The Poison in the Ink Bottle: Poison Cases and the Moral Economy of Knowledge in 1930s Equatoria, Sudan

Abstract

Poison cases – in which people are accused of deliberately administering a toxic substance in food or drink or by touch – have long posed a problem for the government judicial system in Central Equatoria. Although poisoning is potentially provable in court on the basis of a material substance, it also transgresses the boundary between occult practice and ‘real’ criminality. The recent revival in the study of African witchcraft has circumvented the question of ‘reality’ to argue that occult discourse is a sophisticated discussion of modernity and its economic inequalities. This approach risks ignoring the longer history of occult thought and practice and its place in prior economies. This article uses colonial records of mass trials of suspected poisoners and a detailed account of a specific accusation, all in 1930s Kajo Kaji, to demonstrate that poison existed as or derived from physical substances with a history of acquisition and utilisation, and embodied historically dynamic social, economic, and gender relations. The argument is that both the longer-term histories and the individual specificities and local realities of occult expressions can be understood in less isolation if approached through the idea of a moral economy of knowledge. Occult discourse and practice was fundamentally connected to the differentiated introduction or possession of foreign or specialist kinds of knowledge. The article looks at the sources and exchanges of knowledge and materials that formed the deeper historical context for the practice or belief in poison in Kajo Kaji, before turning to the actual cases of the 1930s. The latter reveal how claims and accusations could be wielded as tools of resistance and contestation, as changes in local authority and socio-economic relations were being worked out.

The Chief realised it was not expedient for talk of poison to reach the ears of Government. … It was, he said, very hard. Karibbe had only just come out of prison, the woman had run away from him, he had had great provocation. It would be wiser to await the event of this allegation of magic. Let her brothers keep the suspect bottle (it was a squat ink bottle which had been thrown away up at the police office) and if anything bad came to Gune within the next two moons, then Karibbe should be held responsible. …

Every misfortune has its author, and sudden death, however natural, is always attributed to some human malice, a look, a touch, a word, and lastly poison. Throughout the forest community there is heard a stealthy susurrus of conjecture and suspicion, which, multiplied by chance or mere coincidence, one day bursts out into a surge of panic. In that day murderous torture is abroad.1

‘Poison’ cases have long posed a problem for the government judicial system in Equatoria, as this extract from a detailed account by a British colonial official of one case in Kajo Kaji in 1939 reveals. More recently during the civil wars in Southern Sudan, military personnel or community authorities have practised ‘murderous torture’ on alleged poisoners in Equatoria. In an attempt to overcome the potential for injustice, the rebel Sudan People's Liberation Movement/Army (SPLM/A) encouraged the courts in the ‘liberated areas’ instead to force the accused to swallow the poisonous substance allegedly found in their possession by the accusers, which echoes poison ordeal trials of an earlier period. From a Western legal perspective, poison potentially differs from ‘witchcraft’ in that it appears to be more easily provable in court on the scientific basis of a harmful material substance and witnesses to its administration; often insecticide is produced in such cases, while an overturned fertilizer lorry has rendered one village notorious for poisoning. The belief that certain people have caused illness or death by deliberately administering a toxic substance in food or drink or by touch is virtually universal in this region of Southern Sudan. An account from 1924 described nyania poison in Kajo Kaji as an exclusively female practice, and detailed the actual extraction, preparation and administration of snake venom. But, as the colonial official who wrote up the 1939 case suggested, poison also appears to belong in the realm of ‘magic’ in terms of the nature of accusation, the more common absence of a certifiably toxic substance (let alone proof as to its administration) and the idea that certain families or individuals are born with the propensity to poison. In struggling to differentiate actual criminal substances from ‘superstition’, colonial officials were also unwittingly touching on very real local debates about the attractions of material substances and the dubious – and gendered – morality of the specialised knowledge that developed or deployed them.

Poison thus transgresses the boundary between ‘occult’ practice and evidential criminality. The question of the ‘reality’ of occult practice has long troubled academic approaches to African witchcraft, which have often sought to circumvent the problem by examining the phenomenon of accusations rather than practice. The recently much revived study of the topic has treated witchcraft belief instead as a discourse on political and economic inequalities, and argued for its relevance as a sophisticated and dynamic discussion of modernity.2 The problem with this approach is that it implies that witchcraft discourse is largely a response to, or ‘gauge’ of ‘the impact of global cultural and economic forces’ upon African societies.3 Not surprisingly, given the inaccessibility of earlier ideas about witchcraft, such studies have rarely explored the historical context to these ‘modern’ witchcraft discourses; analyses of the dynamism of occult discourses have largely relied on contemporary accounts to explore ‘memories’ of earlier globalisation in the form of the slave trade, or to recall occult rumours of the colonial period, as ‘new imaginings for new relationships’.4

This article demonstrates the continuing importance of colonial sources to historicising the occult in Africa. It argues that the poison cases of the 1930s in Kajo Kaji reveal the complexity of local ideas about the potential for using material substances for harmful ends, which defy singular explanations or categories. It will be shown that poison existed as or derived from physical substances with a history of acquisition and utilisation, and embodied historically dynamic social, economic and gender relations. The concept of poison also absorbed and expressed the changing economy and socio-political landscape of the colonial period, but not without precedent, and not necessarily in a consistent form. Colonial officials inevitably conflated ‘poison’ as a category, but the 1939 case was very different from mass trials a few years earlier of women accused of poisoning, which had created the impression of a moral ‘panic’.

The argument here is that both the longer-term histories and the individual specificities and local realities of occult expressions can be understood in less isolation if approached through the idea of a moral economy of knowledge. Witchcraft, as we have seen, has increasingly been interpreted as a signifier of (im)moral economies, and particularly of the inequalities and exploitations of production and consumption that have characterised the articulation of African communities with global capitalist markets.5 Knowledge, on the other hand, has been shown to be a key resource in older political economies of pre-colonial patronage and survival.6 The former studies risk depicting witchcraft as only an extractive, consumptive force, as it appeared most obviously in later twentieth-century Africa, ignoring its more ambivalent and even necessary relations to power and production.7 The latter on the other hand treat knowledge as a positive resource without exploring its moral ambiguity and the debates over its composition within a community.8 If the two approaches are integrated however, it is possible to adopt a more historical and a less categorising approach to witchcraft, by seeing it as part of a continuum of ways of discussing knowledge and its practices and productions.

Austen's connection of African conceptions of witches to a world in which production of new wealth depends on appropriating the scarce reproductive resources of others, while collaborating with an arbitrary and destructive external power, is highly relevant to the poison cases in 1930s Kajo Kaji. But it is important to include knowledge in the equation, because it governed the ‘conditions of access to material resources’.9 A changing moral economy was privileging certain kinds of specialist knowledge, at the same time as overall reproduction, both human and agricultural, was struggling in the face of diseases, population density and nutritional deficiency.

The article begins with a brief discussion of Evans-Pritchard's famous categories of witchcraft and sorcery, their relationship to knowledge and how this applies to poison. It revisits the discredited colonial theory of the foreign origin of certain occult activities in Sudan, not to argue that the activities were indeed a direct alien import into discrete tribal societies, but rather to argue that occult discourse and practice was fundamentally connected to the differentiated introduction or possession of foreign or specialist kinds of knowledge. It then looks at the sources and exchanges of knowledge and materials that formed the deeper historical context for the practice or belief in poison in Kajo Kaji, before turning to the actual cases of the 1930s. First, the mass trials of women are discussed; while linked to crisis and moral panic, they also reveal a complexity of resistance and domination as changes in local authority and political economy were being worked out. The detailed account of the 1939 case cited above will then be more fully explored to show how colonial officials conflated different concepts into the category of poison, and how the latter could be adapted to incorporate new symbols of the changing economy. Threats or accusations of poison could be variously wielded by men or women, elders or young men, chiefs or community as part of struggles over wealth and authority in the colonial period. But this should also be seen in the context of a historical and ongoing process by which authorities, elders and established experts sought to build and maintain their vision of the community and the body of knowledge contained within it, in the face of continuous incursions brought from outside by new wives, younger men, migrating newcomers and trading partners, as well as foreign armies, traders and governments.

