BA (UBLS) LLB (Hons) (Glasgow) DPhil (Oxon)
Extraordinary Professor, University of Pretoria; Professor Emeritus, University of Cape Town, South Africa
Edition: AHRLJ Volume 17 No 1 2017
Pages: 1 - 34
Citation: T Nhlapo ‘Homicide in traditional African societies: Customary law and the question of accountability’ (2017) 17 African Human Rights Law Journal 1-34
Download article in PDF
The article discusses the attitudes of traditional African societies towards the taking of human life, aiming to understand the incidence, nature and causes of killing in traditional society. The article explores the responses of these societies to homicide, seeking to unearth legal, religious or other norms, if any, governing the taking of human life. The article interrogates the issue of accountability, to discover whether traditional societies recognised any obligation to ensure that a killer was made to account for his or her act - thereby inevitably raising questions about the right to life. The article concludes that in the customary law of these societies values and norms in respect of killing existed and that notions of accountability were indeed recognised, although (being drawn from strong communitarian foundations and a widespread belief in the supernatural) they differed significantly from modern human rights norms.
Key words: right to life; homicide; traditional Africa; accountability
There is no shortage of written materials dealing with the manner in which traditional African societies went about their daily business and dealt with all types of practical and moral questions, including the killing of human beings by other human beings. There are ethnological, sociological, historical and legal accounts, supple-mented by personal records and narratives of sundry travellers, adventurers, traders and colonial civil servants. 1
What these accounts mostly have in common is their capturing of the circumstances occasioning the taking of human life in various societies at various periods in the development of these societies. Frequently, these accounts will reveal the dynamics underlying such incidents and set out the consequences of the behaviour identified. Occasionally, these descriptions will yield rich insights into the reasons for these dynamics, including the political, economic, religious and sometimes purely pragmatic considerations that come into play. All these are helpful in understanding the context within which the taking of human life occurred in traditional society.
Less accessible is information which points directly to the value that these societies placed on human life in a sense that might be useful in determining whether a right to life was recognised and, consequently, protected and promoted. The purpose of the article is to attempt the task of discovering the ‘mind’ of traditional society on the killing of human beings by other human beings. This is another way of saying that the article will attempt to unearth ethical, legal, religious or other norms, if any, that were applicable to killing in these societies, not for their own sake, but for the purpose of throwing light on present practices and whether these in any way can be explained by reference to past values.
This already hints at several problems inherent in the task. In the first place, the notion of killing needs to be interrogated a bit further. The taking of human life occurs in many forms and at many levels, differing in motivation, intent, scale, intensity and method. It is beyond the scope of the article to deal with all types of killings and their ramifications. Therefore, the article seeks to confine itself to those killings that happen in communities and can be seen in the context of people’s daily lives and interactions. For this reason, it is not possible to give full attention to killings on a massive scale, such as during wars or other upheavals (for example, genocide).2
Clearly included will be those killings perpetrated by political authorities (or on their orders) upon their subjects. Such killings are important, not only as comparisons to violence between ordinary citizens, but also as a measure of whether all lives in the group were considered to be of equal worth and weight. For all of the types of killings included for consideration here, the word ‘homicide’ is used because its dictionary meaning as ‘the killing of a human being by another person’3 properly captures the neutrality that we seek. This neutrality enables us to deal with all instances of killing, regardless of the level of culpability attaching to each and whether they are punishable.
A second preliminary difficulty relates to the notion of traditional societies itself. While the concept of tradition itself is contested, we do not wish to engage with that debate here. Since our aim is to try and fathom the attitudes to killing of indigenous African societies, we will be satisfied with accounts of precolonial society as well as any other evidence of African practices, even if these straddle different historical periods, including the extension of foreign administration to these societies. In the work undertaken by anthropologists in the early parts of the twentieth century, for instance, it was often possible to detect a useful residue of precolonial thinking even when the societies in question were grappling to understand the ‘laws of the white man’. This was sometimes graphically illustrated in reported instances of the persistence of customary practices in the face of harsh penalties, amid general perplexity on the part of the populace as to the reason or need for these curbs.4 Although not dealing exclusively with African societies, Diamond provides a useful description of his use of the adjective ‘traditional’ or ‘small-scale’ as meaning5
past and present societies living at low population densities in small groups ranging from a few dozen to a few thousand people, subsisting by hunting-gathering or by farming or herding, and transformed to a limited degree by contact with large, Westernised, industrial societies.
To put this another way, we accept Diamond’s notion of traditional society as being helpful for purposes of describing the populations reviewed in this article. The definition has the merit of covering the very broad range of living conditions recorded in respect of African societies as well, from very small-scale and isolated, to ‘almost westernised’, and the various permutations in between. The importance of this approach is that it focuses attention on the core task, and avoids scholarly skirmishes about the complexities of the concept of ‘tradition’. That core task is to attempt an understanding of how indigenous African societies dealt with killing and whether any similarities between them can be detected and, with these similarities, some idea of the existence of normative underpinnings, if any. The most direct route to this information are the recorded accounts of populations studied in the eighteenth and nineteenth centuries, as well as pre-colonial social formations where available, right down to groups living in relatively contemporary times but still shielded from thoroughgoing modernisation. In the article, then, reference to ‘African societies’ (or ‘these societies’) means nothing more ambitious than any groups (whether kin or clan, tribe or nation) for whom evidence of customs relating to killing has been recorded.
The third point to be cleared is the question of accountability. Accountability, in the way the concept is used in the article, is the corollary of the right to life as defined in the literature of international law.6 This basically holds that there are two components to the right to life: a prohibition against killing, except for good reason (that is, in self-defence or in defence of another life); and bringing the perpetrator to account when the norm has been violated.7 This theme will be fleshed out in section 7 below.
Finally, a word about the structure of the article is in order. The writer has reviewed, at different levels of depth, a literature on killing which covered some 30 societies spread geographically around the continent in West, East, Central and Southern Africa. As indicated earlier, some of these accounts are classical anthropological studies, while many are attempts at unravelling the legal principles in the communities concerned. It can only be possible here to give a sense of certain commonalities and regularities of thinking distilled from these studies; bombarding the reader with the minutiae of the rules pertaining to homicide in over 24 indigenous communities is neither possible nor desirable. What appears useful is to paint a picture of how African peoples over time have dealt with the taking of human life and whether any norms and values can be discerned from their reactions to these incidents.
2 Homicide in traditional African societies
2.1 The Nuer
There can be no better place to start a review of killing in traditional African society than with Evans-Pritchard’s classical account of the Nuer of Sudan,8 a ‘stateless’ society lending itself well to a study of social mechanisms in respect of death caused by human agency. For a lawyer, it is not easy to fathom whether homicide was considered a crime or a delict or neither among the Nuer: Killing was killing, and was rendered quite frequent because of the Nuer’s fighting culture. Evans-Pritchard describes this culture as follows:9
A Nuer will at once fight if he considers that he has been insulted, and they are very sensitive and easily take offence. When a man feels that he has suffered an injury there is no authority to whom he can make a complaint and from whom he can obtain redress, so he at once challenges the man who has wronged him to a duel and the challenge must be accepted. There is no other way of settling a dispute and a man’s courage is his only immediate protection against aggression.
Evans-Pritchard also describes the phenomenon of the ‘blood-feud’, which he explains as follows:10
Blood-feuds are a tribal institution, for they can only occur where a breach of law is recognized since they are the way in which reparation is obtained. Fear of incurring a blood-feud is, in fact, the most important legal sanction within a tribe and the main guarantee of an individual’s life and property. If a community of one tribe attempts to avenge a homicide on a community of another tribe a state of intertribal war, rather than a state of feud, ensues, and there is no way of settling the dispute by arbitration.
According to Evans-Pritchard, the process following upon a homicide (that is, ‘the procedure for settling a blood-feud’11) is for the slayer to report himself to the leopard-skin ‘chief’12 so that he can be cleansed. This requires certain rituals and is also a period where the sanctuary at the chief’s home offers the killer a respite from immediate retaliation, which it is the duty of the deceased’s family to pursue:13
As soon as the kinsmen of the dead man know that he has been killed they seek to avenge his death on the slayer, for vengeance is the most binding obligation of paternal kinship and an epitome of all its obligations. It would be great shame to the kinsmen were they to make no effort to avenge the homicide.
After an appropriate cooling-off period, the chief commences negotiations between the two sides and, after investigating the assets of the slayer’s people, proposes to the deceased’s family that they accept so many cattle as compensation. After some ritual toing and froing, some 40 to 50 cattle are paid, although normally not all at once. Atonement ceremonies are performed and life for the slayer’s kin returns to a semblance of normality (not complete safety until the whole debt has been paid). The cattle are distributed among the deceased’s kin, including a portion to be used to marry a wife in the name of the deceased in order to bear an heir.
