Religion and Africa are paired in the representational armouries of the social sciences, a pairing often proposed uncritically and without adequate reflection. This essay considers some of the reasons why religion and culture are commonly invoked to explain other social phenomena in African studies, and explores some of the repercussions of this kind of analysis. In arguing for an equivalence in analytical approaches to politics and culture within and outside Africa I call for greater sensitivity to the political consequences of exceptionalism and for an increased commitment to sociological approaches which seeks first to interrogate the categories of social organisation locally, rather than assume them. Finally, some directions of categorical change in Tanzania, and the implications for religion, are explored.
Privileging Religion
Religion and religious thought occupy a special place in African studies. This privileging is not confined to disciplines dealing explicitly with religious issues, such as religious studies or theology. Accounts of life in Africa from the perspectives of political science, sociology and anthropology give consistent prominence to religion as both cultural practice and as a determinant of social action. This privileging is most recently exemplified in Worlds of Power, a book which claims to explain the unique role of religion in African politics and society, and hence to account for the particular attributes affecting African political and economic orders. Ellis and Ter Haar provide an overview of the place of religious ideas in popular and elite cultures across the continent. In the absence of an established media tradition, and where citizens rely on radio trottoir, the endless recycling of rumour and street gossip for their political information, these authors confidently assert that ‘it is largely through religious ideas that Africans think about the world today’ (Ellis & Ter Haar, 2004:2).
Such claims are not restricted to the writings of those keen to maintain that understanding political action in Africa necessitates a special kind of analysis. A broad range of disciplinary positions have contributed to the representation of the African continent as a location and social context where religion not only matters more than in other places, but that the content of its mattering is more entrenched, more meaningful, more enmeshed in cultural practice than virtually anywhere else in the world. Indeed, assertions of an essentialising ‘African spirituality’ as culture continue to be a taken-for-granted aspect of the ways in which Africa is represented in Western discourses about it, representations doubtless influenced genealogically by celebratory histories of exploration and colonial mission (Pels, 1998; Mudimbe, 1988). Such representations underlie media accounts in the United Kingdom of the religious practices of immigrants from African countries, consistently emphasising dealings with spirits and witches at the expense of the more mundane forms of Christian or Islamic worship practiced by the majority. A similar preoccupation with the power of what is represented as ‘African culture’ is found in the accounts in the media and in medical sociology of what are claimed to be the uniquely cultural (hence behavioural) determinants on that continent of the HIV pandemic (Epstein & Packard, 1991).
Despite adherence to constructivism as a dominant theoretical frame within the social sciences and the impact of postcolonial studies, academic writing on Africa has done remarkably little to confront accepted truisms about the assumed place of religion in the political and social life of peoples on the continent. This situation is accentuated by the simultaneous representation of Africa religiously within Africa itself through proliferating media, including popular fiction, videos and religious tracts, which in turn become the basis of academic scholarship about Africa (e.g. Meyer, 2004; Bastian, 2001, 2003; Ellis & Ter Haar, 2004:41). Much work in my discipline of anthropology, my own included, has focused on witchcraft, occult economies and religious movements as comprising the analytical frame through which broader economic and social transformations can be apprehended (1994; 1997; 2003; 2005a,b).1 In political science, two highly influential books have prioritised a largely untheorised concept of culture as the key explanatory variable in determining political behaviours and outcomes across the continent (Chabal & Daloz, 2006; Bayart, 1993). Given the prominence of cultural explanations within scholarship on and about Africa, it is not surprising then that Ellis and Ter Haar prioritise religion as the determinant of culture and politics across contemporary Africa, not only arguing that religion has a special place in African societies but that politics in Africa and, by extension, social practice can only be apprehended through the lens of religiously populated spiritual worlds (2004:23).