Gender, Witchcraft and Sorcery

Kajo Kaji is a high plateau above the west bank of the Nile close to the Ugandan border. The 1930s cases would give its Kuku population a notoriety for poison which has remained up to this day in the Central Equatoria region, although poison cases now reach the courts more widely. Suspicion is not confined to women, but it is believed that certain women have, and pass on to their daughters, a propensity to poison which is inborn, involuntary and indiscriminate. This was also asserted in an account published in Sudan Notes and Records in 1924 by a Syrian medical officer, Negib Yunis, who had served in the Kajo Kaji area.10 He in turn was informed about cases of ‘poisoning’ or nyania by a minor Egyptian or Northern Sudanese official, Ahmed Effendi Abdel Alim, who would have had his own preconceptions of kujur or magical practice, and of women, in Southern Sudan. This publication no doubt influenced subsequent colonial accounts, which largely accepted that women at least intended to poison, although with an oxymoronic caution that ‘it is probably certain that some poisoning has been going on’.11

The official who reported the 1939 case, Yei District Commissioner (DC) Tracey, declared that ‘every woman is a potential poisoner, but magic lies within the powers of few’.12 Comaroff and Comaroff, and others, adopt a gendered version of Evans-Pritchard's categories of ‘witchcraft’ and ‘sorcery’ to argue that the latter, the intentional use of material substances, tools or learnt rituals, is usually a male preserve, while the innate capacity to bewitch is a reflection of the ambivalence inherent in being female.13 This might appear to fit the female poisoning in Kajo Kaji. But there is a danger of oversimplified categorisation. The real distinction in Evans-Pritchard's account of the Azande in Western Equatoria appears to have been between ‘witchcraft’, as the inherited and intimate knowledge objectified within the human body and capable of harm by its location in the social body, and ‘magic’ as knowledge acquired from outside or from experts and objectified as substances and rituals, which could be used harmfully in ‘sorcery’. As one commentator put it, ‘[f]irst and foremost, this was a book about the sociology of knowledge’.14 Witchcraft was an inversion within the community body of familiar, intimate knowledge, whereas sorcery drew on ambiguous knowledge from outside that body.15 Middleton explained the greater fear felt by Lugbara (neighbouring Kajo Kaji) towards ‘sorcerers’ because of their ‘outside’ status, but also that outside magic and medicine conversely had a greater status and attraction.16

Middleton also categorised enyanya poisoning among the Lugbara as a form of ‘sorcery’ practised by both men and women.17 Poison reveals that the categories of witchcraft and sorcery could be more overlapping and complex than has often been recognised in their application; occult activity may combine intimate and foreign knowledge, inborn propensity and deliberate intention. Kuku poisoning did involve an inherited female propensity more akin to ‘witchcraft’, but according to Tracey, Kuku women learnt to poison after hearing ‘tales about the Relli, a tribe away to the south, notorious for their poisonous sorceries’.18 As we shall see, the 1939 case of poison in Kajo Kaji was less concerned with women than with the access of certain men to outside or new kinds of knowledge. So to define poison as either inborn ‘witchcraft’ or as ‘sorcery’ would oversimplify a concept which related to the complexity of the introduction, exchange and moral control both of forms of knowledge and of actual substances or goods.

Johnson has criticised the tendency of colonial officials in Sudan to depict ritual and mantic activities as alien imports which were disrupting the real ‘tribal’ authorities, i.e. government-recognised chiefs.19 But though accurate, his argument may obscure the extent to which the exchange of new and outside forms of knowledge was being discussed through occult discourses. The simultaneous value and ambiguity of such knowledge is apparent in Evans-Pritchard's description of Zande magic: ‘[m]ost Zande medicines come to them from foreign peoples, and Azande believe that foreigners know much more about magic than they do’, and at the same time, ‘people knowing nothing about them [these new medicines] are afraid of them … who can say what the qualities of Baka, Bongo, Mundu and Madi medicines may be?’20 Johnson also criticises Evans-Pritchard's, and colonial officials’ assumption that the Azande ‘closed associations’ were a novel and foreign introduction, arguing instead that they originated before colonial rule, in the need felt by the people for protection from the ruling Avungara clan.21 But Evans-Pritchard's limited account of the Mani association suggests that it was not only about protection, but was also a mechanism for managing and diffusing new kinds of knowledge. An individual would buy a new medicine from ‘foreigners’ and briefly make a profit by attracting new members to his/her association, until the medicine became more widely diffused and hence less sought after.22

There is of course a danger in using the word ‘foreign’ that it may imply that societies were normally closed, living as discrete ethnic groups, unless disrupted by sudden incursions. Allen discusses the more subtle and dynamic Madi and Acholi distinction between what is inside the moral community of close, everyday relationships, and ambiguous outside forces.23 It is in this sense that the word ‘foreign’ is used: not as an ethnic distinction but as an indication of knowledge whose nature was – initially – unknown and outside the existing body of knowledge held within a community. There is a sense in much of the ‘occult’ discourse discussed here of an ongoing need to regulate and evaluate outside knowledge and to define whether it can be incorporated into what James calls the ‘cultural archive’ of ‘moral knowledge’ that defines communities.24

Arguably the advent of international trade and colonial government saw a general increase of ‘sorcery’, because new kinds of commoditised knowledge gained unprecedented availability and could be accessed and owned by anybody. As a result there may have been a shift from the prevalence of ‘witchcraft’ as an almost universal human attribute, to a concern with a distinct and often intolerable category closer to ‘sorcery’, perhaps as the balance of access to outside knowledge and medicine began to swing out of the control of community authorities and elders. Evans-Pritchard reported an increasing availability and individual use of ‘magic’ as former royal ways of dealing with witchcraft were devalued.25 So assertions by chiefs and others about the harsh torture and execution of poisoners or witches in the past in Kajo Kaji (accepted by Yunis) instead reflected current concerns about a loss or lack of collective control over individual access to ‘outside’ knowledge, i.e. an increase of ‘sorcery’.26 But as Johnson argues, it should not be assumed that institutions for dealing with occult practice were something new in the colonial period. The need to manage and regulate the incorporation or monopolisation of knowledge was historically part of the moral economy governing communities, not simply a response to the changing economy of the colonial period.

The Ambiguity of Knowledge as a Resource

Knowledge was a concrete economic resource as well as a cultural and moral one. Anderson and Johnson emphasise the historical importance of networks of diversified economic specialisations to the survival of communities in north-east Africa.27 This is apparent in the history of central Equatoria, which was already involved in long-distance trade networks by the time of the first Egyptian expedition to it in 1841. Knowledge was one of the ‘goods’ exchanged, most obvious in the wide-ranging influence of the Bari rain priests or kings on the Nile. Rain-making in the communities west of the Nile, including Kajo Kaji, is said to have come from the Bari.28 A great variety of experts sold their services as healers, diviners, craftsmen, blacksmiths, and as specialists in protection, revenge, hunting, fishing, cultivation, warfare and so on. But specialist knowledge frequently carried a moral ambiguity, especially if it originated, or remained located, outside or on the margins of the community.

It is from this ambiguity that the attributed ‘foreign’ origin of poison is likely to have derived. The ethnic geography mapped out by colonial governments and the colonial boundaries imposed naturally would not do justice to the history of successive migrations and the economic, social and cultural exchanges between communities in this region. The Kuku were known by colonial officials for their intensive cultivation, using a heavy iron hoe which played a central part in bridewealth payments, along with other iron implements like spears and arrows. ‘Madi’ country to the south was the main source of the iron ore and smithies for Kuku hoes, along with the small community north of Kajo Kaji known as the Nyepo, who are said to be blacksmiths originally from the Bari area. The Nyepo also brought knowledge of rain-making from the Bari, embodied in the rain clan called the ‘Relli’. But DC Tracey's claim that Kuku women learnt poison from the ‘Relli … to the south’ must have indicated the small pocket of Bari-speaking people located in the middle of the Madi-speakers south of Dufile in Uganda, also known as Relli.29 Yet the term nyania or anyanya is a Madi and Lugbara word, the Kuku word for poison being kishum or kisum.30 Despite speaking a Bari dialect, the Kuku were reported to be engaged in closer relations with the Madi to the south and east than with the Bari to the north.31

So while the linguistic term for poison may have been directly appropriated from Madi-speakers to the south, its attributed origin with the ‘Relli’ indicates the ambiguities of relationships between successive groups of migrants from the northern Bari-speaking areas. The second half of the nineteenth century saw the influx of foreign or northern ivory and slave traders into Equatoria. The Kajo Kaji hills may have offered some protection, and certainly lay east of the main slave trade route through Yei. However, from 1861, the area began to be raided for cattle and slaves, and the later movements of the Nubian troops of the Turco-Egyptian government under Emin Pasha and Fadl el Mula reportedly had a particularly harsh impact on the Kajo Kaji area. Patterns of raiding continued into the Belgian period, and the population on the plateau was swelled further by people driven from the north and west by raids.32 It is likely in such a situation that the landowning and rain-priest authorities in Kajo Kaji may have been seeking to forge and maintain a coherent community and defend its values against the influx of ‘strangers’ and migrants.

