Speaking with vampires and angels: the ambivalent afterlives of Christian humanitarianism in rural Zambia

ABSTRACT During the course of fieldwork at a Christian mission hospital in southern Zambia, I discovered that vernacular healers in the surrounding rural area were being visited by ‘angel spirits’ (bangelo) who offered them efficacious advice on how best to treat the patients under their care. According to the healers who encountered them, these angel spirits physically resembled white people (bakuwa), they dressed in white clothing, and their behaviour was inherently unpredictable. In this article, I consider what the presence of these angel spirits can tell us about moral attitudes towards humanitarian biomedicine in the region. But rather than focusing on these angel spirits alone, I situate them alongside a different non-human actor that has also been strongly identified with humanitarian biomedicine in southern Africa: the munyama or ‘vampire’. By describing the behaviour of the human and non-human actors who have been historically associated with medical humanitarianism in southern Zambia – vampires, angels, and European and American medical missionaries – I argue that it is possible to better understand why people in the region, from the mid-twentieth century to the present-day, have developed such a morally ambivalent attitude towards humanitarian biomedicine.

example, drank, smoked or rubbed into an incision). Sometimes angel spirits offered Dr Simamba advice about patients who were already under her care and living within her homestead, while on other occasions bangelo offered advice about patients who were going to visit in the future so that Dr Simamba could prepare in advance for their arrival. Although she tried her best to attract these spirits, Dr Simamba knew that she could not depend entirely on these unpredictable actorsand this is why she also relied on her longstanding knowledge of vernacular medicine during the times when bangelo did not make themselves available.
Dr Simamba was one of many vernacular healers in and around the rural town of Matamba, in southern Zambia, who were visited by bangelo. 1 During fieldwork conducted between 2014 and 2016, I interviewed and spent time with many healers who described their relations with angel spirits in strikingly similar terms. 2 As I outline in more detail below, these healers attempted to attract bangelo using a range of techniques and objects but, like Dr Simamba, they also described the difficulty of entering into predictable or stable relationships with these non-human actors.
My interest in these healers and their relations with angel spirits emerged during the course of fieldwork at the hospital that was nearby to Dr Simamba's home. Matamba Mission Hospital was built in 1957 during the final period of British colonial rule in Zambia (then Northern Rhodesia). Today the hospital is central to the local landscape of healthcare and American missionary doctors have continued to visit to offer biomedical treatment to the surrounding rural population. The small town of Matamba is situated along the central road that runs through Zambia's southern province and the hospital is one of the most prominent features of the town. While I lived there, thousands of people travelled to Matamba Mission Hospital. Some arrived alone, others with family members, friends, or neighbours. The majority were Zambians from the surrounding rural area, seeking treatment for a number of complaintsbroken limbs, tuberculosis, malaria, complications associated with pregnancy, HIV/AIDS-related illnesses, and many other afflictions. These visitors had mostly travelled from poor areas where the majority of the population were small-scale farmers who predominantly grew maize, alongside other crops such as ground nuts, sweet potatoes and sunflowers, ocassionally selling any surplus if the harvest had been a successful one. These people predominantly spoke dialects of Chitonga (although many knew other Zambian languages and some spoke English) and they walked, cycled, travelled by oxcart, or sometimes drove by car or motorbike to reach the hospital.
At the same time, from its establishment in the mid-twentieth century to the present day, the hospital has been visited by hundreds of European and American medical missionaries who have stayed for different lengths of time; some for several weeks, others for several years. I arrived at the hospital intending to study the relationships that were formed among the diverse group of people who encountered one another in this hospital. This was a clinical setting shaped by different expectations, moral ambitions and hopes on the part of patients, staff members, and visiting medical missionaries, and it was a site of profound material inequality. During the course of fieldwork, however, I was told by patients and staff members that many people sought treatment from the vernacular healers in the surrounding area, either before or after seeking hospital care. When I began to explore this, I was surprised (and initially rather troubled) to discover that these patients were visiting healers who were channelling the spirits of 'angels' who physically resembled white people. This led me to consider the place of non-human actors in the history of the relationships between Christian medical humanitarians and Zambians in the region. At a time when anthropologists have begun to critique the association of whiteness with 'biomedical expertise' in many humanitarian organizationsorganizations that continue to be 'dominated by white European and American practitioners' (Benton 2016, 4) it is discomfiting to find healing practices in postcolonial Zambia that involve non-human actors who physically resemble white humanitarians and who offer efficacious healing advice.
But how might we make sense of the presence of these 'angel' spirits within vernacular healing practices? What is the relationship between these angel spirits and the white medical humanitarians they seem to physically resemble? And does their presence in vernacular healing reflect a morally positive attitude towards the ongoing presence of white medical humanitarians in the region?
