Tuesday, 2 October 2012

Human Sacrifice

At Kerma, Nubia, on the present-day border between Sudan and Egypt, there are ruins of an ancient African civilisation that flourished c.1600 BCE. The Kingdom of Kerma, located between the First and Fourth Cataracts of the Nile, was probably the earliest 'black' African complex civilisation, and likely rose in order to control trade along the river from tropical Africa up to the Nile Delta. From the Amarna letters, a collection of diplomatic correspondences found at Akhenaten's capital Akhetaten (now known as Amarna), we know that tropical African products were traded in the Near East, especially ebony, indicating that the trade was extensive by the middle of the second millennium BCE. G. A. Reisner, who excavated the capital city of Kerma in 1913, found 322 human sacrifices in Tumulus X, one of several massive burial mounds at the site. He estimated that there had been over 400 sacrificees in Tumulus X before tomb robbers disturbed the place. This is the greatest concentration of human sacrifices revealed by archaeologists from any tomb in the world.

Human sacrifice has been practiced the world over. Europeans mentioned it with horror wherever they encountered it, and it tends to leave some indications in the archaeological record, so it is a reasonably well-documented behaviour, although it is difficult to separate from simple murder or execution. Some opponents of capital punishment1 liken execution to sacrifice, perhaps because of an instinctive revulsion at the apparent atavism of the practice, but it seems to me that sacrifice is usually different to execution (although they can be one and the same). It correlates with states or, at least, chiefdoms.  It is hard to sacrifice a human being without there being a concentration of power in the group commanding the sacrifice and a corresponding lack of power among another group, so where institutionalised human sacrifice is found, we may be reasonably sure that the society that did the sacrificing was a stratified one - one separated into different classes. The sacrifices at Kerma – in particular, their stunning scale – are part of the evidence attesting to the fact that Kerma was the earliest state in tropical Africa.

Human sacrifice is evident in the earliest Egyptian complex societies. Flinders Petrie, a nineteenth-century archaeologist who specialised in Egypt (and developed seriation to an artform), noted the presence of human sacrifices in the Naqada II (or Gerzean) archaeological culture, found in predynastic Egypt from the middle of the fourth millennium BCE. The people had had their throats slit and their heads cut off. Similar remains were found throughout the Naqada area, indicating a widespread tradition that correlates with the rise of states in Egypt. In early China, human sacrifice is attested in writing and the archaeological record. Wu Ding, king of the Shang state in the second half of the thirteenth century BCE, is recorded as having sacrificed 500 Qiang people in one ritual, and other Shang texts give details of rituals in which different numbers of Qiang might be killed – ten, twenty, thirty. The Qiang were probably speakers of Tibeto-Burman languages – that is to say, non-Sinitic languages of the Sino-Tibetan family (from which Chinese also comes). They may have raised horses, but that is unclear. They appear in inscriptions as enemies and sufferers of sacrifice and little else.

Human sacrifice was also practiced in Europe. The Romans practiced it, although many writers, including Livy, avoided calling the practice 'sacrifice'. It seems as if sacrifice was seen as a practice of Rome's enemies, as a repugnant act conducted by barbarians like the Carthaginians, Germans, and Gauls. Several executions bore a resemblance to sacrifice – including that of Marcus Marius Gratidianus and, of course, Vercingetorix, executed in public during Gaius Iulius Caesar's triumph in 46 BCE. Sacrifice was banned on several occasions, decades before the rise of Christianity. Many of Rome's enemies did practice human sacrifice, of course; Tacitus says that the Germans who defeated Rome's legions at the battle of the Teutoburg Forest in 9 CE sacrificed Roman survivors to their gods, for instance. Strabo claims that various Gaulish tribes sacrificed humans and took the heads of enemies:
The heads of enemies of high repute, however, they used to embalm in cedar-oil and exhibit to strangers, and they would not deign to give them back even for a ransom of an equal weight of gold. But the Romans put a stop to these customs, as well as to all those connected with the sacrifices and divinations that are opposed to our usages. They used to strike a human being, whom they had devoted to death, in the back with a sabre, and then divine from his death-struggle. But they would not sacrifice without the Druids. We are told of still other kinds of human sacrifices; for example, they would shoot victims to death with arrows, or impale them in the temples, or, having devised a colossus of straw and wood, throw into the colossus cattle and wild animals of all sorts and human beings, and then make a burnt-offering of the whole thing.2

Perhaps the death of Edmund of East Anglia ties in here; according to tradition, he was killed by Viking pagans in 869 CE by being shot full of arrows and beheaded. If this accurate – and it is entirely possible that it isn't – then it might be another case where execution and sacrifice overlap.

