The Ndoki Index: Sorcery, Economy, and Invisible Operations in the Angolan Urban Sphere

In this text I propose to discuss local conceptions of spiritual power in the city of Luanda, Angola. More specifically, I explore how the spiritual field in this city is configured through the markers of a plural ethnicity and Christian practice. In this framework, sorcery (ndoki) has become the practical element that connects Christian and traditional spirituality politics within the same discursive regime, addressing problems of urban life and social change. I suggest that both Pentecostal/Charismatic Christianity and traditionalists engage in competing, correlated and mutually defined ideas of invisible agency, associated with the image of financial market activity—the prediction, explanation, and fabrication of causality in social life. From this perspective, ndoki appears as an index of religious activity in urban Angola, much in the same style as the Dow Jones does for global financial markets.

emergence of sorcery is contiguous to the urban experience and participates in its moral and economic dimensions.

InvIsIble notes concernIng Ndoki
Whenever I am debating contemporary Angolan politics with my friends in Luanda, the part that I find more interesting is that the conversation somehow yet inevitably ends up in a discussion on sorcery and its effects. 2 This association of witchcraft and politics, as we know, is not specific to Angola, but is rather a theme that pervades African social life (see, e.g., Englund 1996;Geschiere 1995;Comaroff and Comaroff 1999;Moore and Sanders 2001;Ashforth 2005;Marshall 2009). Thus what seems interesting is the recurrence and social pervasiveness of this politization of witchcraft throughout time and with an intriguing superficial sense of continental continuity in what comes to such perceptions. In this text, in complement to these points of view, I propose to explore their ramifications, such as the outpouring of sorcery into other dimensions of social life-namely morality and economy-in the urban settings of Angola.
In Luanda specifically, considering the significant place of the Bakongo ethnicity in its current social, cultural, and demographic configuration, sorcery is often identified with the kikongo name ndoki (or kindoki), vulgarly translated as feitiçaria (in Portuguese) or sorcery. 3 As I have argued elsewhere (e.g., Sarró et al. 2008;Blanes 2014), the Bakongo play a highly relevant yet problematic role in the contemporary history of Angola. On the one hand, they were pivotal in the Angolan liberation wars (1961)(1962)(1963)(1964)(1965)(1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973)(1974), hosting the guerrilla and military movements that would then proceed to become the political protagonists of Angolan independency (see, e.g., Rocha 2003). They were also instrumental in the development of transnational commercial networks and the introduction of trade goods into Luanda, and in the dramatic urban development of the capital observed since the 1970s (see, e.g., Lukombo 2011;Pereira 2015). On the other, they remain associated to ideas of 'foreignness' and 'backward-ness' (traditionalism) that are antithetical to the sociopolitical project of the ruling political elite (MPLA), based on the enforcement of centralized national integrity and modernistic, iconoclastic ideas of progress (Schubert 2014). This is related to what is widely perceived as the Bakongo's dubious political allegiance-connected to the memory of the ancient Kingdom of Kongo 4 -and the centrality of 'tradition' in their social, religious, economic, and political configurations, where ndoki plays a central role. Within this framework, there are active narrative and semantic configurations that associate this ethnicity with informality, illegality and, ultimately, sorcery.
