The combination of cannibalism and skull-cup production at Gough's Cave is so far unique in the European upper palaeolithic. However, similar skull modifications have been described at the Magdalenian sites of Le Placard (Charente, France) and Isturitz (Pyrénées-Atlantiques, France). At Le Placard archaeologists concluded that most of the skulls had been deliberately shaped to make "goblets" or "skull-neck cups", immediately after the bodies had been defleshed. At Isturitz some of these cups had even been engraved with representations of animals.

The striking similarity of these Magdalenian skull-cups to recent examples suggests they were used as containers or drinking-cups. In the fifth century BC, Herodotus portrayed the Scythians as people who drank from the skulls of their enemies. Similar traditions were described in China by Sima Qian (145/135–86BC) and amongst Vikings by Mágnus Ólafsson writing in 1636. In India human skull-bowls were used as libation vessels for a number of Vajrayana deities; the tantric Buddhist ritual seems to be still practiced by the Aghori sub-sect. In the 19th century, skull-cups were reported amongst Aborigines in south-eastern Australia (where close relatives of the deceased were said to have used them to drink water), and in Fiji and other islands in Oceania. In 1808 Byron wrote a poem about an ancient skull-cup, in respect rather than disgust. "Better to hold the sparkling grape," he said, putting words into the skull's absent mouth, "Than nurse the earth-worm's slimy brood... when, alas! our brains are gone, What nobler substitute than wine?"

Despite this widespread occurrence, archaeological evidence for skull-cup preparation is very rare. Direct radiocarbon determinations on two of the Gough's cups at around 14,700 years ago, make these the oldest known dated examples. These were sophisticated hunter gatherers – people like us. They were able to survive in challenging environments, and must have had detailed knowledge of many plants and animals. They made intricate stone tools and fine carvings in bone, antler and ivory, and they painted and engraved caves with astonishing art.

They also conducted quite complex burial practices, sometimes associated with cannibalism. Making the skull-cups was a painstaking process, and the remains reveal considerable anatomical knowledge. We can only guess why it was done. But what we know about these people, suggests that great thought went into the practice. The bones are just the pieces that happen to survive from a much bigger puzzle.

Silvia Bello, Simon Parfitt and Chris Stringer (Department of Palaeontology, Natural History Museum; Parfitt is also at UCL Institute of Archaeology) are members of AHOB3, the third phase of the Ancient Human Occupation of Britain Project funded by the Leverhulme Trust, of which Stringer is director. Their detailed description of the cranial remains (Earliest directly-dated human skull-cups) can be read in the online science journal PLoS ONE. The toothmark experiments are described in "When humans chew bones", by Y Fernández-Jalvo and P Andrews, Journal of Human Evolution 60 #1 (2011), 117–23. A cast of the adult skull-cup could be seen at the Natural History Museum in London until the end of May. Gough's Cave is open to the public.

<< prev | index | next >>

CBA web:

British Archaeology

Jan/Feb 2005
Mar/Apr 2005
May/Jun 2005
Jul/Aug 2005
Sep/Oct 2005
Nov/Dec 2005
Jan/Feb 2006
Mar/Apr 2006
May/Jun 2006
Jul/Aug 2006
Sep/Oct 2006
Nov/Dec 2006
Jan/Feb 2007
Mar/Apr 2007
May/Jun 2007
Jul/Aug 2007
Sep/Oct 2007
Nov/Dec 2007
Jan/Feb 2008
Mar/Apr 2008
May/Jun 2008
Jul/Aug 2008
Sep/Oct 2008
Nov/Dec 2008
Jan/Feb 2009
Mar/Apr 2009
May/Jun 2009
Jul/Aug 2009
Sep/Oct 2009
Nov/Dec 2009
Jan/Feb 2010
Mar/Apr 2010
May/Jun 2010
Jul/Aug 2010
Sep/Oct 2010
Nov/Dec 2010
Jan/Feb 2011
Mar/Apr 2011
May/Jun 2011
Jul/Aug 2011
Sep/Oct 2011
Nov/Dec 2011
Jan/Feb 2012
Mar/Apr 2012

CBA Briefing

Courses & lectures
CBA Network
Grants & awards

CBA homepage