Quests for therapy in northern Uganda: healing at Laropi revisited

Abstract

This article presents a case of diachronic ethnography. It examines quests for therapy among the Madi people of northern Uganda. It is based on ethnographic fieldwork carried out in and around the small trading centre of Laropi; originally in the late 1980s and again in 2008. By revisiting the same field site at different points in time, and by drawing on related ethnographic material collected in the 1950s and 70s, we are able to examine how such quests have altered and to discuss factors influencing these changes. We also comment on shifts in conceptual approaches of medical anthropology that have influenced perceptions and analysis. Laropi lies close to the border with Sudan and its inhabitants have experienced much upheaval and political isolation. We examine how this has influenced understandings and responses to ill-health and misfortune. Particularly important in recent years has been the increasing availability and accessibility of biomedicine, which the population have embraced and indigenized as a mark of progress and political recognition. On the face of it, this has rendered recourse to more “traditional” forms of healing obsolete. However, as we describe, the situation is more ambiguous. Notions of witchcraft, spirit possession and ancestor veneration are more pervasive than they might seem.

Changing quests for therapy in northern Uganda: healing at Laropi revisited

Diachronic ethnography remains relatively rare, especially in the arena of medical anthropology and in politically fragile locations. Not many researchers have been in a position to return repeatedly to their field sites over protracted periods and assess how things alter over time. Even now, the anthropological lens tends to be focussed on observed moments, and anthropologists escape from the present by reference to other disciplinary approaches, especially those of historians. This article attempts something different: the linking of multiple anthropological presents to a discussion of the differences between them.

Between 1987 and 1991 Tim Allen spent 20 months living at Laropi, a small town on the White Nile, close to where the river crosses from Uganda into Southern Sudan. In a series of publications, he has described the interactive and pluralistic conceptions of causality and healing at this border location, and showed how these characterised the population's response to extreme circumstances.1 Allen returned to Laropi briefly in 2004 and in 2005. Then in 2008 he worked with Laura Storm and other researchers on aspects of health in the vicinity.2 In this article, Allen's earlier findings about local approaches to illness and therapy are summarised, and, drawing on Allen and Storm's more recent fieldwork, comments are made on how quests for therapy have been changing. Remarks are also made about shifts in approaches to analysis in medical anthropology. Those have affected ways of seeing and understanding, so they are part of the story too.

Laropi and the Madi

Twenty years ago, Laropi was located in the middle of Moyo District. It was the site of the ferry linking Moyo Sub-District with Adjumani Sub-District on the east bank of the river. Subsequently, Moyo and Adjumani were made separate districts. Laropi is part of the former, although Adjumani town is slightly easier to reach if the ferry is working, because Moyo town is located over the Metu Mountains. Links with both towns have been greatly enhanced in the last few years by improvements in the roads and regular buses and taxi services. People still complain about being cut off from the rest of the country and ignored by the government, but circumstances are very different to the way they were in the late 1980s.

The people in this part of Uganda speak the Madi language, which is related to the Lugbara language spoken in areas further west around Arua, and is completely different to the Acholi language spoken in the central north of Uganda around Gulu and Kitgum. The word “Madi” itself is an ethnic label which seems to have been introduced towards the end of the nineteenth century by Europeans, and was initially also applied to other groups too. It is probably derived from a local word for “a person” (ma ‘di).

In so far as it is possible to discern in any detail what life was like before the upheavals associated with the arrival of slave and ivory traders and Turco-Egyptian forces from the 1860s, identities were associated with clans or small chiefdoms. What became the Madi and Lugbara “tribes” under the Protectorate were in part a product of deliberate policies of separating and isolating groups settled near the Nile. It was recognised that ways of life and language were similar between the populations later classified as Madi and Lugbara.3 However, there were concerns among British officers that those people living close to the Nile had been affected by the presence of Muslim garrisons so, to restrict the spread of Islam to other areas, the territory was deliberately isolated. Catholic missionaries were also encouraged to concentrate activities in the area, with considerable effect.

Missionaries were similarly encouraged in other parts of northern Uganda too, but the loyalty of the influential Muslim chief, Fadilmulla Ali, to the British during and immediately after the First World War, when the Protectorate administration in the region was stretched to near breaking point, had the paradoxical effect of promoting Islam among people immediately to the west of Moyo. As a result, Fadilmulla Ali's people became a distinctive group of Lugbara – sometimes referred to as Aringa (i.e. as a separate ethnic group) – and, as the population nearer the Nile converted to Catholicism, the distinctiveness of the riverine people from their neighbours was underlined. By the late 1930s, the first Madi priest was ordained, and it was reported by the Italian Comboni Fathers there were around 10,000 Madi Catholics, with an estimated 15,000 “pagans” left to convert. It was claimed in the order's newsletter that the Madi were the “sole fully Catholic tribe in Africa”.4

Another reason for constriction of movement was the presence of sleeping sickness. Large areas were forcibly depopulated and turned into wildlife sanctuaries, concentrating the villages along the Nile and the main roads. This has had long-term effects in terms of attitudes to governmental authority and also disease control programmes.

In Protectorate times, the area was administered as Madi Sub-District, and passed back and forth between the neighbouring district commissioners based in Gulu and Arua towns. It only became a separate district a few years before Ugandan independence. By that point, until the upheavals of the late 1970s, the mainstay of the local economy was cotton, supplemented by the sale of cattle and, in riverine locations, dried fish. In addition, most families had at least one male member who travelled to the south to work on plantations, in factories, or in the armed forces.

At the turn of the 1980s, the Madi and the Lugbara were regarded as having been sympathetic to Idi Amin, and following the Tanzanian invasion and return to power of Milton Obote, atrocities were perpetrated against civilians. Northwest Uganda became the scene of protracted guerrilla warfare, and perhaps a quarter of a million people fled across international borders – the majority going to Sudan where they were mostly settled in refugee camps, and became recipients of relief and development aid.

By 1984, the refugees tended to be self-sufficient in food production, and a few became relatively well off, compared to the host population. A handful also played a role in the complicated and increasingly bitter divisions in the politics of Sudan's autonomous southern region, which was a factor in to the country's return to war. What was often perceived as the support of the Ugandan refugees for the re-division of southern Sudan and the creation of a separate Equatoria region made the refugee camps an obvious target for the Sudanese People's Liberation Army (SPLA). In April–May of 1986, the SPLA drew upon local resentment of the Ugandans, fuelled by the aid that had been channelled in the refugees’ direction, and attacked the camps to the east of the Nile, where many of the Madi were settled, killing several people. Neither the Sudanese authorities nor the aid agencies were in a position to offer protection, so the refugees returned home as fast as possible. They left crops in the fields and, in a matter of three months, thousands had crossed back into Uganda.

Meanwhile, Yoweri Museveni's National Resistance Army had seized power in Kampala at the beginning of 1986 and established a National Resistance Movement government. Although there was still fighting going on in nearby parts of the country, Moyo District was fairly stable. Nevertheless, the returnees found themselves in a devastated environment. Marketing and transport facilities had collapsed. School buildings had been demolished, and government health care services hardly existed outside Moyo town, where the hospital was in a poor state of repair. During this time, Médecins sans Frontières (MSF) (France), Swiss Disaster Relief, and later MSF (Switzerland) were based in Moyo and Adjumani towns. They provided some very basic health services, and were grappling with a serious sleeping sickness outbreak.

When Allen arrived at Laropi in early 1987, the population was struggling to find food on a daily basis. Almost everyone had returned during the preceding months, having fled from the refugee camps in Sudan with what they could carry. Many families were surviving on the ground-up seeds of a particular water lily and on wild roots. Initially, Allen slept at the sub-chief's office in what had once been a small holding centre for prisoners, and often had to be assisted by the sub-chief and his staff because he would be mobbed by angry, and often very drunk people shouting “give me food”. It was several months before he was able to organise the building of a small house, and to be accepted enough to be able to walk around freely. Unsurprisingly, severe misfortune, affliction and death were ever present threats. This was discussed in various ways – sometimes in terms of specific disease, their symptoms and what was known about clinical therapy but often with reference to other perceptions too, grounded in more locally specific idioms. In the following sections, we describe these various perceptions of health and illness, look back to how comparable notions were described by anthropologists in the 1950s and 1970s, and discuss how therapies have evolved in the past two decades.