But there were also important loci of specialised skills and knowledge that lay outside that community, and which had an ambiguous but necessary value. Blacksmiths were universally associated with amoral values and occult abilities, as was said by one Kuku informant recently of the Nyepu: ‘because of their specialised skill … other people think that they live less strictly in moral rules’.33 Larick highlights the role of blacksmiths as brokers of foreign influences, incorporating new styles into their craft.34 Iron also played an important ritual and judicial role, being used in oath-taking. Even the early colonial administrators in Kajo Kaji sought to appropriate its value, keeping the iron rod of a prominent Madi diviner, Tombe Gborron, in the district office to swear upon before giving evidence.35

While ordeal trials and iron-working were male domains of knowledge, Madi women monopolised the processing of red ochre earth in the iron ore areas, which was highly valued both as decoration and medicinal ointment. During the colonial period, as iron-working competed with new imported iron, and as colonial governments sought to undermine the independent authority of diviners, the main market trade between Kuku and Madi involved the exchange of Kuku grain for Madi ochre, which was reportedly conducted mainly by women.36 So specialist knowledge of iron and ochre was an example of a material and moral commodity, whose exchange may have contributed to ambiguous relations between proximate communities, which could be aptly expressed in the belief that poison had also been brought into the community through those relations, particularly if they became gendered.

Poison did of course have a prior or alternative existence as a valuable commodity, used both in hunting and as a judicial tool. Evans-Pritchard attributed the Zande receptivity to foreign influences to their regular long-distance travels to obtain poison for their oracles (usually administered to chickens) from dense Congo forest.37 In 1932 the Kajo Kaji chief, Yenge, called Ajuke, a Madi expert from the east bank of the Nile, to conduct a poison ordeal trial of women.38 Poison thus played a revelatory role which demonstrates how specialist knowledge and substances brought from outside could be productive of knowledge within the community, as a means of divination and investigation.

Poison was also productive in its use for hunting. The Moru in northern Equatoria combined the two functions, making ordeal poison from poisoned arrows. The Nyepo of Kajo Kaji were known as great hunters as well as smiths because of their ‘special spears’. Iron and poison thus had interrelated functions, and hunters shared with blacksmiths an ambiguous moral status. Bari hunters were not fully incorporated into village life: ‘their sphere is in the forest’; while Kuku blacksmiths also lived ‘outside the village’.39 Iron-working and hunting played similarly transformational roles that linked them both to death and to fertility, and both took place outside the moral community in the ‘forest’.40

Yet the forest or bush was also the source of fertility, foods, medicines and divinity: the Rudu sacred groves where rain priests were buried in Kajo Kaji were located in the thickest undergrowth. The issue was one of access to the ‘forest’: only the authorised priests could enter Rudu; anyone else would become infertile.41 The forest was also a source of poison: both the hunting and fishing poisons, and the ordeal poisons used by Ajuke were reported to come from trees.42 Lugbara enyanya poison was usually a resin that could either be bought or collected ‘in the bush’; but Middleton was informed that ‘women prefer to use snakes to resin’.43 Yunis emphasised that Kuku women poisoners had to go into the ‘forest’ to trap snakes to extract venom.44 Snakes were widely associated with divinity, including by the Madi, especially if they lived in sacred groves. In 1919 the nearby Yakan protective ‘cult’ had used a snake oracle, and associated snakes with the maternal line and particularly the grandmother.45 So Kuku nyanya poisoning may have evoked and generated a fear of female access to knowledge in the forest and knowledge embodied in the female line. It also involved the inversion both of snakes as sacred and of poison as a male-controlled commodity for hunting and trials.46

More generally, women were a source of new kinds of knowledge incorporated into patrilineal societies through marriage.47 The Kuku story of a female dog who taught the ancestors how to eat grain, hunt and give birth properly illustrates the value of alien female knowledge to the fertility of the community.48 Allen suggests that one reason why Madi women, especially new wives, would be suspected of poison or witchcraft is because they came from outside their husband's community, bringing unknown quantities. Yet they are also ‘most likely to be sought as diviners and healers when abnormal things are happening’ (such as to deal w ith an increase of ghosts of foreign strangers killed in the 1980s).49 Marriage could act as another mechanism for foreign knowledge being incorporated and contained within kin relations; the danger was usually perceived to come from recently married or widowed women who still had an ambiguous or marginal status.

New Loci of Foreign Knowledge

As well as posing a threat, the influx of ivory and slave traders and the Egyptian army created new centres and opportunities during the nineteenth century. Trading, and later government, stations were established close to Kajo Kaji from 1861, populated by soldiers and slaves from both northern and southern Sudan who came to be known as Nubis.50 This led to the privileging of certain kinds of knowledge, notably the ability to speak Arabic, as local intermediaries attempted to mediate the harsh and extractive relations with the traders and soldiers. Many of these mediators went on to be recognised as ‘chiefs’ in the colonial period. Arabic language and Islamic or Nubi culture spread around the stations, but in the early colonial period it became restricted by British policy to the small towns along the Nile or district headquarters.

The armies and stations had been both an extractive threat and a source of employment or security for those who joined them (whether voluntarily or as slaves) as soldiers, servants or translators. The young men of Kajo Kaji appear to have actively sought to join: under the Belgians, Kukus were reported to ‘supply most of the troops of the Enclave’.51 As Larick writes of the Kenyan Lokop, it was generally the young warrior class who acquired ‘the foreign material evidence for worldliness’ through rustling or warfare, and later through colonial armies, education and urban employment.52 The police in Kajo Kaji by the 1930s had mostly served as soldiers in the Equatorial Corps, including the police sergeant who recorded the details of the 1939 case. So complicated sets of relations and access to foreign experiences and knowledge had developed in the region since the mid-nineteenth century, and no doubt generated rival claims to authority between prior owners of knowledge, like the rain priests, and the younger generations who had adopted guns and Nubi culture and knew how to mediate with the foreigners.

The populations of the stations also expanded as the traders and soldiers acquired local wives, reflecting the way that women could, through marriage, access new kinds of knowledge.53 Later in the early 1920s, the Kajo Kaji chiefs petitioned the colonial government to deal with the women who had been married and abandoned by foreign officials, calling for their enforced prostitution in the merkaz or government station. This was clearly an attempt to control the influx of new knowledge and experiences that could ‘upset the morals of the other women of the district’, by confining these women to the foreign domain of the town or station.54

Colonial Courts and Female Poison

Kajo Kaji became part of the ‘Lado Enclave’, occupied from 1896 by the Belgian Congo government until King Leopold II's death in 1909. It was then incorporated by the Anglo-Egyptian Condominium government of the Sudan into Mongalla Province, later Equatoria Province, and by 1935 it was a sub-district of Yei District. With all the attendant complications of inventing traditions, British administrators had sought to identify ‘tribes’ and ‘chiefs’ in a region whose recent history was shaped by migrations and forced movements, and characterised by fluid and multiple political authority and social affiliation. The first colonial ‘chiefs’ in Kajo Kaji came from the cattle-wealthy Kasura clan of rain priests.55 Chiefs’ courts, established on an ad hoc basis by individual British District Commissioners in the 1920s had been given a more formal status under the 1931 Chiefs Courts Ordinance.56 Captain Stigand was the first British administrator of Kajo Kaji and lived there for some years. Before his death in 1919, he wrote a detailed description of the former Lado Enclave, in which he made no mention of poison.57 While this may reflect simply an absence of information, it suggests that poison accusations gained prominence quite suddenly by the time of Yunis’ account in 1924, when cases of poisoning were being heard and punished in the government courts.