While such accounts are important in sensitising us to the salience of religion across diverse social settings, the privileging of religion as culture and of culture as an analytical tool actually constrains the range of analytical possibilities that could be brought to bear on particular social phenomena. Religion must be apprehended as a social and analytical category, the constitution of which, by the analyst as much as by the society in question, requires sociological explanation (Asad, 2003). The universalising and exoticising effects of these culturalist accounts of religion mask important differences in the range and effect of religious organisations and practices across different countries and communities. And, in directing focus on extreme religious forms which engage adherents within apparently totalising symbolic systems, such approaches may blind us to other significant social changes ongoing across the continent. These include the social effects of increased inequality, increased mobility, migration and economic transformation. Granted, these transformations have religious manifestations, as in the extent to which forms of religious participation come to stand for, and accentuate, social differentiation. However, the apparently religious dimension of social change across the continent does not mean that such changes are religiously motivated, or that ‘religion’ as a category can explain them. It means rather that institutional and categorical separations between different dimensions of the same processes do not apply in the ways conforming to the analytical expectations of Western social science. These categorisations into separate domains corresponding to separate institutions are in any case misleading. Political power is nowhere confined to political institutions, economic activities everywhere transcend the narrow confines of what is defined as the economy. Delimiting the religious and isolating religious institutions is equally problematic. Like the other categories through which we think to understand the world economy, family and society – the ‘religious’ must be grasped as a category of analysis and practice which has origins in the political struggles around delimiting the power of certain institutions (Polanyi, 2001; Asad, 1993; Mitchell, 2002).
Analytical Assymetry
The contemporary privileging of religion within the social science imagination of Africa reiterates the position of the continent as object of the missionary and colonial gaze. Culture and religion, representational antitheses of the rational and scientific, were utilised not only as points of differentiation between Western coloniser and colonised Other (Said, 1991; Scott, 2003). They provided a convenient rationale for Western involvement in African economic, political and social development, an involvement informed by an ideology of conversion, albeit conversion up to a point. Of course, missionary orders and colonial regimes did not share perspectives on the content of the new cultural orders to which colonial peoples should be subjected. Nor did they agree on the importance or otherwise of what were defined as religious or cultural values (e. g. Beidelman, 1982). British colonial rule, informed by Lord Lugard's principle of the Dual Mandate (indirect rule), claimed to valorise local cultural forms of organisation and leadership as the institutional building blocks of rural governance. A firm boundary was established against local cultural content which was viewed as inimical to the aspirations of colonial development, as were some religious aspects of missionary Christianity. Consequently, British colonial policy actively promoted the expansion of Western style education but remained critical of missionary influence where this was seen to impact on indigenous traditions of social organisation, particularly around control over young people and marriage (Mudimbe, 1988:20; Green, 2003:38-42). Christian missions, in turn, sought to select out those elements of cultural practice that seemed most closely to correspond to Western categories of religion in order to prohibit, condemn and replace them, often leaving more or less intact indigenous cultural and organisational forms (Bloch, 1986:26; Bond, 1987).
Narratives of conversion suited the purpose of colonial mission, acknowledging on the one hand the categorical difference between missionary and convert, coloniser and colonised and, on the other, the ultimate potential for this difference to be transcended. Conversion as a cultural process of transformation permitted the recognition of an equal and common humanity, but as a deferred promise. The category of convert continues to bear reference to a previous status and thus to the differentiation of the past (Thomas, 1994). As long as novel Christianity was brought into confrontation with indigenous cultural forms, and hence as long as conversion focused on separating out the religious, the transformative potential of conversion as an equalisation strategy was restricted. Colonial narratives of development were similarly premised on an idea of deferred equality. Equality as an end point of conversion narratives, secular or otherwise, is compromised, in the sense that understandings of Africa remain stubbornly constituted within Western discourse as essentially Other. Africa remains defined by its propensity to the cultural determination of social practice, much of this pathological, with negative social, economic and political consequences. Conversion out of Otherness remains a possibility within this strand of contemporary discourse but, because of the essential power of culture, only up to a point. The report of the Africa Commission emphasises the significance of unique cultural forms in African societies, and the importance of uniquely cultural values, which would render alien organisational forms inappropriate impositions on African communities (2005).
While acknowledging the politics of inequality and colonial history which renders the issue of imposed forms problematic, there is something of a double bind to this politics of difference which further embeds the principle of unequal treatment in relation to Africa, legitimated through its claims to recognition as a special case. This principle of difference not only determines the ways in which social development policy in Africa is designed and implemented; for example, the constitution of social welfare policies with an emphasis on community mechanisms, or the utilisation of traditional birth attendants as weapons in the fight against maternal mortality, as opposed to greater investment in formal sector maternity services or public assistance programmes.2 It informs the ways in which what is culturally claimed as ‘traditional’ medicine is politically valorised in many countries as equivalent to allopathic medicine (Ashforth, 2005:137). The disappointing outcomes of these kinds of approaches to health and social development strongly suggest that there are profound practical limits to the efficacy of special treatment, particularly where this relies on claims to cultural suitability rather than the design of effective institutions and policies.