But what is, after all, ndoki, and how is it perceived from within the specific context of urban, post-war Angola? 5 A commonsensical understanding would refer simply to 'evil spirits', their presence and/or effect. But there are more overarching narratives. My Bakongo friends in Luanda-Christian and non-Christian alike-usually frame it as a system of knowledge that is closely connected to certain aspects of 'tradition', in the sense that it refers to knowledge transmitted within the kanda or lineage (Cuvelier 1934;Van Wing 1959;Thornton 2001), from uncle to nephew, concerning the 'agency of things and places'; it is an 'ancestral property' connected to the 'spirit' of such things and places. According to local traditional epistemologies, there is a recurrent association of ndoki with the idea of ingestion, as something that is introduced inside the body. Eoghan Ballard (2005) notes that historically this has been negatively linked (by Catholic authors, unsurprisingly) to acts and processes of 'poisoning'. This seems to complement what J. Van Wing, the Belgian missionary and folklorist noted, when he tied it to the image of 'clanic blood '-liens du sang, blood connections (1959: 359-360). So there is in such theories an implicit logic of bodily 'internalization' of witchcraft that is simultaneously its concealment from the public gaze while remaining pervasive in the social sphere as a narrative of suspicion. Here, the conflation of 'substance' and 'power' in the body becomes politically active (Graeber 2005). This becomes evident in the way notions of secrecy come to envelope the discourse of ndoki, ultimately rendering it as an 'invisible agency', but also ensuring its endurance and persistence, as I describe below. In other words, if ndoki were not a secret, it would no longer be ndoki. As Gabriele Bortolami (2012: 297) describes, this sense of secrecy and mystery was also in itself a factor of prestige within the traditional social structure, namely for those who could somehow access or master such wisdom. However, the prestige does not eliminate suspicion (Graeber 2005: 418). For instance, figures such as the nganga ('witch doctor') not only refer to a ritual/ spiritual specialization or expertise, but also to political agency (Bortolami 2012). 6 In this respect, a central aspect of this agency is the notion of deferred or remote effect. In other words, it usually involves what Knut Rio (2002) describes, in the case of Vanuatu in Melanesia, as an 'absent third person', an element that remotely intermediates between the cause and effect of sorcery through the establishment of a triadic relationship. From this perspective, if one becomes aware of the effects of ndoki (illness, death, and bad luck), the cause (or causer) will most likely remain invisible or anonymous unless it is revealed by an intermediary with the appropriate knowledge and power. But in any case, the focus always becomes the relationship, the space of connection that conveys meaning and possibilities of action for the actors involved in the process.
But one thing is the knowledge or theory of ndoki, and the other its practical unfolding. From this perspective, there is also 'good' and 'bad' ndoki, which leads to a subsequent distinction between sorcery (feitiçaria) and fetishism (feiticismo), both concepts harboring different levels of moral configuration. At this point, we argue, the urban space of cities such as Luanda introduces the moralizing element into the equation, becoming the third element that totalizes the experience of ndoki and makes it constitutive of the social relationships (Rio 2002: 130).
A case in point is the current 'epidemic' (as it is usually framed in the local media) of child sorcery accusations in Luanda, related to both an idea of a corrupted use of ndoki knowledge and a heightened sense of insecurity and paranoia in the particularly harsh context of the city's neighborhoods. As Luena Pereira describes in her ethnography of the Bakongo regressados 7 in the Palanca neighborhood of Luanda, child sorcery accusations, although not new in local history, have been object of the 'social construction of a problem' (2008: 31) that involves traditional authorities, the state and concurrent institutions, and local media, within a progressive narrative that invokes the destructuration of the traditional familial model, the consequences of war, the perils of urban life and economic crisis (see also Pereira 2016;Soares 2016). This has reinforced the historical production, on behalf of both colonial Portuguese and postcolonial Angolan political and religious authorities, of ndoki as an inherently negative phenomenon, one that illustrates the kind of past that needs to be erased, be it for the sake of a 'modern' colonial empire or for a 'developed' postcolonial nation (on this see, e.g., Milando 2008). Simultaneously, it reveals the agency of moralized political discourse in its configuration (on this see Brinkman 2003).
Mario, 8 a good friend of mine from Luanda, told me once how he interfered in a case of accusation of child sorcerers in the neighborhood of Palanca, using his authority as an elder and spiritual savant to solve one such case. He attended an interrogation session, where they were asking children what they had dreamed about, in order to decide if they were sorcerers or not. Mario contested this process, arguing (in a somewhat Freudian fashion) that it was very normal for children to dream about people they know, since it was part of the process of infantile socialization.
Mario thus managed to abort what would have eventually made yet another media headline. In the newspapers, every now and then we hear gruesome stories about feitiçaria that end up in a moralizing epilogue about how it represents a 'backward Angola' that has no place in the new, modern country. One particularly striking case became a running joke in Luanda, back in 2011: the national television channel, TPA, ran a story about a woman who had given birth, after a 21-month pregnancy, to a turtle. In the news piece broadcasted in the television, André Manuel, a traditional healer she had resorted to, could not hide his amazement: "this is sorcery. I have treated many people and never seen this before in my life". The undertone of ridicule and abjection in the news piece was evident, and several people I talked to shared and commented on this story as an example of how far people can go down the road of irrationality in what concerns sorcery and its effects.