Health and healing among the Madi and their neighbours

The term used to translate the word “health” among the Madi is cwe, which basically suggests something which is “good”, so it is not confined to conventional notions of being healthy. In a way it is closer to the World Health Organisation's well-known but idealistic and rarely applied definition of health as “a complete sense of mental, physical and social wellbeing, and not just the absence of disease”. It is also difficult to translate precisely the words “illness” or “disease”. In Madi both are translated as laza, but this can also refer to almost any kind of suffering, misfortune or pain. This does not mean that Madis who speak only their own language cannot understand what a disease is, or how it might be cured with a clinically formulated, manufactured medicine. There have been several attempts by governments to control particular diseases both during and after British rule. Additionally, various kinds of medicines have long been available at trading centres or at clinics, and there has been some basic science teaching in schools. Catholic missionaries have also tended to promote the use of manufactured medicines in preference to local remedies.

In the 1980s, Tim Allen found that in many cases, Madi labels for particular sets of symptoms were linked to specific disease control programmes in the past (e.g. mongoto, sleeping sickness; loboto, yaws; njuku, syphilis/gonorrhoea). There were also certain general terms, such as jue (boils or swelling), that were used to refer to many illnesses, with the implication that specific cure of the symptoms was required, either from a clinic or from a local herbalist. In other words an interpersonal or metaphysical explanation was not being sought, at least not at that point of time, by the person using the term. In addition, there were other diseases, including bilharzia (schistosomiasis), malaria, measles and HIV/AIDS, where the biomedical terms had been adapted as a Madi word. Although the local understandings of these diseases could vary from clinical interpretations, their use reflected a widespread recognition that certain kinds of laza existed that could potentially be cured by biomedical therapies. Sometimes a Madi might use an expression like laza Rubanga dri'i (affliction [is] given by God), or say something like “we are just born to suffer”. They did so to suggest that the causes of laza are not to be found in the immoral actions of neighbours, or the interventions of ghosts in the lives of the living. However, such explanations were a matter of debate. Others might mention the Madi proverb awo otu kwe ku, which might be translated literally as “crying does not climb a tree”. It suggests a rejection of coincidental causation, and a need to find out who is responsible.

When Allen was living in Laropi after the population had returned from Sudan there was considerable emphasis, especially among older people, on the need to re-establish ancestral shrines, and to ritually cleanse the area. In so doing they were evoking ideas about the moral order associated with notions about how things were done before the political upheavals that had forced the population to become refugees. Doubtless there was some romanticism about the way things should have been, and there was perhaps an ulterior motive in elderly people promoting ceremonies that underlined their social importance. To give some idea of what debates about therapy might actually have been like before the flight of the population to Sudan or earlier in the twentieth century, it is necessary to turn to previous ethnographic accounts. Unfortunately, the Madi people were never closely studied before by an anthropologist, and with the exception of one article by John Middleton based on a brief visit sometime around 1950, there is very little to turn to. However, Middleton's work on the linguistically related Lugbara, as well as that of Barnes-Dean in the 1970s, provides some idea of how things may have been. In broad terms, their insights resonated with the recollections of those Madi elders interviewed in the 1980s, especially those of old women – who perhaps had less at stake in reinterpreting the past in an effort to assert a degree of authority in the present.

Changing Lugbara healing between the 1950s and 1970s

The Lugbara that Middleton lived with in the early 1950s explained to him that azo (a term for misfortune/affliction and illness which is equivalent with the Madi word laza), should be explained in terms of an ideally unchanging socio-moral order.5 Unsurprisingly, this was a view that was promoted most clearly by ritually important elders, at least in part because it was an important aspect of their own authority. Although livelihoods were rapidly changing, at that time most of the Lugbara lived in small, exogamous, patrilineal/patrilocal groups, each headed by an elder. The lineage was the primary sphere of direct social relations, and neighbours, who were also kin, were the core of the moral community. Within the community there was intense competition among male elders for seniority, which eventually led to lineage fission. Competition between elders and the control they exercised or sought to exercise over their families was expressed through a cult of the dead.6 Deceased ancestors, usually males, were believed capable of sending azo to people whom they considered harmful to the well-being of their lineage. Virtually anyone who felt wronged could invoke ancestral ghosts to inflict azo on their behalf, but a senior male was expected to do so. It was part of his work, to “cleanse” the lineage home. In addition, ancestors sent azo without invocation if they felt the living were neglecting them.

Thus, azo showed the living when ancestors were displeased. Ideally, therapy proceeded in the following manner.7 The ritual guardian of the afflicted person consulted male-operated, God-empowered oracles at ancestral shrines (tumi) to discover the identity of the ghost concerned as well as the nature of the sacrifice to be made when the patient recovered. If recovery did not occur, it was said that “God refuses”, indicating that it was futile to do more, because God had decided to take that person away in death. According to Middleton, death was ultimately thought of as God's will.

However, sometimes when a patient failed to recover, other suggestions were made about the real cause. Perhaps the oracles had made a mistake or they did not “know” or understand the particular azo. The oracles were either consulted again or alternative explanations for sickness were sought. For Middleton's informants, these alternatives reflected recognition among the Lugbara that social order was not as stable as it ought to be. Quite apart from the upheavals of the early colonial period, there had been commoditisation by the 1950s as a consequence of short-term male labour migration to the south of Uganda and the introduction of cash cropping (mainly tobacco in the Arua area). In addition, the authority of the elders was being directly challenged by the protégés of the government and the missionaries. Middleton describes various aetiologies that could not be known by oracles.

One of these sorts of azo were associated with spirits and linked to specific symptoms, such as those of cerebrospinal meningitis. Another kind were said to have “self-evident” causes, such as venereal diseases. Doubtless, there was some connection between these ideas and limited exposure to biomedicine via dispensaries and government control programmes for particular diseases, notably sleeping sickness and yaws. Significantly, treatment was given by an ojo (diviner) and administered to effect a cure for the specific symptoms in the patient's body (i.e., for these illnesses, no metaphysical or interpersonal explanation was sought – other than that the affliction might have been caught from someone else). Ojo were almost invariably women who, at puberty, had been “seized” by the much-feared aspect of God, which dwelt in the bush and near streams. They were diviners and often also local healers. As a healer, an ojo would remove objects from the bodies of patients, attend births, and administer herbal remedies. Her central role, however, was in mediating that which was “outside” (amve) the social and moral order. She could identify what caused those illnesses not sent by ancestors, for she “knew the words” of the feared aspect of God, and also those of witches and sorcerers.

According to Middleton, the Lugbara believed that onzi (evil/amorality) was in the world because of ambition and envy. In particular, there were certain people who could cause azo and death in mysterious ways. These were people with ole, a term that encompassed both notions of witchcraft and sorcery. In his analysis, Middleton follows the old anthropological convention of distinguishing between witches and sorcerers on the grounds that the former have an inherent power that, possibly unintentionally, can harm others, whereas the latter use technical means, termed enyanya (poison), in full deliberation. What Middleton describes as “witchcraft” was associated only with men. The accused was usually an elder who was losing authority, a person with physical disabilities, or someone marginal in the lineage. When a diviner indicated that witchcraft was the cause of the affliction, an attempt might be made to cool the envy of the witch, perhaps by inviting him to a meal.

Although it was considered a bad thing, witchcraft could be comprehended in that it occurred as a result of normal masculine ambitions within a lineage. In contrast, what Middleton describes as “sorcery” was a heinous crime. If it occurred between agnatic kin, it amounted to fratricide, for enyanya killed victims, something that ghost invocation and witchcraft rarely did. Sorcery also cut across spheres of ritual authority in a manner that threatened the very fabric of moral interaction. Accusations often occurred where obligations of kinship and neighbourhood were ambiguous. Sorcerers in these cases tended to be men from the outside, like returning labour migrants or other migrants living within the home. If an ojo revealed that such a person was involved, he might be beaten and have to flee the area.

Sorcery, in addition, was believed to have been practised by both sexes, and punishment for females could be more violent. Middleton explains that the female sorcerer was commonly thought to act out of sexual jealousy. She poisoned co-wives and their children, who were, of course, children of her husband's patrilineage (unlike herself). If caught, she might be put to death by her husband's agnates by cutting off her limbs, burning, or spearing, the latter a practice that seems to have been quite common in the 1920s and 1930s.8 Middleton argues that this violence toward women and the driving away of strangers were responses to rapid social change. The concentration of the population in towns or market centres, cash cropping (which placed wives in direct competition with their husbands over land), together with such factors as labour migration had, according to Middleton, greatly exaggerated the normal tensions of daily life.