There is sometimes a tendency to generalise about the colonial government suppression of all witchcraft detection practices.58 As Waller shows, there were more complex struggles to reconcile colonial law with administrative concerns, which variously sought to suppress occult activity, or to undermine the witch-finders, depending on which appeared the greatest threat or rival to the government.59 In the case of Kajo Kaji, colonial officials clearly believed that women did practise, or deliberately intended, poisoning. This is unsurprising in the context of the belief among officials that native women were ‘whores at heart’, embodied in the early unofficial court regulations of the 1920s, which sought to prevent divorce and consequent social breakdown.60 Female poisoners were reported to poison only the ‘menfolk’, using this threat in the ‘battle of the sexes’.61 Officials puzzled by their apparent willingness to confess to poisoning recognised that, ‘the power wielded by a reputed poisoner is a bait she cannot resist’.62 Colonial officials themselves were thus interpreting the paradox of women's proud admissions that they were poisoners as a form of resistance to male control, a struggle to which their own patriarchal court regulations were contributing a new factor.

However, the discomfort of British officials about the evidence in poison cases was exacerbated by episodes in Kajo Kaji in 1932 and 1937. In 1932, the chiefs called a Madi expert to conduct a poison ordeal trial in which four women died; the ‘poisoner-hunter’ was later imprisoned for manslaughter.63 In 1937 the (mostly new) chiefs used physical torture to try to obtain confessions of poisoning from over two hundred women. The Province Diary asserted that ‘[t]here is no doubt that the public panic has little foundation in direct evidence and probably little in fact’; although it also acknowledged the possibility of a few cases of actual poisoning. A heavy collective fine was imposed on the community and the government afterwards publicised its ban on torture and ordeals, to which the Kajo Kaji chiefs publicly voiced their assent.64 We have seen that the belief or claim that certain women used snake venom to poison others reflected a concern about the access of women to outside forms of knowledge, brought with them in marriage, gained in the forest or passed from mother to daughter. It also provided women with a potential source of power. The mass trials in 1932 and 1937 suggest that gendered relations of dominance and resistance were being worked out in new arenas: the chiefs’ courts, which we will return to below. But the women singled out were also made the scapegoats for a wider sense of crisis and loss of fertility.

Crisis and Panic

The CMS missionary in Kajo Kaji played an interesting role in the 1937 crisis; he was the one who initially sent a tin of suspected poison to the DC in Yei, which, perhaps by appearing to lend official credibility to suspicions, sparked off the ‘mass hysteria’. The missionary then also persuaded the DC to ameliorate the fine and punishment because the chiefs had acted according to ‘tribal custom’. His accounts also emphasised the suffering of the Kajo Kaji population at this time, as a result of a new clan taxation policy and sleeping sickness regulations, both of which necessitated the forced relocation of settlements (including burning of huts), accompanied by successive droughts. He thus contributed to the colonial impression of the poison trials as a mass ‘panic’ in response to crisis, as well as to the belief that it was ‘tribal custom’ to hold such trials.65

The reported crisis also had roots however in the impact of colonial rule upon the wider social and economic relations in the region. Concerns over the spread of sleeping sickness into Sudan, together with administrative confusions over the ethnic geography, had led to the forced relocation of settlements away from the 1915 Uganda boundary, which sought to divide the ‘mixed’ Kuku and Madi population by ‘transplanting’ villages.66 Cross-border movement was entirely forbidden until a pass system was introduced in the 1930s and, while never entirely effective, interaction must have nevertheless been reduced. Formerly close neighbours may have increasingly become strangers; some disputes may have gone unresolved. Ethnic identities were forming in new ways in the colonial period generally, and the sudden isolation of a ‘Kuku’ territory and people may have hardened the distinction between what was inside and outside the community.67 Cross-border movement increasingly had to take place under the cover of dark, which contravened the normal dislike of ‘night-walking’ associated with occult activities.68

Kuku villages were forced to resettle away from the boundary and streams, onto the Kajo Kaji plateau, increasing the already dense population there.69 The concentrated villagisation, as elsewhere, was likely to increase witchcraft accusations, because it created unprecedented intimacy and made economic disparities more apparent.70 The sleeping sickness campaign also involved unpopular examinations of neck glands and lumbar punctures, and infected individuals were removed to special camps, where in the early days they were treated with atoxyl injections that could cause blindness or death.71 In addition, Kajo Kaji suffered heavily from cerebro-spinal meningitis, especially in the 1930s, which prompted another unpopular medical campaign of lumbar punctures and injections, which were reportedly held responsible for the high mortality, and connected to poison.72 Poison was all about something foreign being ingested or implanted in the body, making it an obvious metaphor, rooted in local practice, for the colonial medical impositions.73

‘A Case of Poison’, 1939

The mass trials of women might appear to fit long-standing interpretations of witchcraft accusations as a reaction to crisis and an example of the assertion of power by elders and chiefs who were able to take advantage of the colonial courts.74 But the dynamics of authority in 1930s Kajo Kaji were more complicated than this, as revealed in the subsequent case from 1939 written up at length and in flowery prose by DC Tracey.75 In the absence of any surviving court records, it provides a rare source of the detail behind a case, based on the evidence collected by the local police sergeant over the previous week. It is summarised here:

A retired policeman, Karibbe, had married a fourth and very young wife, Gune. Despite having one son from his elder wife, it became apparent that he was now sterile and, in order to pay a local doctor to cure him, he traded tobacco in Uganda and was imprisoned for breaking sleeping sickness regulations by failing to obtain the necessary border pass. In the meantime, Gune met and eventually eloped with a younger, mission-educated man, Paulo. Her male relatives, including Chief Abina, opposed this because Paulo could not afford as high a bridewealth amount as Karibbe had paid, and was now owed back, but Gune's female relatives supported her and she was not forced to return to Karibbe. Karibbe on his release consulted the doctor – a quasi-Islamic practitioner known as the Khalifa or Moallim – and one night went to Gune's hut and painted some kind of potion on her body from an ink-bottle. She accused him of trying to poison her with chinyi, and the next day Karibbe was taken to court. The chiefs, keen not to attract Government attention, decided that they would wait to see if Gune showed any ill-effects from the magic.

Soon afterwards, Karibbe was attacked by his other wife's relatives, now suspicious of him, and had to be hospitalised. His hospital report reached the D.C., who ordered the attackers to come before the chiefs’ court. Chief Tete ensured again that the word poison was not mentioned in the record, and later Gune was granted a divorce, aided by the continuing rumblings about poison. During the case her brothers produced the ink bottle, though Karibbe dramatically demonstrated his innocence by drinking the contents himself. Paulo was unusually allowed to pay the bridewealth in cash rather than cattle. But, while he was away working in Juba town, the pregnant Gune claimed that Karibbe and the Khalifa had used occult means to make her return to Karibbe. Paulo demanded his bridewealth back, and the case went to court yet again, to the dismay of Chief Tete: ‘the case was a dangerous and persistent nuisance’. Gune was remarried to Karibbe, but she haemorrhaged during the birth of Paulo's child, and died, despite the attentions of the Khalifa. Her brothers accused Karibbe and the Khalifa of poisoning her and Paulo went to the local missionary, who informed the government. The case was finally heard by the chiefs in the presence of the D.C., but the court members were reluctant to pass sentence, particularly on the Khalifa. The D.C. imprisoned Karibbe for six months ‘because he had frightened his wife with a pretence of magic’, and exiled the Khalifa from the district for taking ‘money from simple folk by his fraudulent talk of sorcery’. He condemned both for disturbing the peace by ‘tampering with the powers of darkness’, whilst also emphasising that such powers did not really exist and that Gune had died of natural causes.

This story reveals a number of different factors and debates which found their expression in the versatile concept of poison, concerning reproduction and the impact of migrant labour, money, writing and Christian missions on the moral economy.