In this article, I argue that the presence of angel spirits in vernacular healing practices can tell us a great deal about attitudes towards humanitarian biomedicine in the region. But I do not identify the figure of the angel spirit as evidence that people have a straightforwardly morally positive attitude towards humanitarian biomedicine. Instead, these angel spirits need to be positioned alongside another figure that has been historically identified with biomedical practice: namely, the munyama or 'vampire'. Anthropologists and historians have written extensively about how medical missionaries, medical humanitarians and biomedical researchers have been associated with the figure of the vampire who is known to extract the vital life forces of Africans, such as blood and body organs (for example , Geissler 2005;Last 2013;Weiss 1998;White 1993White , 1995White , 2000. On the face of it, the figure of the vampire and the angel do not seem to have much in commonindeed, they appear to be morally antithetical to one another. And yet, when we consider their behaviour and their relationships with people in southern Zambia, it becomes possible to identify a number of striking similarities. As detached outsiders who do not conform to ordinary rules of sociality, neither angels nor vampires enter into predictable or stable relationships with human beings. Indeed, like many white people in the history of the region, these entities do not seem to be 'inhibited by human obligations' (Colson 2000, 335). They do not enter into relationships of mutual dependence or reciprocity with others and they are alarmingly self-sufficient. Furthermore, they are concerned with tangible and intangible goods (that is, blood and secret healing knowledge) that are outside of the realm of ordinary everyday exchange. The central difference between them, however, is that vampires take and do not give, while angels give but do not takea difference that I will explain further below. 3 The behaviour of these non-human actors can help us to understand how people in rural southern Zambia have historically understood their relationships with the diverse array of humanitarian outsiders who have been providing biomedical care in the area since the mid-twentieth century. And this analysis helps to correct the idea that we find in much of the regional and anthropological literature that foreign biomedical practitioners have been perceived largely in morally negative terms. Instead, in this article, I demonstrate that the attitude of many people in southern Zambia towards humanitarian biomedicine has been profoundly ambivalent. Outsiders who do not conform to ordinary social relations can be a threat when they take without giving anything back, but they can also be valued (albeit with certain qualifications) when they give without taking. This analysis can help us to better understand the morally ambivalent 'afterlives' (McKay 2012) that humanitarian biomedicine has produced in the region.
In making this argument, I draw on Scherz's (2018) recent suggestion that anthropologistsparticularly those working in Africaneed to pay more attention to vernacular healing practices and should situate these alongside biomedical forms of healing. Scherz points out that in recent years vernacular healing has been largely neglected by anthropologists who have tended to concentrate far more on biomedicine and global health. In southern Zambia, biomedical and vernacular healing practices have strongly influenced each other and adopting this approach enables me to situate bangelo within a longer regional history of biomedical and vernacular healing and enables me to draw on ethnographic material from sites of biomedical and vernacular healing.
In the first part of this article, I consider the anthropological literature on vampires and other associated figures. This literature has been criticized in recent years for its 'reductionism'. While these critiques are valuable, I suggest that they have diverted our attention away from a different issue with this literature: namely, that it paints a picture in which people are always expressing moral criticisms of the social relations that surround biomedicine. While this is important, it leaves little room to understand other attitudes towards humanitarian biomedicine that have also been present. Following this analysis, in the second section of the article, I narrate the history of Christian humanitarian biomedicine in southern Zambia in order to situate the presence of vampires in this history, before demonstrating how the presence of angels complicates this picture. If these nonhuman actors (both angels and vampires) are so often present alongside medical missionaries, humanitarians and other white outsidersand may even have arrived with them, as several healers suggested to methen it is necessary to describe their contradictory behaviour in order to fully understand how people in the region have regarded biomedical humanitarianism from the mid-twentieth century to the present.

Speaking with vampires
During the past 20 years, anthropologists and historians have described and analysed how a wide variety of colonial and postcolonial biomedical interventions have been perceived by African populations. There is now a large anthropological literature on the relation between biomedical practices and rumours, idioms, and discourses about vampires (for example, Fairhead, Leach, and Small 2006;Geissler 2005;Last 2013;Weiss 1998). Arguably the most influential scholar of this history is Luise White, who has traced the anxieties among colonized African populations as they encountered not only white European biomedical practitioners but also other white Europeans, including game rangers, fire fighters, mine workers and missionaries (see White 1993White , 1995White , 2000. This research culminated in her celebrated book Speaking with Vampires (2000), which had an enormous impact on debates about vampires and witchcraft in the following decade. In her work, White gives many examples of European settlers in Africa being associated with vampire rumours. The vernacular terms that White translates as 'vampire'which include mumiani, kachinja and munyamahave slightly different etymologies and histories, but they all refer to malevolent beings who extract the vital life force of others, whether this is their blood, flesh, or internal organs. As early as the nineteenth century, in parts of East Africa the term mumiani was used to refer to a powerful medicine that Europeans created using human blood and in colonial Kenya and Tanzania, by the 1920s, the term mumiani was used alongside kachinja to refer to Europeans (or Africans controlled by Europeans) who stole blood in order to create these medicines (2000, 10-11). 