Underpinning what Middleton writes about witchcraft and sorcery is the notion that women, like the dangerous forces that, as ojo, they might know, could be thought of as “outside” and onzi themselves.9 A woman was never truly “of the home”, because she left that of her father at marriage and became part of her husband's lineage only through her children. It was therefore predictable that, when ritual or other institutionalised vengeance became untenable, women would end up as scapegoats. More positively, the spiritual powers and symbolic attributes vested in women might become legitimising avenues of traumatic change.10

In the 1950s and 1960s, Evans-Pritchard's great work on Zande witchcraft was a pervasive influence on East African anthropological studies. Although what Evans-Pritchard actually wrote is more nuanced and ambiguous than most summaries suggest, his book was open to criticism for not adequately recognising processes of social change. His book was written in the present tense, and seemed to imply that therapy was constrained by a closed-system whereby the Zande are “caught in a web of belief”11 and unable to incorporate new ideas, such as scientific thinking, empirical causality and allopathic medicines. Middleton drew inspiration from Evans-Pritchard's approach to the internal logic of what might be called a healing system, but was also acutely aware that daily life for the Lugbara was being transformed under the Protectorate administration. So, while he had much to say about the competition between elders for control of shrines, and the ways in which they would draw on moral ideas to make arguments in their own personal interests, he was also well aware that their position was being challenged. He commented on the influence of the Christian missions, as well as the effects of education and public health programmes, and he discussed witchcraft and sorcery as a response to perceived breakdowns in social norms. It becomes clear from close reading of his work that in the early 1950s, there were contradictory, conflictual and overlapping approaches to therapy for the Lugbara, rather than a single, stable system. Also things were changing rapidly. It is these elements in Middleton's work that Virginia Barnes-Dean built on in the 1970s and early 1980s. Essentially her argument is that the changes Middleton noted had, by the time of her fieldwork, made ritual elders largely irrelevant. In addition, her analysis drew on newer analytical models, notably those of Arthur Kleinman and John Janzen.

During the 1970s, Kleinman had developed an influential theoretical taxonomy of health systems. In what he termed the professional sector, biomedicine has become dominant, but tends to be adapted to the local environment. Biomedicine is also affected by popularisation, as notions derived from it become an aspect of local popular beliefs and ways of thinking, although not necessarily in a straightforward way. Kleinman additionally included in his taxonomy a folk or traditional sector, made up of non-professionalised specialists, and the popular sector, which he argues is where various therapeutic options are evaluated and choices are made. For Kleinman, all these sectors interrelate and overlap, and in combination make up a health care system. Contradicting the conventional view of medical professionals that health care is organised and provided to a population, a key point Kleinman makes is that people activate their health care by deciding when and whom to consult, whether to adhere, whether treatment is effective, and when to switch to another treatment.12

This issue of combining options was taken up and elaborated in detail by John Janzen in his seminal book, The Quest of Therapy: Medical Pluralism in Lower Zaire, published in 1978. The starting point for Janzen's analysis was the therapy management group of the patient – made up of relatives and friends as well as the afflicted person. The quest for therapy was negotiated between them, and various options were adopted in diverse ways. According to Janzen, sometimes there is a hierarchy of resort, whereby one therapy is tried as a starting point and others only brought in to play if it that fails. At other times, multiple therapies are adopted at the same time, or there may be specific symptoms that are always taken to a particular kind of specialist and never to anyone else. Thus Janzen avoided conceptualising health care as some sort of structural system by prioritising the endlessly negotiated strategies of afflicted people themselves.13

Writing in the early 1980s, drawing on these theoretical contributions, and leaning rather more towards Kleinman than Janzen, Barnes-Dean describes Lugbara healing in the 1970s as being made up of sectors, between which patients make choices.14 In 1973, she found that azo was diagnosed either as enyanya, now including some practices that Middleton would have termed witchcraft, or as “other illness”. She also noted that the authority of elders had been eroded and that ancestor invocation was, it seemed, no longer practised. Azo understood as “other illness” had impersonal aetiologies and could be contracted by Europeans. Local remedies were known for many of them, but it was recognised that cures were also possible at mission or government hospitals. Azo from enyanya, on the other hand, could only be contracted by Africans and could not be cured by biomedicine. Local treatments for both enyanya and some “other illnesses” were administered by ojo. They cured a range of specifically named ailments, from headaches and itchy skin to the effects of poisons. Several of these treatments involved making small cuts on the patient's body and rubbing in the concocted remedy. Interestingly, treatment for venereal diseases involved drinking a liquid containing certain pounded leaves in combination with going to the hospital for an injection. Barnes-Dean maintains that the work of ojo as herbalists was linked to the perception of women as “outside” and of the bush, from where herbal remedies were collected. But whereas in the 1950s it was this dangerous aspect of women that enabled ojo to mediate the evil forces that impinged upon the sphere of moral action, they had become the healers of “true Lugbara sickness”, now understood exclusively as enyanya. The former medical system had “in a sense been turned inside out by culture contact”.15

According to Barnes-Dean, the therapies she describes had arisen out of processes occurring in the context of, and largely as a consequence of, the emergence of a relatively stable, colonial then postcolonial state. The preceding two decades had been marked by increased provision of biomedical facilities and the promotion of empirical causality in schools, churches, and government offices. But there was perhaps another factor influencing her findings too: she worked in the vicinity of a functioning hospital with missionary staff. The broadening of a general category of impersonal affliction in the area was surely linked to this provision of biomedicine and Christian influence. It is therefore likely that notions of ancestor veneration had not been so completely abandoned as she suggests, even if collective rituals seemed to have been set aside.

Interviews conducted by Tim Allen with elderly missionaries who had worked in Lugbara and Madi areas during the 1960s and 1970s, suggest that there was much criticism of pagan rituals by Christian groups at the time. Educated young people, chiefs recognised by the government and Christian converts were actively encouraged to abandon them. The effect was probably to weaken their significance at the time, rather than to eradicate them completely. Visiting Lugbara areas in the 1980s, it was not difficult to locate shrines. They were often to be found in long grass or disguised as a grinding stone or seat. Returning from refugee settlements in Sudan, people were often eager to reveal them, as a way of showing that this was their ancestral home. As we shall see, a similar process occurred among the Madi, where, more recently, shrines have again been targeted by Christian activists and have again been hidden away.

It is probable, furthermore, that Barnes-Dean's observation that enyanya had become the only local non-biomedical aetiology had rather less to do with the cognitive inversion she posits, than with an increased recourse to local-level courts. Records were kept in English, and, although it was usually impossible to make accusations of “witchcraft”, enyanya could be introduced into proceedings because it was translated as “poison”. Court records from the time examined by Allen in the 1980s indicate that this was common.

Madi healing in the 1980s

Living at Laropi in the 1980s, Allen found a very different situation to that described by Barnes-Dean. Although the division between interpersonal and impersonal aetiologies she had observed in the early 1970s was broadly evident, those seeking therapy would frequently combine contradictory conceptions of causality. Healers themselves, including those attached to the health centre, would commonly cross refer patients. The importance of biomedicine was recognised, and there was widespread self-treatment with syringes for what was perceived to be “malaria”, but there was very little conventional clinical therapy available.

Moreover, a situation prevailed in which acute competition for scarce resources and extreme misfortune were commonplace, in which provision of government services of all kinds was very weak and in which those formerly employed found themselves forced to farm and fish to feed themselves and their families. In the absence of an adequate system for resolving disputes and regulating social order, the population had to negotiate rules internally, including attempts to reinvigorate ancestor veneration. Conceptions of affliction causality were much more vigorously pluralistic than is suggested by Barnes-Dean, and could be grounds for heated argument. Concerns about inyinya – the Madi pronunciation of the Lugbara enyanya – were particularly intense and now there were few of the constraints on accusation in place of the kinds that Barnes-Dean noted in the 1970s. Occasionally, when therapy appeared to fail, there could be appalling interpersonal violence. Some idea of the range of healing on offer may be given by briefly describing the range of therapists available to the population of Laropi, and commenting on the sorts of affliction they dealt with. We mostly keep to the way Allen described these when he was writing about them in the 1990s. It will be clear that his work was, like that of Barnes-Dean, influenced by the work of Kleinman and Janzen, but with the latter's idea of quests for therapy being more central to his analysis.