Likewise, child sorcery accusations also become part of the same discourse of abjection. In one such example, in early 2008, several news pieces circulated in the local media concerning how members of a Catholic congregation rescued several children in the Palanca who had been beaten and abandoned by their parents, accused of sorcery. One of the Catholic nuns involved in the rescue, sister Rita, declared to the local media that what was beneath such phenomena was "social disorder and extreme poverty, and the main victims are the children" (Ecclesia, 30 January 2008).
However, Mario was not recalling the episode he intervened in to argue how 'wrong' these accusations were from a moral perspective, but rather from a methodological one. For him, there were traditional structures and procedures within the Bakongo tradition that secured the collective regulation of such phenomena. From this perspective, as Bortolami (2012) and Milando (2013) describe in the case of rural northern Angola, ndoki acted in principle as a regulator of collective interaction. However, in Luanda, due to the dramatic societal change experienced by the Bakongo, such procedures were no longer being respected. 'Tradition' had somehow been corrupted by the process of extreme urbanization.
This example shows us how, through the distinction of positive and negative ndoki, and of ndoki in theory and in practice, a process of moralization emerges. As I will explain below, this is the result of plural and competing understandings of ndoki on behalf of different religious (mostly Christian) and political agents. But it is also the outcome of what has been described in the field as process of commodification in the stricter Marxian sense, of insertion of ndoki within a capitalist, moneydriven system that is proper of urban life (see also Pype, this volume). Both processes are, undeniably, interconnected.

on capItalIst sorcery
As I tried to argue above, perhaps the most defining characteristic of ndoki practice is its invisibility, or the fact that it exists insofar as it is concealed from public sight and therefore public knowledge. It only becomes manifest through what Philippe Pignarre and Isabelle Stengers, in their book on the sorcellerie capitaliste (2004) called 'the power of the event'(2004: 11), 9 or in abstract, the agency of fabricating effects. From this perspective, one could argue that sorcery theories and practices such as ndoki work with a 'cloaking device', an imagined technology 10 by which objects, agents, or actions are necessarily invisible or concealed, although its consequences are not necessarily so. In fact, as I suggested above, despite its inherent secretive condition, it is fundamental that ndoki's effects become visible in order for it to 'be' ndoki. 11 Such consequences and mechanisms of 'turning visible', I argue, involve a process of capitalization.
Taking this argument further, as a realm of knowledge that is an art of producing effects, ndoki can be seen as resembling other forms of sorcery. I am obviously thinking, from Pignarre and Stengers' suggestion concerning capitalism as the ultimate form of creating relations and effects. This obviously resonates with the Comaroffs' notion of 'occult economies ' (1999), where they uncover the 'big business' (ibid.: 286) behind the postcolonial enchantments-or, in other words, how sorcery is a means to material ends produced within the millennial capitalism context (see also Myhre, this volume). Similarly, Harry West (2005), looking at the case of Mozambique, acutely described how the invisible realm is appended to processes of postcolonial governance in neoliberal reform contexts, explaining how, transcending the analytical prism of encapsulating sorcery (uwavi) with the past (see also Geschiere 1997), we can detect its agency as a contemporary social technology that orders, secures, explains, and empowers. Within this particular framework, uwavi works as a counter-critique of the neoliberal reform project of the Mozambican state, and reveals an alternative understanding of economy in the Mueda Planalto where West conducted his fieldwork. I am particularly interested in this last point: understanding sorcery as an economy in itself, and not just an expression of, or reaction to, a millennial capitalism (on this, see Austen 1993).