4 In later years, anthropologists began to encounter these rumours. For instance, Wilson (1959) found men and women in colonial southern Tanzania who spoke of Europeans who 'go around killing people and taking their blood for the Government to use for medicines ' (1959, 150). In colonial Zambia, the term banyama was more commonly used. This word derives from the noun for flesh or meat (nyama), which is common to several Bantu languages (including Chitonga, which is spoken in southern Zambia) and banyama therefore refers to people who extract the flesh and blood of others to enrich themselves. 5 In some of her earliest work on vampire rumours in Africa, White was keen to stress that we should not seek to offer any single overarching interpretation of these colonial-era rumours. As White put this in discussing banyama rumours in colonial Zambia: These images and characters had the power to terrify and explain because they touched on so many Zambian … experiences and concerns. They had intense meaning because they were told and retold in the vocabularies of peoples' daily lives and conflicts. For this reason no one interpretation fits all banyama rumours, no single analysis can explain how banyama accusations develop and then fade. Banyama rumours, like the blood extracted and abstracted in them, had the fluidity to describe many situations. (1993,772) White here suggested that their very 'fluidity' was a part of what made these stories so compelling for Zambians in a range of different contextsfrom their encounters with missionaries, whose behaviour seemed puzzlingly asocial, to their experiences with new biomedical technologies, such as intravenous blood extraction. However, when White subsequently discussed these rumours in her book, she suggested that, although these rumours could be used to describe a variety of situations, the similarity of these rumours (and their presence across such a wide geographical area) was because they enabled Africans to describe one situation in particular: namely, the colonial situation. As White suggested, given the violence and dispossession that defined colonial rule, it is unsurprising that 'Africans might have thought that colonial powers took precious substances from African bodies ' (2000, 18). 6 White's suggestion that vampire rumours were often metaphoric commentaries on the extractive violence of colonial rule became a central idea within anthropology in the subsequent decade and generated a great deal of debate. These debates were not only inspired by White's work on vampires but also coincided with Comaroff and Comaroff's (1999) similar argument about 'occult economies' and Peter Geschiere's (1997) work on the 'modernity' of witchcraft. While White argued that vampire rumours emerged in the midst of the extractive violence of colonial rule, Jean and John Comaroff attempted to explain the rise of extraordinary witchcraft and zombie rumours in post-apartheid South Africa in similar terms. For the Comaroffs, these rumours were best interpreted as a response toand metaphoric critique ofthe invisible forces of neoliberal capitalism (or 'millennial capitalism') that were generating such visible economic inequality in South Africa during the late 1990s. Similarly for Geschiere, it was important to stress that witchcraft was not an enduring element of 'traditional' life, but was often a recognizable response to 'modern' political and economic processes and dynamics (see also Moore and Sanders 2001;Meyer and Pels 2003;West and Sanders 2003).
These arguments have been criticized by a number of anthropologists. At the time, some argued that looking for such large-scale processesfrom colonial rule to neoliberalism to modernitywas a way of turning a 'general context into [a] particular explanation' (Moore 1999, 306;cf. Englund and Leach 2000). Instead, anthropologists should contextualize rumours in relation to local social situations and practices. This is an enduring and understandable anthropological response to large-scale analyses and, consequently, a number of anthropologists have attempted to situate the emergence of vampire rumours and witchcraft accusations in relation to smaller-scale dynamics such as local forms of inequality and 'social tensions' of various kinds (see, for example, Geissler 2005). However, even these more fine-grained ethnographic analyses have not satisfied critics who have taken up a different line of argument. In recent years, many anthropologists have argued that these approaches are 'reductionist' because they attempt to explain the presence of vampires, zombies, witches (and other entities) in people's lives in relation to other more 'real' processes, structures and political realities: whether this is colonial rule, neoliberalism, or modernity (see, for example, Pedersen 2011;Scherz 2018;West 2007). Instead, when we encounter people discussing the presence of vampires (or other entities) in their lives, we should take seriously their ontological claims about the reality of these actors rather than assuming that these are figurative or metaphorical claims that must ultimately be about something else (see also Holbraad and Pedersen 2018).
These persuasive criticisms have been influential within the discipline. Indeed, it does seem that scholars of vampires, witches, and other entities have failed to take seriously the reality of these non-human actors in people's lives by regarding them principally as metaphorical commentaries on other realities. However, one of the consequences of these debates is that anthropologists have arguably become preoccupied with theoretically convoluted discussions about reductionism and ontology (for example, Ellis 2015). And these kinds of anthropological debates can obscure a differentand I think more interestingproblem with much of the scholarship on vampires and biomedicine in Africa: namely, that it paints a picture in which African populations have responded to biomedical interventions in uniformly morally negative terms. Whether vampires are a reality in people's lives or a metaphorical language for describing other realities, the fact is that they are troubling and morally harmful entities. Historians and anthropologists have tended to focus their attention on the figure of the vampire alonewhether this is the mumiani, kachinja or munyamaand therefore the reception to new biomedical interventions seems to be relentlessly and enduringly negative. This ultimately prevents us from capturing the profound moral ambivalence at stake in responses to humanitarian biomedicine in the region. However, when we consider that vampires have not been the only non-human entities associated with biomedicine, it becomes possible instead to identify the more interestingly ambivalent attitudes that people have adopted towards humanitarian biomedicine. In the next section, I outline the history of southern Zambia and the presence of banyama within this history before considering how the figure of the angel might complicate this history.