Turning first to the healing of interpersonal aetiologies, there were the elders, who seem to have been the moral arbitrators of the past, much as they were among the Lugbara at the time of Middleton's fieldwork in the 1950s. In exile, the people of Laropi had often lived among non-kin in refugee settlements. In 1986, they had little option but to return to live in their ancestral lands, something that a number of them had not been doing before they left, since they had been working in other parts of Uganda. Old people, not necessarily those who are elders in terms of inherited lineage authority, tried to use this enforced recongregation to assert themselves in the fraught process of establishing a degree of community cohesion. They did this by drawing upon old beliefs, of which younger people often professed ignorance. They argued that harvests were poor because ceremonies should be performed to placate the ancestors and to cleanse the land of the blood spilled upon it. Similarly, laza, especially when the afflicted person was a former soldier, was linked to past antisocial behaviour, now being punished by outraged patrilineal ghosts. Stories were told of individual men dying or becoming deranged with remorse in the vicinity of ancestral shrines known as tumi. In these cases, however, elders were less involved in therapy than in explaining misfortune with reference to a morality they sought to promote. Oracles were not consulted, but the affliction showed justice being done.

In all the time Allen lived in Laropi, he never encountered an instance where a promise to perform a sacrifice at a tumi by elders was the only therapy adopted. Occasionally, it was combined with a biomedical cure. Thus, in the case of boy with sleeping sickness, his aged step grandmother insisted on a commitment to “feed” the ancestors before she would agree to his father taking him for treatment at Moyo Hospital. More often, elders were called upon to play a role only when a specialised healer divined that ancestral ghosts were involved. These specialists were often called “witch doctors”, the English word having been incorporated into the vernacular, and included “traditional” ojo, a Zande refugee from Sudan living in Moyo town, and new kinds of spirit mediums, sometimes also referred to as ojo, but more usually called ajwaki (singular: ajwaka).

When writing about the Lugbara in the 1950s and 1970s, Middleton and Barnes-Dean used the term ojo to refer only to those healers who were possessed by God as an adolescent, and who acted as diviner/herbalists. In the late 1980s, such figures were largely absent. Their work as midwives and herbalists had been taken over by semi-professionalised traditional birth attendants (TBAs) and daktaris (local doctors), and their work as diviners by ajwaki. Allen met only one elderly woman who claimed to be an ojo in the old sense. She dismissed her local competitors as fraudulent but respected the conscientious, trained midwife employed at the government health centre situated close to her home and was in fact registered as a TBA herself.

In the 1980s, it was normally the ajwaki that dealt with laza when an interpersonal cause is suspected. Unlike ojo, ajwaki were women permanently possessed by named, usually non-kin, spirits. One in Laropi was possessed by five spirits, including that of a murdered bishop. Ajwaki did not treat laza from ancestors or the retributions of the wronged deceased. These cases were referred to elders. But they could be efficacious in the treatment of inyinya and possession by wild spirits, particularly a kind of wild spirit usually referred to as jok (an Acholi word for spirit/ghost). Both inyinya and jok-possession were so common that many thought of them as daily hazards.

Since the people of Laropi did not live together in exile, even closely related neighbours might be strangers to each other upon their return there. However, the attempted reassertion of the moral and economic values associated with patrilineal relations meant that those not of the patrilineage were more suspect of having ole, the motive for inyinya, than others. Those who fell into this category included Muslim Lugbara from Aringa County, who operated as traders; migrants, like the Zande healer in Moyo town; men who were living with their wives’ families (and are therefore not of the patrilineage); and married women whose status was in some way problematic (perhaps because she had not produced children with her current husband or because the marriage had not been properly negotiated with her brothers). When inquiry was made into knowledge of inyinya, a frequent response was that the Muslims use it. But when it came to direct accusation, most instances involved recently married women for whom virtually no bridewealth payments had been made.

Talk of inyinya came up at most funerals, being initiated by maternal uncles of the deceased who use it as a lever in negotiations for compensation. Efforts were made by the men of the patrilineage to quash it, and angry scenes were common. Occasionally the suggestion was taken up within the patrilineage, and a woman thought of as angwe (the Madi equivalent of the Lugbara word amve [outside]) was blamed. In a neighbouring home to Allen's (on a day he was visiting Moyo town), a man whose son had died, accused the daughter (by a former husband) of his father's second wife. The girl and her mother were tortured horrifically and eventually beaten to death.

Violence toward women was not new in West Nile. Apart from Middleton's references to it, Rowley writes of the Madi “addiction to the use of various kinds of poison” and gives an instance of an old woman being severely beaten to extract a confession.16 In the 1980s, it was hard to establish if accusation was more common than it was in the past, but it is was certainly widespread. Moreover, ideas about inyinya had been influenced by the promotion of empirical causality, the introduction of biomedicine, the teaching of biology in schools, and experiences in exile. As “poison”, it readily allowed a scientific explanation to coexist with an interpersonal one, and even the compound of the MSF team in Moyo town was not exempt. A driver was convinced that one of the new female cooks had “poisoned” him, and a long meeting involving the French administrator was necessary to sort the problem out. Unlike the situation described by Barnes-Dean, inyinya was no longer thought of as an aetiology specific to local people. Harrell-Bond17 provided shocking examples of violence linked to “poisoning” among Ugandan refugees in West Bank camps; and when Allen had a severe hangover after drinking locally distilled liquor, it was speculated that Allen too had succumbed. A further issue that was becoming significant at the turn of the 1990s, was that inyinya was associated with HIV/AIDS. Government efforts to promote behavioural change, and in particular constraints on sexuality, were linked by the newly formed and directly elected local councils with exerting constraints on individuals deemed to be acting unacceptably or suspiciously.

Ajwaki interviewed denied being consulted over allegations made at funerals, but it was known that they could divine if inyinya was causing laza. When it was, they always claimed that they refused to reveal the name of the sorcerer, but that would not stop them making hints and offering advice about what to do. They did not, as a rule, administer herbal remedies to a patient (this was done by a daktari), but they could find “poison” put on a path or buried in a field that might afflict an entire neighbourhood, and received considerable rewards for doing so. They were also paid well for successfully dealing with jok possession, again a phenomenon closely associated with women.

The terms jok and ajwaka were recent imports from the neighbouring Acholi to the east. In Madiland, jok referred specifically to one or more named spirits that seized a victim, almost always a woman, and caused her to do peculiar things like sleep in a tree or dance wildly in the bush. Her male relatives would take her to see an ajwaka, who would arrange a ceremony at which she would go into a trance-like state and dance wildly with the patient and other women who came along for the occasion. Young men played drums, and quantities of strong alcoholic drinks were consumed. It could be very exciting. The spirits of the ajwaka would speak through her and call upon the spirits possessing the patient to reveal themselves and explain what they wanted. When they did so, in strange high-pitched voices, sometimes a grievance concerning the home was revealed. In one case, a woman turned out to be possessed by her dead father, who, through her, castigated her brothers as drunkards. The ceremony was relatively expensive, and often further ones had to be organised, particularly if a jok “loved” its medium and decided to stay permanently. If this was so, the woman might be initiated as an ajwaka herself.

It is tempting to see many cases of jok possession in the 1980s as a form of female expression or even resistance, rather than as sickness. It highlights the weakened control men had over women as a consequence of the refugee experience and the chronic instability of marriage. It also confirmed to men the irrational nature of women. Several times Allen heard men complain that jok dancing had become an epidemic simply because women enjoyed it. Men could be possessed themselves but, it seems, less benignly. This was so in the case of an ex-seminarian who attempted to murder his mother. He had formerly gone to Kampala for psychiatric treatment, which failed, and was taken to see an ajwaka. After three days of exorcism ceremonies, his therapy-managing group was instructed to tell his elders to ensure the success of the cure by performing a sacrifice at the lineage shrine.

Turning to those healers specialising in curing ailments with impersonal or God-given aetiologies, therapy for these afflictions was of two kinds: local or herbal remedies, and biomedicine. Local remedies were usually administered by daktari (who were normally men), or, in the case of pregnant women, by TBAs (who were women). Daktaris and TBAs were individuals who had responded to the half-hearted attempts to professionalise traditional healers since the late 1970s. TBAs were again being registered at the time of Allen's fieldwork, and the trained midwife at the health centre was supposed to improve their skills. The basic problem she faced was that the concept of a “traditional birth attendant”, in the sense of a woman who specialises in antenatal care and child delivery, was a fiction. Severe suffering in childbirth could be interpreted as symptomatic of unrevealed wickedness, and the few women who give birth at the health centre were thought to have done so because they had something to hide. The calling of ojo to attend births in the past was to discover the reason for the difficult delivery, to find out if the woman had committed adultery, or to determine if she had inyinya. In the late 1980s, TBAs generally denied being ojo, but in this respect they did sometimes fulfil a comparable role, particularly in cases of “illegal” pregnancy, that is, when the impregnator is not known and therefore cannot be fined by the woman's patrilineage. However, TBAs mainly prided themselves on herbal remedies, which, like those of the daktari, were revealed by God in dreams.

