Archive | Cannibalism RSS feed for this section

Gough’s Cave Skull Cups

9 Apr

The surgery went very well, and I’m glad to be out of hospital so soon.  As I entered the house again, my eyes fell greedily upon the latest edition of the British Archaeology magazine.  An article that caught my eye in particular was the latest developments concerning the various excavations, and new modern scientific investigations analysis at the Upper Palaeolithic site of Gough’s Cave located in the Mendip Hills, Somerset, southern England.  It has turned out that the cave has the earliest directly dated human skull cups.  This was reported in the media a few months ago, but now an in-depth article has come out on the online PLoS ONE journal.  The 2011 article, written by Silvia M. Bello, Simon A. Parfitt and Chris Stringer, can be found here and is open access.

Palaeogeography At The Time of use of Goughs Cave (Figure 1, Bello et al. 2011).

The Upper Palaeolithic Site of Gough’s Cave

Now, this is big news.  The site of the butchered animal and human remains is dated to the Magdalenian period of the Upper Palaeolithic, around 14,700 BP (BP simply stands for Before Present) during the end last of the glacial period (Last Glacial Maximum).  The artefacts that have been found in this particular period at Gough’s Cave include flint tools, carved reindeer antler and mammoth ivory; a particular key find is the reindeer antler baton, a fine example of the craftmanship of the humans that used to live and roam this area.  Now the finds we are most concerned with are the evidence of the defleshing of the human remains, and the very probably use of human skulls as drinking vessels.  According to the articles stated above, the human remains were found with fauna including (in descending commonest order) wild horse, red deer, wolf, brown bear, lynx, saiga antelope, arctic fox and arctic hare (Stringer Et al, in BA magazine 2011:16).  The animal remains showed evidence of butchery, in accordance with using the flesh for food.  The co-mingled human remains, mostly cranial elements with post cranial elements also showed butchery marks, and do not seem to be deliberately buried.

A selection of the human cranial elements found, highlighting the breaking and fracturing of the cranial elements during reshaping (Source: Natural History Museum).

Human Cranial Remains and Modifications

The remains subjected to new scientific analysis included 41 elements, 37 from skulls and the rest from mandibles (lower jaw).  From the study of remains it has been suggested that they represent at least 5 individuals, including a young child, two adolescents, a young adult and an older adult (Stringer et al 2011: 19).  There were three complete mandibles alongside three skulls caps present (see above).   Although it had been suggested from earlier excavations, it is now thought that the bones did not suffer much from post-depositional effects (ie weathering or trampling).  Many of the elements have evidence of stone cut marks; most were done by slicing, some chopping but signs of scraping were seen as rare.  The skulls had less evidence of percussion marks whilst cut marks were particularly evident.  Importantly they showed no sign of fire damage (such as colour changes or flaking) and all cuts are ectocranial (Bello et al 2011).

Highlighting the main points of reshaping of the human crania (Figure 8 in Bello et al. 2011).

Carefully placed ectocranial percussion marks on the vault of the crania (Source: Natural History Museum).

The processing of the head can be clearly discerned.  A) The head was detached from the body, probably whilst the body was either frozen or in the grip of rigor mortis.  Cuts at the base of the skulls and on the cervical vertebrae indicate this took place shortly after death.  B) The mandible was removed next, evidence is seen by post-mortem scratches on teeth of both mandible and maxilla alongside percussion fractures (Bello et al 2011).  C) The major muscles of the skull were removed next (Temporalis & Masseter muscles in anatomical position) alongside the removal of the lips, ears, tongue, and the possible extraction of eyes and cheeks.  D) Cut marks along the parietal and occipital elements indicate scalping as well. E) Finally, ‘the face and base of the skull was struck off with minimum damage to the vault, and the broken edges were chipped away to make the more regular’ (Stringer et al 2011).

Key Points

Evidence for cut marks on human bones in the Magdalenian period have also been found in the Rhine Valley in Germany, Dordogne area in France.  Sites such as Le Placard in Charente & Isturitz in Oyrenees-Atlantiques (see above location map), both in France have evidence for similar skull modification and processing.  Strikingly at Isturitz, one example even has carvings of animals in the skull elements.

However, as pointed out in an earlier article on cannibalism, post cranial elements found (including metatarsals with evidence of being chewed by humans) are thought to be an example of ‘nutritional cannibalism’, even with the large amount of faunal remains co-mingled with the human remains.  The slicing marks present on these post cranial elements are consistent with the striking of ‘green’ (fresh) bone.  An interesting experimental archaeological test involved two researchers having their students chew fresh sheep and  pig bones.  This was carried out in order to test if the bite marks found were similar to bite marks on human metatarsal and radius elements found, amongst other bones (Fernandez-Javlo & Andrews 2011).  The results helped to provide evidence that the chewing marks on the human bones (including a distal rib fragment) were probably caused by human teeth themselves.