White Europeans and banyama
The first Europeans to settle permanently in the region of present-day southern Zambia were agents of the British South Africa Company (BSAC) who governed from the 1890s until 1924, when the British colonial office began to rule the territory. In southern Zambia, it seems that some people initially regarded the behaviour of these Europeans favourably in comparison with the more aggressive Ndebele and Lozi raiders of former years, who had attacked many people and stolen their cattle (Colson 1962, 22). But these newly arrived Europeans soon implemented their own profoundly unpopular and violent policies. Discontent coalesced around the decision to remove local people from their land and resettle them in 'native reserves' so that the most fertile areas could be given to white settler farmers. When local people were evicted from their homes by BSAC officials (between 1918 and 1921) there were reports of 'the wholesale burning of entire villages, destruction of crops and floggings' (Vickery 1986, 124). Historians have identified this period as one that left a 'deep and lasting impression' (Vickery 1986, 124) and became a central memory in the formation of anti-colonial sentiments that persisted until independence (Dixon-Fyle 1977;Macola 2011). In the decades after British government rule began (from 1924 onwards) the people of the region were on the receiving end of a wide variety of colonial interventions. The most well-known and significant of these was the creation of the Kariba dam in the Gwembe Valley. This project to build a hydro-electric dam was conceived during the period of Federation (when present-day Zambia, Zimbabwe and Malawi were ruled together between 1953 and 1963) and it required 57,000 inhabitants of the area to be resettled so that their land could be flooded. 7 During this period, the people of the region also began to encounter American and European Christian missionaries for the first time. Although European missionaries had attempted, without much success, to settle in southern and central Africa from the midnineteenth century onwards, it was at the turn of the twentieth centuryand after the BSAC began governing the territorythat European and American missionaries started to establish themselves in greater numbers in the region. Catholic Jesuits and Seventh-Day Adventists (SDAs) were the first to arrive in the area of present-day Southern Province in 1905 and they have remained central to the Christian landscape of the area ever since (Colson 2006;Kirsch 2008). These missionaries began to intervene in the lives of rural Zambians by building clinics and hospitals and offering various forms of biomedical treatment. During the colonial period, Christian missionaries provided far more biomedical treatment than colonial governments ever did, particularly in rural areas (see Vaughan 1991, 56). After independence, relations between medical missionaries and the new Zambian government were 'far from cordial' (Kalusa 2014: 232), but the government continued to subsidize missionary biomedicine and in many areas of the country missionaries continued to play a prominent role in the delivery of healthcare. In many parts of southern Zambia these mission hospitals are still in placeand many of them, including Matamba Mission Hospital, are still visited by significant numbers of American and European medical missionaries today. This is a history in which outside actors have intervened in different ways in various domains of Zambian life. And it is not surprising that scholars of the region have suggested that 'the ever-present "project" from external actors' has come to profoundly shape the identity of the people of southern Zambia (Cliggett, Bond, and Siamwiza 2013, xxvi). It is within this kind of colonial history that scholars have situated the emergence of vampire rumours. White describes how all of the actors I have mentioned abovecolonial officials, white settler farmers, and Christian missionarieshave been accused of being banyama at different moments. White gives many examples of Europeans being accused of being vampires: shop owners who treated locals poorly; Catholic missionaries (the White Fathers) in the north of colonial Zambia who made their students perform agricultural work to pay for their lessons; and white settler farmers who were accused of abusing African children (1993; 1995; 2000). And banyama have also been present in the lives of people within southern Zambia, as I will describe in the next section.

Banyama in southern Zambia
During my fieldwork, I heard several stories of people being killed in the area for their organs. One of my friends who worked at the hospital, a man called Chipo who worked in the dispensary, told me on a quiet afternoon about how people in Matamba were sometimes killed for their body parts. When these people were found, they were missing internal organs. According to Chipoand several other people who told me about these casesthere were wealthy foreign businessmen in South Africa who enlisted local residents to kill their relatives and harvest their organs in exchange for large amounts of money. This is why some people in the town were sometimes able, mysteriously, to buy cars and other expensive luxury items.
Some of the earliest descriptions of banyama in the Southern Province are from the 1970s and 1980s at which time banyama were described as entities who stole people's body organs and bloodor even turned others into 'zombie labourers'in order to enrich themselves (Colson 2000). White settler farmers and Indian shopkeepers were accused of being vampires or, as Elizabeth Colson prefers to translate the term banyama, 'witch cannibals' (2000,340). During the mid-1990s in the Southern Province, banyama were described as people who stole others' body organs in order to sell them to buyers in other parts of the world. Colson suggests that these rumours were informed by global news stories about the trade in human organs (2000,340).
But banyama were also associated with the work of the hospital in Matambaand the Christian medical missionaries who could be found there. At Matamba Mission Hospital, Zambian staff members and American missionaries would occasionally travel to rural health postssome of which were remote and took many hours to reachin order to collect blood. These 'blood drives' were necessary because a large number of operations were performed at the hospital, particularly caesarean sections, and the laboratory was perpetually short of blood. When staff members organized blood collections, they arranged for community health workers to request that people in the area assemble at their nearest health post. This procedure was also used when clinical staff organized vaccinations or health promotion campaigns. One of the difficulties, however, was that the communication between volunteer community health workers and clinical staff members at the hospital was often poor or involved misunderstandings and confusions about when and where to meet. On several occasions when I visited rural health posts with hospital staff members and visiting missionaries, we found that nobody was expecting the visit because of a misunderstanding about the date or time. Unexpected arrivals of this kind often seemed to incite a great deal of anxiety and this was heightened when missionaries and staff members were arriving with the aim of taking blood.