The best known daktari in Laropi equated himself with district medical staff and saw himself as working in conjunction with a small village dispensary. Patients were referred to him by the nurse working there, and he referred other cases to her. However, in spite of his high reputation, he seemed to earn very little from his clinic, and he complained to me that the government ought to pay him a salary. He also vehemently rejected any association between his practice and the work of “witch doctors”. Although he did not practise any form of divination, his medicine (like those of TBAs) was in fact quite similar to remedies used by the old, “traditional” ojo and to cures that, according to Barnes-Dean,18 were used by Lugbara ojo in 1973. His medicines were derived from dried leaves and roots, which he collected himself. Among other ailments, he treated various body pains, boils, tropical ulcers, worm and intestinal problems, hernia, measles, throat infection, and infertility. He also had different sorts of purges to cause vomiting or bowel movements, as well as potions to alleviate these symptoms. Purges were used in cases of inyinya where the victim had swallowed something alleged to be poisonous, and he would sometimes assist if inyinya had been inflicted externally by touching or by leaving it on a path. It was not his role, however, to locate a culprit. He treated inyinya simply as another sickness. Most medicines were drunk, but others were rubbed into scores of tiny cuts made with a razor blade on the skin of the patient in the manner to that described by Barnes-Dean among the Lugbara. It is possible that several of these cures could be assessed to be biomedically effective. Others, like the cutting out of the lower canine teeth of small children to cure diarrhoea, surely were not.

Part of the appeal of the daktari was that the availability of manufactured biomedicines was very limited. They were mostly brought into Moyo District either as supplies from the Ministry of Health store in Entebbe or as part of the medical aid programs administered by international non-governmental organisations, theoretically under Ministry of Health auspices. Unfortunately, district medical staff were so poorly paid that the only way that they could make ends meet was to find additional sources of income. The midwife at Laropi Health Centre made a living by selling a special sort of beer on Sundays. Others charged for services that were supposed to be free and, one way or another, large quantities of manufactured drugs ended up on the open market. There was a never-ending demand for aspirin, chloroquine, and antibiotics, all of which were generally available in the small shops of the Laropi market, tucked behind piles of matches, soap, and batteries.

In certain situations, an afflicted person would seek assistance from trained medical staff at an early stage. This was always the case with severe cuts or a broken bone. In instances where aetiology was more open to debate, there tended to be dissent between therapy managers over when, or if, a biomedical remedy should be sought. A few equated the use of biomedicine with being progressive and adopted it invariably as a first resort. This amounted to a public rejection of belief in inyinya, which could be dangerous if the patient's health deteriorated, because it might lead to suspicion that information about who was responsible was being hidden. Others would talk of hospitals as places where people go to die in the hands of strangers (as indeed they often did). However, in mild cases, usually with symptoms of bacterial infection, headache, or fever, which did not require a group of therapy managers, a visit to the health centre was unproblematic and common. Patients would not generally go there for diagnosis but rather to try to obtain drugs free of charge, and if informed that supplies had run out or that they were not really needed, there was the alternative of purchasing them for self-treatment from the market. Home treatments included the external use of tetracycline powder, derived from tablets, the consumption of chloroquine for malaria (a term used to refer to a wide range of symptoms), and the widespread abuse of penicillin injections, administered with reused disposable syringes.

The fundamental problem associated with biomedicine was perceived as one of access. The popularity of daktari and recourse to home treatments reflected this. People would say things like, “In your country, you do not grow old quickly because you have plenty of medicine.” If drugs were readily available, it was suspected that most laza with impersonal aetiologies could be cured. It was, after all, well known that the white expatriates had their own private supplies with which they regularly dosed themselves; so why did they not simply bring more and give them out?

MSF cars came to Laropi every other day when Allen was living there, but on many occasions people expressed confusion about what the French expatriates were doing. It seemed peculiar that they were so eager to inoculate children, which made them feel ill, but were reluctant to provide injections for adults. Rumours even circulated at one point that they were making a lot of money from experiments carried out on patients in Moyo Hospital. Personal relationships with them were, as a rule, impossible, because they stayed only briefly on visits outside of Moyo town, lived in an enclosure, were often not fluent in English, and were on short-term contracts. By and large, they were thought of not as real people at all but as a resource, like the UN refugee agency (UNHCR), from which things ought to be forthcoming on a regular basis.

Twenty years later

Returning to Laropi in 2004, Allen was able to reconnect with individuals he had known 20 years earlier, or in several cases visit their graves. In the 1980s, HIV/AIDS was becoming feared and, as noted above, was linked to ideas about inyinya and the enforcement of moral behavioural change. In retrospect it became apparent that several of those taking a lead in such activities were themselves at risk. Three of the brothers whose compound Allen had shared in the 1980s were said to have died of the disease.

Another development was the increased influence of the Catholic Church. That was in many respects a surprising observation. Attending church on Sundays had been quite an event in the 1980s, and the Catholic priests had been figures of enormous respect. At that time most were still Italian or European, so the ordination of a group of Madi priests was viewed as very significant. Two decades later, the Europeans had left and the church had become much more overtly Africanised. One aspect of that has been re-invigorated targeting of “satanic” rituals. Whereas among the Acholi of Gulu distict, where the Catholic Church had in various ways supported traditional reconciliation as a means of re-integrating those returning from the Lord's Resistance Army (LRA), in Moyo and Adjumani Districts, both ancestral shrines (tumi) and spirit mediums had become targets for Catholic activists.

Few ojo, let alone Acholi-style ajwaki, were now prepared to have noisy séances, or even to openly admit what they were doing. Individuals who Allen knew were possessed in the 1980s now were a little embarrassed to be asked about it. One of the best known ajwaki in the 1980s said that she has “left all those things behind”. She also explained that instances of spirit possession were rarer these days. She gave the following as a reason:

The spirits were those of the soldiers. All over this land there was bloodshed and the spirits of those killed lived in the bush. The women went to the bush to collect wood and the spirits would take them over. Now there is peace there are no new spirits, no more blood. We have chased away the ones that remained.

Another former spirit medium said that she had been forced to diversify her activities, offering local herbs for the treatment of barrenness and other afflictions in order to maintain her business, while also working in her neighbours’ gardens to subsidise her income.

Madi healing in 2008

In 2008, Allen co-directed a research programme in Moyo and Adjumani districts. One group of fieldworkers was based at Laropi. Laura Storm lived there for three months and worked with Allen on a restudy of local healing. Initially working independently, Storm found a situation that was in marked contrast to that described by Allen for the 1980s. Therapy appeared to have become much less complex and pluralistic, divided along the lines of the “traditional” practices of ojo and herbalists, who addressed laza caused by inyinya, and the medicines of the health centres, clinics and drug distributors. Many of those interviewed claimed that the two therapeutic pathways should not be used in conjunction, because there would be complications or the medicines would fail to work. With rare exceptions it was explained that the latter now took precedence.

When asked why this development had occurred, it tended to be explained by residents as the result of a three main factors. Those may be summarised as the consequence of relative social and political stability since the early 1990s, a corresponding moderate improvement in overall health, and an increased provision of biomedical therapies championed in particular by a new generation of educated Madis. Thus, Storm encountered an apparent return to the situation Barnes-Dean had alluded to from her fieldwork in the early 1970s among the Lugbara, in which the authority of elders was much diminished and ancestor invocation appeared no longer to be practised. Instead, people drew attention to their acute needs for biomedicine, and often professed to have completely set aside old ways.

Closer investigation together with Allen, drawing on his detailed knowledge of where shrines had been located in the 1980s and who had been involved in providing local therapies, revealed that things were a little more complicated. In the years he had been away, ancestral shrines had been hidden or desecrated with the help of local Catholic priests, who preached against ancestor veneration in their sermons and willingly responded to requests to destroy the shrines. The Laropi parish priest explained his reasons for encouraging this destruction:

[Ancestral shrines] tie people to the past, they remove their freedom. They block the people from knowing new things, from modernising. For example, an obvious disease, they will not go to the hospital but will perform rituals at the shrine instead. If such things are attributed to ancestors it is bad for health, people will delay in seeking medical help.

