Cannibalism in the Early Pleistocene of Europe


Although the term cannibalism derives from Caribbean peoples, references to cannibal practices have been mentioned all over the world in both prehistoric and historic periods. It is a controversial topic in anthropology and palaeontology that provokes contradictory reactions. The first report of cannibalism was made soon after the discovery of hominid remains at Krapina (Croatia 1895-1905, c. 130 ka). Claims were also gradually linked with ‘cults of skulls’ in the 1930s with discoveries in Steinheim (Germany, c. 250 ka), Monte Circeo (Italy, c. 50 ka) and Zhoukoudian (China, c. 400-500 ka). In many instances it has been later proven that the cause of damage was taphonomic rather than cannibalistic.


Cutmarks are not sufficient to establish cannibalism. Remains from sites such as Bodo (Ethiopia, c. 600 ka) and Gough’s Cave (England, c. 12 ka) have undeniable cutmarks, indicating the skeletons were intentionally defleshed, although not necessarily eaten. Some of the functional types of potential human cannibalism are:

  1. Nutritional
    1. incidental: survival (periods of food scarcity or due to catastrophes
    2. long duration: gastronomic or dietary (humans as part of diet of other humans)
  2. Ritual, magic, funerary (in relation to beliefs or religion)
  3. Pathological (mental disease: parapathic; for political reasons)

These functional types have also been sub-divided into social divisions that include aggressive (consuming enemies) vs. affectionate (consuming friends or relatives), or endocannibalism (consumption of individuals within the group) vs. exocannibalism (consumption of outsiders).

The identification of nutritional, as opposed to ritual, cannibalism, is based on a combination of indicators, the main criterion being the comparison of human and animal remains from the same context. According to Villa et al (1986), these indicators are:

  1. similar butchery techniques in human and animal remains
  2. similar patterns of long bone breakage that might facilitate marrow extraction
  3. identical patterns of post-processing discarding of human and animal remains
  4. when applicable, evidence of cooking.

However, when human and animal remains are found in separate contexts, with different patterns of exploitation and distribution, ritual or some other explanation should be considered.

The Site:

The site of Gran Dolina belongs to the southern part of the karstic site complex of the Sierra de Atapuerca in northern central Spain. It is an 18m thick cave infilling, of which eleven sedimentary layers have been distinguished. Human remains have been recovered from a distinctive stratum of unit TD6 named ‘Aurora’.

The Human Remains:

Six individuals were found mixed with stone tools and non-human faunal remains. Their age is more than 780 ka and they have been assigned to Homo antecessor.


Each human fossil was identified as follows:

  1. body part
  2. segment and portion (diaphysis, proximal end, distal end, complete, lateral, body, process, arch)
  3. age (juvenile/adult/infantile) determined from dental eruption and wear, as well as epiphyseal fusion and bone texture.

The relative abundances of skeletal elements were calculated in comparison with the expected numbers of each element multiplied by the minimum number of individuals.

The following features were also recorded:

  1. Fracture
    1. Length/width/thickness
    2. Peeling
    3. Percussion pits
    4. Adhering flakes
    5. Conchoidal percussion scars
  2. Tool-induced surface modification
  3. Cutmarks
  4. Chopmarks
  5. Scraping marks


The ages based on dental traits are as follows: two infants of 3-4 years old, two adolescents, one of about 14 years and one of about 11 years, and two young adults about 16-18 years old.

Human anatomical elements are representative of all major skeleton areas, although they are not representative of the entire skeleton, element-by element. Some elements are scarce or absent. This may be due to the small area of excavation rather than by selective butchery.

Elements that have a small diameter with little marrow content appear almost unbroken. The most damaged are the skulls, mandibles, maxillae, the femur fragment, and vertebrae. This patterning is also observed on the non-human animal remains from the Aurora stratum and is consistent with those bones that held the most nutritional value.

Cutmarks have been observed on radii, phalanges and metapodials, suggesting dismembering. They are also frequent at strong muscle attachment points on the crania (face muscles, temporalis and sternocleidomastoid). This suggests detachment of the cheeks. Cutmarks on the temporal bones indicate separation of the head from the trunk. Vertebral damage is frequent and is considered to be mainly due to dismembering, defleshing and crushing the spongy bone portions.

Comparison with Other Sites:

Cutmarks are more abundant in the Aurora Stratum than at many other sites, probably because most anatomical elements recovered are bones with little meat and strong muscle attachments. Conchoidal scars, adhering flakes and peeling also appear more abundant, in contrast to the greater abundance of percussion scars seen on bones from Mancos and Yellow Jacket in Colorado. This is likely to be due to the lack of fire among early Pleistocene hominids in comparison to the Anasazi culture.

Type of Cannibalism:

The Aurora stratum is characterised by:

  1. analogous butchering techniques in human and non-human animals
  2. similar breakage patterns to extract the marrow
  3. identical pattern of post-processing discard of humans and animals

In summary, the techniques observed were aimed at meat and marrow extraction. No ritual treatment can be suggested in this assemblage. It is, therefore, presumed that nutritional purposes were the cause of this case of cannibalism.

If it is assumed that the stratum represents a single incidental and short event, then the temperate environment indicated by pollen evidence and the high diversity of faunal remains do not justify a starvation period survival strategy. It is therefore suggested this represents gastronomic cannibalism.

Equally, if the event represents a biologically long period of time, then the distribution indicates cannibalism was taking place throughout this period of time. This also can be modelled as gastronomic cannibalism, indicating humans were part of the diet of other humans.


Fernández-Jalvo, Y, Díez, J.C, Cáceres, I and Rosell, J. 1999. Human cannibalism in the Early Pleistocene of Europe (Gran Dolina, Sierra de Atapuerca, Burgos, Spain). Journal of Human Evolution 37: 591-622.


7 responses to “Cannibalism in the Early Pleistocene of Europe

  1. Pingback: Oops! I Forgot Know Your Primate [Afarensis] · Articles

  2. Pingback: Oops! I Forgot Know Your Primate [Afarensis] · New York Articles

  3. Very interesting post – the idea of gastronomic cannibalism raises the question of whether humans were killing each other for food, or politely waiting until their intended meals died of other causes, before tucking in to their remains. I like the way in which the presences of other faunal remains, along with the pollen data at Gran Dolina, enabled the conclusion to be drawn that starvation evasion wasn’t a factor, although it could be argued that we don’t know with absolute certainty that no ritual was involved somewhere along the way – or indeed, how other traces of ritual activity could even be detected.

  4. That is very true. ‘Ritual’ can be such a difficult thing to define. It’s probably where the joke that if archaeologists don’t know what it is, it’s ritual, comes from. Also, I suspect in many instances in the past the line between ritual and prosaic was less rigid than we might like it to be, which makes it even harder to detect.

  5. Hi Archaeozoo- I’m very sorry to bother you, but I was wondering if there was any chance you had access to the paper you reference, ‘Fernández-Jalvo, Y, Díez, J.C, Cáceres, I and Rosell, J. 1999’, as I don’t have a sub to the journal, and I could really do with reading up on something detailed, because I have a feeling I’m not going to be able to get a lot of what they’re saying when I visit, and if I can get my printer working I’d like to take some notes, or of not some written ones – apologies for interrupting your weekend, best, Tim

  6. Unfortunately my copy is a paper photocopy, not a pdf copy, I think, although I will certainly check for you.

  7. Thanks a lot, but don’t worry, as I found a few other bits and pieces to read, and they should be enough, hopefully, for the time being – best, Tim

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s