Poison and Fertility

On the basis of the story of Karibbe, it would be tempting to define chinyi poison in relation to venereal disease: ‘a liquid which, applied externally, raises fierce blisters; applied internally the flame of lust’. Tracey ascribed Karibbe's sterility to gonorrhoea picked up while in Uganda.76 But venereal disease had a longer history in Kajo Kaji, as elsewhere, in the activities of the northern or slave soldiers of the Egyptian army or traders. In Bunyoro in Uganda, Doyle highlights ‘an increasing prevalence of venereal disease in the late nineteenth century, most probably caused by traders and soldiers introducing new strains of syphilis and gonorrhoea’.77 By the 1920s, Yunis reported that male infertility was the main reason given by women for divorcing their husbands; this tendency to divorce in turn contributed to the further spread of disease, as well as to the sense of diminishing male control over women. At the same time female nyanya poisoning was reportedly passed from mother to daughter, with the threat that if a daughter refused to practise she would become infertile. So it also involved fears about women taking control of their own fertility.78

In the face of new or increased disease, people also sought new knowledge to treat it, as Karibbe's decision to consult the expensive Khalifa showed. The Khalifa was also associated with the Nubian troops who may have brought the very disease he was treating, having reportedly been a ‘waif’ in the Egyptian army during the re-conquest and adopted Islamic practices while living in the towns along the Nile. A Moru doctor in northern Equatoria told Evans-Pritchard that his expertise originated in a new medicine brought from people in the west, apparently used to treat syphilis. In the symbiotic relationship attributed to disease and cure, doctors usually began practising after having developed sores themselves, and were also capable of inflicting the disease on others.79

Fertility was also diminished by food shortages as well as by diseases. Tracey attributed Gune's death to her narrow frame, which was likely to have resulted from protein malnutrition.80 Stigand had reported that sheep and goats ‘did badly’ in Kajo Kaji, and also that milk was not a staple of the Kuku diet. Cattle losses from raiding and rinderpest in the late nineteenth century had been partially recovered in the early colonial period, by buying ivory from the Lugbare and Kakwa to the west with sheep and goats, and then selling the ivory to traders from the south for cattle. In 1914 however, a new epidemic killed half the Kuku herds, so that the slow process of rebuilding had to begin again.81 The density of population on the plateau had also led to increasing problems of erosion and chronic food shortages by the late 1930s.82 It was therefore very apt of the Khalifa to tell Karibbe to use the gugu (grain storage) pole from his original ancestral homestead in rituals to cure his own infertility: a clear statement of the negative effects of the sleeping sickness relocations on agricultural and human fertility.

For Austen, the defining feature of African witchcraft is its relation to scarce reproductive resources, in contrast to European witchcraft.83 The idea that women passed poison onto their daughters in order to prevent sterility was an expression of the way that an individual woman might use privileged knowledge to guarantee her own fertility. Similarly Karibbe's pursuit of individual reproduction was perceived to have cost the life of Gune. Indeed the practice of polygamy, dependent on wealth, made visible this pursuit and increased the spread of venereal disease. It also inflated bridewealth, making it more difficult for younger men to marry, adding to low fertility.84 Poison in this sense was a powerful explanation of the cost to the overall fertility of the community of a selfish recourse to outside knowledge and wealth. Those who went out from the community to marry foreigners or to work for them accessed foreign knowledge, but could also bring back disease.

Migrant Labour

What happened was that many people were going there to work in sugar or cotton, they got their money, came back and settled, got married. It is during this migration, and interaction with poisoning in Uganda, especially among men, that some men also acquired this practice and brought it back.85

Middleton similarly reported the introduction of new elojua poison into Lugbara, bought by labour migrants for cash from Congolese migrants in southern Uganda.86 From the early 1920s, labour migration was beginning on a significant scale from northern Uganda and the Sudan borders to the sugar, cotton and sisal plantations further south, in order to pay taxes in cash.87 Migrant labour evoked and generated another occult fear, that in dusk or darkness, ‘European, Arab or even Indian cannibals’ would come across the Uganda or Congo frontiers in saloon cars ‘and carry off the local people for the sake of their meat’.88 This referred both to the Indian recruiters for the Ugandan plantations who regularly visited the district, and to memories of the slave traders.89 But poison was less concerned with the extraction of the young workers than with what they might bring back with them. Historically affected less by slave raids and more by disease, human and bovine, the primary concern on the Kajo Kaji plateau seems to have been to keep out harmful incursions.

Poison could also, like other occult discourses, reflect mutual mistrust between rural communities and migrant workers in the towns. While the Kajo Kaji communities blamed migrant labour for the introduction of poison, migrant workers discussed poison as a feature of the home community. In 1937, near Masindi in Uganda, Col. Lilley met some Kuku youths who had been three years in Uganda, who referred to Kajo Kaji as ‘the place in which you could never be sure of surviving any meal’.90 This in turn reflected the mutual resentments and struggles over the moral economy that wage labour generated, and the belief that each was ‘feeding off’ – or in this case poisoning the food of – the other.91

Paper, Cash and the Moral Economy

The colonial economy was thus generating new opportunities for earning wealth, embodied in the new medium of cash, and which required and imparted new kinds of knowledge. The alleged poison in the ink bottle from the police station was a powerful symbol of such knowledge, and of the relative wealth that Karibbe had earned from it. The Khalifa was known to require payment in cash, and he used written Islamic charms in his practice, probably washing off the ink words into a pot to make a medicine.92 Papers and ink would have been first encountered in the hands of the Turco-Egyptian government officials, soldiers and traders in the nineteenth century. By the 1930s, Christian missionaries were teaching basic vernacular and then English literacy, and gradually a new generation of semi-literate chiefs were succeeding their fathers or uncles.

The deep historical memories and the ambiguity of ink and paper in South Sudan have been demonstrated by Hutchinson's study of the Nuer in the 1980s. Associated with the turuk – the generic name for the foreign personnel of government and towns – knowledge of paper conveyed both power and mistrust. Above all it was seen as a knowledge that was located outside the rural communities.93 Not surprising then, that Karibbe's ink bottle should have been such a suspect object. The police, the missions, the hospital were all linked to the hakuma, the distant and alien government, whose knowledge and power were embodied in the ‘mystery of writing’. This phrase was used by Tracey in reference to the Kajo Kaji court clerk, whom Karibbe knew well enough to avoid having to bribe him for a border pass.94 Like other forms of specialist knowledge, writing could bring personal profit to the chiefs, clerks and police who were able to use it to their advantage. But Karibbe's connection to the knowledge and networks of the hakuma also made his status within the community ambiguous. Similarly the police sergeant who wrote down his case in detail was a Nyepo – the small ethnic group associated with blacksmithing – and said to be the main patron of the Khalifa, who also claimed to be a Nyepo. The central role played in the changing political economy by people with a marginal or ambiguous status was stimulating debates about authority, which we will come to shortly.

More generally, it was younger men (and gradually younger women too) who were associated with education and government. Migrant workers also returned from working in Uganda with new clothes and new Christian names. Older people blamed education and migration for the disrespectful behaviour of younger people. There was reportedly particular disappointment that cash wages disappeared into clothes or beer, so that the workers brought home little benefit to their families. Growing enthusiasm among parents for education as a route to waged employment was thus countered by concern about the often unproductive and potentially harmful nature of the new knowledge.95 It earned new kinds of wealth outside of social and kin relations, and required new discourses about the moral economy of knowledge. Cash was thus as much a symbol of new knowledge as the poison in the ink bottle, particularly as it was largely available only to younger people who could undertake waged labour, in the general absence of profitable cash crops.

Younger people were also challenging elders over marriage and bridewealth, using elopement to make their own choices. Girls’ education was said to reduce parental control over marriage and to increase adultery and divorce; once again women's access to knowledge was mistrusted.96 Gune's poison accusation against Karibbe reveals that occult discourse could be used in radical ways, against even the powerful senior men. In another idiom, young men and women asserted an alternative moral economy by criticising the practice of marrying young girls to the wealthiest older men, winning support from some elders and chiefs – and British officials – by arguing that it increased the likelihood of adultery.97 The success of Paulo and Gune in persuading her kin to accept bridewealth in cash rather than cattle is highly indicative of the changes that were occurring, and which involved profound adjustments in a moral economy that had centred on cattle wealth. Cash bridewealth payments were noted in Yei District earlier than in most areas, no doubt due to migrant labour and market trade in Uganda.98 Cattle were tied tightly to kinship relations; each cow was earmarked for a specific relative in bridewealth payments, so that very few were freely disposable by an individual.99 The shift to individual wealth in cash was therefore generating deep debate about moral and social obligations.