Rib chewing-archaeology style.  In experimental tests archaeologists found that volunteers chewing ribs replicated the marks made on archaeological material human rib samples at Upper Palaeolithic sites (Fernandez- Jalvo & Peters 2011).

As stated above, the skull elements was treated remarkably different with careful processes present.  There was a distinctly high number of cut marks on the cranial elements present.  Alongside this, a lack of trauma indicates that this is not for mutilation purposes, as seen at some American sites (Stringer et al 2011: 20/Larsen 1997).  At sites where nutritional cannibalism has been documented, the skull is often fractured and broken in aiding access to the brain tissues within.  At Gough’s cave, the skulls have been carefully prepared with flints and carefully processed.

This hints at possible uses of the skull-caps as containers for liquids or holders for other objects.  Ethnographic and historical sources have pointed to various cultures preparing and using human skulls as containers, war trophies or as libation instruments.  Classically, Herodotus portrayed the Scythians as people who drank from the skulls of their enemies, whilst in ‘Buddhism human skull bowls have been used as libation vessels.  In India, the use of skull cups seems to be still practiced by the Agori sub-sect’ (Stringer et all 2011: 20).  Very interestingly, the article by Bello et al (2011) remarks that there are few archaeological finds for skull-caps, in consideration of the wide temporal and geographical spread of ethnographic and historical evidence.  One example is the Neolithic site at Herxheim in Germany, previously discussed in a blog post by Kate Brown.

In conclusion, the Gough’s Cave skulls caps have been securely dated, and are the only ones found so far in the British Isles.  The mystery still remains why they took part in this painstaking process.


Bello, S. M. Parfitt, S. A. & Stringer, C. B. 2011. ‘Earliest Directly Dated Skull-Cups‘. PLoS ONE. (Open Access Article).

Bones Don’t Lie. 2011. Cheddar’s Cranial Cups.  Blog Site.

Fernandez-Jalvo, Y. & Andrews, P. 2011. When Humans Chew Bones. Journal of Human Evolution. 60 (1): 117-123.

Larsen, C. 1997. Bioarchaeology: Interpreting Behaviour From The Human Skeleton. Cambridge: Cambridge University Press.

Stringer et al. 2011. Gough’s Cave, SomersetBritish Archaeology. May-June.

Guest Blog: ‘Cannibalism In Archaeology Part 2: Mancos Canyon And Herxheim Case Studies’ by Kate Brown.

3 Apr

Kate Brown is a current archaeological undergraduate student at the University of Sheffield.  Her research interests include Osteology, Zooarchaeology, Mesoamerican archaeology, and Scandinavian archaeology alongside the general study of funerary rituals in human culture.


Following my previous post on cannibalism in archaeology, I would like to discuss a few archaeological case studies in more detail.

Mancos 5MTUMR-2346 (White 1992)

The Mancos site, of the Anasazi or Ancient Peublo Peoples culture,  is located is located on Ute Mountain in Montezuma County, Colorado. The bone assemblage consisting of 2106 bone fragments was excavated in 1973 by Larry Nordby. Although the site is based on high ground, it is only 75m North East of the Mancos river, giving it access to both security and a reliable water source. Mancos 5MTUMR-2346 is held as a site of significant importance when referring to cannibalism, due to its excellent levels of preservation, owing to the lack of evidence for either pre or post depositional disturbance of the assemblage. This is imperative to reach such conclusions as it reduces the possibility of confusion in interpretations.

Location of the Mancos 5MTUMR-2346 site (White 1992).

In the 1973 excavations, a multi-room habitation was found, which had been built over the remains of an earlier dwelling. Primary interments (burials 1, 2, 4 and 12) were fairly typical, and were mostly contained within the rooms of the earliest structure. The rest of the skeletal remains were found in the room fill as well as on the floor surface, and because of this they were originally interpreted as either secondary or disturbed burials. The fragments in these ‘bone beds’ at the site could not be found to have any association or articulation, and individuals appear to be mixed together indiscriminately across the assemblage.

There are many indications of possible cannibalistic consumption happening at the site.

Thin long bones, most notably fibula, were found mainly intact, however, larger more robust bones, such as tibias and femurs were found highly fragmented; this points to impact being inflicted on the bones in order to reach the bone marrow, known as percussion. Percussion marks can be seen on many of the skeletal remains recovered.