A visiting American missionary named Karen described to me what happened when she once visited a rural health post in order to collect blood. After travelling for several hours by car with several men from the hospital's laboratory, Karen found that there had been a misunderstanding and there were very few people at the health postalthough there were enough to try and take some blood donations. The Zambian staff members explained to the community health workers that they wanted to assemble volunteers to give blood to help people at the hospital. According to Karen, this seemed to cause a considerable amount of anxiety among several of the patients who were visiting the health post. Karen told me that there were lengthy discussions between the Zambian hospital staff members and the community health workers before it was decided that the staff members should give a speech to the people who were available. One of the staff members from the laboratory gave a speech in which he explained that they had not arrived to steal people's blood and they were not 'vampires' (or working on behalf of vampires). They were not going to sell the blood or use it to enrich themselves, rather, they needed the blood to help people at the hospital. On this occasion, few people were prepared to give blood and Karen and the Zambian staff members moved on to another health post where people had been informed of the plan and were more willing to give blood.
From the perspective of the people at the health post, who were unprepared for the arrival of this white visitor and her Zambian colleagues, it would have seemed that these outsiders had arrived unexpectedly in order to take blood. Furthermore, during these blood drives, there was often no real screening before people were asked to give blood. The policy of the hospital was to take blood before testing whether it was safe to use. So the blood would be taken back to the hospital where it would be tested for HIV, Hepatitis C, and so on, before it was stored and subsequently used. It is perhaps unsurprising that this kind of unexpected and non-selective form of blood extraction was regarded with suspicion.
Medical anthropologists have found similar suspicions and anxieties surrounding the taking of blood for the purposes of medical research (see Geissler 2005;Fairhead, Leach, and Small 2006). One of the suggestions made by anthropologists who have studied the relation between biomedicine and vampires is that biomedical practitioners whether they are humanitarian doctors or medical researchers taking blood samplestend to unsettle the expectations and rules of sociality of those with whom they engage. In other words, arriving somewhere, taking blood samples and then leaving is not a normal way of conducting oneself and is, understandably, viewed with suspicion.
And this is what makes the vampire similar to a range of other malevolent figures, including the witch, the zombie, and so on. These entities tend to violate certain central ideas about how one should conduct oneself within moral relationships of mutual obligation and dependence (Englund 2008;Ferguson 2013). In the Lowveld area of South Africa where Isak Niehaus worked, there was a 'perception that witches used zombies to compensate for their lack of helpful dependents' (2005,202). In other words, the seemingly self-sustaining and asocial position of certain people provoked a considerable amount of anxiety that they were not real social persons. Not only are witches selfsufficient without appearing to have any dependents who care for them, but they cannot eat with other people (Green 2005, 225); they cannot have children (Comaroff and Comaroff 1999, 289); they are active at night time instead of during the day (Green 2005, 251); and, ultimately, in various ways, they demonstrate 'the inversion of normal human attributes of sociality' (Green 2005, 251).
These are the behavioural characteristics associated with witches, vampires and other malevolent actors who are capable of causing profound harm to human beings. But these are not the only non-human actors who behave in an asocial or unusual manner. And this is why it is important to consider, in the next part of this article, the figure of the angel and its strong association with humanitarian biomedicine.

Bangelo in southern Zambia
The word bangelo was probably introduced in the region with the first missionary preaching in Chitonga and translations of the Bible in the early twentieth century. But, as often happened in missionary encounters, the term came to be used in ways that exceeded the expectations and understanding of missionaries themselves (cf. Comaroff 1991, 1997;Keane 2007;Pels 1992). It was during the 1970s that healers and diviners first began to be visited by the spirits of 'angels' (see Colson 2007;Keller 2007). Bangelo spirits gave diviners the ability to identify the cause of illness or misfortune, whether this was an evil spirit (masabe), witchcraft (bulozi), or the displeasure of ancestral spirits (mizimu). In one sense, the history of angel spirits is a history of the way in whichto use the words of the historian David Gordon -'Christianity [has] populated the invisible world with new spirits' (2012, 12;cf. Robbins 2007), but the angel spirits who visited healers were quite distinct from other non-human actors in the region. Indeed, it is important to consider that bangelo occupy a distinctive position relative to other entities, both Christian and non-Christian. Firstly, angel spirits remain distinct from other spirits, such as ancestral spirits (mizimu) because they do not abide by the same rules of interaction as ancestral spirits and they are present principally in sites of healing and divination. But, secondly and perhaps more surprisingly, they are also quite unlike many of the other nonhuman actors that are often present in Christian contexts. This is worth pointing out because Christianity is pervasive in Zambian social life (see, for example, Cheyeka,Hinfelaar, and Udelhoven 2014;Haynes 2015;Gordon 2012). I have already described the early arrival of Christian missionaries in the region, but in the postindependence period Christianity continued to influence social and political life. In the post-independence period, the new President, Kenneth Kaunda, promoted a political ideology he called 'Humanism', whichdespite the nameinvolved the promotion of Christian ideas and values. As Megan Vaughan has described it, Kaunda's 'Humanism' was a mixture of 'Fabian socialism, nineteenth-century liberalism, Christian morality and [the] idealisation of the communal values of Zambia's pre-capitalist past ' (1998, 178). From the 1980s and 1990s onwards, the rise of Pentecostal and charismatic forms of Christianity has had a significant influence on Zambian politics and society and, in 1991, the newly elected President declared the country to be a 'Christian Nation' (see Haynes 2015). As historians of Christianity in Zambia have recently pointed out, however, it is important to note that the mission churches were not supplanted or eclipsed by the rise of Pentecostal and charismatic forms of Christianity and Christian practice within Zambia today remains extraordinarily diverse (Cheyeka et al. 2014).