The discourse of cultural difference as essentially African is not confined to colonial and development representation of African possibilities and limitations, nor to the culturalist politics of African post-nationalist alternatives (Ashforth, 2005:149-153). It is perpetuated by academic discourses about Africa, particularly in the social sciences which consistently interpret African social phenomena in terms other than those accorded equivalent phenomena in Euro American or ‘international’ settings (Green & Mesaki, 2005). The anthropologist Bruno Latour has referred to this process of analytical displacement as the imposition of asymmetry, whereby substantially different analytical standards are applied to the same phenomena depending on where these phenomena are situated (1990:20). This situation is not geographical or cultural, although it is often represented as such, but is actually constituted within Euro American classificatory systems and hence social ordering as deterministic of the assumed break between those societies categorised as ‘modern’ and those which are deemed not to merit this categorisation. The result is an inequitable analysis, where systems and practices which are fundamentally incomparable are brought into an unfavourable comparison (1993:91-104). An instance of this kind of approach is the well known paper by the anthropologist Robin Horton on ‘African Traditional Thought and Western Science’ (1967a, b). Horton's paper claims to compare equivalent ‘systems of thought’ in order to demonstrate certain similarities between them. Horton's argument is that ‘Western’ science and African ‘traditional’ – that is religious – cultures are equivalents as theoretical systems which are the domain of specialists. In this sense what Horton terms ‘African traditional thought’ is ‘scientific’ because its objectives are concerned with what he represents as the core values of Science, that is the search for explanation and prediction of phenomena within the natural world (1967b:161). There are strict limits to this equivalence. Traditional thought may be scientific in one sense, but its explanatory potential is limited if confined within what Horton calls ‘closed’ systems, epistemological echo chambers in which ideas are neither subject to falsification nor free to make new associations. Western science as a system of knowledge based on the experimental method enables its progression into newer, yet decreasingly stable, realms of truth. African traditional thought remains stuck temporally and conceptually within the same, that is Other, paradigm. Equivalence up to a point.3
Social scientific accounts of ‘scientific’ cultures, actually the cultural practice of scientists, challenge this characterisation of ‘Western’ science (Latour & Woolgar, 1987; Law, 1994). Latour's analysis enables us to perceive that what Horton's article actually achieves is to compare two distinct institutional traditions of authoritative knowledge construction, both legitimated by powerful institutional complexes and both restricted to particular social contexts and personnel. African ‘traditional thought’ is not in actuality a mode of thinking, any more than laboratory science in Euro America is equivalent to popular rationality. A more balanced comparison between Africa and elsewhere would have been to take two equivalent systems, comparing, say, religious ideology in European tradition with the ideologies of certain religious specialists within an African one; or cross-cultural systems of divination like mediumship. A comparison of this sort would not only reveal similarities of structure and content. It would reframe discussions about purported difference within a new political context in which the conditions of sameness or equivalence become not merely possibilities but the springboard for analysis.
Studies by anthropologists, and others, of cults and sects, extreme cultural practices around witchcraft and the occult, and the tendency of some analysts to perceive political intentions in the symbolic meanings of ritual practices performed by some people in Africa has perpetuated the notion that the religious is the world for the majority of African citizens. Whether reduced to religion, culture or pathology, African political and economic situations become seemingly and authoritatively explicable, while Africa spins out of control on an axis of Otherness pushed into ever faster orbit by the contributions of Western claims to expertise. Here religion as explanation serves a dual purpose. In presenting an analytical category as explanation it renders what is apparently happening in Africa inexplicable sociologically. Society itself becomes comprehensible only in religious terms. Moreover, in asserting the redundancy of social analysis religion is sacralized, in the Durkheimian sense of the sacred, as set apart (Durkheim, 1995:44) with important consequences. As with other categorisations within extant Western discourses of social ordering, for example the private, the ‘result is to enclave certain matters in specialised discursive arenas and thereby to shield them from broadly based debate and contestation’ (Fraser, 1995:133).