Evans-Pritchard makes a strong point about the significance of social distance in the Nuer attitude to homicide: ‘What happens when a man kills another depends on the relationship between the persons concerned and on their structural positions.’14
Settlement is easier the closer the lineages are to each other because ‘corporate life is incompatible with a state of feud’.15 Where nearby villages are involved an accommodation is quickly reached. In his discussion of the Barotse, Gluckman16 elaborates on Evans-Pritchard’s account of the Nuer and teases out important questions of liability in such stateless societies where vengeance is the primary method of redress. He states:17
The operation of rules of vengeance brings up in sharp form key concepts in the law of wrongs: recognition of intention to kill as against killing by negligence or accident, questions of responsibility and liability, and provisions of restitution or punishment.
In respect of the relationships involved, he continues, intention appeared to be irrelevant in cases of the killing of one member of a kinship group by an outsider. What was important was that a social relationship existed where the rules dictated that ‘murder’ must be compensated by blood money.18 Evans-Pritchard also makes the point that any killing disturbs the social equilibrium by disrupting the ‘balance of blood’ between the groups, which must now be redressed as part of the interplay of the rights they have against each other.19
Strongly corroborating Evans-Pritchard, Gluckman points out the two extremes representing the position where settlement is less likely in distant relationships though liability is recognised, and more intimate relationships where redress is not possible (because the givers of blood money would be the very same people to receive it) and liability - though recognised - is virtually ‘waived’, the whole incident being characterised ‘as a sin, subject only to ritual or religious sanction’.20 He concludes: ‘At this extreme there is no legal liability, while at the other extreme compensation is not likely to be paid. In the middle range compensation may be offered and accepted.’21
Gluckman also makes an interesting point about the Nuer and the question of guilt. It appeared to him that guilt did not arise as an issue where the moral question lay not in the incident but in its concealment:22
Seemingly, among the Nuer the issue of guilt for a particular killing was unlikely to arise. It was a heinous breach of moral duty for a man to conceal that he had killed another, since killing set up a spiritual barrier between the two sets of kin. For example, if they are together illness would assail them. The killer himself had to be cleansed from his spilling of blood.
We shall return to the lessons learnt in observing how a stateless society, like the Nuer, deals with homicide.
2.2 The Barotse
Two decades after the appearance of Evans-Pritchard’s work on the Nuer, Gluckman published his study on the Barotse of Northern Rhodesia, as it was at the time. Homicide amongst the Barotse, conceived without distinction between murder and culpable homicide, occasioned a range of response options, from actually killing the offender, subjecting him to a fine, taking him as a slave, to letting the matter quietly drop when it was ‘politic to do so’.23 When payable, compensation to the deceased’s family was set by the court, as was the case with enslavement. Still pursuing his interest in the question of liability, Gluckman concludes that in situations of feud, liability attaches regardless of intention if the killing occurs within a certain range of relationships.24 This discussion leads him to the work of Elias, who asserts that ‘there is a notion of mens rea in African law’.25 In the quoted passage, Elias cites the work of Penwill on the Kamba of Eastern Kenya, who asserts that Kamba law does not distinguish between murder, manslaughter and death caused by accident and, yet, concedes that a portion of the fine (the cow of the accident) appears to acknowledge special consideration of the incident if it was accidentally caused.26 Elias believes that the paradox is resolved by keeping in mind ‘the confusion between liability and the qualification of damages’ and that27
in homicide cases ... the elders naturally regard the consequences of the children and other dependants of the victim being suddenly left destitute as more important than the manner of the encompassing of the death by the killer.
Gluckman disputes this reasoning, arguing that the more accurate position is ‘to see the law as applying a strict liability arising out of duty to avoid harming others, and that the extent of liability can be reduced if the defendant can adduce mitigating circumstances’,28 where the onus lies on the defendant (based in part on certain social relationships which raise a presumption that the harm was maliciously caused).
Gluckman’s discussion of strict liability (sometimes inferred from circumstances) flows into an interesting consideration of witchcraft, which he contrasts with the class of wrongs where the mental state of the perpetrator is irrelevant. He writes that in some societies the offence of harming others by witchcraft is believed ‘to flow from the mere feeling of malice, envy, spite, hate, anger, jealousy, or greed, without the volition or action of the alleged witch, provided that he has the power of witchcraft in his body’.29
Arguing that in societies holding these beliefs ‘any misfortune suffered by a tribesman ... can be converted into a tort or a crime’, Gluckman explains his thesis as follows:30
Groups of kin have to co-operate to achieve multifarious purposes of all kinds. The effect is that every action by any member of a group, which departs from norms, influences adversely many purposive activities. A quarrel between two brothers has wider effects than in our industrial society; it may imperil productive work, affect the solidarity of political action, or disturb the unity of a religious congregation. Hence each breach of norm has a spreading moral disturbance: many relationships and activities are affected. Conversely, if outside events do not run normally, this points to secret disturbances in the moral relations of the members of the group.
His conclusion is that in such societies, where opportunities for individual advancement are few, the damage is done to a kinsman merely by wishing him ill, even though one does not act on one’s ill will.31
This bears a striking resemblance to the views of Metz, whose search for the content of ubuntu leads him to the conclusion that it is a moral principle animated by a sense of goodwill towards one’s community.32 (We return to this theme later in the article.)
2.3 The Tiv
Bohannan’s study of the Tiv of Central Nigeria adds to the stock of knowledge of societies where ‘there was no hierarchy of “responsible offices” and the “law” was kept through ad hoc meetings of groups of elders associated with lineages, or by self-help’.33 The Tiv recognise different levels of blameworthiness in respect of homicide: Killings are divided into those that happen accidentally, or ‘through ignorance’ or are done purposely. Murder is, thus, recognised as a crime, as is culpable homicide. Other common types of killing include the killing of a wife’s lover; the killing of people accused of witchcraft (or, more interestingly, killing for fear of witchcraft - including suicide by those accused of witchcraft); the killing of thieves; and accidental killing during the communal hunting season.
The death penalty is often prescribed, mostly for premeditated murder and theft when the thief is caught in the act. Killing because of adultery has also been known to attract the death penalty (apparently when the killing showed evidence of premeditation) although, in many instances, a fine or sentence of hard labour appears to suffice. Bohannan’s example showed noticeably high levels of ‘guilty but insane’ verdicts in the courts, mostly linked to various killings in the heat of the moment (for instance, those of thieves caught red-handed and suspected witches). Insanity (including epilepsy and sleeping sickness) appeared to match self-defence as a legitimate defence to a charge of murder.34
Two significant aspects of Tiv culture may be highlighted. The first is the Tiv paradox, noted by Bohannan, relating to witchcraft, where the Tiv believe that one’s closest kin (agnates and one’s mother) is simultaneously one’s refuge and support system, and also the people most likely to harm one.
Referring to this as ‘a very overt Tiv belief’, Bohannan explains that ‘it is one’s agnates who bewitch one, but not the agnates of one’s own generation who are, rather, one’s protectors against the elders of one’s agnatic lineage’.35 He gives the background as follows: 36
Tiv explain much of their cosmography and most of their notions of illness and misfortune by reference to tsav, which is sometimes translated into English as ‘witch-craft substance’. Tiv say some of the elders of the community, who have tsav and form the mbatsav, meet at night in order to carry out rituals which are to the advantage of the lineage and all people who live within its area. However, in order to carry on their work they must make human sacrifices from time to time, and it is as mbatsav that the elders decide who is to be sacrificed. All illness and death is ultimately attributable to tsav, usually it is the mbatsav among one’s agnates who have reason and power to kill one.
Reviewing a court case in which a man was found guilty but insane for killing a kinsman whom he thought was bewitching him, Bohannan concludes that the accused had acted according to sound Tiv values, and summarises the paradox as follows:37
Agnatic kinsmen are loved and respected because they are kinsmen and neighbours, but they are also feared because they share a common patrimony with you and because, as a group, they must renew themselves spiritually and ceremonially through spilling the blood of one of themselves as a sacrifice.
The second cultural feature of note is significant in that it links to a general point sought to be made in the article, namely, the importance of ritual killings. The Tiv believe in human sacrifice, not only for maintaining prosperity, but also for establishing what Bohannan reports as the ‘great fetishes’.38 These sacrifices or ‘fortifying ceremonies’ require bits of the human anatomy for their potency. A further upshot of such killings (which are considered murder in modern Tivland and punishable by hanging) is related to an instance where a man, in fear of being sacrificed in this way, pre-emptively killed the suspected perpetrator first in order to protect himself. 39
Finally on the Tiv, Bohannan notes a matter of some interest, which is that the material reveals a glaring difference between this society and societies in the developed West: the almost total absence of killings motivated by economic gain. He explains: 40
Lack of ‘economic motivations’ for murder is associated with the sort of community life which Tiv live. One would find it impossible to keep or enjoy anything which he might have gained by murder.