This makes it tempting to situate these angel spirits squarely within this historical and religious context. However, it is important to note that they were quite distinct from the angels who are spoken about by many Christians in Zambia. Indeed, it is illuminating here to compare the physical appearance and behaviour of these angel spirits to other Christian spirits. In a compelling essay on the presence of the Holy Spirit (muya usalala) in the lives of Christians in southern Zambia, Kirsch (2013) describes in great detail the 'multifarious and wondrous ways in which the Holy Spirit is here experienced to locate itself in and move through space ' (2013, 34).
As Kirsch points out, the Holy Spirit is an entity that, for the most part, 'human beings cannot control ' (2013, 41). This is very similar to the bangelo spirits who visited healers. However, as Kirsch also points out, the Holy Spirit is an 'inevitably unbound and evanescent entity' (2013) and the 'morphology and scale' of this entity is radically different to that of human individuals. The Holy Spirit is defined by its 'quintessential unboundedness and evanescence' and it 'perceptibly changes its position in space by making itself present at different locations at the same time' (Kirsch 2013, 39, 47). This is very similar to the way in which Christians speak about angels in their ordinary lives. One of my friends at the hospital, a man called Thomas who drove the hospital's vehicle, described to me how an angel once prevented him from dying in a fatal car crash by guiding him, at the last moment, to steer away from an oncoming vehicle. Thomas did not see this angel, but described how he had felt its presence as it took control of the situation. This is similar to the Holy Spirit. This is why Kirsch is correct to note that when describing Christian spirits it is often crucial to avoid 'the problem of methodological individualism ' (2013, 48). It is unhelpful to think of them as bounded individual entities whose morphology resembles that of human beings. By contrast, when it comes to understanding the bangelo spirits who visited diviners, they very closely physically resembled human individuals in their morphology (indeed, they were much like vampires in this respect). This can be seen in more ethnographic detail by describing the behaviour of bangelo as they encountered healers.

Healing and bangelo
How did these angel spirits behave and appear when healers encountered them? As we have seen, Dr Simamba placed a white feather in her shrine and, like other bangelo healers, she dressed in white. Another bangelo healeran elderly man named Matthew told me that he also dressed in white when he attracted angel spirits. But he did not use white objects; instead, he would play a guitar because this is an instrument associated with white people: 'the whites came with guitars, that is their instrument'. Another healer named Mutinta, who had also been a bangelo healer for many years, told me that she also dressed in white, but when she wanted to attract angel spirits she would make sure that she ate bread instead of nsima (the staple maize porridge). She said, 'when I wanted the white person [mukuwa] to come I would not eat nsima because I know that they don't eat nsima, so I would be eating bread instead.' Not only were these entities attracted to objects associated with white peoplewhite objects and clothing, guitars, and breadbut they also spoke in English (Cikuwa) to these healers. None of the healers whom I interviewed could speak very much English but they each explained that the angel spirits with whom they communicated would speak to them in English and they were able to understand it. As Matthew said, '[the angel spirit] speaks English [Cikuwa] and when he speaks to me in English he is able to speak through me in Chitonga'. Mutinta said something similar when she pointed out that, 'I don't know English, but whenever this man came he would speak to me in English and I would be able to understand him'. These angel spirits resembled white people, they spoke in English, and they were attracted to objects that the healers themselves tended to associate with white people. This is quite unlike the way in which Christians in Zambia describe angels in religious contexts.
The healers I encountered were all keen to point out that the advice that bangelo offered to them was highly effective. Matthew explained that his healing was so effective that nurses from the mission hospital would send patients to him that they could not heal within the hospital. Matthew explained that he had a house next to his own for patients to stay while he was treating them. Once he had healed them he would tell them to go back to the hospital for the nurses to confirm that the patient had been properly healed. Mutinta said that her healing had been so effective that she was invited to move to Zimbabwe in order to cure people there. She said she would have been paid a great deal of money to cure people there, but she turned it down in order to look after her mother in Zambia. The first time I met Dr Simamba she told me that she could prove how effective the advice from one of the angel spirits had been. She instructed a young woman who was staying with her to bring her son to see us. The young woman arrived with a healthy-looking little boy and Mrs Simamba explained how close to death this boy had been before she treated him. 'The boy was taken to UTH [the University Teaching Hospital in the capital city, Lusaka] and even those doctors over that side could not cure the boy', she said.