Here, we can perhaps relate to the notion of the 'market as God', as Harvey Cox (1999) eloquently suggested in his discussion of a 'business theology'-ideas conveyed by economics gurus about the selfregulatory capacity of the financial market, and ultimately the inherent justice of the capitalist system, precisely due to its detachment from the effect of individual, subjective behavior. This, Cox argues, is parallel to religious utopian eschatology, and inherits the history of an equally religious ascension of The Market (as Polanyi would have it) into a state of 'Olympic supremacy ' (ibid.: 20). This ascension, to be fair, is at the core of Max Weber's argument on the 'spirit of capitalism', the transformation of a household oikonomia (Agamben 2011) into a public financial emporium (Booth 1994). Thus economy becomes part of a theology, imbued with a spiritual component. Within this framework, 'Capital' becomes prophecy and millennium. 12 From this perspective, this association with capitalism is not just a recursive strategy on my behalf. In fact, it refers to a subsequent narrative that pervades interactions in Luanda such as those described above: the idea that the current transformation of ndoki as a 'social problem', displayed in the public sphere and involving different sectors of civil society in Angola, is the outcome of its display within an urban environment, with the subsequent dramas that it entails: disconnection, individualism, hyperconsumption, accumulation, etc. In other words, many religious folk I encounter in Luanda argue that, in contrast with the idea of immobility or stability of rural life, there is a process of social and spiritual corruption associated with urban life, which produces a metadiscourse concerning tradition, social change, and the power of spirits (see Bortolami 2012). Some, in fact, trace the 'arrival' of ndoki to Luanda from the rural areas to the 1960s, with the beginning of the war in the northern provinces of Angola and the subsequent social upheaval, which in turn coincided with the process of the Angolan capital's urban growth and economic development. This process of corruption refers to the conflict that emerges from the disconnection, in the urban sphere, with traditional forms of social, familial, and territorial organization, by which the regra tradicional, the traditional rule, that binds knowledge, relationality, and place, is broken. 13 The overarching association of witchcraft with problems of modernity (Geschiere 1997;Eves 2000), development (Moore and Sanders 2001), and morality (Ruel 1965;Austen 1993;Englund 1996) has been extensively debated and is obviously pertinent in this discussion. But another underlying narrative in discourses pertaining to witchcraft is its association with wealth, accumulation, and other financial operations. Here the association of money and morality is obvious (Parry and Bloch 1989), but we can also speculate around the idea of other, economic activities that are not so morally charged. Jane Parish (2001) has described this in her depiction of witchcraft among the Akan in Ghana, where the emergence of a market of anti-witchcraft shrines and talismans is a response to tourists seeking an 'authentic Africa ' (2001: 133). But there is also a necessary point about the vicissitudes of urban life, the perceptions of danger and insecurity that often invoke the emergence of ndoki.
Here, I understand the recurrent references I hear in Luanda about the idea of 'insurance' (assegurado)-i.e., the need to ensure spiritual protection in order to achieve success and prevent the effects of inveja (envy)-as part of a wider perception of spiritual agency that incorporates, detects, and delineates, but is not exhausted in, the idea of morality. One example was given to me by the mais velho (elder) Malungo, who built the first multi-story building in the Palanca neighborhoodnow used as a mechanical garage, hostel for Chinese workers, and headquarters of the Union des Traditions Kongo (UTK)-back in the early 1980s. The construction was recurrently attacked and sabotaged, object of ndoki, and thus took ages before it was 'finalized' (the building still looks unfinished and somewhat ghostly). All this because the neighbors did not agree with such a display of wealth on behalf of one of their own. Malungo never explicitly told me so, but I quickly assumed that this was only possible after obtaining some kind of 'guarantee' at a spiritual level.
A similar story was told concerning a neighbor who, back in the 1970s, attempted to build the first supermarket in the neighborhood. The supermarket was constantly attacked due to its symbolic and visual association to an idea of accumulation and storage that was antithetical to Bakongo logics of circulation and redistribution. But unlike Malungo, the neighbor gave up on his idea and left the Palanca.
From this perspective, ndoki is necessarily a response to questions of spiritual and material insecurity (Ashforth 1998;Rio 2010), but it is also part of what Filip de Boeck (2011) called a 'spectral topography', a way of inhabiting the city through its processualities and evanescences. Insurance, as any broker would tell you, can cover misfortune, but it can also create conditions for a protected future. This balance between action and prevention within the spectral topography thus becomes part of the index and regulatory mechanisms displayed through ndoki. And with the example of Mavungo's house, we can also see a constant balancing between mechanisms of individual accumulation and, per contra, social regulation towards redistribution. Ndoki can thus be seen as a regulatory mechanism that responds and simultaneously anticipates and creates.