The Power of Religion
Just as religious ideologies sacralise the practices they promote, religion as an analytical category is accorded power in setting apart from the religious. This is not only because of the conflation of religion with thought, which we have already considered, but also because of entrenched ideas about African spirituality and community, ideas which have been perpetuated by social science studies of religion and ritual which have tended towards the theological. Such studies have succeeded in making sense of the complex cosmological representations and ritual practices which inform religious sociality in Africa. At the same time, they have presented African social worlds as situated within totalising cosmologies and in which all social relations are not merely informed but determined by cosmological power. From books such as Placide Tempels' Bantu Philosophy (1959) through Godfrey Leinhardt's Divinity and Experience among the Dinka (1961) to Worlds of Power (2004), the mode of being human in African society is presented as not merely religious, but ontologically so. It is this, perhaps, combined with the search for theological equivalence, which renders the idiom of conversion persistent within academic accounts of religious transformation on the continent, in which effects and drivers of transition are once again understood in primarily religious terms.4 Conversion to world religions, Islam or Christianity, is explained in terms of the appeal of ideas and the functional utility of world views permitting adaptation to incorporation within global social relations. The tenacity of cosmological accounts of social transformation in Africa in religious terms is explicable not so much because it is convincing, but because it speaks to those who already share to some extent its convictions about religious power and African culture. The concept of conversion itself, lest we forget, is derived from images of miraculous transformation in Protestant evangelical thinking (Weber 1985:143). Even where power differentials are acknowledged as in the case of Christianity, the catechist/teacher and the class of pliant children desiring access to the word is often for the practical purpose of reading and writing rather than salvation (Carmody, 1988); not only is conversion assumed to have occurred, its implications are claimed as cognitive. Adoption of missionary Christianity in southern African Tswana communities in the nineteenth century amounted to, in the words of the anthropologists John and Jean Comaroff, a ‘colonization of consciousness’ (1989).
This idealisation of religion as the driving force in social transformation on the continent has effects in other fields. It is evident in current initiatives promoted by international development agencies and the governments which finance them to foster ‘faith based organisations’ in sub-Saharan Africa. Of course, this drive to incorporate religion is also part and parcel of the armoury invoked to fight what is referred to as the ‘war on terror’ and the assumed role of religious organisations in promoting fundamentalism. The notion that there are organisations based primarily on faith is part of the same strand of idealist thinking, thinking which is blind to recognition of the other attributes of all human organisations, however categorised, including resources, leadership, political significance, power structures. It was the brute materiality of Christian power combined with political relations with colonial regimes which gave missionary Christianity an initial foothold in much of Africa – a foothold consolidated through the policies of colonial governments which gave missionary churches considerable control over education and health services (e.g. Green, 2003; Comaroff & Comaroff, 1992; Vaughan, 1991).
The appeal of Pentecostal and Evangelical Christianities and some of the more extreme cults derived from them is equally grounded as much in materiality as the power of ideas. Engagement with these forms of Christianity may be more visible now than previously, but it is a novel phenomenon. The mass popularity of these movements was attracting the attention of anthropologists and other commentators in the 1960s (e.g. Sundkler, 1976, 2004). Early innovative social manifestations of spirit-inspired Christianity, albeit derived from Catholicism, appeared in the 1950s, as in the charismatic Jamaa movement of Zaire, a movement in part inspired by the theological writings of Placide Tempels, a Catholic priest and scholar of African religion (Fabian, 1971; De Craemer, 1977). Such movements then, as now, might have dealt with spirit but, like many forms of Christianity, they were and are structured through and around materiality.5 This takes the form of experiences as in embodied healing through possession (the word made flesh), as well as through intense personal relationships with things, including biblical texts and tracts, uniforms and buildings (Green, 1996; Luedke, 2003; Hoehler-Fatton, 1996).
Related arguments concerning the centrality of experiences and institutions can be proposed to account for at least some of the tenacity associated with beliefs in the powers of witchcraft and associated practices and, by extension, for the continued institutionalisation of witchcraft in much of Africa. As the anthropologist Mary Douglas has shown, witchcraft beliefs do not simply exist in an ether of ideas which populations somehow believe because they believe them. For ideas about witchcraft to have social force, and hence social significance, they need to be embedded within the social institutions which support them. Ideas about witchcraft depend for their institutionalisation on the social consequences of accusations for those alleged to be witches or to practice witchcraft. In medieval Europe, the social traction of witchcraft gradually eroded once it become disarticulated from systems of ecclesiastical justice (Douglas, 1991). My own work in Tanzania has shown how the institutionalisation of witchcraft suppression practices necessarily contributes to the institutionalisation of witchcraft, just as academic writing on witchcraft has contributed to expectations about its normativity in African social settings (Green, 2005).