Breaks and fracture with subsequent polishing marks on the humerii (White 1992).

Scratches on some skull fragments are likely indicative of scalping rather than attempts to crack open the skull as in other cases, due to the thickness of the skull in this case as a result of osteoporosis.

The bones themselves were bleached quite light, and this is seen frequently in cases of cannibalism, as a result of being interred without any flesh adhering to the bones. Evidence of burning is evident on a large amount of remains, and because the pattern of burning on the bones is so varied it is possible to assume that they were heated whilst some flesh was still attached. Pattern fracturing and fragmentation of long bone shafts in the assemblage are strongly evident of marrow extraction, which is common across cannibalism sites.

Crushing evident on anterior alveolar region of the mandible (White 1992).

The evidence for cannibalism at the site of Mancos 5MTUMR-2346 is extensive, and includes high frequencies of most of the standardised factors for recognising such activity- the polishing of the ends of long bones as a result of cooking in coarse pottery, splintering and shaft breakage of long bones to facilitate marrow extraction; clear percussion scars, hammerstone abrasion, fracturing and crushing of bones; cutmarks indicating skin peeling and butchery; crushed skulls, chopmarks and peeling on lumbar vertebrae as well as a high frequency of rib breakage.

Peeling marks on thoracic vertebrae (White 1992).

In terms of the pathology of the Mancos Canyon assemblages, it is quite typical of an Anasazi population. At Mancos MTUMR-2346 there are at least seven individuals with cranial deformation, and this is found to be present in all skeletal assemblages from Mancos Canyon. Cases of caries and abscessing are identifiable on two mandibles, and dental enamel hypoplasiawould appear to be quite prevalent throughout the population. This is also typical of Anasazi populations, who often suffered significant nutritional stress.

Overall, the high number of young adult individuals far outweighs the instances of older individuals, which is unusual for a cemetery population.

Nordby (1974) interpreted that the site of Mancos 5MTUMR-2346 was either attacked, with its inhabitants being killed, dismembered and consumed at the kill site, or that the inhabitants of Mancos 5MTUMR-2346 attacked a larger site elsewhere and brought dismembered bodies back to their own site for consumption.

Herxheim (Boulestin et al. 2009)

Located in the South of the German Federal State of Rhineland-Palinate, above a loess soil plateau, Herxheim is an early Neolithic Linearbandkeramic (LBK) site with compelling evidence for cannibalistic activity. Excavations have found evidence of a village that was inhabited between 5300 and 4950 BC. At the site there is a non-continuous (pseudo) ditch, which is rare in the Neolithic period, and is thought to have served as a symbolic boundary rather than as a physical defence. This is evidence of the sites importance, and demonstrative of a central position at a regional level. This could also serve as an explanation of the sites importance through to the final linear pottery period despite the change in function it underwent at this time.

Location of Herxheim site (Boulestin et al. 2009).

During the final linear pottery period, no new pits were dug, instead previously existing ones were re used to allow for the deposition of human remains, along with some fauna, ceramics, and tools made of both stone and bone. Scatters of bone fragments, some numbering up to 2000 fragments, have been recovered from these pits, and are representative of a minimum number of 500 individuals. However, with only half of the enclosure having being excavated at this point, it is hypothesised that there could be up to 1000 individuals within the entire area. In the assemblage, there is a notably high proportion of both skull fragments and leg bones compared to fragments from elsewhere in the skeletal system. Deposition occurred over a maximum of 50 years, but was probably a lot less than this.

Deposit 9 at Herxheim (Boulestin et al. 2009).

Deposit 9 was excavated in 2007, and contained a much higher density of human remains than anywhere else on site. In the assemblage recovered from deposit 9, breakage was common, especially that of long limb bones. Short shallow cut marks are typically indicative of defleshing, which is common in cases of cannibalism. There is also evidence of butchery and skinning on fragments, shown by deeper varied cutmarks. Across the skull fragments found in the deposit, cracks and fracturing occurred often. Spongy bones were also often found to have been crushed, and peeling marks were frequently seen on both vertebrae and ribs, showing a butchery technique similar to that used in the butchery of animals to separate the ribs from the vertebral column.

Rib breakage and peeling marks on vertebrae (Boulestin et al. 2009).

As I have previously discussed in my last post, this is one of the standard indicators of cannibalism. Defleshing of long bones and marrow extraction are visible through scrape marks on the bones, and marrow cavities, and is another common manifestation of cannibalistic activity. Differential breakage of long bones can be observed, with bones housing larger volumes of marrow being far more likely to have been broken or fractured. This could be a result of the relative nutritional value to be gained from the differing bones. Finger bones were also preferentially broken, although foot bones seem to have been left more often intact.