It is at this point where I see a parallel with financial markets and indexes, such as the Dow Jones Industrial Average (Dow, in short). In Wall Street and elsewhere, the Dow acts simultaneously as a necessarily morality-free reflection and definition of economic activity (financial exchange, speculation), 14 configured through an idea of 'standard', i.e., a measure of reference. Based on a statistic analysis of the trading activity of a major set of companies in the USA, the Dow 'objectively' creates a reference that sets the standard for daily stock exchange and economic transaction. This objectivity is attained by the alleged 'freedom' of numbers, whose material existence would, a priori, not depend upon what Noam Yuran (2014) called an 'economy of desire'. However, as the same author notes, this assumption is based on a 'professional blindness' that is oblivious to how money and its circulation relies on expectation and trust, on performance and repertoire (Guyer 2004), and ultimately on values and relationships (Hart 2010;Graeber 2011;Moya 2015), beyond the idea of an inherent, self-contained rationality (Godelier 1972). The Dow, from this perspective, has its own, long, socialized history of fraud, suspicion, and scandal. 15 Likewise, in Luanda ndoki appears with a similar quality, as an operation that produces a commentary of social (urban) life and simultaneously effects upon it. The same conclusion emerges from the opposite perspective: the Dow, as a mathematical operation that, through an analytical device, produces a signal to the financial markets, thus effecting upon its subsequent behavior, reveals a sorcery quality; likewise, ndoki becomes an index, a moral marker that participates in the religious agents' spiritual and political mappings of the city, pushing individuals into processes of prevention, insurance, or accusation. Again, as many of my interlocutors stressed, the use of ndoki in the urban space is related to problems of reputation and scapegoating, and to the personal ambition of 'becoming big men' or 'big women', proper of an urban lifestyle where the possession of money can entail an increase in individual authority. It also becomes ambiguous and part of a politics of suspicion, as the practice of ndoki often becomes confused with its accusation (you never know if the denouncer is, in fact, the sorcerer). A friend from Luanda once told me the following story, which I transcribed in my own notes: Once, an elder comerciante (tradesman) accused his own children of doing witchcraft against him. A makanguilu (communal meeting) was called, and my friend acted as an intermediary, interrogating both sides in order to understand what had happened and what were their motivations. They found out that the children were only trying to find a way to keep their father close to them and not leave for business trips so much. So he managed to 'free' them from the accusation and possibly murder… This story is telling of how ndoki is object of an 'excess of definition', in the sense that it often encompasses more than the actual (or eventual) act of sorcery: it also includes an apparatus of moral and economic statements that envelope it within a wider understanding of spiritual affairs. Here, the role of Christianity plays an active role.

chrIstIan Ndokis
Having worked with different churches within the 'charismatic specter' in Luanda, 16 I recognized several discursive continuities and distinctions that offered concomitant and complementary perceptions on ndoki, as if trying to delineate different surfaces of a hypothetical object called ndoki, discernible only in the three-dimensional stereogram form, when we acquire the necessary technique or technology to 'recognize' it in the otherwise chaotic picture. In doing so, they detect and/or create different patterns and infrastructures within ndoki.