These debates in turn were drawing on new or shifting idioms. At the same time as the poison cases in Kajo Kaji, young men from the nearby missionary secondary school had started a ‘Revival’ in Moru District, organising marriages and preaching against government, chiefs and ‘witchdoctors’; a discourse influenced by Christian revivalism in Uganda.100 Christians were also reported to be heavily involved in the 1937 poison trials in Kajo Kaji, and Paulo utilised the local missionary in his efforts to accuse Karibbe of poison.101 The missionary association of all kinds of religious belief, medicine and ritual with ‘evil’ and the ‘devil’, offered opportunities to attack established authority figures as well as feeding into discourse on witchcraft and poison. It may have encouraged public ‘witch-hunts’ as people sought protection by displaying their own membership of new communities of Christian morality.102 Young men were using their new kinds of knowledge, of Christianity, literacy and distant workplaces, to assert themselves against the older holders of specialist knowledge like doctors, diviners, rain-makers, and elders.

Poison and Authority

Chanock argues that accusations of witchcraft, and the use of ordeals to detect it, were a powerful way of enforcing political power in nineteenth-century central Africa.103 In Kajo Kaji however the chiefs do not appear to have driven the accusations. In the ordeal trials of women, the chiefs were reportedly pressured by the whole tribe to take action, and in the 1939 case they were clearly trying to defuse and conceal the issue of poison.104 As Fields argues, the significance of the ordeal in central Africa had been its expression of the power of the collectivity; the chief or diviner had merely been the agent of the latter.105

The chiefs in the 1939 case of poison were both young and had recently succeeded much older chiefs with knowledge of the rain. Tete had become ‘familiar with the ideas and manners of the Government’ while in prison in Juba for dealing arms with ‘Abyssinians’; he is also said to have discretely become Muslim. He promoted cash crops and mission education, built a European-style house and later became one of the early Southern ‘politicians’, elected to the 1947 Juba Conference and to the Province Council.106 But his youth and foreign knowledge marked a sudden change in authority at a time of hardship and economic change. Kajo Kaji was in a sense integrated relatively late into the patterns of authority that had developed in central Equatoria, in which rain priesthood had usually become divided from a chiefship specialising in knowledge of the government. The poison cases therefore represented an attempt to work out what this meant for chiefship and the moral economy in practice.

The elders and hereditary experts were perhaps more likely to be accusing women or younger men of poison, as they sought to protect their body of knowledge and custom from foreign influences and rival knowledge. They were certainly struggling to control the moral economy, as we have seen in relation to wage labour. In 1940, Chief Abina's elder brother – 'the man who really mattered’, according to Tracey – went to Abina's court to demand that he receive half the chief's salary for his services as rain priest, to which the court assented until the DC intervened.107 This is revealing in a number of ways: first, in that the rain priest clearly had great authority, even in Abina's own court. Second, he was attempting to translate older moral economies, in which his specialist expertise would have earned gifts and labour services, into the new cash economy. The DC's ruling that he could not receive a salary for ‘holy offices’ thus confirmed a new differentiation in the economy of knowledge: it was literacy and knowledge of government and urban life that earned cash. Finally, this episode illustrates the indigenous continuum of knowledge, in which there was certainly no distinction between ‘secular’ and ‘holy’ (or occult) offices; instead each form of knowledge provided a service for which reward was expected. But each kind of knowledge was also highly personalised, and accessed through personal relations, ensuring that local epistemologies of authority were at odds with the notion of the modern bureaucratic state, which was supposed to rule through impersonal relations and centralised knowledge.

The challenge for the chiefs was to translate between these two idioms of authority and morality, and to manage conflicting demands for secrecy and publicity. The pressure on them to conduct poison trials perhaps reflected a desire to create public spaces in which to ‘unveil’ secret activities and hidden enmities.108 Cases of poison and witchcraft were increasingly brought to the chiefs’ courts from the 1930s.109 Yet in the ‘B’ courts, which were regularly inspected by DC's, occult accusations had to be somehow translated into judicial ‘proof’ without the benefit of divination or ordeals. And even if proven, remedies could only take the form of fines or imprisonment. Prison was in itself an interesting new domain. Far from curing witchcraft, it functioned as a source of new knowledge of all kinds, situated in the uncertain realm of the hakuma and the town. It was, after all, where Chief Tete had learned the ways of the government, and Tracey's ‘Case of Poison’ story ended with Karibbe attending a Christian evangelist's service in Yei prison.

So it is not surprising that the chiefs in Kajo Kaji were trying to prevent discussions of poison from reaching the level of the B court, the primary interface with the hakuma. Pels discusses the emergence of Tanzanian nationalism in Uluguru in the secret domain of ‘backstage’ local politics and rumour.110 The difference in Kajo Kaji was that the chiefs were deliberately complicit in the secrecy. Within this local arena, it is actually impossible to know who was driving the poison accusations. Tracey emphasised that the ‘real poison’ lay in gossip and rumour. It reflected a popular demand for protection from harmful incursions as intercourse with the ‘outside’ intensified and expanded, and as elders and experts lost some of their control over the influx of new kinds of knowledge. In fact, public arenas for the discussion of witchcraft or poison were not only, or necessarily, about uncovering the secret, but about the need for a forum in which to debate knowledge and define its morality. Geschiere describes village palavers in Cameroon in which elders combined their ‘ancestral secret knowledge’ or ‘witchcraft of authority’ with oratorical skill and public knowledge to achieve consensus, but in which younger people might also challenge them through alternative claims to knowledge.111 It is this kind of forum for the socialisation of knowledge that was partially undermined in Equatoria during the colonial period.

The limitations on chiefs’ ability to deal effectively with concerns over witchcraft and poison – i.e. over the changing moral economy of knowledge – helped to drive occult discussion into churches or revivalist meetings, which offered dichotomous moral definitions of good and evil to overcome the historical ambiguities of specialist knowledge; or into unregulated channels of rumour, from which it tended to erupt suddenly and violently. This had the effect of altering such discussion from a more normal aspect of everyday discourse (as Azande witchcraft appears in Evans-Pritchard's account) about the ambiguities of access to foreign skills and substances, to an attempt to define and eradicate a distinct category of people who were ‘witches’ or poisoners. In other words, there was an intensified concern, or conflict, about who could access outside sources of knowledge for selfish or harmful interests, later represented by Satan or the cannibals and vampires of the global economy.

But on the other hand, and despite the difficulties, the engagement of the chiefs with popular concerns about poison also helped to locate them inside the local moral community rather than in the realm of the hakuma. ‘It is difficult … even to depict African witchcraft idioms as a weapon of African resistance’, declares Austen.112 But the eruption of backstage rumours about poison into the colonial courts as a way of discussing changes in the moral economy of knowledge did signify resistance to the colonial attempts to impose new epistemologies and medicines and to prevent discussions or divinations outside the officially public arenas of courts and churches. Occult discourse functioned as a local language that excluded the information-centralising project of the colonial state and vociferously argued for the personalisation of knowledge and power. It is perhaps then no coincidence that in 1963 the southern rebels against the Khartoum government in the first Sudanese civil war came to be called Anyanya, or ‘poison’, or that they upheld the chiefs as ‘the cherished social heritage of the people’.113

Conclusion

This article has situated the 1930s Kajo Kaji poison cases recorded by colonial officials in the context of a historically dynamic moral economy of knowledge. Because poison encompasses the potential for using actual material substances in both productive or harmful ways, it has formed a versatile concept for the expression of contemporary concerns about new materials such as ink (or more recently, insecticide and fertilizers), and about access to them, which is in turn determined by access to specialist knowledge. The 1930s cases demonstrate the adaptability and contested deployment of occult accusations or claims and practices. Poison appeared to act as a focus for a kind of moral panic at a time of crisis, which targeted women, and yet it also offered opportunities for different people to claim power or to accuse others. In 1939 a woman used a specific accusation of poison in her own marriage struggles, which received attention because it touched on bigger debates about new kinds of knowledge and their role in the changing economy. But to see poison simply as a new occult response to colonial modernity would ignore its historical use and potential for production by hunting, reproduction by female inheritance, and revelation by divination and ordeal justice. The 1930s cases demonstrate a specific concern about the dangers of substances being brought into the community which was not new. The concept and acquisition of poison were located above all in the ambiguous relations and exchanges between different communities or distant regions with their own specialised knowledge and material commodities. As those relations changed or expanded, so the category of poison could absorb new materials, manifest in different situations and express different concerns about the morality of new or specialist kinds of knowledge and their use for individual or community gain or harm. It materialised as Arabic words in a colonial police ink bottle in 1939 as an ultimate expression of the moral ambiguity of foreign knowledge and as part of perennial and historical debates over access to it. And in the midst of a changing moral economy, it revealed deeper concerns that knowledge gained by individuals should be incorporated into a fertile body of expertise in the community rather than directed into personal profit and reproduction.