Bohannan relates an instance of a robbery where the Tiv attacked a Hausa trader and took his goods, and makes the point that when the murderers were caught, the goods were found to have been distributed among kinsmen.41
2.4 The Igbo
Seeking to understand the nature of punishment among the Igbo of Eastern Nigeria, Igwe42 classifies Igbo laws into two broad categories: divine and human.43 The former pertains ‘to God, divinities, spirits and ancestors’44 and offenders need to be punished ‘either during their life time or at the end of it, or even during their next life cycle’.45 Murder straddles both categories, being regarded not only as an offence against society but also as a violation of divine law.
Crimes in Igbo law include homicide, incest, suicide, arson and adultery. Murder was considered a crime against society as well as a violation of divine law, since all life comes from God.46 The killer is expected to hang himself, failing which he is banished. His property is forfeited, and it appears that ‘his family is excluded from most community privileges and also have their properties confiscated’, although it is not clear whether this happens upon banishment or only if he flees.47 Even a killing during war calls for ritual cleansing.
Where the killing is not perpetrated by kinsman upon kinsman, as in the case where the murderer is from another village, retributive justice calls for a life-for-a-life to stave off war,48 as well as compensation, the latter being payable even in the case of accidental killing.49 An unproven suspicion of witchcraft leads to the suspect being required to swear his innocence before a deity and, if over a period no harm comes to him, he is presumed innocent - his guilt is deemed proven if he suffers harm or illness.
An unusual crime found in the Igbo list is the crime of ‘unmasking a masquerade’, which the writer describes as the removal of the mask of a performer at an important ceremony, where the masked performers are believed to embody the spirits of the ancestors. Unmasking, thus, is tantamount to ‘killing an ancestral spirit’, with dire consequences for the group.50 The penalty for such behaviour is death. Penalties for other offences on the list include banishment, ostracism, a fine or forfeiture of property.
2.5 The Joluo
Wilson’s account of the Joluo (sometimes known as the Luo) of Kenya starts with a revealing statement: 51
The Nilotic Joluo regard violence as an unnecessary loss of control, or at best a calculated necessary last resort to secure goals which have been carefully weighed and assessed as to the possible supernatural and social consequences. The Joluo believe that a goal does not justify the means if the means require violence, unless all methods of arbitration, conciliation, appeal to authority, both temporal and supernatural have been tried and proved to be unsuccessful or inadequate.
Nowhere in the other materials studied has there been such an upfront description of an anti-violence principle as background to the understanding of homicide in the society in question. Apparently the reason is that ‘violence requires cleansing’ in order to ward off supernatural consequences. This belief in the supernatural permeates the Joluo response to homicide:52
Deeply ingrained is the notion that life goes on after death on earth - that ancestors in the spirit world are able to influence the lives, conduct, fortunes, for good or evil, of their descendants. Ghosts are more likely to be benevolent if they can count numerous descendants on earth and extremely malevolent if life on earth was cut short by violence or unnatural death by war, accident, or suicide. The more descendants a ghost has, the higher its status in the after-world, and therefore the greater its pleasure and benevolence.
The Joluo distinguish between unpremeditated, heat-of-the-moment killings arising from quarrels during drinking parties or incidents of infidelity or sexual jealousy, which are regarded as less serious (and having consequences only for the perpetrator) and premeditated murder, which is believed to affect the whole group.
As is the case with other studies reviewed, social (and sometimes geographical) distance plays a role. Murder by ‘outsiders’ leads to immediate sanction, which may take the form of confiscation of cattle, or the killing or banishment of the slayer. Sometimes the slayer’s group will be required to give up a woman to bear an heir in the name of the deceased.53 Linked to the Joluo belief in the supernatural, the placation of the ‘ghost’ of the deceased appears to be important. This will be further discussed below.
2.6 The Basotho
Closer to home, we begin with a scan of the approaches to homicide among the Basotho. According to Ashton, homicide is considered a civil wrong, as are assault and rape, and is punishable by the requirement of compensation.54 Compensation was in the range of ten cattle for the killing of a male or female adult and four or five for a child, or a figure set at the court’s discretion. No compensation was payable upon the killing of a kinsman, for the same reasons as in other communities studied, namely, that the givers and the recipients are the same legal entity. Compensation was calibrated according to the degree of culpability and, controversially according to Duncan, the value of the victim.55 This was later elucidated by Palmer56 as meaning social position rather than monetary value. Defences to homicide included killings deemed to be lawful, such as those on witches, enemies during wartime or in cattle raids, thieves caught in the act and an honour killing by a father on a son who has brought dishonour to the family.57 Abortion and infanticide were dealt with by the family even though they were not classed as lawful.58 Accidental deaths were not prosecutable.
2.7 The Zulu
Krige starts off her chapter on the political organisation of the Zulus with a useful description of ‘the tribe’, which is reproduced here because in many ways it mirrors the social organisation of indigenous communities in South Africa:59
Among the Bantu, political organisation has not yet been divorced from principles of kinship, and the true national unit appears to be the sib. In pre-Shakan days, Natal was peopled by small exogamous sibs. But the sib is not a stable unit; it grows, sub-divides and is reinforced by sections of people who have, for political reasons or because of an accusation of witchcraft, fled from their own kin. In this manner it develops into what is known as a ‘tribe,’ consisting of people belonging to many different sibs.
The social organisation described above was underpinned by the ascription of great powers to the King, who had both legislative and judicial powers, although Krige warns that ‘in neither was he as autocratic as the powers exercised by the great Zulu kings have led us to suppose’,60 pointing to the role of elders and councillors. Krige refers to another limitation on the powers of the King, which was that no killing could take place unless an offence had been committed, but she concedes that it was easy enough for trumped-up charges to be levelled by the King, especially against a rich man whose estate the King may covet.61
Homicide, then, was closely tied up with the powers of the King as far as both its perpetration and its punishment were concerned. Recognised crimes were witchcraft, desertion, incest and treason, all of which were punishable by death. In many cases, the death of the culprit was accompanied by the destruction of his homestead and confiscation of his cattle, which were ‘eaten up’ by the King. Cases of homicide were heard by the traditional courts of law and were classified into offences against the individual and offences against the King.62 Some regard was had to mens rea for purposes of punishment, where a fine or compensation may suffice for unpremeditated killings. A fine was also exacted for adultery and rape and, in aggravated cases, sometimes the death penalty. During the reign of Shaka, there were a slew of offences specially created by him which were punishable by death: trespassing into the royal harem; eating the new harvest before the King had performed the first fruits ceremony; and cattle rustling.63
Apart from the death penalty, killings perpetrated by the rulers were also known in traditional Zulu society. Krige points out that the King was both the group’s representative to the ancestors, and the nation’s chief medicine man, ‘the centre of all agricultural and war ritual and representative of the unity of the tribe’.64 In these roles, the King needed to be strengthened in order to be able to harness forces in the universe for the benefit of the nation. Krige continues:65
Furthermore, he must be strong, not only physically but magically. As representative of the tribe, he must be a man who has been specially fortified against all manner of harm, for so closely bound up with the welfare of the tribe is the life of the King, that enemies can strike a blow of the life of the tribe through him. By hurting the King, they can break up the whole tribe.
For these reasons, the King undergoes strict strengthening procedures: 66
On his accession, the Zulu King has to be spiritually prepared and fortified for his office and for his fortification parts of the human body are essential. The human body is the strongest and most powerful of all medicines, and for doctoring a King the most effective parts must be used.
Finally, Krige reports on burial rituals which again involve the loss of human life in the form of close household servants who are ritually killed and buried with the King.67 This aspect will be discussed further below.
2.8 The Xhosa
According to Soga,68 AmaXhosa did not recognise the civil-criminal distinction and recognised the death penalty only in the case of witchcraft. He emphasises the point that ‘[t]he primary object of Xhosa law ... is to preserve tribal equilibrium’,69 and that the task of the law is to guide the individual in maintaining the cohesion and integrity of the group. In addition to this, the task of the courts was to restore equilibrium rather than to dwell on guilt: ‘[T]he ethical question scarcely counts, restoration is the principal thing.’70 Incest, witchcraft and rebellion fall into this category: They are objectionable not because of moral reasons, but because they threaten the very existence of the group.
Soga also mounts a persuasive argument about the Xhosa view of murder, which he reports as being quite contrary to the Mosaic law of an eye-for-an-eye. AmaXhosa avoid capital punishment for a reason, says Soga, and such reasoning is ‘characteristic of the value placed upon the life of an individual as part of the tribe’.71 The stock answer of the Xhosa to the question as to why they eschew capital punishment, according to Soga, is ‘Why sacrifice a second life for one already lost’.72 Instead, a heavy fine is paid.
2.9 Southern African peoples in general
There is an abundance of materials on homicide among the indigenous communities in South Africa73 and its neighbours, and perhaps all that can be done here is to summarise some of the data, pulling out the common threads and any other information of special interest or relevance.