Limits to Totalisation
Asymmetry, in Latour's sense, can be said to characterise academic and popular representations of religion in Africa. Social phenomena are reduced to religion as a means of interpretation with interpretation itself and the search for meaning becoming explanations for African social forms. The kinds of analyses of social institutions and politics which would pertain in other settings are not applied. The result is a swathe of studies privileging the religious as explanation for politics, culture and ideology in Africa today, and numerous studies of religious practices as coded commentaries on political and economic transformations. These accounts rarely consider why the religious should be accorded such importance for adherents in such places; that is how the category of the religious comes to be dominant at particular locations and particular times. This is partly because of the tendency to fetishise the religious in African studies, rather than to consider why such practices and representations should come to prominence. It is doubtless reinforced by the tendency, in anthropology at any rate, to study communities in which cultural systems appear totalising, where meanings become accessible to the outsider because they are endlessly reiterated through practices across all dimensions of the social domain, the kind of social systems subject to the systematised impacts of doxa in Bourdieu's sense of the term (1977:168). Indeed, this apparent uniformity of hegemonic order intellectually is what Horton, and others, have claimed characterises ‘traditional’ thought ‘systems’, and hence provides the basis for their apparent differentiation from ‘non-traditional’ ones (e.g. Horton, 1967; Levi Strauss, 1977:332, 340). In my view such assumptions about closed or open systems can no longer be maintained as genuine sociological analysis. I doubt whether such ideological uniformity was ever effectively imposed across social groups, whatever their scale, and this seems less and less likely in the contemporary global milieu. What may be maintained is the power of this model of uniformity and conformity, a model sought by those seeking to establish communities of control which are inward looking and bounded. Doxa ceases to be the default state for whatever social unit is significant and becomes instead the desired end constituted in differentiation to the ideologies and cosmologies of other groups. Such strategies are invoked by leaders of ethnic factions, by fundamentalist movements of one sort and another, and often by small-scale religious congregations, all of which are at particular pains to systematically police the boundaries between themselves and others. For members of religious congregations, such as, for example, adherants of Zionist type Christian churches in southern Africa, this separation may be achieved through ritual prohibitions, most often on the consumption of various types of food and drink; on styles of dress; times and mode of ritual engagement, and explicit rules about socialisation (Comaroff, 1985; Ashworth, 2005).
Totalising cults and sects, small community based churches, and the communities of practice which coalesce within the client groups of particular healers and diviners conform to these sociological expectations. Such communities are not necessarily longstanding, but come into being around particular ritual agendas. In southern Tanzania clients seeking the services of one of the country's leading specialists in the suppression of witchcraft come to his homestead from all parts of Tanzania. They stay at his place for a few days only, but for this period of time they become members of a sacred community-in-waiting subject to the restrictions and regulations imposed by the diviner's ultimately spirit-sanctioned authority. People waiting for their witchcraft powers to be suppressed through a complicated ritual centred on shaving and purification adopt a series of restrictions around dress, the preparation of food and the use of fire and water – separating them from normal society and creating through external signification and internal orientation a community of dependants around the diviner's power (Green, 1994, 1996; Green & Mesaki, 2005).
Totalising practices and the communities they bring into being are appealing to social researchers, but they present only a partial snapshot of religious organisation on the continent. Although participation in practices around healing, witchcraft and divination are important for large numbers of individuals, and although minority Christian churches and Islamic sects attract considerable followings, the vast majority of religious adherents in African countries are formally affiliated to majority Christian churches and to mainstream branches of Islam. While these same individuals may also participate in non-sanctioned religious practices around healing and divination, many do not. Closer study of mainstream religious communities and of less totalising involvements would reveal enormous differentiation in the extent to which people are religiously engaged, not only between different social categories and locations, but between different genders, life stages and religious affiliations.
Religious Realignment: Southern Tanzania
As institutional realignment continues apace in states subject to successive waves of liberalisation efforts – in the economy, political organisation and market – the monopolies once held by established religious organisations give way to fragmentation and increasing diversity of religious organisations and institutional possibilities. What seems to characterise religious affiliation and practice in contemporary Africa is its considerable diversity and the pace and scope of change. This situation is not solely a matter of ‘religious’ choices, although these are significant. Such choices are rarely made on strictly religious dimensions but are related to social, economic and political factors. The institutional landscape of many countries in sub-Saharan Africa is shifting rapidly. What once consolidated the influence of religious organisations is increasingly being passed to other institutions and dispersed. Tanzania provides a good example of these kinds of processes in operation, as the socialist monopolies associated with the single party state under the Chama Cha Mapinduzi (CCM) weaken and dissolve in a liberalised marketplace. In remote rural areas where majority Christian churches once provided the only alternative to state services and power, operating health and education facilities, providing access to employment and transport, market alternatives now come into play for those who can afford them. Ex-missionary churches now no longer hold the only keys to a better world. Their once totalising grip, an effect of the absence of alternatives under colonialism and socialism, loosens as the vacuum seal put in place by the single party socialist state is broken.