When I asked healers about when they had encountered angel spirits for the first time, they often explained that it was during a period in which they themselves were ill. Dr Simamba described the time when she first encountered angel spirits: I became sick and during the sickness I began to see and dream about a white person who was standing next to a particular tree, which would indicate to me that the tree was where the right medication was to be found. When I was sick I didn't know that I was possessed, but during the illness I came to realise that the spirits within me wanted me to start healing the community.
Dr Simamba's account resembles the descriptions of diviners and vernacular healers throughout the region. In her work on vernacular healing in Tanzania, Stacey Langwick notes that many people who become healers are 'called into relationship with a variety of [new] actors ' (2011, 94) for the first time during their illness, including spirits and other non-human actors. And for bangelo diviners, one of the non-human actors with whom they were 'called into relationship' was the angel spirit.
But the bangelo diviners I interviewed were keen to point out that the relationships they entered into with bangelo were not like the relationships into which humans typically entered with other spiritssuch as, for example, ancestral spirits (mizimu). Historically, people in the region have had a number of clear and enduring obligations to ancestral spirits. They have historically taken their relationships with ancestral spirits (mizimu) seriously, although this is less common today, particularly among Christians from certain churches, such as SDAs and many Pentecostals. These relationships with ancestral spirits were essentially hierarchical patron-client relationships, but people were not utterly dependent on particular ancestral spirits. When they felt that certain ancestral spirits were failing to meet their needs, people would switch their allegiance to other ancestral spirits at alternative local shrines in much the same way as dependent clients might have switched their allegiance to alternative patrons in certain circumstances (for example, Miers and Kopytoff 1977;Ferguson 2013). As Colson described these relationships, 'spirits, like people, were expected to observe the rules of reciprocity … [and] if they failed to make return when approached … they were told in no uncertain terms that they would be forgotten and people would seek more responsive patrons ' (2006, 29;cf. Langwick 2011, 111). Ancestral spirits had formerly been human and adhered to human-like social behaviour. But bangelo spirits did not.
Much like the Christian missionaries whom these angel spirits often physically resemble, bangelo spirits have stood enduringly outside of local relationships of mutual obligation and dependence. The erratic and unpredictable behaviour of these bangelo was something that healers responded to in different ways. When I first met Dr Simamba she was actively working as a healer and was certified by the Traditional Health Practitioner Authority of Zambia (THPAZ). For Dr Simamba, it was important that bangelo spirits visited her and she found it troubling when they did not arrive. Other healers described their relationships with bangelo in similar terms, although they had different attitudes to the fundamental unpredictability of these non-human actorssome of the healers who were elderly or who had largely stopped practising did not mind that they were not visited very often, while others felt frustrated and confused by the sudden departure of the bangelo who used to visit them.
Matthew explained that the angel spirit who used to visit him no longer arrives very often. But, for Matthew, this was not a problem because he was no longer regularly practising as a healer. A middle-aged healer named Rosie explained to me that she had not encountered any of the bangelo spirits who used to visit her for several years. For Rosie, this was deeply disappointing. She explained that she used to be visited regularly. Like the other healers, she had first encountered bangelo during the course of her own illness. When I asked Rosie when she had last encountered bangelo, she told me about the last dream which involved one of the angel spirits (or the 'white man' as she called him at this point in the discussion) who used to visit her often: 'I was flying with the white man [mukuwa] in the dream. I put on white clothing and we flew until we reached the roof of the sky … Then the white man went through the roof but I couldn't go through it and I started to fall down on my own'. In this dream Rosie seemed to be returning with the angel spirit, but this was cut short at the last possible moment. She had not encountered any bangelo in person or in dreams since this time.
In certain important respects, the behaviours of vampires are not so different from those of angel spirits. Vampires and angel spirits arrive and leave unexpectedly, they have their own pre-existing ambitions, and they are not dependent on anybodyindeed, they are both disconcertingly asocial and self-sufficient. Angel spirits arrive unexpectedly, dispense medical advice and assistance, and then they leave without requesting anything in return. But they decide when and why they are going to arrive. Vampires behave in a similar way, although their role is reversed. They arrive unexpectedly in order to steal the blood and body parts of others and therefore enrich themselves. In the final section of this article, I suggest some of the ways in which the behaviour of vampires and angels can help us to understand attitudes to humanitarian biomedicine in the region.
Giving, taking, and postcolonial humanitarian biomedicine As many anthropologists have noted, the behaviour of humanitarian actors is regarded as unusual in many contexts. In her work on humanitarianism in India, Erica Bornstein notes that the expectation is that 'one helps those one has relationships with, not abstract others … That humanitarians help others with whom they have no connection is what makes them distinctive' (2012,(146)(147). The same is true of life in rural Zambia. In their everyday lives, people have tended to receive help from those with whom they have pre-existing relationshipswhether these are other humans or even the spirits of ancestors (mizimu). Anthropologists of southern Africa have long known that people in the region have made use of patron-client relationships and networks of mutual support in order to sustain themselves and their relatives, often under conditions of profound economic change and material hardship (on this work see Evens and Handelman 2006;Schumaker 2001;Werbner 1984). Anthropologists working in the region more recentlywho have been shaped by this anthropological traditionhave begun to use the concept of 'dependence' in order to describe people's reliance on others within enduring relationships and networks. In her work in Botswana, for instance, Julie Livingston (2007) has described how elderly women actively work to 'maintain' their 'dependencies' on their children and neighbours in order to continue to receive care and assistance during their later years (cf. Cliggett 2005). Harri Englund, writing about street vendors and their clients in urban Malawi, refers to their relationships of mutual reliance and 'loyalty' to one another as 'deliberate dependencies' (2006, 189;see also Englund 2008; see also Ferguson 2013).