Strongly echoing the Xhosa approach, AmaBhaca, also of the Eastern Cape in South Africa, have a notion of crime which distinguishes those offences that compromise the solidarity of the group from inter-personal disputes. The former are seen as an attack upon the King, who embodies the nation: Such is the Bhaca view of murder, assault, witchcraft and incest. Again showing similarities with the Xhosa, the response to murder is restitutive and not necessarily the death penalty.74 The Swazi also differentiate between private wrongs and cases ‘with blood’,75 which include murder and witchcraft, punishable by death coupled with the confiscation of the slayer’s property.
Mönnig76 and Harries77 on BaPedi, Prinsloo78 on BaLobedu, Cook79 on the Bomvana and Junod80 on the Tsonga all speak of various kinds of compensation for a killing, whether the death penalty is an option (as in the case of BaPedi) or not. The models of reparation range from giving cattle to enable a male member of the victim’s family to marry a wife or the giving of a daughter as procreator, to the delivery of a token number of beasts to the deceased’s kin ‘to dry the tears’, sometimes allocated by the chief from the fine received. BaTlokwa exact the death penalty (including euthanasia) for murder, accompanied by a fine to the chief, a portion of which is used in the purification rites.81 Myburgh and Prinsloo82 report that the Ndebele prescribe the death sentence for murder but, in certain circumstances, a fine or even corporal punishment.
In these responses to homicide, it is interesting to note the defences that are allowed to a charge of murder, as they throw some light on which killings are viewed as ‘lawful’ or in any other way excusable. Self-defence is the most common exculpatory factor, and is found among the Pedi of Phalaborwa, BaTlokwa, Batswana and Bakwena83 and the Ndebele. This is closely followed by the killing of a witch or wizard (someone identified as such by a divination) or a thief caught in the act. (It does appear that, for many indigenous communities, the killing of a suspected witch mitigates the sentence, is treated as a version of self-defence or, as in the case of the Tiv of Nigeria, is characterised as a form of insanity.) Other common defences include the killing of a wife’s lover, or a killing during conflict with outsiders or in obedience to a lawful command. Amongst BaTlokwa, provocation is a recognised defence while among indigenous societies, generally, drunkenness is hardly ever acceptable as a complete defence.
This review of Southern African societies is concluded with accounts of two significant neighbours: the Swazi of Swaziland and the Ndebele of Zimbabwe. The Swazi account is significant in that, of all the countries in the Southern African Development Community (SADC) region, the Kingdom of Swaziland arguably is the one nation state in the neighbourhood which survived the colonial experience with its indigenous culture virtually intact.84 Kuper85 describes the richness of Swazi culture, from precolonial times to the present day, painting a picture of a nation in which the monarchy plays an important role. Crime among the Swazi is divided into private wrongs, and cases ‘with blood’.86 Private wrongs include theft, slander, adultery and property disputes - all punishable by fines or compensation, or both. ‘Blood’ cases are murder and witchcraft, both punishable by death with confiscation of property, although banishment is often substituted in the case of witchcraft. The heavier penalties are invariably exacted where the offence is against the rulers as opposed to commoners. However, for our purposes the real significance of the Swazi lies in their rituals of kingship, which will be discussed later in the article.87
As for the Ndebele of Zimbabwe, their attitude to homicide mirrors much of the values of other Nguni groups in the region. They divided crime into two categories: amacala amakhulu (serious crimes which included murder, treason and witchcraft); and amacala amancane (less serious offences such as theft and domestic conflicts). The former were heard before the King, while the latter could be dealt with by any tribunal within the hierarchy of household heads, headmen and chiefs.88
Both murder and witchcraft were punishable by death, with banishment being the norm in the case of unproven allegations of witchcraft.89 Death was also the penalty for treason and other political crimes, and in succession upheavals there was much evidence of purges and massacres.90 However, as will appear below, our interest in the Ndebele again lies in their model of governance.
3 Homicide in traditional African societies: Some common themes
A quick review such as the one above throws up some commonalities and a number of discernible trends, many at a fairly superficial level but some having the potential for deeper analysis. For our purposes, we shall concentrate on those themes that provide some insights into our core question, which is that of seeking norms relating to the taking of human life. From this perspective, recurring themes appear to be the following: the group and its influence on the individual; religion, especially notions of life after death; belief in the supernatural; kinship as the basis of social relationships; and, in selected cases, the issue of violence at the hands of those in authority.
3.1 The group and the position of the individual
In all the traditional communities reviewed, some version of the communal ethic is in evidence in a stronger or weaker form, depending on the group surveyed. It is assumed here that it is beyond debate that African traditional societies displayed or were perceived to display strong group solidarity in their social organisation. Wiredu captures this well in the following statement:91
In Africa it is anthropologically verifiable that generally, the operative ethic is communalism. This is a kind of social formation in which kinship relations are of the last consequence. People are brought up early in life to develop a sense of bonding with large kinship circles. This solidarity starts from the household and radiates outward to the lineage and, with some diminution of intensity, to the clan at large. The normative meaning of this bonding for an individual is that she has obligations to large groups of kith and kin. This relationship is balanced in the converse by rights due to her from a corresponding multitude of relatives.
As will be seen in the conclusion below, the ethic of group solidarity is the single most important consideration in understanding the approach to homicide in traditional African societies.
It is possible to proceed on the basis of this assumption even as debates continue as to the existence, then and now, of this notion and whether its influence persists. At least two ideas appear to be embedded in the notion of group solidarity: kinship as the basis of social organisation; and survival in the face of harsh economic realities.
3.1.1 Importance of kinship
As pointed out by Wiredu, African societies are generally organised around lineages, regardless of whether they remain small and close-knit or have expanded into nations, nation states or indeed even empires.92 One immediate offshoot of this in respect of homicide is the distinction drawn between a killing involving close kin, distant kin, and ‘outsiders’. As a general rule, killings within the family were considered serious transgressions which, however, did not lead to the death penalty, and sometimes did not necessarily lead to compensation either. In the former instance, staying the death penalty appeared to be based partly on the matter being ‘somewhat private’ and subject primarily to the jurisdiction of family heads and elders and supernatural sanction, and partly on a general reluctance to compound the loss of a life with the taking of another. 93
In the latter case, compensation was not payable simply because of the logic of role overlap: The people paying the compensation would be the same people due to receive it, using the same pool of resources.94 As a result, the matter tended to be left to the ancestors or God to punish, being considered a sin rather than a crime.95 Where a fine was payable to the chief, the matter fell into the category below.
3.1.2 Economic survival
There is an obvious overlap between the various reasons for the strong hold of group dynamics in traditional society (for instance, both religious and moral considerations play a part in mediating the behaviour of members as kin or as part of the community). The imperative of survival is one such consideration, which cuts across many identity boundaries. In respect of homicide, the survival of the group as an economic entity is key. Generally, members are considered to ‘belong’ to the chief or the king in his role as the embodiment of the unity and the well-being of the group.96 The issue of the death penalty thus comes in, underlining the seriousness of depriving the group of a productive unit in the struggle for survival, whether the deceased’s role was that of warrior, agricultural producer, child bearer or medicine man.97 The same logic underpinned the rules regarding compensation to the victim’s family (usually in the form of cattle to enable them to marry a wife, or the surrender of a female, as procreator) in addition to any fines payable to the ‘state’.
The question that really cries out for an answer is whether power and influence tilted in favour of the group and away from the individual. On the evidence of the societies reviewed, there seems little doubt that this question should be answered in the affirmative: The primacy of the group comes through at all levels. A good example is the question of ritual cleansing which, in all the societies studied, underpinned a clear distinction between individual actions, on the one hand, and their impact on the group and its well-being and survival, on the other. A clear distinction was made between punishments specific to the perpetrator (such as a court-imposed fine) and group activity, usually involving a slaughtering to appease the ancestors, as expiation for the deed lest the group be visited with calamity. 98
3.1.3 Religion, life and the hereafter
It is beyond the scope of this article to do a treatise on African religion and its influence on the thinking of traditional communities. All that can be managed here is an attempt to sketch the basic shape of those belief systems that may fall under the rubric of religion, always with the aim of determining their relevance to the issue of homicide.
There is no better place to start than with the seminal work of Mbiti, who begins his treatise with the following words:99
Africans are notoriously religious, and each people has its own religious system with a set of beliefs and practices. Religion permeates into all the departments of life so fully that it is not easy or possible always to isolate it. A study of these religious systems is, therefore, ultimately a study of the peoples themselves in all the complexities of both traditional and modern life.
He goes on to assert that for the African, religion permeates all areas of life and that ‘there is no formal distinction between the sacred and the secular, between the religious and non-religious, between the spiritual and the material ...’100
Linking these thoughts with the so-called community principle, Mbiti concludes: 101
Traditional religions are not primarily for the individual, but for his community of which he is part. Chapters of African religions are written everywhere in the life of the community, and in traditional society there are no irreligious people. To be human is to belong to the whole community, and to do so involves participating in the beliefs, ceremonies, rituals and festivals of that community.