These social shifts are evident in Ulanga District, Morogoro Region, in south-central Tanzania, where I studied the effects of Catholic Christianity in the early 1990s. Then, in the final days before liberalisation, at a time when the Tanzanian economy was in severe crisis, where access to imports was restricted and basic goods often in short supply, when the health service lacked sufficient drugs and equipment to perform even basic functions and where the Catholic church controlled more vehicles in the district than all the government offices combined, the church maintained a powerful influence over the local population, the vast majority of whom were at least formally affiliated to it and for whom being recognised as a good Christian offered some kind of security in uncertain times. During the period when local enterprise was severely restricted, ventures operated by the church provided services in vehicle repair, milk production and high quality accommodation for official travellers. In addition to its network of dispensaries and clinics, the Church ran trade schools in carpentry, shoe repair and tailoring, a couple of farms, a boys' secondary school cum junior seminary and a domestic science establishment for girls (Green, 1995). Popular religiosity was very evident in the ostentatious display of rosary beads and sacred hearts of Jesus worn mostly by women, and in the packed masses held on Sundays and at the main Christian festivals. Priests and sisters were spoken of with enormous respect, to the extent that mystical powers were popularly associated with senior male clerics. Being recognised as a good Christian had religious status for parishioners, but it also situated people in relationships of potential support and assistance from church personnel, profoundly important in the absence of other economic and social safety nets and at a time of economic crisis (Green, 2003).
The effective monopoly of the Catholic church was at that time challenged only by an established Lutheran congregation, and a small Muslim community. This situation was a legacy of colonial policies which had divided different parts of the country among different Christian missions, an allocation to some extent solidified by socialist disinterest in further promoting missionisation. Today Ulanga district is still largely Catholic and the church remains influential, but there is no doubt that the content of religious influence has changed. First, the Catholic church no longer enjoys an uncontested monopoly on Christian practice in the area. As migrants from across southern Tanzania move into the fertile valleys of the western parts of the district, they bring their churches with them. Seventh Day Adventist and Pentecostalist churches, among others, are establishing themselves in the district. In the aftermath of liberalisation, the Catholic church no longer holds monopolies on transport and other services. Private enterprise is flourishing, transport services are available and the shops are stocked with basic supplies as well as previously unimaginable luxuries that had once been available only in the capital. Several private sector chemists sell medicines and equipment, the district hospital now has its own vehicle to transport the sick to regional and national referral facilities, and private sector businesses provide services in car repair, tailoring, furniture making and accommodation. The trades once offered by the mission vocational schools are no longer viable in the context of globalisation where high volumes of new and second hand manufactured clothes and footwear flood the country's markets. The church has recognised that times have changed and is diversifying. Although it continues to operate dispensaries, it has curtailed its vocational training programme and the range of services it provides to the general public. It has recognised that education services are still popular, but that these must reflect new markets and new economic demands. Domestic science for girls has given way to private academically oriented secondary education available at two sites within the diocese.
Institutional Transitions
In many ways, the Catholic church in becoming less of a totalising institution within the district is being forced to operate like Christian churches outside postcolonial Africa do, where the institutional separation of church, state and enterprise is pronounced. What has happened is that the forced totalisation under socialism ensured that the church operated much as it had done during the colonial period, when in the absence of other institutions and accorded a degree of power by government through control over health and education services, the church was involved, and in a position of power, in every dimension of social and economic life in the locality. It did not confine itself to activities which could be categorised as religious, although it was keen to maintain its identity, and hence political autonomy, by insisting on its privileged role as a religious institution. As the institutional landscape diversifies and goods and services become available elsewhere, the Catholic church in Tanzania loses this totalising dimension and is reduced to becoming a more narrowly focused organisation, left with its core areas of responsibility. Perhaps for the first time since colonial mission, the Catholic church in Ulanga, and places like it, is functioning more like a ‘religious’ institution, focusing on core activities such as the organisation of masses, the Christian calendar of ritual action and mediating life crisis events like baptisms, confirmations, weddings and funerals.