People in the region have typically needed to maintain a series of diverse relationships with various relatives, neighbours and acquaintances upon whom they depend, in different ways, and towards whom they have many ongoing obligations and responsibilities. Some of these relationshipsthat is, those with elders and with ancestral spiritsare more hierarchical, while others involve a greater degree of egalitarianism and mutuality. These relationships also involve economies of exchange in which money, food and other resources (both tangible and intangible) are central. But these are also networks of exchange and redistribution from which certain objects and goods are excluded. For example, body parts and blood (which are associated with sorcery and witchcraft) and also certain forms of esoteric and secret knowledge, which are acquired by healers and diviners. 8 These are not ordinarily given or taken within everyday relationships of obligation and dependence.
Vampires and angels are entities who simultaneously stand outside of these everyday relations and engage in the giving and taking of goods that are typically excluded from everyday networks of exchange and redistribution. This makes their positionand their giving and takinghighly distinctive within the local context. Angel spirits, for instance, have no real connection to the healers whom they assist: they arrive when they wish to, they leave, and they are difficult to attract.
As I have been attempting to show in this article, bangelo and banyama are more similar than we might initially imagine. Humanitarian action, like malevolent action, offers 'little scope for reciprocity' (Keane 2016, 257; see also Wintrup 2019).
But what is the relationship, then, between these non-human actors and human actors whom they seem to resemble? One of my central claims in this article has been that in order to understand people's attitudes to biomedicine in the region, we need to understand their attitudes to the various non-human actors who have been associated with biomedicine. And, in doing so, we need to move beyond the preoccupation with vampires and malevolent entities in order to recognize that other less morally troubling entities have also been associated with humanitarian biomedicine. In this sense, we do not need to have a theory of how these entities are connected; it is enough to describe that they are a part of an assemblage of human and non-human actors who are often present in contexts of humanitarian biomedicine.
But some healers had ideas about how angel spirits and white people were related. One healer told me that it was possible that these angel spirits first arrived with white people. He suggested that just as Zambians have ancestral spirits, white people have their own spirits. And if we follow this idea then white people bring with them certain entities and these entities tell us something about white people; they are not metaphors for the behaviour of white people, but they behave in ways that resemble them.
The various humanitarians who have been visiting the region have tended to avoid becoming incorporated within locally recognizable relationships of obligation or dependence. The medical missionaries I encountered in Zambia lived highly isolated lives while they were in the country. And they have historically given something that has stood outside of local networks of exchangea kind of healing knowledge and skill. At the same time, they have often arrived in order to take the vital life forces of Africans: blood and body parts. In this sense, it would not be surprising if some of the nonhuman entities who these medical humanitarians brought with them exhibited similarly unpredictable and surprising behaviour. And the moral attitudes that people in the region have developed towards human humanitarians can be better understood by considering, at the same time, the moral attitudes that people have adopted to the many nonhuman entities with whom they have been associated.

Conclusion
By attending to the behaviour of the non-human actors who have arrived in rural Zambia, I hope to have shown that we can learn something about people's moral attitudes to the various human actors who have been arriving in the area to offer biomedical assistance since the 1950s. These moral attitudes are profoundly ambivalent. Like the human actors whom they resemble, the angels and vampires who have been present in the area have tended to behave in a way that is troubling: they arrive from a place far away and do not fully participate in ordinary social life. Even when these actors provide valuable assistance, there is still a suspicion that they cannot be relied upon. During a conversation with Matthew, the elderly healer, I asked him about the missionary doctors who have been working at the hospital. I asked if he thought they had been doing good work: 'I feel that [they] should have been staying longer. We want a good doctor here who the people develop confidence with. But the problem is that people come and then the missionary doctor leaves and then the people lose confidence again.' Like bangelo, these missionaries could not be relied upon.
To return to the question that I raised at the beginning of this article, it is worth considering what we are to make of the presence of white angel spirits in postcolonial rural Zambia. They were certainly not regarded as 'white saviours' in the old nineteenthcentury model (Vaughan 1991). Rather they were actors who offered assistance that was sporadic, intermittent, and essentially unreliable. The desire of many of the healers and patients whom I encountered was that these American visitors might develop relationships with local people that were more enduring or predictable. Angels are not simply morally praiseworthy figures. The idea that medical humanitarians in Zambia are somehow outside of local relationships, and therefore not bound by obligations to Zambians, is one that might explain their association with vampires and angels. While I am not saying that these non-human actors are metaphoric representations of the 'real' white people whom they represent, it is important to consider the behaviour of these human and non-human actors alongside one another. People's moral attitudes towards vampires and angels can tell us something about their attitudes to the human actors whom these non-human actors have, at times, so closely resembled in various ways.