Very generally speaking, many traditional religions will exhibit some or all of the attributes listed by Igwe102 in his discussion of the Igbo, as follows:
- a belief in God - a supreme being, called by various names relating to the roles of creator, protector, ruler of earth, sea and sky, and generally perceived to reside ‘above’ (that is, in the sky, the clouds or atop some fearsome mountain);
- a belief in divinities, who are spiritual beings who ‘act as servants of God in His theocratic government of the universe’.103 The supreme being is usually approached via the divinities;
- a belief in spirits - sometimes an overlapping concept with divinities, sometimes separate. Spirits, according to traditional Igbo, are ‘good or evil, benevolent or malevolent’ and may be used as a channel to approach God;
- a belief in ancestors - who are ‘departed spirits who stand in close relation to their family or tribe’104 and who are regarded as members, to varying degrees, of the families and communities to which they belonged during their lifetime, a concept very familiar to local southern African traditional societies and systems of customary law;
- a belief in magical forces - believed to emanate from the universe and capable of being harnessed, for good or for evil, by those with the skills or powers to do so. This is the basis of the widespread African belief in witchcraft,105 an element that is important enough to discuss separately (below).
The implications of these beliefs on the issue of homicide are readily apparent, as are the obvious overlaps between them. Hopefully, the influence of each will emerge in the discussion which follows.
Conscious of these overlapping elements, I propose to highlight two aspects, which I believe are key to understanding traditional attitudes towards life and death: belief in the hereafter; and belief in the supernatural.
3.1.4 Belief in life after death
The belief among African traditional societies that life continues in some form after a person’s physical death is so pervasive that it can be categorised under headings as divergent as religion and family law. According to Mbiti, ‘[d]eath stands between the world of human beings and the world of the spirits, between the visible and the invisible’.106 Using the example of a Ndebele funeral, Mbiti sketches the procedures and rituals that are performed, and explains their symbolism in relation to the Ndebele belief in the hereafter.
Thus, the spear with which the heir of the deceased strikes the grave is for defence and protection ‘on the way to ... the new country’; personal belongings buried with the body are for use in the hereafter; and the animal slaughtered after the ceremony is likewise to provide the deceased with food.107 Post-funeral ceremonies and rituals are equally redolent with meaning, cleansing the living from the stain of death, mystically connecting them with the deceased and, finally, ‘summoning back’ his spirit so that the unity of the living and the departed can be established. Mbiti states:108
It is a ritual celebration of man’s conquest over death: For death has only disrupted and not destroyed the rhythm of life. It indicates also that the departed is not really dead: His is a living-dead, and can be contacted, invited back and drawn into the human circles. The new ‘beast of the ancestors’ symbolizes the continuing presence of the living-dead in the family and among his people.
Wilson takes up the story, relating similar funeral rites and the reasoning behind them directly to the Joluo attitude to homicide. 109
Lee and Vaughn echo these sentiments:110
To simplify, the dead could only find their place as ancestors, rather than vengeful ghosts, if their loss had been properly registered, not only by the individuals closest to them, but by the social groups of which they were members.
The message embedded in these examples is that each family considers that it has members in the present as well as in the past; that those in the past are the living dead whose condition in the afterlife is largely dependent on the actions and piety of those they left behind. They are thus likely to be well-disposed towards their earthly kin if they are happy with their performance, and angry and ill-disposed if they are not. The task of the living is to keep them happy and well-disposed.
3.1.5 Belief in the supernatural
Allott111 refers to these beliefs and practices as ‘religio-magical’ when emphasising their significance in customary law, thus confirming the overlaps mentioned above. Whether considered as an aspect of religion or some other kind of spirituality, the supernatural has always played a key role in the lives of traditional African societies, explaining daily phenomena, such as illness, or natural disasters, such as droughts and floods. However, its most visible form in day-to-day social interactions is the issue of sorcery or witchcraft.
All the societies reviewed in the article share a preoccupation with witchcraft, and invariably reserve some of their severest penalties for perpetrators, real or imagined.112 Krige reports that ‘witchcraft is looked upon as the most terrible crime by the Zulus, and as such cannot be tried by judicial process’113 because it is due to unforeseen forces discoverable only through divination, which appeals to the same forces. Witchcraft is perceived not only as hurting individuals and families, but also as destabilising the group.114 It is thus classified variously in many societies as a crime and/or a delict and even treason.115 The same multifaceted interpretations lead to differences in approach to sentencing, ranging from ‘exemplary’ harshness (Zulu); death or banishment (Swazi); to a strong link to insanity (Tiv), which has the effect of mitigating the sentence or providing a defence in the case of one who kills a suspected witch as a pre-emptive strike.
3.3 Killings at the hands of authorities
Before the central issue of accountability is discussed, it is important to treat the issue of homicide perpetrated by rulers or at their behest, or in consequence of customs and rituals surrounding them. In the previous discussion we have seen how traditional societies have viewed, and dealt with, homicide amongst their members, with the role of the rulers (whether an aristocracy, a militaristic leadership or kinship hierarchy) being confined to adjudication and sentencing.
What has not received attention is the question of the taking of human life in the absence of any wrongdoing on the part of the victim, particularly in those cases where the killing is on the direct orders of the ruler (but not in execution of a sentence for an offence), or is done on his behalf or in pursuit of custom relating to him. For convenience, I shall refer to the ruler as the king, and to the phenomenon under discussion as human sacrifice.
Excluded from this discussion, as mentioned above, are killings in execution of sentence for wrongdoing, self-defence and killings in war or other conflict situations. Also excluded is the crime of ritual murder as committed by ordinary citizens, and the issue of regicide, which is the other side of the coin in the current discussion. The issue of regicide will be discussed later in the article.
A final word of caution before the discussion: Although there is ample anthropological evidence of the practice of ritual killing by or on behalf of kings, in contemporary times the matter remains mostly shrouded in secrecy. In this respect, it is no different from many other traditional practices where there is a cult of secrecy, such as in respect of the exact details of initiation rites.
A useful starting point is an understanding of the place and role of kings in African traditional society. In the section of his book headed ‘Kings, queens and rulers’, Mbiti has the following to say about kingship:116
Where these rulers are found, they are not simply political heads: they are the mystical and religious heads, the divine symbol of their people’s health and welfare. The individuals as such may not have outstanding talents or abilities, but their office is the link between human rule and spiritual government. They are therefore, divine or sacral rulers, the shadow or reflection of God’s rule in the universe. People regard them as God’s earthly viceroys.
This is echoed by Krige, who describes the Zulu King as both representative to the ancestors and chief medicine man, ‘the centre of all agricultural and war ritual and representative of the unity of the tribe’.117 This explains the need for the King to be fortified for his office, not only so that he can have powers to harness forces in the universe for the benefit of the group, but also in order that he becomes strong enough physically and magically to withstand attempts to harm him, for to harm the King is to harm the nation.118
For these reasons, the King undergoes strict strengthening procedures.119 Krige also reports on burial rituals which again involve the loss of human life in the form of close household servants who are ritually killed and buried with the King.120 Human sacrifice for similar purposes is reported as far afield as Asante in the nineteenth century.121
There is also evidence of ritual ‘fortifications’ involving human sacrifice among the Tiv, although it is not clear if this was in all cases at the behest of royalty. 122
4 Incidence and types of homicide: A summary
From the materials reviewed, it is evident that homicide occurred significantly in traditional society, even frequently. What is difficult to assess is the rate of the occurrences, because of the lack of uniformity in the incidence of killing in the various traditional contexts, ranging from societies like the Nuer, where it can legitimately be described as ‘over the top’, to systems such as those of the Joluo, where a considered and apparently successful set of restraints operated to keep the incidence of homicide muted.123 However, in all the societies studied, homicide presents as a human urge that they all struggled to deal with.
An attempt to list the types of killing that took place would produce something like the following:
- hot-blood fights and disagreements (the ubiquitous drinking party);
- sexual jealousy (crimes of passion including those triggered by marital infidelity);
- accidental killings (eg the hunting party);
- negligent and/or reckless killings (to be distinguished from accidents);
- premeditated killing (encompassing murder);
- war, cattle raids and revenge attacks;
- killings driven by ‘lawful instructions’ (these would include the execution of a lawful death sentence, as well as ritual homicide ordered by authority of custom);
- self-defence or other necessity which precluded mens rea (including, in some societies, pre-emptive killing in the belief by the slayer that he was the target of sorcery);
- violence involving family issues (infanticide, matricide, patricide, sacrifice, euthanasia);
- killings driven by superstition and belief (for instance, deformities at birth, including the issue of twins); and
- regicide, assassination and other political ‘solutions’.