Sacrifice was common in southeast Asia. In the Malay archipelago it was often associated with house construction. In several places, children were smashed beneath house-posts – specifically, the right-front post of a newly-built house, often considered to be the 'sacred' pole of the house in Indonesia. Signe Howell says that this occurred in Lio communities on the island of Flores, where the child of a chief would be given a final meal before being killed by the post. Slaves were killed in Borneo to accompany the building of a house – Kayan people killed their slaves to accompany construction, as did several other groups (according to missionaries and early ethnographers, of course).3 Almost all of these groups practiced headhunting as well, and in some cases – as in the 'war of the mat and pillow' (Lamaholot: ohã bělone, Malay: tikar bantal), a violent, ritualised conflict in east Flores – it was explicitly connected to placating deities and 'filling [their] barns' with heads.

'Civilised', Indicised, southeast Asia was not lacking in sacrifice, either: it was found in Java, where vajrayana (or 'tantric') Buddhism flourished and where local pre-Indic cults continued until recently, both of which encouraged the sacrifice of human beings; and it was also present in Angkor, the Khmer empire, where (again) sacrifice was associated with construction. Humans were often crushed beneath buildings in Angkor to placate deities and to provide the building with protective spirits - the elegance of Angkor Wat conceals the skeletons of humans murdered to provide a supernatural foundation for it.  Also associated with vajrayana Buddhism, by the way, is the rkang-gling, a Tibetan flute made from a human thighbone.  It is preferred that the human whose bone it was died violently.

Sacrifice in the Austronesian world was probably taken to an extreme by the kings of Hawai'i, who continued to sacrifice humans into the nineteenth century. Victims were usually slaves or captives, and they would be killed in some way before their bodies were burnt on a pyre set on a large sacrificial rock inside a luakini heiau, a temple of war and human sacrifice (dedicated to Kūka'ilimoku, the martial manifestation of Kū, a Hawaiian god and counterpart of Lono).4 Statues of Kū were also thrust into holes in which the body of a sacrificed man had been placed, a similar practice to those found in Indonesia. Hundreds of people would be killed at such occasions – such that the rock became shiny with human grease, according to Hawaiian mo'olelo (oral historian) Kamakau:

'There were many heiaus made famous because of the shine of the grease (hinu) of burnt offerings from Hawaii to Kauai; they were offered up by the hundreds and thousands (he lau, he kini, he lehu). The soil became fertile and saturated with slime from the grease of the heavenly chiefs and from the burnt offerings. It is impossible to count the hundreds and thousands of years of sacrificing.'

It is often claimed that the Aztecs sacrificed an unusually large number of people, but actually large numbers were sacrificed in West Africa and Hawai'i as well, and the hundreds found at Kerma and attested in Shang dynasty China show that scale was an important variable elsewhere too. What marks Aztec sacrifice out as peculiar, perhaps, is that it was so public, and so connected to blood. Autosacrifice was common, as it was throughout Mesoamerica; blood would be drained from various parts of the body, including the tongue and penis, as offerings.  A maguey needle was the most common blood-letting instrument. Aztec priests were expected to do this frequently, as were men in general, although women seem to have had less compulsion to perform it, finding their inspiration in conscience rather than social pressure. Blood was the gift provided by the gods themselves in the creation of humankind, and therefore it constituted the holiest gift to offer back to them, both from the wounds of living humans themselves and from humans sacrificed by the priests.

Most of the people sacrificed were men5, often captives in battle, and they died in plain sight. If they had been captured in battle, then the man who had captured them would forge with them a special and enduring relationship. They would eat together, and the captive would be known as the 'son' of the warrior. Prior to the sacrifice, the victim was often expected to fight in a mock battle while armed only with a club lined with feathers in place of obsidian blades. He was expected to show masculine resolve throughout – to fight hard and then run up the bloodied steps of the temple (not around the back) to offer his body for sacrifice. This show of courage was valued highly. The priests, wearing black robes, their long black hair matted by the victims' blood, would perform the sacrifice, stretching out the victim's body over the sacrificial stone and using an obsidian blade to cut through the chest and pull out the still-beating heart.  The body would then be pushed down the steps of the temple, the very steps the victim had just climbed, and the warrior whose 'son' had been sacrificed would take blood from the victim to daub on statues of the gods throughout the city.6