One such politics of recognition came from the Catholic Church. Traditionally, since colonial times, the official discourse on behalf of the Catholic leadership signaled ndoki as part of a superstition complex, which should be combatted inasmuch as it is the product of backward, wrongful beliefs, and therefore is ultimately inexistent. This configuration of ndoki as a virtual or imaginary object can be object of multiple interpretations. Taking a political economy perspective, there has been an evolution in the perception of ndoki on behalf of the Catholic Church that is in many ways concomitant of the transition from rural to urban spaces, and from a colonial to a postcolonial state: from its insertion within the colonial apparatus of domination, in which witchcraft appeared as one of the several constructions of moral alterity of the project; to its insertion within the postcolonial construction of an Angolan independent 'modern man' (Blanes 2013;Blanes and Paxe 2015); and finally its post-war insertion within civil and state 'reconstruction' discourses, where witchcraft becomes a problem of social welfare, as I suggest above. From this perspective, the Catholic discourse appears aligned with the secular state's project of nationhood. A good example of this was the creation, in early 2006, of the Inter-Sectorial Committee for the Fight Against Sorcery in Angola (Comité Intersectorial de Luta contra a Feitiçaria em Angola), championed by members of the Catholic and mainline Protestant churches in collaboration with government authorities (see, e.g., Angonotícias, January 24, 2006). In 2010, the then leader of Committee, bishop Anastácio Cahango, explained that, upon a request to the committee, a group of experts conducted scientific research on the subject of sorcery culture and uncovered the existence of 'merciless traditional and ancestral pirates' who, thanks to the absence of structural measures, (…), 'seduce and entice many towards committing crimes against human life (Jornal de Angola, September 21, 2010; my translation).
A second narrative construction appears within the so-called Mpeve ya nlongo movements, which represent, in the Angolan urban setting, what could be described as syncretic movements, in the sense that they convey an explicit process of adaptation of traditional healing practices (kimbanda) into a Christian template, and consequent transformation of former healers into pastors and prophets in the urban setting. These churches are usually portrayed as traditionalist, but in fact incorporate highly charismatic and Pentecostal elements (gifts of healing, vision, speaking in tongues, etc.). However, if the old kimbandeiros were service providers and could engage in sorcery to solve sorcery, so too are the pastors and prophets, who are specialists who are 'fortunate' enough to be working with the Holy Spirit on their side. From this perspective, among these movements there is a recognition of ndoki's dual (good/bad) nature, and that it can be 'corrected'. One example is the story of a man called Alfredo, a Bakongo involved in political and human rights activism, and who also attends mpeve ya nlongo churches, told me: Alfredo once knew a professor who had knowledge of ndoki, and was known in the Malanjinha neighborhood for 'combining psychology and magic'. At a certain point, Alfredo dreamt that this person was attacking him, so he summoned him for a makanguilu, but the man refused to participate. Alfredo again dreamed of the man, this time seeing him spreading some dust in his bathroom. A few days later, the professor was eventually seen again and Alfredo and his buddies decided to follow him, finding him spreading some dust in his backyard. When they confronted the professor, he was confused and left in a hurry. Later Alfredo dreamt of him once again, except this time he was tied to an imdondeiro (baobab) tree. Again Alfredo summoned the professor for a meeting, but to no avail. Until eventually the man grew seriously ill, with kidney or liver problems. Alfredo, who attended a Mpeve ya Nlongo church, then decided to pray for him, and some time later the professor sought him begging for forgiveness, explaining that it was all an act of the devil (something related to an aunt of his who had died)… He asked to be taken as friend again; Alfredo accepted, although his wife wasn't very pleased with his decision. But he stressed that the best way to attack the feitiço (sorcery) is to pray to God.
Thus, for Alfredo as for many members of Mpeve ya Nongo movements, it is the church that has the Holy Spirit, and the Holy Spirit is the only power that can counteract the negative effects of corrupted uses of ndoki.
The third narrative stems from the Pentecostal, charismatic, and reavivamento (revival) churches, a very strong movement in Angola that nevertheless covers a wide array of phenomena-from the local secessions of international branches (Josafat, etc.) to Congolese-originated églises du réveil and to the Brazilian-originated neopentecostal churches, such as the Igreja Universal do Reino de Deus (IURD) or the Igreja Mundial do Poder Deus (IMPD, seceded from the former). 17 From outside these churches, as with the Mpeve ya Nlongo, there is the recurring notion that such churches engage with ndoki through the perspective of healing practices, which have become commoditized within what is deemed as a neoliberal logic of accumulation of wealth-but does not differ, from the point of view of the believer, from the kind of 'healing service' offered by traditional kimbandeiros. From this perspective, as Clara Mafra et al. (2012) have described pertaining the IURD, ndoki, and witchcraft in general, has become part of the discourse of this church's 'dialectics of persecution', through which it is recognized as existent, agentive, and powerful, but inherently negative and corrupt, thus occupying a central role in the church's content and demand. The IURD has been particularly successful in the adoption of a 'politics of proximity', working within micro, neighborhood settings, identifying their social and moral particularities and developing a narrative that addresses them within such dialectics (see van de Kamp 2011 for a similar conclusion in Mozambique, and Bratrud, this volume for the case of Vanuatu). Within this policy, sorcery becomes a particular problem that 'explains' microsocial relationships-from nuclear family relationships to love/romanticism, work careers, etc. One can speculate if this is part of the so-called process of modernization and individualization, as has been argued pertaining the Pentecostal personhood complex. But for this discussion, I prefer to look at it from within the urban issue, as something specific to urban culture. From this perspective, sorcery becomes part of a public space used by Pentecostal churches within their attempts to establish their own theo-political messages of community enforcement (see also Andersen, this volume) in a highly competitive, plural religious setting. Within this framework, sorcery and deliverance play a fundamental role.