All these kinds of killings come into sharp relief when, in each society, they are compared to the defences accepted as being adequate to exculpate completely or at least to mitigate sentence on the basis that moral blameworthiness was reduced. Thus it is interesting to note that, in addition to self-defence proper, many societies were prepared to countenance the killing of a suspected sorcerer124 and a thief caught in the act, almost on the same basis of self-help. Also interesting are the lengths to which some societies were willing to go to extend the blanket protection of insanity and epilepsy, even sleeping sickness, to certain categories of slayers.125
For our purposes, the deployment of defences and other mechanisms of excusability begin to hint at the values of these societies and the norms they observed in relation to the taking of human life, a subject to which we now turn.
5 Homicide in traditional African societies: Norms and sanctions
The question of sanctions is important in throwing some light on the disapproval of certain kinds of behaviour, and how strongly or weakly such actions were disapproved of. At the apex of the hierarchy of disapproval stands the death penalty. In common with some modern legal systems, the death penalty was considered appropriate for premeditated killing (namely, murder), although in traditional systems this highest form of disapproval was also extended to homicides that contemporary systems would balk at. Indeed, some kinds of killing, as we have seen, were considered a duty.126
Generally speaking, traditional society reserved the death penalty for murder, witchcraft, red-handed theft and treason.127 The line between the death sentence as an order of the court, the King and other judicial process, on the one hand, and permission for a member of the group to take the law into his own hands in certain circumstances, is blurred.128 One may thus qualify the earlier statement by referring not only to the death penalty, but also ‘to any death of a wrongdoer occurring at the hands of ordinary citizens for which there were no repercussions’.129 This has implications for the discussion of accountability which follows, but for now it should suffice that the sanctioned killing of a wrongdoer marked the pinnacle of social disapproval.
There were other punishments. The taking of a life may result in a fine, especially (but not exclusively) in those societies where the death penalty was not recognised.130 The fine may be coupled with a requirement for compensation to the victim’s family. A fine could expand to the confiscation of the slayer’s assets (which itself may be executed in parallel with the death penalty). Banishment was another mainstream punishment.131 In earlier societies, the list of sanctions may include an entitlement to revenge (blood-for-blood) by the victim’s kin. Other forms of compensation included enslavement132 and various versions of providing offspring to swell the ranks of the victim’s lineage. 133
The philosophical question here is a challenging one. It is easy to discern a very pragmatic motive in these arrangements, animated, on the one hand, by a strong disapproval of wanton killing and, on the other, by the practical need to ‘make good’ the loss to the deceased’s kin. At a cursory glance, no evidence of ethical codes such as ubuntu readily avails itself, but this perception will be interrogated at the conclusion of the article.
It should suffice here to conclude that, on the question of traditional values with regard to homicide, one clear norm which comes to the surface is this: Killing was not allowed except for good reason. The reality is simply that the list of reasons considered to be good reasons was, in the circumstances of those societies at the time, somewhat longer than would be countenanced in modern law. Any taboos that supported this norm appear to have been based on religion and on the supernatural - namely, that wanton killing defiled the group and consequently attracted the wrath of God, or the ancestors, hence the need for ritual cleansing.134 Compensation and other forms of reparation were the secular side of the coin. Self-defence and its variants are helpful in understanding the outer limits of the norm, indicating when it was considered to have been breached and when it was not.
6 Question of accountability
The Collins dictionary135 gives one of the meanings of ‘account’ as ‘an explanation of conduct, especially one made to someone in authority’, and describes ‘accountable’ as being ‘responsible to someone for some action; answerable’. There is a stricter, more formalised meaning in international law, a working version of which for our present purposes can be formulated as follows, based on the notion of the right to life.136
The formulation concedes that there are various forms of accountability, ranging from imprisonment and fines to non-criminal forms, such as formal disciplinary processes and reprimands. The question is whether any of these forms of accountability existed in traditional society. The answer, after the review we have gone through, must surely be in the affirmative. Unfortunately, that is not the end of the matter. Several questions remain, such as whether such forms of accountability actually imply the recognition of a right to life as known in human rights law or by any other yardstick. That interesting discussion is beyond the scope of this article, but the question of human rights is raised vividly in one particular context, which provides a fitting conclusion to these reflections on traditional society.
This is the attempt to squarely confront the elephant in the room: the killing of ordinary people by those in authority. I confine myself to the ritual killing of members of the group by their leaders, specifically kings. It was noted earlier that kings were considered to be the embodiment of the group,137 perceived to have both earthly and magical qualities necessary for the protection of the collective. The discussion is thus about that complex of issues which implicates the so-called communal ethnic and questions of the right to life.
Time does not permit a full exposition of the terrain, but several pointers to the argument should suffice in making the point. In the first place, I am persuaded that the communal ethic or solidarity principle is a reality for African societies and that continuing doubts about this are the result of weak argumentation on the part of its supporters or wilful blindness on the part of doubters. I will attempt to make the point by referring to the work of African scholars who argue, persuasively in my view, that the failure to acknowledge the significance of kinship in the make-up of pre-colonial African societies is a missed opportunity to understand Africans and the world view that animates their response, for instance, to human rights.138
In brief, these writers make the point that a human rights dispensation for Africa that consciously accepts and works with, rather than against, these insights will find more resonance with African populations. Second, many of these scholars question anthropological and historical versions of pre-colonial societies which emphasise the ‘single despot’ view of African kingship.139 Taken together, these views paint a picture of political accountability by traditional leaders that went far beyond anything found in modern constitutions.
Simply stated, the argument is that there was far more real accountability in pre-colonial traditional society than there is in contemporary African society. Asserting this argument, Ayittey observes:140
The indigenous form may be different from the Western. Nonetheless, the important fact is traditional African rulers - chiefs and kings - are held accountable for their actions and were removed from office or killed (regicide) for dereliction of duty. By contrast, most modern African leaders cannot be held accountable and commit crimes with impunity.
Ayittey details various accounts of West African societies where the duties and responsibilities of a king were so onerous that many candidates (where choice was possible) declined the honour when offered because, in addition to a life of genuine powerlessness except as a mouthpiece of the council of elders,141 their lives were forfeited if their stewardship of the group did not pass muster. Both the Yoruba and the Akan had mechanisms to trigger regicide, royal suicide or abdication.142 Family-sanctioned assassination of the leader was also included.143
In this regard, Ayittey is corroborated by Ndlovu-Gatsheni who describes the development of the Ndebele kingdom from the militaristic period of state formation to a ‘settled phase’ where the contours of an accountable kingship started to emerge.144 He rejects the model of a single despot, explaining his position as follows:145
By a single-despot model I mean the emphasis on precolonial forms of governance as centralised pyramidal monarchies with all-powerful kings at their apexes with power over both the life and death of their subjects ... In the first place it is too simplistic to use a single-despot model to explain African governance systems because the diverse nature of African societies defies such generalizations. Each precolonial society had a unique set of rules, laws and traditions suitable for particular contexts. These rules, laws and traditions, commonly termed customs, formed the basis of how people lived together as part of a community or state.
Elaborating on the settled phase, Ndlovu-Gatsheni maintains that it was during this time that issues of governance, accountability, legitimacy and human rights (long ignored during the turbulence of the formative years) now began to receive attention in state politics.146
It was a phase marked by ‘royalisation’ where, as the kingship took on more shape and power, so also did the constraints on that power increase. Ndlovu-Gatsheni explains:147
In theory, the king was the head of state, head of government, religious chief, commander-in-chief of the armed forces and the supreme judge of all criminal cases. In practice, however, the king was basically a ceremonial head of state in all these posts and a source of unity in the state.
This aligns with the view of Ayittey who states:148
Generally, African natives accepted the king as a necessary evil. He was necessary for the preservation of the social order but a potential danger. He could abuse his powers and be intrusive, trampling on the independence and freedom of his people. To resolve this dilemma, African ethnic groups sought, with various degrees of success, to create some personality hidden from public view but whose awe-inspiring authority could be invoked to maintain order and harmony.
Such analyses support the assertion that African kings in early times were not only accountable, but the things they were accountable for covered a much wider spectrum. So, for instance, as Ndlovu-Gatsheni explains,149 pre-colonial kings were accountable even for natural disasters.150
Two points emerge from this. The first is that it would be difficult to deny that the notion of accountability existed in traditional society, even in its contemporary sense of requiring that consequences be visited upon the taker of human life. The second is that, unfortunately, this does not dispose of the matter of whether such accountability was exacted in pursuit of a recognised right to life. When aides close to the king arranged for people to ‘accompany’ the king to the other side when he died, they believed they were acting in the common good. As my KwaZulu-Natal informant said, ‘if not, it was just murder’.151 It is difficult to decide whether or not this is evidence that a right to life was recognised in traditional society. On the one hand, an argument may be made that, generally speaking, life was respected and killings of this type were an exception to the rule (that is, that taking life in pursuit of the common good was excusable), or it may be contended that an example such as this proves the absence of a notion of the right to life, on the other.
On balance, I believe that there is enough material to support an assertion that traditional African society recognised a right to life; it was in its operationalisation that the practice of these societies differs markedly from modern practice.