Example of differential breakage (Boulestin et al. 2009).

Green bone breakage is another requisite for proof of cannibalism, and there is strong evidence of this taking place at Herxheim from the form of fragments as well as fracture outlines on bones.

Skulls seem to have been the subject of particular attention, with many showing evidence of skinning following a repetitive method. In many cases, the tongue was removed, which is evident by cut and scrape marks on the lingual surface of the mandible. In some instances, the mandible was also removed from the skull following this.

A distinct distribution of chew marks support the interpretation of cannibalism occurring at Herxheim; if the result of carnivore activity it would tend to have a much more random distribution across the remains than what is evident. However, because the cause of death is, at this point, undetermined, it is difficult to say whether this instance of cannibalism was a result of war, ritual activity, or a response to nutritional stress or starvation. Current interpretations view it to most likely be either a result of sacrificial ritual or revenge related to warfare, perhaps as an element of possible crisis at the end of the LBK period. This would also be supported by the evidence of increased violence at this time.

Instances of cannibalism in the Neolithic is often underestimated, largely because of the difficulties in recognising it following the current set of criteria, and in defending such interpretations, which are the subject of high amounts of controversy. However, these two sites, along with many more, have provided at least the possibility of cannibalism happening within past societies for varying reasons, and hopefully with more research, more stable interpretations can be reached and agreed upon.


Boulestin, B., Zeeb-Lanz, A., Jeunesse, C., Haack, F., Arbogast, R., Denaire, A. 2009. Mass Cannibalism in the Linear Pottery Culture at Herxheim. Antiquity 83 (German langauge).

Nordby, L.V. 1974. The excavation of sites 5MTUMR-2343, -2345 AND -2346, Mancos Canyon, Ute Mountain, Ute Homelands, Colorado. Bereau Indian Affairs, Contract MOOC14201337 Report.

White, T.D. 1992. Prehistoric Cannibalism at Mancos 5MTUMR-2346. Princeton: University Press.

Guest Blog: ‘Cannibalism In Archaeology Part 1: Recognition and Debate’ by Kate Brown.

12 Mar

Kate Brown is a current archaeological undergraduate student at the University of Sheffield.  Her research interests include Osteology, Zooarchaeology, Mesoamerican archaeology, and Scandinavian archaeology alongside the study of funerary rituals in human culture.


Cannibalism in Archaeology

Cannibalism is generally defined as the conspecific consumption of human flesh (White 1992). It is often used to support perceptions of savagery or primitiveness; however, the reasons for cannibalistic activity are often complex, and indicative of a basis in more than simply hunger, with evidence for this based across a long time period around the world (Hogg 1958).

A Still From Cannibal Holocaust (1980)

There are two major classifications of cannibalism; exocannibalism, the eating of persons outside the cultural or social group, and endocannibalism, where members within the social or cultural group are consumed by other members (White 1992). These can further be broken down into the respective reasons behind the act of cannibalism, or the method of consumption:

Survival cannibalism – Also referred to as obligatory or emergency ration cannibalism. Actual or perceived starvation leads to cannibalistic consumption.

Aggressive cannibalism – Consumption of enemies. Can be interpreted as a form of reveng

Affectionate cannibalism – Consumption of friends or relatives. Thought to ‘keep them close’.

Ritual cannibalism – Also known as ceremonial cannibalism. The consumption of human flesh as a part of spiritual belief or ritual undertaking.

Gastronomic cannibalism – Cases of cannibalism that are neither starvation nor ritually motivated.

Auto-cannibalism – self consumption

How is it Recognised?


Hammerstone abrasions from impact (White 1992, 152)

In an archaeological context, cannibalism can be very difficult to recognise. A majority of the following archaeological standards must be met to prove the presence of cannibalistic consumption at a site (Villa et al. 1986):

–          Skull modification in order to expose the brain

–          Facial mutilation

–          Evidence of cooking, including burnt bone and fragment end polishing, which is a result of cooking in course  ceramic pots

–          Dismemberment or butchery marks. Similar to that seen on animal remains on the site if present

–          Pattern of missing elements. Post processing discard again similar to the treatment of animal bones if present

–          Green-stick splintering of long bones. This facilitates the extraction and consumption of bone marrow, which is highly nutritious

–          Cut marks

–          Bone breakage

–          Anvil and hammerstone abrasions

–          A significant number of missing vertebrae

Shaft breakage types (White 1992, 135)   Cutmarks on front of skull (White 1992, 170)

The Cannibalism Debate

When discussing cannibalism, the argument against such interpretations cannot be ignored. As well as the evidence and interpretations supporting cannibalism, there are, as always, other schools of thought.  Because archaeological evidence of cannibalistic activity is so varied and often circumstantial, this has been used to discredit any interpretations of cannibalism. Paul Bahn (1990) is well known for his work on cannibalism, and scepticism of interpretations of such activity at a site. Even with a protein only occurring in humans being found in a human coprolite at the Anasazi site of Mancos, in the South West of the USA (Whittell 1998), Bahn remains unconvinced.