Many people talked of the importance of “opening minds”, or about “moving forward”, “development” and “modernisation”. Biomedicine – particularly pharmaceuticals – was strongly associated with these ideas. The “commoditisation” of health driven by the expansion of the private pharmaceutical sector in Laropi was cited as an example of progress. One man remarked, “Now we have education and medicines, we are no longer left behind or stuck in the past. Now we are in the future.” People had persistently expressed aspirations of this kind to Allen in the 1980s, but a difference now was that that they were a bit more than aspiration. There was also an eagerness to express to outsiders how much such developments were welcome, and how much more still needed to be done.

Similarly, on the surface, much of the recourse to interpersonal explanations seemed to have been lost to notions of empirical causality. As a grandmother on her way to take her grandson to the health centre reasoned:

Sometimes people connect their illnesses to social problems at home. They will seek help from a witch doctor. But the worms and the flies are here, along the river, these are what cause the sicknesses. So people should come [to the health centre] first.

Such a comment was not unusual; people were concerned about vectors such as mosquitoes, worms and lice, and attributed illnesses to them. Moreover, the attribution of empirical causation was not limited to instances of ill health. In 2008 the rains had not arrived and crops had failed leaving villagers desperately hungry. Rather than blame this failure on the neglect of rain shrines, in most conversations it was expressed in terms of a modern, scientific misfortune – global warming.

There was little doubt that this kind of thinking has grown in conjunction with the increase in provision of biomedical healthcare during the previous twenty years. New government health centres had been established in the sub-county, and private drug shops and clinics had proliferated in the trading centres rendering biomedical healthcare more widely available, accessible, and understood than ever before.

Also, relative peace and stability had influenced illness behaviour and healing choices. The transformation of social structures away from clan membership and the ritual moral authority of clan elders, and towards a focus on smaller “nuclear” style families and the authority of government-associated local councils, had encouraged a parallel move away from interpersonal explanations and an orientation to collective healing, towards individualised biomedical and pharmaceutical practices.

Government provision of primary healthcare services had improved and there were now six health centres in the sub-county, two of these in Laropi parish. The main Laropi Health Centre was long established, operating from its site in Ubbi South since the population returned from Sudan. In 2008, new, improved buildings served a daily crowd of women and children. This was the only health centre with a resident Clinical Officer, although personal and professional responsibilities often meant that in reality he was often absent. The nurses had between them many years of experience and training but some of the women waiting for treatment were doubtful of the adequacy of their knowledge and skills, complaining that “the doctor” was “always away”. Medical education was valued highly and for many people interviewed, the years that a healthcare professional had spent studying and training counted far more than clinical experience.

Attitudes to the midwife at Laropi is an exception. She had first worked in the main Laropi Health Centre back in the 1980s, and was an old friend of Allen. Although not a Madi by origin, over the years, she had established herself as a well-loved and respected member of the community. She could be found at the centre on most days, running antenatal clinics and voluntary counselling and testing programmes for HIV. Like other public servants (including school teachers), she had spent periods at other posts as part of a government policy of rotation in rural areas. She explained:

I have moved around a lot. The community will first study you and look at you. Then they will listen when you talk. They watch the way you work. Then they will discuss you: are you efficient? Do you do the right things? Do you get annoyed? Does your advice work? They will spread this around the community and eventually decide that you are to be trusted. If they do not they will not listen to your advice or come to you for treatment.

She had become so well liked in Laropi that the villagers wrote letters to the district medical officer protesting her transfer, although she did not return until a few years later. They general view was that they deserved, indeed were owed, someone of her dedication and training.

Despite the popularity of the midwife, however, the chief reason most potential patients visited the health centre was to obtain medicines, rather than to establish a specifically biomedical diagnosis. Also, while increased government provision of biomedical services and pharmaceuticals had been widely welcomed, services inevitably fall far short of rising expectations. On most occasions, after a brief examination, a person attending the dispensary would receive a prescription and instructions to purchase the named medicines at a private drug shop. The fact that the dispensaries were frequently out of stock of the most common and basic medicines, such as pain relievers, chloroquine and septrin (co-trimoxazole), caused much frustration within the community. There was some suspicion that staff at the health centres or at the district headquarters were selling drugs through privately owned enterprises that should really be given free. Nonetheless, paying for drugs has become established as a norm, and any free treatment available something of a bonus.

These privately owned drug shops can now be found in most neighbourhoods. There were five operating in Laropi trading centre alone, of which four also offered others services, such as laboratory tests and consultations. These establishments largely usurped the role of herbalists in providing accessible first line therapies, and were particularly widely used by people living nearby. The transactions and practices within the shops, which held government licenses, ascribed to biomedical approaches and were almost exclusively owned by healthcare professionals. The day-to-day running of the shop, however, was often taken care of by untrained family members or affiliated health workers such as laboratory assistants, who held no pharmacy training.

The varying degrees of training and expertise resulted in transactions characterised by a mix of medical and lay knowledge. After a customer had described their complaint, the assistant behind the counter would invariably focus on one symptom in order to formulate a prescription. If the complaint involved the stomach the customer would be given albendazole or “worm” medicine; if a fever or headache were present, chloroquine was offered for malaria. Antibiotics are often sold, particularly metronidazole and septrin and were thus well known to the customers. Bound up in such prescriptions were lay perceptions about symptomatology, aetiology and likely diagnoses and these encounters in turn contributed to an increase in popular biomedical knowledge. For example, a headache was considered a cardinal feature of “malaria” and rarely was the customer satisfied unless anti-malarials were offered alongside analgesia. Malaria was also associated with “infected blood”, which led some customers to demand injections rather than tablets so that the blood would be treated directly.

Allopathic medicines were now widely available and as such commonly used. They were considered to be a suitable first line treatment for almost all sicknesses. Indeed when Storm fell sick during her stay in Laropi the first thing people enquired was whether she “had tablets” and if so that she must swallow them in order to recover quickly. In comparison with herbal remedies, allopathic medicines were praised for being “strong” and quick to work: “These tablets reach the blood more quickly than the herbs because they dissolve in you. If you are really sick then you will be injected so that the medicine reaches the blood immediately.” People working within the trading centre would frequently pop into a drug shop to purchase single doses of pain relief capsules or anti-malarial tablets. This was more convenient than walking to the Laropi Health Centre, four kilometres along a hot dusty road.

Another component of the burgeoning private sector was a clinic held each Sunday at the largest of the drug shops at Laropi trading centre. The service proved immensely popular, with many citing its convenient time, location and guarantee of treatment. A large proportion of the population of Dufile sub-county (and others from further afield in Adjumani and Yumbe districts) gathered in Laropi on Sundays for the established market, which was housed in purpose built buildings funded by the European Union. Patients could buy flour, onions or groundnuts while waiting for their turn to be seen. A Clinical Officer born in Laropi, but now working in Metu, provided consultations in a designated room at the back of the shop, while the front remained a functioning pharmacy, serving a steady stream of customers. He described the provision of this service as a “duty” he felt he owed to “his people” in Laropi; it also bolstered his government wage. Curtains had been put up to divide the small back room into a consultation area containing a bed, desk and two chairs, and a “lab” where patients' blood, stool and urine were tested for diseases such as syphilis, typhoid, malaria and intestinal worms.

A queue of people began to form at about 10 o' clock, not long after the Catholic Church service finished and the market started to bustle. It was most often women and children waiting to be seen, although a father might accompany a child with a fever while the mother was busy selling or buying in the market. It was common for thirty or forty people to be seen in one day. Everyone was offered one investigation or another, usually a blood test to check for the presence of malaria or a stool sample to confirm intestinal parasites. Very rarely did anyone question the necessity of these investigations, or the prescription that always followed. De-worming medication, for example, was prescribed – as therapy or prophylaxis – irrespective of the stool test result. The fee for the investigations and prescription was similarly rarely disputed, and never observed to be waived, even for close family members.

Pharmaceutical therapy has also become available through the annual mass administration of drugs to combat neglected tropical diseases endemic in Moyo district including schistosomiasis (bilharzia), intestinal helminth infections ad lymphatic filariasis. The tablets were distributed freely under the auspices of the Vector Control Division of the Ugandan Ministry of Health and funding from international donors.