Sacrifice in Tawantinsuyu (the Inka empire) was not public, and it differed considerably from Mesoamerican styles. The Inka practice was documented by chroniclers after the Spanish conquest, and there is also considerable corroborating archaeological evidence for it. In Tawantinsuyu, most of the victims were children and young women. The latter, known as aqlla-kuna, were selected by an official known as the apu panaka from lands under Inka control. They were chosen for their beauty and elegance, and they had to be virgins.7 The victims, known collectively as qhapaq hucha (sometimes seen as 'capacocha') would be taken to Qosqo (Cuzco), were they would be kept in what amounted to a nunnery. The aqlla-kuna would fast – forbidden meat, alcohol, salt, chilli, and sex – and be kept in relative seclusion, meeting only one another and the mama-kuna, women who dedicated their lives to the state and to training the aqlla. They would chew quids of maize which they would then spit into containers to ferment into chicha, the corn beer of the Andes and a product of religious significance. Some of these aqlla would then be taken from Qosqo to the provinces along siq'e, lines radiating from Qosqo connecting waqa, sacred sites, and also the sites of sacrifice. The waqa were almost invariably in out-of-the-way places – mountains, caves, and so on – although there were of course several in Qosqo itself. The aqlla would be led out to waqa well above the treeline, sometimes covered in snow, accompanied by a couple of attendants and a high priest. They would be given some food and then they would be killed, either by burning or live burial.


Clearly, these are all different practices. Sacrifice in Tawantinsuyu was private and involved young women and little boys, while sacrifice in the Aztec world was primarily of men – warriors, even. There is some overlap with capital punishment, and an unclear division between sacrifice, headhunting, divination, and the public execution of enemy tribes. Human sacrifice is therefore a polythetic class. That means that the different instantiations of it are not entirely similar, and may even be different in almost every respect, but they overlap with one another – they share what the philosopher Ludwig Wittgenstein called a family resemblance. We group them together more for convenience than anything else. While religion is clearly key to all of these practices, and they all involve the deliberate killing of human beings, it would be hard to separate any definition based on these features from support for capital punishment among American conservative Christians, who frequently justify their support for the murder of human beings on a religious basis.

On the other hand, we can look at each individual case and work out the psychological basis for it. In the case of Kerma, the scale of the sacrifice and the association with public architecture gives a good indication of the power of the state. We can use this kind of large-scale sacrifice as part of a set of diagnostics for social complexity. This is because it is hard to get people in an egalitarian society to engage in the deliberate murder of large numbers of human beings for little reason; sacrifice is pre-meditated and relies not on a mob mentality but on the cold-blooded decision to kill a human being of a certain class or origin, often relying on a state apparatus to perform the act. In order for 400 people to be killed as part of a funeral, there has to be a concentration of power among a certain class of people. There has to be the delegation of power to certain individuals. Likewise, the qhapaq hucha system could only operate while a state existed that could exert violence on individuals who disobeyed its precepts.  (We know that the qhapaq hucha system relied on coercion from the fact that families in Tawantinsuyu would encourage their daughters to lose their virginity at a young age, so as to avoid the peril of being chosen for sacrifice.)

The basic principle of human sacrifice seems to be the same as with other forms of sacrifice: provide an offering to a supernatural entity, and it may grant you favours or improve your life in some way. This seems to be the basis for it. Humans are some of the most valuable things – more useful and cleverer than animals, and more beloved by other humans – and therefore ought to have great value to supernatural entities. Sacrificing something so valuable requires the concentration of power in someone capable of commanding it, and this is especially true if the victim is from the same society as the killers – because what that usually means is that there is a class of people who can be used as victims, and that implies a stratification of society and the concentration of power in an upper class.

There also has to be a violent ideology. You'll notice, I hope, that human sacrifice is no longer found anywhere on earth (except perhaps in some parts of Africa, where traditional religious practices continue under the surface). It hasn't been practiced in Europe for a long time. The Lottery is a shocking story precisely because human sacrifice is so rare and seems so absurd to us. The episode of South Park about sacrificing Britney Spears to ensure a good harvest is funny and satirical because human sacrifice is not a possibility anymore. People don't kill one another in this way these days.

Many of these cases involve the sacrifice of outsiders or a proscribed class. This isn't invariable – in Indonesia, of course, the children of chiefs were sometimes killed – but usually slaves, commoners, foreigners, and hated ethnic groups are the subjects of sacrifice. The Shang kings don't seem to have sacrificed their own people, but instead killed the Qiang. When Rome was in trouble, Romans weren't sacrificed to the gods – Greeks and Gauls were. The victims of Aztec sacrifice sometimes included citizens of the state in which they were sacrificed, but normally they were citizens of other states, warriors from rival cities. The Inkas appear to have sacrificed 'their own' people, but in fact many of the qhapaq hucha were from far outside Qosqo and seldom included people of the Inka ethnicity; Thomas Besom suggests that qhapaq hucha was especially common in the southern part of the empire, then recently conquered by the Inkas. It is difficult to tell from the archaeological evidence which ethnic groups or classes were involved in sacrifices at Kerma and in the Naqada sites, but I would be willing to bet that many of those killed were enemies or members of an underclass.