In the first place, ‘life’ would have meant something different to these societies. As mentioned above, life would have been framed in communal terms, both from a practical standpoint as well as in a moral sense. A conception of a right to life would perforce conform to this ethic, a notion that is perhaps already hinted at in various versions of the ‘wealth in people’ idea.152
Additionally, as a human rights principle, the right to life would need to be understood in terms of the possibility that an African conception of human rights may be different from the Western conception.
Ake explains it as follows:153
First, we have to understand that the idea of legal rights presupposes social atomization and individualism, and a conflict model of society for which legal rights are the necessary mediation. However, in most of Africa, the extent of social atomization is very limited mainly because of the limited penetration of capitalism and commodity relations. Many people are still locked into natural economies and have a sense of belonging to an organic whole, be it a family, a clan, a lineage or an ethnic group. The phenomenon of the legal subject, the largely autonomous individual conceived as a bundle of rights which are asserted against all comers has not really developed much especially outside the urban areas.
Ake elaborates further:154
It is necessary to extend the idea of human rights to include collective human rights for corporate social groups such as the family, the lineage, the ethnic group. Our people still think largely in terms of collective rights and express their commitment to it constantly in their behaviour. [This disposition] ... underlies the so-called tribalists voting pattern of our people, the willingness of the poor villager to believe that the minister from his village somehow represents his share of the national cake, our traditional land tenure systems, the high incidence of cooperative labour and relations of production in the rural areas. These forms of consciousness remain very important features of our lives. If the idea of human rights is to make any sense at all in the African context, it has to incorporate them in a concept of communal human rights.
If there is substance to these assertions, then such a conception of human rights is bound to provide a context one cannot ignore in assessing how the right to life would have been perceived in traditional African society.
In the second place, traditional African societies were facing exigencies in their daily lives which were quite different from contemporary life. There were categories of killings then prevalent that have no counterpart in modern conditions (or, at least, the justifications of which were unique to those circumstances). Infanticide, euthanasia, killing of the elderly and regicide would largely fall in these categories.155 The reasons are varied, ranging from inhospitable climate and terrain making it impossible for a nomadic group to travel with the infirm, to superstition. 156
Interrogating the link between the community principle and human rights naturally leads to a reflection on the impact of this ethic on morality. Here the work of Metz is instructive. Metz argues, to my mind convincingly, for a link between the communal ethic and ubuntu as a moral theory.157 Setting out to refute criticisms of ubuntu as vague (that is, without ascertainable content), collectivist (and, therefore, hostile to individual freedom) and anachronistic (because of its traditional origins), Metz articulates a normative theory based on the capacity for community represented by what he calls ‘identity’ and ‘solidarity’.158 This leads him to a more nuanced interpretation of the popular rendition of ubuntu as meaning that ‘a person is a person through other people’. According to his version, ‘one becomes a moral person insofar as one honours communal relationships’.159
According to this moral theory, grounded in a salient Southern African valuation of community, actions are wrong not merely insofar as they harm people (utilitarianism) or degrade an individual’s autonomy (Kantianism), but rather just to the extent that they are unfriendly or, more carefully, fail to respect friendship or the capacity for it. Actions such as deception, coercion and exploitation fail to honour communal relations in that the actor is distancing himself from the person acted upon, instead of enjoying a sense of togetherness ...
Arguing on the basis of dignity being the keystone of human rights and rejecting the Kantian view that persons have worth because they have the capacity for autonomy, Metz describes his account as asserting that persons have superlative worth ‘because they have the capacity to relate to others in a communal way’.161 This is an attractive account of human rights from an African perspective as it does not deny the resilience of the communal ethic as it, similarly, does not deny the importance of human rights. This may yet prove useful in untangling the Gordian knot posed above, namely, the issue of accountability for homicide in situations where an assessment of action as right or wrong centrally involves the well-being (however perceived) of the group.
Having reviewed the practices and attitudes of a number of traditional African societies on the issue of homicide, we are left with a sense that we do not have to go very far to establish that these societies did recognise and understand the notion of accountability. We also find that such understanding encompassed aspects of accountability that are recognised in modern society, namely, the subjection of a perpetrator to sanctions. But there the similarity ends.
To understand this, we need to go back to the task at hand, which is to see if there is anything in traditional practice that has resonance in contemporary life relative to the taking of human life and the consequences of such action. Heyns sets out the basics:162
The right to life under international human rights law has two distinct components. The first sets out the substantive norm, and provides that no one shall be subjected to an arbitrary deprivation of his or her life. It is thus preventative.
In respect of accountability, Heyns states:163
The second component of the right to life is that there shall be accountability for arbitrary deprivations of life. Accountability entails notions of justice, which is traditionally seen as requiring investigations and where applicable, prosecutions, but increasingly also truth seeking and reparations.
Importantly, Heyns emphasises that the lack of accountability for an arbitrary deprivation of life, itself, is ‘a stand-alone violation of the right to life’164 in international law.
Finally, observing that the right to life is not unlimited and may in certain circumstances be curbed, Heyns discusses the ‘protect life’ principle, concluding that in current international law the right to life calls for the taking of life only to protect another life. ‘Life may not be taken for example purely to protect property.’165
This then puts into perspective the question posed earlier, namely, whether the undoubted examples of traditional accountability imply that a right to life was recognised in traditional African society. It also puts into perspective the statement made that there are similarities between the practices of traditional society and the requirements of modern law, but that these similarities do not go far.
To make sense of the traditional African approach to homicide, one has to understand that it is, predictably, pervaded by the community principle, which itself is no stand-alone factor but part of a complex set of interlinked considerations: religion; belief in the supernatural; and the survival imperative. Inevitably, ‘life’ in the sense of the right to life must have meant something to traditional society that is fundamentally different from the meaning it has in contemporary society. There is a sense in which it is not overstating the case to suggest that a person’s life ‘belonged’ to her but to others as well. Identifying her individual ‘right’ to this life was, therefore, not a single linear calculation. A homicide in traditional society (whether accidental, negligent or intentional) could invoke justifications, or produce consequences, that are demonstrably different from anything that one might associate with, for instance, a killing in a South African suburb or township today. In a traditional small-scale nomadic society of hunter-gatherers or small-scale farmers, for instance, the taking of a life might invoke explanations such as that she was a witch intent on harming us or our crops; he was a thief endangering our livelihood; she was a deformed baby representing a curse by the ancestors; he was old and infirm, unable to travel with us when the snows come; he was plotting against the king, destabilising the group and courting the wrath of the ancestors. The list is long. Even in the case of a straightforward murder, there were consequences not only for the perpetrator but for the group (be it the clan or the tribe), and these were differentiated according to whether the life taken was that of a kinsman or an outsider.
Modern notions of accountability are only partly accommodated in such a scenario, as are notions of life and notions of right, indeed even of justice. That the contemporary notion of accountability is tied up with justice is not surprising: It operates at a level where the nation state, acting through elected representatives, requires to maintain order among large populations, the members of which are largely strangers to each other.166 In small societies where everybody knows everybody else (and are most likely to be kin), the priorities may be different - restoration of relationships may be more important than justice in an abstract sense.167 The question whether restoration may itself be an aspect of the idea of justice in such societies suggests that further work needs to be done to understand the ‘mind’ of traditional African societies on the matter of the taking of human life.
1. See, eg, F Boyle Through Fanteeland to Coomassie: A diary of the Ashantee expedition (1874); TB Freeman Journal of various visits of the Kingdom of Ashanti, Aku, and Dahomi (1844); W Hutton A voyage to Africa (1821). For a sample on writings on Southern African societies, see J Shooter The kafirs of Natal and the Zulu country (1857); AT Bryant Olden times in Zululand and Natal (1929); AF Gardiner A journey to the Zoolu country (1836); BA Marwick The Swazi (1940); HA Junod The life of a South African tribe Vol 1 (1927); HC Lugg Life under a Zulu shield (1975); M Hunter Reaction to conquest (1936).
2 War, in particular, would have provided a useful counterpoint in the current discussion. Not only would it introduce a different scale of assessment on the question of the taking of human life, but it would also supply a helpful context against which to understand modern developments such as the growth of international humanitarian law. Despite these attractions, however, the subject of war is vast and the results yielded by the excursus may not throw enough light on the core concerns of this article to justify the digression. One notes, eg, that in a 15-page section devoted solely to war and warfare, and covering everything from mobilisation and doctoring the army to tactics and strategy and weapons of war, Soga hardly touches upon the issue of norms around killing. One has to try and deduce attitudes from headings such as ‘Mutilation of the dead’ and ‘The chief’s address to the warriors’, neither of which is really helpful to the quest: The former is about the fear of backlash from the enemy’s medications, and the latter is largely about chants and exhortations (see JH Soga The Ama-Xosa: Life and customs (1932) 65-81). It is similarly difficult to extract any deep understandings about the right to life from Kuper’s rich account of warfare among the Swazi (see H Kuper An African aristocracy: Rank among the Swazi (1947) 123-126).