It was striking that the sometimes high level of fear and suspicion about the ‘real’ purpose of distributing free drugs in this way, found by Allen and the research team among other groups, such as the Alur population living close to Lake Albert, was mostly absent in Laropi, as it was in Madi area more generally.19 Almost all of those asked in Laropi about the distributions welcomed the intervention, and considered it to be a small step towards the help they felt the government should be providing. Increasing acceptance of biomedicine and familiarity with pharmaceutical therapy among the Madi contributed to the success of the programme: the tablets were readily accepted to be the best treatment available and the distribution proved a vast improvement on the alternative of visiting a health centre only to be told de-worming medicines were out of stock. The number of tablets an individual received was dependent on their height; the taller you were, the more tablets you were issued. It was not the taller Madis who complained about the number of bitter tablets they were ordered to swallow, but rather those who were shorter and felt they were missing out who demanded extra tablets from the distributors. Certainly there was little fear of overdosing or the side effects of the medication in comparison to the fear of sub-optimal therapy.

The distributions attracted large numbers of villagers of all ages, and queues quickly formed at parish distribution points. Often parents had kept children out of school to ensure they received treatment (although there was a parallel distribution in schools). Distributions also attracted people from further afield. Storm heard reports of people who had travelled from neighbouring districts, including Yumbe where the scheme is not in place, and Adjumani where some had missed out despite tablets being distrbuted. The livelihoods of people in Laropi revolve around the Nile River: farming its fertile banks or fishing its depths. Both of these activities were widely known to put people at particular risk of contracting “bilharzia” and “worms”, and it was common for concerns to be expressed about the detrimental health effects, and the need for regular tablets and medical treatments. Even fisherman who spent most of the year living on marshy islands in the Nile, largely cut off from the mainland during the heavy rains, were found to have obtained the distributed drugs at some point.

The mass drug administration (MDA) was viewed as a sign “that the government cared about the well being of the people of Laropi”. A woman in Logubu parish, who had postponed harvesting her cassava crop to come and receive the tablets explained, “[The government] knows that for us to be able to work we must be healthy and strong so they give us these tablets to prevent the common diseases found here.” A man at the same distribution told Storm he took the medicines he was offered because “it is what I deserve, and my family too, we deserve to be healthy and strong”. These comments reflect the notion of “pharmaceutical citizenship” as identified by Ecks, whereby marginalised people are afforded a higher degree of inclusion through the availability and attainability of effective pharmacological treatments.20 For the Madi, access to allopathic medicines represents the recognition of their needs by the state and the act of swallowing the small white tablets embodies their ties to a “modern”, “Western” and “global” world.

The expansion of the private sector, and the introduction of the MDA, had rendered biomedicine more accessible, but not universally so. Allopathic medicine was ideal, but not always available or affordable. In many circumstances, using local herbs was still the only option. A few families grew medicinal herbs in their compounds, and many knew where to find the source of commonly used natural remedies, such as the moringa tree, close to home. Herbalists were also still operating around Laropi, although on a much smaller scale. Some provided a service comparable to that of the daktaris Allen had encountered, although this term was no longer in use. They treated common biomedical symptoms with herbal tonics and topical pastes, the latter being applied through small nicks made in the customer's skin. Interestingly, although these herbalists did not condone the simultaneous use of allopathic medicines, many had adopted practices more commonly associated with biomedical therapy, such as using disposable latex gloves or administering the tonics with syringes.

Further investigation additionally revealed that there were other local healers too, also generally referred to as herbalists, who were consulted when biomedical therapy was deemed to have failed. Indeed, it became apparent that although biomedical knowledge and therapies had been embraced, local concepts of misfortune were far from forgotten. Indeed, a little prodding and a few leading questions revealed that inyinya remained a pervasively worrying issue. Those who had become relatively affluent in local terms felt they were particularly at risk. A teacher in Laropi explained,

People will see you prosper when they are struggling to eat. Of course they will want to take something from you.” Similarly, the owner of a drug shop in the trading centre who claimed to have been ‘poisoned’ in the local market commented: “I was her victim because she saw my success. I have a business and I earn some money … she wanted to punish me for having things that she could not have.

Just as Allen had found before, after spending time in Laropi people started opening up in this way. It did not take long before statements about inyinya being a thing of the past were replaced in any prolonged conversations about health and wellbeing with points about the risks inherent in forgetting or denying the its existence. Although here too modern ways were being explored, and a new ‘democratic’ mechanism had been introduced to reveal who were responsible. The process involved a voting procedure. The names of those suspected of inyinya were written on slips of paper and collected. Votes were then added up and a kind of trial was held. Careful records were kept of the proceedings and the decisions made. Those with the most ‘votes’ were confronted by all their neighbours, with no specific individual being responsible for the accusation of witchcraft. In the cases investigated, a handful of individuals ended up being expelled from the area. Their photographs were taken and put up on a tree, so everyone could know who they were. A few were also severely beaten; nevertheless, the procedure was openly discussed in terms of development and progress, and contrasted with the violence towards alleged witches in the 1980s.

Thus, in spite of statements about embracing biomedicine, witchcraft was an explanation shared by patients and health workers alike, especially when first line biomedical intervention seemed to fail. For example, when Anastasia's 18-month-old son developed a fever and diarrhoea she first visited Laropi health centre, a short walk from her home in the neighbouring village. She had experienced these symptoms herself many times and knew that “artificial” erwa Mundro (literally, White medicine) from the dispensary was effective. A nurse briefly examined her son and prescribed syrup for malaria and tablets for de-worming. Anastasia administered the medicine but returned to the centre six days later when the tablets had finished and her son's condition had worsened. The child was admitted to the paediatric ward, a small room containing four beds, and intravenous quinine treatment was commenced. When her son still cried and refused to feed after two days at the health centre, Anastasia consulted her family. Her sister advised her to take the boy to a herbalist in Dufile; “if the medicine at the health centre is not helping, perhaps he is not suffering from malaria” and the nurse at Laropi endorsed this plan. Anastasia explained that her son's sickness was “not of the medical people” but perhaps it had come from “the practices of a jealous Madi” and so required erwa Madi (local herbal medicines). If the herbs worked, she would know for sure. Here were echoes of the therapy management groups encountered by Allen in the 1980s.

Ancestor veneration too, it gradually became clear, had not been so entirely set aside as was suggested. Espousing a development agenda came at a price, and while everyone accepted the rhetoric, by no means everyone was so comfortable about the abandoning of the tumi. Several elders, both male and female, expressed a wish that there could be meetings at the old shrines. They lamented what they described as the loss of morality and setting aside of social responsibility. Now, they complained, everyone looks after themselves. Some said they feared calling the ancestral ghosts because it would incur their wrath. The ancestors had been neglected and their living relatives had done too many bad things since the tumi were last active. To reactivate them now would surely guarantee misery and misfortune. Meanwhile, when questioned closely on why old customs were no longer wanted, men with young families tended to become a bit more equivocal, or more honest. At one level they seemed to still recognise their potential importance, but found the talk of ending pagan practices useful. Partly confirming the views of elders, they said that they need to concentrate on paying their children's school fees, and were resistant to reinstating ancestor veneration, because the cost of the practices, including the provision of livestock for sacrifice, would fall upon them.

These ambiguities in attitudes to shrines was all too apparent when Allen and Storm inquired further about their destruction. Far from having disappeared completely, tumi had often moved quite literally underground. In many compounds the flat stones, which comprised the shrine, remained in situ, although cracked and covered with dusty earth. In one home, the clan shrine had not been broken but buried under the earth in a field now full of corn. Following the directions of the eldest living family member, his male relatives spent an afternoon digging up the field to eventually locate the shrine, upon which a large crowd gathered to reminisce and share stories. They agreed it was important that future generations should be told of the existence of tumi and understand their importance. Another day was spent pushing through overgrown vegetation to reach Laropi's abandoned rain shrine. This shrine, which had been so important in the years immediately following the return from Sudan, when the ancestral lands had to be cleansed and reclaimed, lay disused, concealed in long grass. Although the pots were broken, they still contained the implements used in the rainmaking rituals, including the precious rain stones. It seemed it was not that the rituals of the shrines were really thought to be wholly evil. It was that they no longer fitted with current priorities. But they are still there in reserve.

Conclusion

Approaches to healing in Laropi now echo those described by Barnes-Dean among the Lugbara groups she studied in the 1970s. A comparable process has occurred, whereby relative stability, Christian evangelism and increased provision of basic clinical therapy have simplified quests for therapy. More often than not, those interviewed stated that they would try a herbal or ideally biomedical cure before recourse is made to ideas of witchcraft, while possession cults and concerns about ancestral ghosts are talked about as things of the past. It would be easy for a visitor to conclude that clinical therapy and the use of manufactured medicine is becoming increasingly hegemonic, and traditional ways are being set aside. At one church service in 2008, a Madi priest even chastised the congregation for going too far, and being too willing to swallow any drugs on offer from district health staff or aid agencies without knowing what they were for. However, further inquires reveal that the situation is more complex.