The role of priests and sisters is also changing. Priests find themselves confined to more religious roles under a new regime of diocesan authority, sensitive to the changing responsibilities for the church in the postcolonial world. Popular religiosity is altered. The church and Catholicism seem less important in daily life, less imbued with mystical power. Priests, reduced to church officials who wear uniform on a Sunday, are perceived as ordinary men, not the powerful people of the past who occupied special status endowed with both material and magical power. The performance of Christian practice takes place on Sundays and holy days, an activity increasingly confined not only temporally but institutionally within the church, and, as this shrinks away from engagement in other sectors, increasingly limited to a narrower sphere of social action. These transformations in the location of the religious are not confined to Christianity. The place of non-Christian religious practice is also shifting, again towards less totalising forms. These trends are clearly evident in relation to ancestral practices such as first-fruit rituals (when a portion of the crop is presented to the ancestors before it can be eaten by the people on whose territory in has been produced) and the performance of blessings and in relation to the traditional healing sector. The former have become in many communities a minority activity or are performed by a few individuals on behalf of their families. The latter is diversifiying and expanding, at the same time as user engagement increasingly adheres to the rules of the marketplace in constituting the basis of relations between healers and clients. Ulanga's famous anti-witchcraft institutions have been affected by these transitions. Whereas during the 1980s and 1990s those seeking anti-witchcraft services would voluntarily assent to becoming part of a residential community around the diviner's homestead, staying for several days or longer, today's services are carried out over a limited two day period. The proliferation of ‘branches’ (matawi) where witchcraft suppression is available closer to people's homes means that the long journeys and protracted stays of the past are no longer usual, although people are still expected to stay at least one night at the homestead of the diviner. Obtaining witchcraft suppression services has become efficient, functional even. In fact, modernisation and convenience were cited by Shaibu Magungu, the diviner responsible for initiating innovations after inheriting the practice after the death of his famous grandmother, Bibi Kalembwana (Green & Mesaki, 2005). Magungu is aware that totalisation is no longer the way in which all people operate religiously or culturally in contemporary Tanzania. Like many middle class Tanzanians, he sends his young son to an English language school in preparation for his future of secondary education and professional employment. Increasingly aware of and pressured into the new social conventions of differentiation, Magungu sees the world as changing. These changes have implications not only for his client base but for the interaction between people and traditional healing and divination services.
While modes of traditional healing and divination remain entrenched in Tanzania, often associated with the possession status of individuals and territorial shrines, the traditional healing sector is characterised by innovation in terms of its organisation and the ways in which its services are accessed and delivered. Well known healers like Magungu provide new forms of service delivery, strive to establish new specialist environments for the delivery of their services – for example wards and clinics – and self consciously adapt the signifiers of traditional, hence ancestral, authority. Elsewhere in Ulanga, shrine diviners unable to obtain the white kaniki cloth once demanded by the sprits must make do with, and must obtain the spirits acceptance of, synthetic colourful manufactured cloths, imported from Indonesia and China. Such innovations are not new in Tanzania (Redmayne, 1970; Feireman, 1986), but were probably put on hold or were less widely advertised, formally and informally, during the height of single party control and economic austerity. Similar processes are at work in other countries in Africa as they emerge from the constraints of statism into the confusion of competition and different kinds of uncertainty. These changes have created unparalleled opportunities for some, notably urban elites and well connected rural families who are able to capitalise on economic openings in the public and private sectors, to access improved education for the children and to travel outside their own countries with fewer barriers than previously. Their situation leads us to consider the implications of another asymmetry, the vast and expanding differential between rich and poor evident across the continent, and which will become more accentuated in the future.
It is undoubtedly the case that religious affiliations and modes of religiosity are becoming increasingly indicative of social status and social opportunity in many African countries, and that this too is a consequence of detotalisation. Certain forms of healing ministry and Pentacostal miracle churches have the largest followings amongst the poor and marginal, a following which is expanding as the number of disaffected grows and these churches split apart and reform endlessly around new prophets. In contrast, more mainstream evangelical and ‘Born Again’ churches offer spiritual succour of a more rational sort to the expanding middle classes (Marshall, 1991; Gifford, 1998). Lifestyle evangelism in Africa as in the US speaks with conviction to the values of self-made urban middle classes and even, as in the case of the Redeemed Christian Church of God in Nigeria (Obadare, 2006), the elites. Religious affiliation here, certainly for the poor and followers of evangelical community churches, aspires to a return to totalisation, an attribute identified by Fernandez in his classic study of the Bwiti sect among the Fang of Gabon some twenty-five years ago (1982). Adherents of Bwiti pursued a syncretistic incarnation of Christianity and ancestral practice, renowned for its utilisation of a hallucinogenic drug, eboga, as the basis of communion with the divine. Although the withdrawal of Bwiti adherents from the world is more extreme than practices by other sects, they have much in common in seeking to create a totalising moral community which, like representations of a lost traditional sociality, is animated by divine power. The return to totalisation in this instance is voluntary, but seems also to characterise the situation of the excluded and the poorest, even within mainstream Christianities. In districts like Ulanga, while the rich and middle class can opt out of religiosity because they do not depend on the church for access to support and emergency assistance, the situation is more ambiguous for the poor. Indigenous religious practices and conformity operate in similar ways. To return to the example of witchcraft, in Ulanga and neighbouring Kilombero the poor are most likely to be publicly accused of witchcraft and to be taken for cleansing than elite people are. This is not only because the better off who have some education and status with district officials can access the protective powers of the state, but because the vulnerable would only increase their vulnerability by refusing to participate in what are widely accepted to be effective means of suppressing the powers of witchcraft, hence in refusing to be cleansed of their imputed witchcraft powers (Green, 2005). In this instance, as in others, it is clearly evident how social institutions work to create the effects of religious totalisation, and that the most significant institutions are relations of differential power.