One example of this strategy comes from an event that took place on October 27, 2013, during my fieldwork in Luanda. The Catholic Church was celebrating the closure of the Year of Faith (Ano da Fé) with a multitudinous event in the church of São Domingos in Luanda. A similar event was taking place in the pilgrimage site of Muxima, situated 130 kilometers southeast of Luanda. The sanctuary of Our Lady of Muxima is today perhaps the most iconic Catholic landmark, considering its spontaneous emergence as a pilgrimage site in the last decades. Today, it is sought by Angolan Catholics from all around the country.
On that particular day, the sanctuary was attacked by a group of youngsters, who attempted to destroy the statue of the Virgin Mary that presided the chapel. They were able to cause some damage, and only made it out alive from the compound due to the ready intervention of the police, who prevented a public lynching. Soon later, it was disclosed that these youngsters belonged to the Igreja Profética Arca de Noé (Prophetic Church Noah's Ark), a Pentecostal movement that promoted an iconoclastic anti-idolism against the Catholic Church by accusing the statue in itself of being a product of sorcery 18 (compare similar processes of iconoclasm with Strong, this volume). They claimed to have received the prophetic message to do so in the Muxima.
Similarly, another particularly notorious and 'extreme' version of this kind of appropriations of sorcery takes place in a church located in the Hoji Ya Henda neighborhood, known as Combat Spirituel (the French name deriving from the fact that it is a movement originated in the DRC, and indicating that its public is mostly composed of regressados). This church is led by two pastors that are a married couple. As a Bakongo friend of mine explained, their success is based on their strong emphasis on demonology and the organization of 'spiritual crusades' (cruzadas espirituais) that engage in deliverance. Through a 'confessional regime' (Badstuebner 2003), believers in this church are required to 'leave everything behind', including their material possessions, in order to be released from the effects of ndoki. According to several comments I collected from people from other churches, the point of abandoning your money and possessions (and handing them to the pastors) in order to be healed was central. The heightened expression of such a logic takes place in their well-known extreme fasting sessions (so long that many people are reported to have died in them). Through this operation of bodily and material dispossession promoted by the Combat Spirituel, we see a kind of appropriation of the traditional conception of ndoki as an object inside the body, complemented by the fact that the believer who wants to be released must 'become invisible', in the sense of abandoning his material possessions and entering the church compound. Interestingly, I was never 'allowed' by my Angolan friends to visit the church.
The interesting point here is that, more than seeing ndoki as an agent of corruption, Pentecostal churches seem to insert it within a logic of production of alterity, of 'cosmological strangeness' (Kwon 2008: 29; see also Bertelsen, this volume), through a Manichean reasoning (whereby ndoki is a necessary and convenient corruption) that also emphasizes the economic transaction, similar to what the traditional kimbandeiros performed (so there is more a continuity than a rupture, from this perspective). But here, the economic paradigm is inverted: instead of the traditional redistribution, there is a logic of dispossession versus accumulation.
Thus, we can speculate about the material and symbolic role of money in the process. Heonik Kwon (2007) talks about 'dollarized ghost money' to describe how, in Vietnam, mourning rituals became affected by changing political and economic relations in the country.