This paper is drawn from the largely archival research for my PhD thesis: ‘Knowing Authority’. I would like to thank the AHRB for the PhD funding and Justin Willis for his supervision and his suggestions for this paper. I am also very grateful for the comments on an earlier version of this paper from the members of the African History seminar at SOAS in March 2005.

Notes

1. DC Tracey, ‘A Case of Poison’, Yei, 31 Dec. 1939, Rhodes House Library, Oxford, MSS Perham 549/5 ff. 1–2, pp. 13–14.

2. Geschiere, Modernity; Comaroff and Comaroff, Modernity and its Malcontents; White, Speaking with Vampires.

3. Comaroff and Comaroff, Introduction.

4. Shaw, ‘Production of Witchcraft’, 856–76. Both Englund and Green criticise White for locating the origin of vampire stories in the colonial period as ‘new imaginings for new relationships’ (White, Speaking with Vampires, 22): Englund, ‘Perils of Transparency’; Green, Priests, Witches and Power, 73.

5. Also Austen, ‘Moral Economy’

6. Guyer and Belinga, ‘Wealth in People’, 91–120.

7. Geschiere focuses much more on the ambivalence of ‘the witchcraft of authority’: Modernity, 199, also 92–94.

8. Lambek highlights the moral ambiguity and dangers of knowledge: Knowledge and Practice, esp. 220–26.

9. Austen, ‘Moral Economy’, 94, 104.

10. Yunis, ‘Notes on the Kuku’, 1–41.

11. Equatoria Province Monthly Diaries Sep. 1937, National Records Office, Khartoum (hereafter NRO) Civ Sec 57/4/17.

12. Tracey, ‘A Case of Poison’, 14.

13. Comaroff and Comaroff, ‘Introduction’, xxvii; Douglas, ‘Thirty Years’, xxix–xxx.

14. Douglas, ‘Thirty Years’, xiv.

15. Middleton depicts witchcraft as the inversion of ideal social relations and kinship authority: Middleton, ‘Witchcraft and Sorcery’, 257–75, at 271–73.

16. Middleton, Lugbara Religion, 247.

17. Middleton, ‘Witchcraft and Sorcery’, 264.

18. Tracey, ‘A Case of Poison’, 14.

19. Johnson, ‘Criminal Secrecy’, 170–200, at 179–81, and Nuer Prophets, 27–29.

20. Evans-Pritchard, Witchcraft, 477, 417.

21. Johnson, ‘Criminal Secrecy’, 200.

22. Evans-Pritchard, Witchcraft, 519–20.

23. Allen, ‘Ethnicity and Tribalism’, 112–39, at 128–31.

24. James, Listening Ebony.

25. Evans-Pritchard, Witchcraft, 425–32, 445–46. See also Buckner, ‘Modern Zande Prophetesses’, 102–21.

26. Evans-Pritchard found no proof of the claim that sorcerers used to be executed: Witchcraft, 190–93. Delius warns against projecting ideas of witchcraft back into the pre-colonial past, by showing that while fears of witchcraft were pervasive in nineteenth-century Sotho and Tswana communities, actual accusations or executions were rare, unlike in the later twentieth century. Delius, ‘Witches and Missionaries’, 429–43, at 430, 440–43.

27. Johnson and Anderson, Introduction to their Ecology of Survival, 1–24.

28. C. H. Stigand, Inspector Rejaf Merkaz, to Governor, 6 March 1911, NRO Intelligence 2/29/239.

29. John Winder, ‘Fifty Years On: Service in Mongalla Province, 1930–33’ (1979), Sudan Archive, Durham (hereafter SAD) 541/7/1-32; Stigand, Equatoria, 72, 92–93, 100; Nalder, Tribal Survey, 222.

30. Elias Nyamlell Wakosan, ‘Origin and Development’,127–204, at 127–28; Middleton, Lugbara Religion, 245; Yunis, ‘Notes on the Kuku’, 30.

31. Jackson, Governor of Uganda, to Harcourt, 14 March 1912, Public Record Office, London (hereafter PRO) WO 181 236.

32. Leopold, Inside West Nile, 126–27; Gray, History of the Southern Sudan, 57; Governor Nalder, ‘Mongalla Province Summary of Information’, Nov. 1933, NRO Civ Sec 57/35/131; C. H. Stigand, Inspector Rejaf Merkaz, to Governor, 6 March 1911, NRO Intelligence 2/29/239; Yunis, ‘Notes on the Kuku’, 35.

33. Interview with Barnaba Dumo Wani, 17 Feb. 2003, Khartoum.

34. Larick, ‘Warriors and Blacksmiths’, 301, 313, 318.

35. Yunis, ‘Notes on the Kuku’, 27; Williams, ‘Pagan Religion of the Madi’, 202–10, at 209.

36. Stigand, Equatoria, 92–93; Birch, ‘Metu People of West Madi’, 134–36; John Winder, ‘Fifty Years On: Service in Mongalla Province, 1930–33’ (1979), SAD 541/7/1-32; Governor Nalder, ‘Mongalla Province Summary of Information’, Nov. 1933, NRO Civ Sec 57/35/131.

37. Evans-Pritchard, Witchcraft, 271–80.

38. John Winder, ‘Fifty Years On: Service in Mongalla Province, 1930–33’ (1979), SAD 541/7/1-32.

39. Larick, ‘Warriors and Blacksmiths’, 310; Stigand, Equatoria, 157; Whitehead, ‘Suppressed Classes among the Bari’, 265–80, at 269, 271–72; Evans-Pritchard, ‘Preliminary Draft of An account of the Moro’, n.d. (1930s), NRO Dakhlia 112/14/95.

40. Herbert, Iron, Gender and Power. Blacksmiths also played a unique role in Kuku funeral ceremonies: Interview with Barnaba Dumo Wani, 17 Feb. 2003, Khartoum.

41. Stigand, Equatoria, p. 74.

42. John Winder, ‘Fifty Years On: Service in Mongalla Province, 1930–33’ (1979), SAD 541/7/1-32; Baxter and Butt, The Azande, 116, 121.

43. Middleton, ‘Witchcraft and Sorcery’, 264.

44. Yunis, ‘Notes on the Kuku’, 30–31

45. Williams, ‘Pagan Religion of the Madi’, 204–05; Middleton, Lugbara Religion, 260–61; Baxter and Butt, The Azandi, 125.

46. Evans-Pritchard detailed the Zande opinion that women should not use magic and medicines associated with male activities like hunting: Witchcraft, 427.

47. Guyer and Belinga, ‘Wealth in People’, 115–16; Interview with Simon Wani Ramba, 10 Jan. 2003, Khartoum: a woman brought rain-making to the Fajelu from the Bari in marriage.

48. Yunis, ‘Notes on the Kuku’, 4–6.

49. Allen, ‘Ethnicity and Tribalism’, 131–32.

50. History of the Southern Sudan, 57; Johnson, ‘Structure of a Legacy’, 72–88; Inside West Nile, 118–28.

51. Governor Owen ‘Notes on Enclave’, 1910, NRO Mongalla 1/8/51.

52. Larick, ‘Warriors and Blacksmiths’, 309.

53. History of the Southern Sudan, p. 113.

54. Whalley, A/D. C. Opari and Kajo Kaji, to Governor, ‘Resolutions from Kuku chiefs of Kajo Kaji Lukiko’, 19 Nov. 1929 and 22 Nov. 1929, NRO Civ Sec 1/39/104.