Religious institutions are more likely to achieve coverage and influence where they are not competing with other institutions, or where social pressures work to encourage compliance. This is not to suggest that religious adherence is necessarily forced, although this has often been the case, as during the height of Catholic hegemony in central and southern Europe in the middle ages (Ginzburg, 1992; Moore, 1987), but merely to point out that the totalising impact of a single institution is unlikely to be achieved without considerable social support (Douglas, 1991). This may take the form of lack of alternatives, as was arguably the case in Ulanga during the 1980s. Conversely, where alternatives come into play, totalisation becomes a potential option for those individuals who either lack access to alternative social worlds or, as with the Nigerian elite, for whom new and morally enclosed social universes are perceived as an asset, a means of consolidating networks of influence and power (Obadare, 2006).
The options for the rich and the options for the poor have different social consequences. Lifestyle evangelism and membership of middle class Protestant churches among ‘Born Agains’ ventures close to the realm of the spirit, but never loses its grip on the rationality of modern forms of power (Marshall, 1991). As Obadare demonstrates (2006) elite members of these churches retain ties to professional associations, engage in mainstream educational institutions and remain involved in transnational networks of study and employment. The rewards offered by membership of contemporary spirit churches such as the neneri cult in Mozambique are strictly spiritual (Luedke, 2003) although, given the importance of healing, the spiritual gains are experienced materially within the body. Remaining enmeshed within totalised religious worlds, traditional or otherwise, may erect barriers to encroachments that could offer other forms of succour (cf Fernandez, 1978:216; Farmer, 1999:154). In Ulanga District in 2005 several traditional healers were openly claiming to have available not merely treatments but actual cures for AIDS, a claim seemingly more plausible to many than the formally free availability of anti-retroviral medicines at the neighbouring hospital in Kilombero District.
Future Directions
In contemporary Africa, caught up in the economic and political reordering accompanying the consolidation of postcolonial geopolitics and the forces of liberalisation, religion seems to be becoming for some privileged individuals a matter of choice between competing options and alternatives. As disengagement or at least reduced commitment is now an option, religion becomes, for some at least, less a matter of contingency than of faith, in the process seeming to replicate the institutional arrangements and categorical divisions of a secular order. This ordering, depending as it does on fragile institutional realignments across state and economy as much as on the existence of otherwise of religious competition is not necessarily indicative of a future of secularism as predicted by modernisation theorists. Such orderings are, on the contrary, the result of specific institutional and political efforts to establish an institutional separation of kinds of power between different forms, creating the impression that the secular and religious are actually distinct spheres of influence and activity (Asad, 1993). The future alignments of various institutions and forms of power is uncertain in much of Africa. The current situation in Tanzania, and elsewhere, merely reflects the a current balance of power at a particular place and time. As power shifts this balance changes. The increased politicisation of religious movements in some countries and regions is paralleled by decreases elsewhere. There is no inherent inevitability about the triumph of the religious in some countries in Africa, or in some regions within countries. Rather, the question becomes what made this triumph possible? What social and institutional factors created the vortex religious organisations could fill? Often this vortex is created through the polices and interventions of agencies which in seeking to disestablish political institutions in Africa, either because they fail to conform to democratic forms or because they embody a different vision of the state and political morality than the current market ethos will tolerate, seek to in effect re-establish formal religious organisations as parastatal political institutions which can be represented as community or cultural organisations.