If traditionally the ritual burning of money signaled the transformation of one economic order to the other (from the living to the dead), with the introduction of foreign currencies in the process made it an explicit 'moral economy'. Similarly, the 'capitalization' of ndoki in urban Angola rendered it a Dow Jones in its own terms, operating in the market of the Angolan religious public sphere (Pereira 2016) and affecting the ritual and ideological configurations of the Pentecostal churches. Within this setting, however, Pentecostal churches do not merely engage in the production of prediction and indexicality, but themselves become its objects-for instance, in the recurring accusations that circulate of such churches being hubs of extortion, money laundering, and illicit accumulation.

conclusIon: Ndoki and dow Jones
My speculative proposal of seeing sorcery as an index stems from my previous interrogations on the 'agency of intangibles' (Blanes and Espírito Santo 2013). In that book, we discussed traces and effects of invisible and intangible objects in social life. Here I argue that, beyond the recognition of objects per se, those intangible effects often become shifting and fluctuating 'stock market indexes', objects of measurement of value. Thus, an index is created-not necessarily a mathematical construct like in Wall Street, but a multifarious ideological construct that is able to describe and regulate the 'spiritual market' in modes that may or may not include capitalist modes of economic activity. Therefore, the index necessarily shifts according to who is performing the measurement, and ultimately produces as much regulation as it does deregulation, considering Pentecostalism's particular and continuous moral demand on the person and community, as is noted by Eriksen and Rio (this volume) in the case of Vanuatu.
Thus the point here is that, as an art of invisible agency, ndoki becomes object of what Alfred Gell called, via Peirce, 'abductive reasoning', or inference. One recognizes the agency and identifies the effects, and thus finds the best plausible explanation for the connection between two events. So in conclusion, this kind of reasoning produces an index, which varies according to the agents who produce it: catholic, prophetic, Pentecostal. But this index emerges as a moral index precisely because it is perceived as describing and affecting economic but also political and spiritual activity.
In what concerns Pentecostals in particular, we appreciate how many churches under this umbrella term engage in different approaches to the index, from processes of service-providing to the public production of symbolic action and, finally, extreme dispossession. Therefore, two concurring tendencies can be observed: the externalizing and connectivity (see Pype this volume) of ndoki through its publicization; and the internalizing of ndoki through acts of disconnection and dispossession. From this perspective, the traditional argument of capitalist accumulation does not fully encompass what is observed. notes 1. As we will see throughout the text, I am using a deinstitutionalized conception of 'Pentecostalism' as more of a spectrum of movements, places, and activities in which the holy Spirit plays a central role. In this respect, I will be referring to different manifestations, traditionally categorized as Neopentecostal, Charismatic, Prophetic, and Evangelical, as well as to other, locally conceived trends. 2. I will henceforth use the word sorcery instead of witchcraft, following the operative empirical distinction between such concepts in the working language of Angola, Portuguese, where there is a distinction between a witch (bruxo) and a sorcerer (feiticeiro). 3. As part of the Bantu family, the Bakongo (literally, 'the Kongo' in its plural form) are, along with the Ambundu and Ovimbundu, one of the major ethnic groups in Angola, populating the north-western territory from Luanda to the border with the DR Congo. Historically, they formed part of the ancient Kingdom of Kongo-which was subsequently divided, throughout the nineteenth and twentieth centuries, into Portuguese, Belgian, and French colonial administrations-and find themselves divided today between Angola, RD Congo, and Congo-Brazzaville. Although sharing a common language known as kikongo, the Bakongo of these three countries speak in different dialects within the same linguistic family, as well as lingala (a lingua franca). For a classic study of the Bakongo, see e.g., Van Wing (1959). 4. As Ramon Sarró (e.g., 2009;Sarró et al. 2008) has noted, the memory of the Kingdom, one of the greatest existing African kingdoms before its slow demise at the hands of the European colonial forces, is still politically very relevant, and conflicts with the postcolonial mapping of this region of Africa, posing questions of identity and national integrity.
references memory, heritage, expectations) and knowledge. He has published articles in several international journals and edited volumes at Berghahn, Brill and University of Chicago Press, among others. He is the author of A Prophetic Trajectory (2014, Berghahn). He is also co-editor of the journal Advances in Research: Religion and Society.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.