Human sacrifice has been eradicated by the expansion of moral sentiment to include even other ethnic groups and members of lower classes. It has resulted from what the Australian philosopher Peter Singer called the expanding circle.  Instead of confining our moral judgements to those within our own circle - our village, our city, our nation, our race - we now apply them (or most of us do) to humans in general, and in many cases even to non-human animals. It is all part of a trend in the reduction of violence worldwide, one influenced by many variables, best documented by Steven Pinker in his book, The Better Angels of Our Nature.  (Pinker's treatment of human sacrifice is too brief, but the book as a whole is fantastic.)

In any case, human sacrifice was once common, and now it isn't. In order to grasp the diversity of human life, and to understand human lives and their potential actions – in other words, to understand 'human nature' – clearly this needs to be taken into account. Any understanding of humankind has to take into account the fact that humans the world over have, in the past, killed one another in grand social spectacles and for religious reasons that we now consider spurious, unnecessarily violent, and ridiculous. This is precisely why anthropology, the study of human beings, cannot involve only ethnography. It has to include those scientific methods that allow us to look at these past practices: ethnohistory, archaeology, linguistics, and so on, because only these can shed light on practices that were once common and which must form a key part of any assessment of Homo sapiens sapiens. Ethnography today cannot tell you anything about human sacrifice because it isn't around anymore. Human diversity is now constrained by a moral imperative, and that seems to me like a good thing.

Further reading:

Adams, W. Y. 1977. Nubia: Corridor to Africa. London: Allen Lane.

Besom, Thomas. 2009. Of Summits and Sacrifice. Austin: University of Texas Press.

Connah, Graham. 1987. African Civilizations. Cambridge: CUP.

Dodds Pennock, Caroline. 2011. Bonds of Blood. New York: Palgrave Macmillan.

Hansen, Valerie. 2000. The Open Empire. London: W. W. Norton & Company.

Howell, Signe. 1995. 'The Lio House: Building, Category, Idea, Value.' In Carsten & Hugh-Jones (eds). About the House. Cambridge: CUP.

Kirch, P. V. 2010. How Chiefs Became Kings. Berkeley: University of California Press.

Midant-Reynes, B. 2002. 'The Naqada Period (c.4000-3200 BC)'. In Shaw, I. (ed). The Oxford History of Ancient Egypt. Oxford: OUP.

Pinker, Steven. 2011. The Better Angels of Our Nature. London: Allen Lane.

Singer, Peter. 1981. The Expanding Circle: Ethics and Sociobiology. Oxford: OUP.

Smith, Michael E. 1998.8 The Aztecs. Oxford: Blackwell.

1I am, of course, an opponent of capital punishment (not that it's such a hot topic in the UK!), but I don't feel the need to conflate it with human sacrifice. It's bad enough as it is, and it isn't usually conducted for religious reasons (which is, kind of, the hallmark of sacrifice, even if it is difficult to separate sacrifice and execution formally).
2Strabo, Geography. (4. 4:5).
3Similar practices were also found in South America and the Pacific. Probably elsewhere, too.
4Lono, a god associated primarily with agriculture and found elsewhere in Polynesia as Rongo, was the god James Cook is often claimed to have been mistaken for by the Hawaiians.
5There were exceptions; in the legends of the foundation of the Aztec state, the sacrifice of a woman and the wearing of her skin is a key event, and this appears to have been acted out on occasion.
6This, at least, seems to have been the case in Tenochtitlan. I'm not sure about other Aztec capitals.  And there were, of course, several other methods of sacrifice, including being shot to death by arrows and atlatls.
7It is related that many families ensured that their daughters lost their virginity early on so that they would not be sacrificed.
8There are more recent editions, but that's the one I've got.


  1. "...capital punishment among American conservative Christians, who frequently justify their support for the murder of human beings on a religious basis."

    Are these the only people who consider capital punishment justified?

  2. Of course not - but as at least part of their justification is religious, it blurs the line slightly between execution (a legal punishment) and sacrifice (a death intended to satisfy religious or supernatural aims). There are people who support execution, however misguidedly, for other reasons, include the (false) belief that it lowers the crime rate.


You can post anonymously if you really want to, but I would appreciate it if you could provide some means of identifying who you are, if only for the purpose of knowing who has written what.