4. See, eg, GM Wilson ‘Homicide and suicide among the Joluo of Kenya’ in P Bohannan (ed) African homicide and suicide (1960) 179, who says that ‘[t]he Joluo people - although on the surface they have accepted Christianity - still retain the ceremonial, religious, and ritual aspects of their culture to a very large degree. Their family system remains strong and largely unimpaired by the impact of Western civilisation.’ P Bohannan ‘Homicide among the Tiv of Central Nigeria’ in Bohannan (above) 31 discusses the introduction of British legal arrangements and the Tiv response as one of ‘amazed wonder that Europeans execute a man whose only fault was killing another’.
6. Art 6(1) of the International Covenant on Civil and Political Rights provides: ‘Every human being has the inherent right to life. This right shall be protected by law. No one shall be arbitrarily deprived of his life.’
7. C Heyns ‘The right to life under international law: Introductory remarks’, unpublished paper delivered at the Seminar on Unlawful Killing and Accountability in African Societies, University of Pretoria, 21 September 2015 (used with permission).
8. EE Evans-Pritchard The Nuer: A description of the modes of livelihood and political institutions of a Nilotic people (1940).
12. Conceding that the word ‘chief’ may be a misnomer, Evans-Pritchard describes these members as sacred persons without political authority who are central in the settlement of feuds and are arbiters in issues involving cattle. In earlier times they enjoyed the power to bless or curse.
21. Gluckman, citing I Schapera ‘The sin of Cain’ in Royal Anthropological Institute of Great Britain and Ireland (1955) 33-43.
32. T Metz ‘Ubuntu as a moral theory and human rights in South Africa’ (2011) 11 African Human Rights Law Journal 532.
42 ED Igwe ‘Igbo jurisprudence: A discourse on the nature of punishment in
42. traditional African society’ (2011) 1 Filosofia Theoretica: Journal of African Philosophy, Culture and Religions 119.
46. ‘The Igbo believe in the existence of a supreme being - God whom they call different names according to subculture area groups’ (Igwe (n 42 above) 23). The supreme being sits at the apex of the divine hierarchy which also comprises divinities, spirits and ancestors.
48. As above. The author cites the famous example from Chinua Achebe’s Things fall apart as an illustration. A citizen of Umuofia village, a woman, was killed by the people of Mbano, a neighbouring village. Umuofia demanded compensation in the form of a young man and a virgin, failing which war with the Mbano village would ensue. The virgin was given to the husband of the deceased, and the young man was killed. As the author observes: ‘This takes care of both the human and divine disharmony created by the offence of murder’ (126).
53. Wilson 183. ‘The woman would be regarded as the wife of the dead man; her children would placate his ghost as father ...’
70. As above. The ethical question here presumably is the issue of the moral turpitude of the killing itself.
73. JMT Labuschagne & JA van den Heever ‘Liability arising from the killing of a fellow human being in South African indigenous law’ (1995) 38 Comparative and International Law Journal of Southern Africa 3, to whom I am indebted for most of the citations in this section.
78. MW Prinsloo Inheemse publiekreg in Lebowa (1938), cited in Labuschagne & Van der Heever (n 73 above) 423.
81. ASP Botha Inheemse strafreg in Qwaqwa (1986), cited in Labuschagne & Van den Heever (n 73 above).
83. J Church ‘Murder and culpable homicide’ in AC Myburgh Indigenous criminal law in Bophuthatswana (1980) 78-80.
88. SJ Ndlovu-Gatsheni ‘Inkosi Yinkosi Ngabantu: An interrogation of governance in pre-colonial Africa - The case of the Ndebele of Zimbabwe’ (2008) 20 Southern African Humanities 386.
91. K Wiredu ‘An oral philosophy of personhood: Comments on philosophy and orality’ (2009) 40 Research in African Literatures 15.
96. Among the Zulu, this is captured in the saying ‘all blood belongs to the King’ (Krige (n 59 above) 228). Soga (n 2 above) refers to the chief as ‘proprietor of every individual in the tribe’ (43).
98. See n 138 below. While there is evidence of the practice of group purification, this matter is not in reality straightforward. Between the belief that death itself ‘casts a shadow’ or ‘brings darkness’ over the living, and the religious need to avoid the wrath of the ancestors at all costs, it is sometimes difficult to unravel motives underlying various cleansing rituals, from the washing of hands by funeral attendees before they re-enter the homestead, to the purification of the king as the embodiment of the group, to routine ancestor appeasement ceremonies in times of family misfortune. All these responses to death, in general, tend to cloud the specific issue of homicide as a stain on the group necessitating corporate cleansing. (See, generally, Kuper (n 2 above) 183-187; Wilson (n 4 above); R Lee & M Vaughn ‘Death and dying in the history of Africa since 1800’ (2008) 49 Journal of African History 341.
104. Igwe (n 42 above), quoting HO Anyanwu African traditional religion from the grassroots (1999) 112.
112. The death penalty is invariably invoked, sometimes preceded by torture and, occasionally, involving the slaughter of the whole family and the confiscation of property.
121. C Williams ‘Asante: Human sacrifice or capital punishment? An assessment of the period 1807-1874’ (1988) 21 International Journal of African Historical Studies 441.
123. To cite a non-African example, the Cheyenne of North America had in place a set of restraints against homicide (banishment and the need for ritual cleansing) which were effective because they merged personal punishment with group renewal, thus deterring any hasty resort to killing. KN Llewellyn & E Adamson Hoebel The Cheyenne way: Conflict and case law in primitive jurisprudence (1941) 132-135.
124. My informant on the Zulu explained the extended definition of a witch as anyone who was caught lurking around someone else’s homestead in the dead of night (Zondi interview, Durban, 3 September 2015).
127. Treason had a wide definition, including disobedience to the king and various kinds of disloyalty.
128. As in the case of discovering a ‘witch’ ‘around one’s household at night or catching a thief in the act’ (Zondi interview, Durban 3 September 2015).
129. As above. Methods of execution varied, ranging from clubbing to death to throwing off a cliff. The harshest, involving torture, were reserved for sorcerers; Krige (n 59 above) 227.
130. On the Xhosa, see Soga (n 2 above). It should be remembered that a fine payable to the chief or king indicated that the offence was a crime, not a delict, or both.
133. By providing cattle for lobolo to enable the deceased’s brother or other relative to take a bride, or by providing a young girl as a procreator.
134. For examples of ceremonies of atonement, see Evans-Pritchard (n 8 above) 154; Krige (n 59 above) 228.
136. ‘The failure of the state to take effective measures to identify and hold to account individuals or groups responsible for violations of the right to life can itself be a violation by the state of that right, even more so where there is tolerance of a culture of impunity. Draft General Comment 3 on Article 4 of the African Charter on Human and Peoples’ Rights (The right to life) para 7, commenting on para 3 of the text.
138. GBN Ayittey ‘Traditional institutions and the state of accountability in Africa’ (2010) 77 Social Research 1183-1210; C Ake ‘The unique case of African democracy’ (1993) 69 International Affairs 239-244; Ndlovu-Gatsheni (n 88 above) 375-397; C Ake ‘The African context of human rights’ (1987) 34 Africa Today 5-12; Wiredu (n 91 above); Mbiti (n 91 above).
141. The Zulu description of a king as umlomo ongathethi manga (the mouth that tells no lies) embodies this notion in its implication that words uttered by a king are in any case not his own (Zondi interview, Durban, 3 September 2015).
149. Citing C Ake ‘Rethinking African democracy’ (1991) 21 Journal of Democracy 43-78 and J Cobbah ‘African values and human rights: An African perspective’ (1987) 93 Human Rights Quarterly 309-331.
150. Ndlovu-Gatsheni (n 88 above) 388. He also discusses accountability through patronage, made possible by control of production (390-391). See also Ayittey (n 138 above), who mentions other forms of accountability: regicide; fear of secession; and traditional civil society watchdogs (1196-1200).
152. Whether this is in the form of the Xhosa reluctance to deplete the numbers of the group by capital punishment, or the Swazi notion that people belong to the King. See H Hannum ‘The Butare Colloquium on Human Rights and Development in Francophone Africa’ (1979) Universal Human Rights 65, who asserts that the right to life was much wider in traditional culture, entailing not only a prohibition against killing but also an obligation to maintain needy members of the community (69).
155. Although not exclusively discussing African societies, Diamond (n 5 above) 179 has a useful discussion of infanticide where, among the !Kung of the Kalahari, the decision of whether to kill, eg, a deformed newly born is left to the mother. Interestingly, Diamond points out that no ethical issue attaches to this act, since the !Kung believe that life begins not at birth, but when the child is given a name.
165. Heyns (n 7 above) 4. One can already detect important divergences here between international law and the norms prevalent in traditional society.
166. Diamond (n 5 above) has an illuminating discussion about the evolution of nation states from small-scale origins in which he lists many of the features that changed in the process (10-12).