Acceptance of biomedicine in Laropi is less dominant than it initially appears to be. While there certainly was a degree of awkwardness now associated with discussing the activities of ojo (at least to outsiders), which was not the case in the 1980s, other alternative explanations for illness remain prevalent. Publically, the view expressed is usually that the government and aid agencies are providing treatments, and it is appropriate for those treatments to be accepted. Yet, interviews with non-biomedical healers, and analysis of individual quests for therapy suggest that pluralism remains much more common than is suggested. Also, worries about witchcraft, while ostensibly more moderate than they were 20 years before, have not really abated. Indeed, the obvious inability to deal with afflictions arising from interpersonal causality by those trained in clinical therapy confirms that therapy for such problems needs to be found elsewhere. Those working in the dispensaries themselves recognised that.

For the Madi it is also important to understand that enthusiasm for biomedicine is not simply recognition of superior scientific knowledge and practice. As medical terms and treatments have been introduced since early Protectorate times, they have been linked to other ideas and given local meanings and interpretations. This is not in any way unique to the people of Laropi, nor is the fact that for the Madi, biomedicine has deep cultural associations. In certain respects, biomedicine is as traditional a body of knowledge as any other local approach to therapy. In various forms it has had a potent presence for almost a century. However, among the Madi, biomedicine probably carries more fundamental associations than their neighbours. It is integral to collective identity. This is a point that Allen highlighted in his earlier work, but has become much more apparent in recent years.

Before returning to that point, it is worth noting that while the observations about healing in this article reflect changing approaches to therapy among the Madi and Lugbara, that is only one part of the story. The publications of Middleton, Barnes-Dean and Allen additionally reflect changing analytical approaches in anthropology, and particularly in the sub-discipline of medical anthropology. In different ways, all three adopted fashionable analytical lenses to try to make coherent the muddled situations they confronted in their fieldwork. In each case, they simplified things in such a way as to clarify key issues, but at the same time tried to avoid making their ethnographic accounts seem as functionalist as those before them. The emphasis on the agency of those seeking therapy and on medical pluralism was intended to do this.

However, taxonomies of health sectors or therapeutic pathways can become misleading. They may imply that divisions are as clear as some healers themselves suggest; or they may imply that people always have the capacity or will to make informed choices. Therapies do not really exist separately any more than they form part of a structured health system. They borrow from each other and in practice syncretism is likely to be the norm, rather than the exception. In the 1980s, herbalists and spirit mediums in Laropi, whose remedies were revealed in dreams and by ghosts, would also engage in discussion about drugs from the dispensary and would use them on themselves and their loved ones. At the same time, trained clinic staff could be as concerned as anyone else about witchcraft. To some extent that is still the case. The healing of biomedical practitioners, let alone witchdoctors, is deeply affected by context, and therapy is often not based entirely on discrete and logically coherent bodies of knowledge. Ideas about healing tend to draw from bits of information derived from different places. Logical coherence is not necessarily prioritised and for clinicians, herbalists and mediums alike, there is a considerable amount of “suck it and see” in what they do.

In Laropi now, attitudes to therapy have certainly been affected by the increased availability of clinical medicine and manufactured drugs. Nevertheless, a multitude of treatments continue to be used and attitudes are more complex than they may first appear. It becomes apparent that medical pluralism remains not simply because a certain technique is thought to be the best way of dealing with particular problems, but because therapies are inextricably linked to broader issues and concerns – including who is to be recognised as a moral person and who needs to be socially excluded. Moreover, perceptions of biomedicine itself – or rather local conceptions of biomedicine – are far from straightforward. No doubt this applies to neighbouring groups too, but for the Madi ideas about biomedical therapy do have particular qualities and connotations.

The dreadful upheavals of the 1980s were more than matched by the effects of slave and ivory traders and invading armies a century before. Assessing what life was like in this part of Africa before the 1870s is little more than guesswork, but the evidence available suggests that there were numerous clans, speaking various languages. A Madi “tribal” identity is largely the product of governmental and missionary activities since the 1920s, and is associated as much with Catholicism sleeping sickness control measures and a desire to be recognised as legitimate citizens of the state as any cohesive pre-colonial ethnicity. For historical reasons, all these characteristics – religion, experience of public health controls and openness to administration – are perhaps more pronounced than among their neighbours in northern Uganda. A collective sense of political belonging and integration with the state, as well as a notion of Catholic social progress, has long been linked to the provision of biomedical health care. Those connections have been enhanced by the moderate degree of prosperity now enjoyed compared with the acute deprivations of 20 years ago, and rising expectations of young people.

A basic level of formal health care is increasingly viewed as a right by people in Laropi, and a service which the population is duty-bound, or even morally bound, to accept when it is offered. It was remarkable, for example, to see the enthusiastic response to mass treatment for the control of bilharzia (schistosomiasis) in the area. Scores of people were eager to be treated. They dropped what they were doing and immediately queued up to take tablets. This was not only about the need for individual biomedical therapy, the good humoured jostling for position was a way of collectively claiming access to services and publicly demonstrating enthusiasm. Elsewhere in Uganda, responses have been rather more mixed.21 The epistemological privileging of biomedicine by state and foreign aid agencies, with which the Madi have had much contact from the time that their distinctive “tribal” identity was forged under the Uganda Protectorate, has probably had more impact than among other groups. For the Madi, provision of biomedicine represents progress, modernity, and development more than just about any other governmental service. The indigenising of biomedicine is an aspect of their ethnicity – as much part of their culture as Catholicism, or their experience and understandings of witchcraft and spirit possession. It is part of what makes them special.

Notes

1. Allen, “Coming Home”; Allen, “Upheaval, Affliction and Health”; Allen, “Closed Minds, Open Systems”; Allen, “A Flight from Refuge”; Allen, “The Violence of Healing.”

2. Allen, “AIDS and Evidence”; Parker, Allen and Hastings, “Resisting Control of Neglected Tropical Diseases”; Parker and Allen, “Does Mass Drug Administration … Really Work?”; Allen and Parker, “The ‘Other Diseases’ of the Millennium Development Goals.”

3. Baxter and Butt, The Azande and Related Peoples, 104.

4. “La Nigrizia”, December 1937.

5. Middleton, Lugbara Religion, 252; Middleton, “Witchcraft and Sorcery in Lugbara,” 261.

6. Middleton, Lugbara Religion; Middleton, The Lugbara, 73–86.

7. Middleton, “Witchcraft and Sorcery in Lugbara,” 261–71; Middleton, The Lugbara, 75–83, 89–92; Middleton, “The Concept of ‘Bewitching’ in Lugbara,” 57–67; Middleton, “Spirit Possession among the Lugbara,” 220–31.

8. Middleton, “Witchcraft and Sorcery in Lugbara,” 266, 274.

9. Middleton, Lugbara Religion, 248–50.

10. Middleton, “Spirit Possession among the Lugbara,” 230; Casale, “Women, Power, and Change in Lugbara (Uganda) Cosmology,” 385, 395; Barnes-Dean, “Lugbara Illness Beliefs and Social Change,” 339.

11. Evans-Pritchard, Witchcraft Oracles and Magic among the Azande, 194.

12. Kleinman (1980), 51, cited in Pool and Geissler, Medical Anthropology, 42.

13. Janzen, “The Need for a Taxonomy of Health in African Therapeutics”; Janzen, The Quest for Therapy.

14. Barnes-Dean, “Lugbara Illness Beliefs and Social Change.”

15. Barnes-Dean, “Lugbara Illness Beliefs and Social Change.” 344.

16. Rowley, “Notes on the Madi of Equatoria Province,” 282.

17. Harrell-Bond, Imposing Aid, 309–12.

18. Harrell-Bond, Imposing Aid.

19. Parker and Allen, “Does Mass Drug Administration for the Integrated Treatment of Neglected Tropical Diseases Really Work?” For broader discussion of treatment programmes for parasitic diseases in the region, including among the Madi populations of Moyo and Adjumani districts, see Parker and Allen, “The ‘other diseases’ of the Millennium Development Goals”; Parker et al, “Resisting Control of Neglected Tropical Diseases”; Parker el al, “Border Parasites”.

20. Ecks, “Pharmaceutical Citizenship.”

21. Parker and Allen, “Does Mass Drug Administration … Really Work?”