55. Nalder, Tribal Survey, 216–18.

56. For a history of local authority in central Equatoria, see Leonardi, ‘Knowing Authority’. For general accounts of Condominium administrative policy, see Collins, Shadows in the Grass; Daly, Empire on the Nile, and Imperial Sudan.

57. Stigand, Equatoria.

58. E.g. Gray, ‘Witches, Oracles and Colonial Law’, 339–63, at 340; Fields, Revival, 72–73, 248; Crais, Politics of Evil, 13.

59. Waller, ‘Witchcraft and Colonial Law in Kenya’, 241–75.

60. Brock, Governor Bahr El Ghazal, to Civil Secretary, 4 July 1929, NRO Civ Sec 1/13/43; Holland, ‘Lukiko District Order no. 3’, 30 Sept. 1923, NRO Mongalla 1/1/2.

61. John Winder, ‘Fifty Years On: Service in Mongalla Province, 1930–33’ (1979), SAD 541/7/1-32; Tracey, ‘A Case of Poison’, 14.

62. Equatoria Province Annual Report 1937, NRO CS 57/24/99; Yei District Monthly Diary Sept. 1944, NRO EP 2/24/87.

63. Mrs Selwyn, Kajo Kaji, Aug. 1932, Church Missionary Society Archives, University of Birmingham Library (hereafter CMS) G3 AL; Winder, ‘Fifty Years On: Service in Mongalla Province, 1930–33’, (1979) SAD 541/7/1-32.

64. Equatoria Province Monthly Diaries Sep.–Oct. 1937, NRO Civ Sec 57/4/17; and Feb.–March 1938, NRO Civ Sec 57/7/29.

65. F. J. Finch, Kajo Kaji, 10 July 1935 and 8 Nov. 1938, CMS G3 Annual Letters, 1935–39.

66. Sudan Intelligence Report 228 (July 1913), PRO WO 106 6225.

67. See, for example, Allen's discussion of the colonial origins of the ethnic labels ‘Acholi’ and ‘Madi’: ‘Ethnicity and Tribalism’, 123–25.

68. Equatoria Province Monthly Diary Jan.–March 1937, NRO Civ Sec 57/4/17; Interview with Barnaba Dumo Wani, 17 Feb. 2003, Khartoum.

69. Sleeping Sickness Annual Report 1924, NRO Mongalla 1/6/39.

70. Gillies, Introduction, xvii; Allen, ‘Ethnicity and Tribalism’, 128; Green, Priests, Witches and Power, 137. But though jealousy was an aspect of poison discourse, there were other named categories of occult activity in Kajo Kaji more obviously related to ‘property’ and envy: Interview with Barnaba Dumo Wani, 17 Feb. 2003, Khartoum. Jealousy and the location of witchcraft within close, intimate relations have been frequently discussed: Geschiere, Modernity; White, Speaking with Vampires, 20; Evans-Pritchard, Witchcraft, 55, 73.

71. See also Lyons, Colonial Disease, 198.

72. Yei District Monthly Diary July 1942, Nov. 1943, NRO Equatoria 2/24/87.

73. White also highlights the profound and contradictory understandings of the ability of Western medical techniques to penetrate the skin: White, Speaking with Vampires, 99.

74. Homeostatic social control or crisis interpretations are particularly associated with the ‘Manchester School’ led by Max Gluckman in the 1950s and 1960s: e.g. Gluckman, Custom and Conflict in Africa, 93–108.

75. Tracey, ‘A Case of Poison’. The original spellings have been retained.

76. Tracey, ‘A Case of Poison’. The original spellings have been retained, 1, 14.

77. Doyle, ‘Population Decline’, 436; Kaufmann, ‘The White Nile Valley’, 140–95, at 189.

78. Yunis, ‘Notes on the Kuku’, 16, 30; Doyle, ‘Population Decline’, 454–55; Interview with Barnaba Dumo Wani, 17 Feb. 2003, Khartoum. See Berger, ‘Fertility as Power’, 65–82.

79. Evans-Pritchard, ‘Preliminary Draft of An account of the Moro’, n.d. (1930s), NRO Dakhlia 112/14/95.

80. Doyle, ‘Population Decline’, 453.

81. Vanden Plas, J. ‘Les Kutu’: Van Overbergh, Brussels, 1910, cited in translation by Huntingford, The Northern Nilo-Hamites, 44–45C. H. Stigand, Inspector Rejaf Merkaz, to Governor, 6 March 1911, NRO Intelligence 2/29/239.

82. Finch, Kajo Kaji, 10 July 1935 and 8 Nov. 1938, CMS G3 Annual Letters, 1935–39.

83. Austen, ‘Moral Economy’, 104.

84. Doyle, ‘Population Decline’, 455.

85. Interview with Barnaba Dumo Wani, 17 Feb. 2003, Khartoum.

86. Middleton, Lugbara Religion, 246.

87. See Leopold, Inside West Nile, 77.

88. Equatoria Province Monthly Diaries Feb. 1951, NRO Dakhlia 57/9/24; and Jan. 1953, NRO Civ Sec 2 30/3/6.

89. See White, Speaking with Vampires, 279–80; Shaw, Memories of the Slave Trade, 11.

90. Equatoria Province Monthly Diaries Oct. 1937, NRO Civ Sec 57/4/17.

91. Comaroff and Comaroff, ‘Introduction’, xxvi.

92. E.g. Macphail, ‘The Bandala Method of Hunting’, 279–83, at 283.

93. Hutchinson, Nuer Dilemmas, 270–98.

94. Tracey, ‘A Case of Poison’, 4.

95. Juba Mission Report 1942, CMS G3 Sg2; Governor Parr to Hickson, Inspector Education, 20 March 1939, NRO Equatoria 1/4/17; Yei District Monthly Diary March 1945, NRO Equatoria 2/24/87.

96. Janson-Smith, ‘Report on Juba training school Final Examination, 1941’, NRO Dakhlia 3 1/6/29.

97. E.g. Equatoria Province Monthly Diary Sept. 1946, NRO Dakhlia 57/2/5.

98. Nalder, ‘Mongalla Province Summary of Information’, Nov. 1933, NRO Civ Sec 57/35/131. See also Shadle, ‘Bridewealth and Female Consent’, 241–62.

99. Whitehead, ‘Crops and Cattle among the Bari’, 140–41.

100. Reports in NRO Equatoria 1/5/26.

101. F. J. Finch, Kajo Kaji, 8 Nov. 1938, CMS G3 Annual Letters, 1935–39.

102. Meyer, Translating the Devil; Green, Priests, Witches and Power; Fields, Revival.

103. Chanock, Law, Custom and Social Order, 89–92.

104. F. J. Finch, Kajo Kaji, 8 Nov. 1938, CMS G3 Annual Letters, 1935–39.

105. Fields, Revival, 264.

106. Interview with Barnaba Dumo Wani, 17 Feb. 2003, Khartoum; Tracey, ‘A Case of Poison’, 15.

107. Equatoria Province Monthly Diaries June 1940, NRO Civ Sec 57/12/46; Tracey, ‘A Case of Poison’, 10.

108. Geschiere, Modernity, 204–05, 275 fn 8.

109. In the 1920s, witchcraft and magic do not appear in lists of court cases (NRO Mongalla 1/1/2), but in 1945 there were 39 recorded witchcraft cases in Yei District B courts: Yei District Annual Report 1945, NRO Equatoria 2/27/98. There are recorded discussions in courts of the morality of ‘magic’ in Yei District Monthly Diary June 1942, NRO Equatoria 2/24/87, and Moru District Annual Report 1944, NRO Equatoria 2/27/97.

110. Pels, ‘Creolisation in Secret’, 1–28.

111. Geschiere, Modernity, 92–4.

112. Austen, ‘Moral Economy’, 105.

113. Wakosan, ‘Origin and Development’, 136.