Most opposition stems from the reliability of the evidence, both archaeological and ethnohistorical.

Cannibalism in Brazil described by Hans Staden (1557)

Because it is so circumstantial and subject to interpretation, it can be seen as inaccurate to derive interpretations of cannibalism from. Especially in ethnohistorical accounts of cannibalism, prejudices and the desire to promote themselves above ‘savages’ are relatively clear, and this is used to discredit them as an archaeological source (Arens 1979). Some have argued that this completely removes them from being used in terms of research into cannibalism, because such biases could have caused them to fabricate stories of the natives in order to elevate themselves above them. However, even though they may be subject to personal views and opinions, they are still a valid description of cannibalistic activities.

Recent research may yet put to rest the constant debate around cannibalism in archaeology. Hannah Koon (2003) of York University has conducted extensive research on the effects that cooking can have on bones, and how this can be visible in the archaeological record.

What began as research into heat induced morphological changes in bone collagen based on earlier research by Jane Richter (1986) has become one of the most high profile advancements in the cannibalism debate in recent years. Although the initial use of her work was in forensics, and not archaeological, it has been demonstrated to be particularly important in the debate surrounding cannibalism. In observing that the collagen structure of bones changes and deteriorates when heated, or more specifically boiled, it can be inferred within reasonable doubt that cannibalism must happen in at least some cases, as the cooking of human remains is extremely unlikely unless there is the intent of consumption.

Analysis using this new technique is currently being carried out on some of the human remains that have previously been recovered from Herxheim, a site in Germany with evidence of what has been interpreted as cannibalistic consumption (Boulestin et al. 2009), which I will cover in more detail in a later post. However, to my knowledge the results of this analysis is as of yet unpublished.

The Problem With Cannibalism

The main problem surrounding the interpretation of any cannibalistic consumption is that it is such a sensational subject, both within archaeology and outside it. There is the significant potential for any modern attitudes, semantics and social constructions we have created around the word cannibalism to affect any interpretations and research based around it. The current approach regarding and leading to conclusions of cannibalism can be quite restrictive and leading, with judgements based on associated archaeological interpretations as well as ethnohistoric accounts being used to both prove and disprove instances of cannibalism (White 1992). Following this approach can lead to the exclusion of many of the necessary indicative features of cannibalism, because under such an approach they become inconsistent with such instances.

Ideally, sites with suspected episodes of cannibalism should be approached on an individual basis, which would ensure an objective approach to something that can differ so dramatically across the archaeological record both in manifestation and survival of evidence.

Part two can be found here.


Arens, W. 1979. The Man-Eating Myth. Oxford: University Press

Bahn, P. 1990. Eating People Is Wrong. Nature 348.

Boulestin, B., Zeeb-Lanz, A., Jeunesse, C., Haack, F., Arbogast, R., Denaire, A. 2009. Mass Cannibalism in the Linear Pottery Culture at Herxheim. Antiquity 83

Cannibal Holocaust. 1980. Online image available at last accessed 12th March 2011

Cannibalism in brazil described by Hans Staden. 1557. Online image available at last accessed 12th March 2011

Hogg, G. 1958. Cannibalism and Human Sacrifice. London: Hale.

Koon, H., Nicholson, R., Collins, M. 2003. A practical approach to the identification of low temperature heated bone using TEM. Journal of Archaeological Science 30, 11

Richter, J. 1986. Experimental study of heat induced morphological changes in fish bone collagen. Journal of Archaeological Science 13, 5

White, T.D. 1992. Prehistoric Cannibalism at Mancos 5MTUMR-2346. Princeton: University Press

Whittell, G. 1998. Tell-tale protein exposes truth about cannibals. The Times 8th November 1998.

Villa, P., Bouville, C., Courtin, J., Helmer, D., Mahieu, E., Shipman, P., Belluomini, G., Branca, M. 1986. Cannibalism in the Neolithic. Science 233