Marburg Journal of Religion Vol. 25 No.

1 (2024) – Article 6

Spirit Possession and Exorcism in Tanzania

Reflections on Participation, Perception and Spirituality

Katharina Wilkens
Institute for the Study of Religions, University of Tübingen, Germany
and Anna Daniel Sanga

Abstract
This article deals with two things. One, the process of ethnographic fieldwork on spiritual healing and
exorcism in Tanzania is recounted from a personal perspective including some discussion of methods
employed in participant observation and interviewing. Two, these methods and approaches are
reflected upon critically on several levels. How does a German scholar of religion learn to do things and
perceive things as a Tanzanian would? How does cross-cultural friendship help to navigate this process?
The members of the group in focus, the Marian Faith Healing Ministry, were excommunicated by the
Episcopal Conference of the Catholic Church in Tanzania in the 1990s, which makes it necessary to reflect
on the researcher’s position in intra-ecclesial politics. And lastly, the disconnect of spirit voices, sensory
perceptions and storytelling, on the one hand, and psychiatric explanations of multiple personalities,
on the other hand, is reflected upon within the framework of current approaches in the aesthetics of
religion.

Keywords
Ethnographic fieldwork, reflexivity, friendship, perception.

1. Introduction
Ethnographic fieldwork is about learning other people’s perception and understanding of
their life. Applying this approach in the study of religion means taking a step back from your
own religious upbringing and reflecting on different styles of religiosity and spiritual
realities. In this paper, the authors Anna and Katharina, from Tanzania and Germany
respectively, come from two very different backgrounds and combine their perspectives to
study spiritual healing, spirit interventions and exorcism in a specific prayer group foreign
to them both. The project in focus here was Katharina’s PhD research in the study of religion
(Religionswissenschaft), supervised and financed by a German university and undertaken from
2003-2007. Anna and Katharina knew each other from a decade earlier when they had met
during a church youth meeting. In 2004/05 Anna housed her friend, assisted her during some
interviews and together they discussed the topic for hours on end. At the time, Anna was
studying for a diploma in nursing and later went on to obtain a Bachelor degree in Sociology
and a Master’s degree in Human Resources Management. Our conversations on religion,

1
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

spirits and traditional rituals have continued to the present day as we both keep learning new
things about us and the people around us.
In this article, we present our field experiences, including our expectations, our
learning process and our reflections on the topic of spiritual healing and spirit interventions
in Tanzania and Germany. In this article, Anna’s thoughts are written in italics; the italic “I”
represents Anna’s voice. She opens up a vista to life in Tanzania for our European readers. As
idiomatic language was an important part of our research (and Katharina’s learning process,
in particular), we chose to keep the colloquialisms in an effort to highlight various fields of
knowing, including religious bias, and speaking about things that are often unspeakable.
Katharina focusses more on methodological issues of conducting fieldwork and theorising
that are situated in the disciplinary context of the study of religion (text in roman type; the
“I” in standard script represents Katharina’s voice). We lay a particular emphasis on the
process of translating cultural and spiritual experiences between Germany and Tanzania as
well as between personal perceptions and academic theorising. As we will show below, this
isn’t primarily about a linguistic translation, but rather about learning how to speak about
spirit experiences, their immediacy, their opaqueness or their fictiveness. Reflexivity is a key
term here.
As ethnographic and intercultural researchers, we have to reflect on our position in
the field vis-à-vis our ‘objects of study’, that is the people who allow us to observe them and
take part in their ritual life. We have to be aware of our personal perceptions of body imagery
and ritual based on the conventions of our upbringing. And we must take into account that
while an academic presentation of ethnographic events renders visible social structures,
complex power plays and implicit meanings, it also loses the immediacy of the experience.
Furthermore, the quest is ongoing how to decolonise fieldwork methodology and theory in
our continued conversations about religion and spirituality in Germany and Tanzania.
The focus of the research project under scrutiny here was ‘religious healing in
Tanzania’. Conceptualised first as a comparative study, it ended as a case study on one
particular group, the Catholic Marian Faith Healing Ministry (MFHM), or Huduma za Maombezi
in Kiswahili, Tanzania’s national language (Wilkens 2011). In Tanzania, Christian churches of
all denominations (missionary, independent and Pentecostal) offer liturgical healing services
for the ill and afflicted, in particular for those ills thought to be caused by afflictive demons
or witchcraft. Differently from Europe, rituals addressed to spirits abound in Tanzania in
Islam, Christianity and local religions. Anthropologists have studied spirit possession in
Africa at length, both in the context of symbolic culture and in the context of medical
anthropology more broadly (Behrend and Luig 1999; Nisula 1999). In Christian contexts, both
theologians and anthropologists study the impact charismatic demonology and local spirit
practices have on one another (Meyer 1999; Brown 2011). So-called ‘fire preachers’ denounce
the work of the devil and often perform exorcisms on afflicted patients. In local traditions,
spirits are usually seen in a more positive light. Though they initially cause afflictions, they
are then appeased with sacrifices and cultivated through trancing.

2
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

Research projects are not developed solely on the basis of academic questions (though this
certainly is of primary importance), but also out of personal interest and curiosity. Thus, we
begin our story with some basic information on our own cultural, religious and academic
backgrounds. For Katharina, the encounter with spirits was new, for example, while Anna
found the academic approach to them intriguing. A methodological interlude then provides
some information on the process of fieldwork and data gathering as practices in
anthropology of religion. The following three sections centre on the relationship between
researchers and their fields beginning with our thoughts on the experience of doing
fieldwork among a religious group that was new to us both. Second, researchers have to
reflect on their position amid the politics of identity among their counterparts; in our case
the situation was indeed quite tense between the Church and the healing group. Third, while
theorising field data follows established rules, it is necessary to reflect on the specific
relationship between the reality of spirits in the field as experienced though language and
ritual performances and the secular explanations offered in the academic treatment of
religious experiences. We conclude with some thoughts on the enduring friendship between
Anna and Katharina and the meaning this has for our working relationship.

2. Perspectives: The personal background of the researchers

[Katharina] At university, I studied religions and anthropology of Africa. During a Kiswhili
language course on Zanzibar I became interested in the spirits people kept talking about,
particularly the kibuki spirits (Larsen 2008). Successes in football tournaments were
attributed to spirits and nightly attacks of the popobawa spirit were discussed in the local
newspapers (Thompson 2017). One day, some of us German language students were taken to
watch a kibuki trance performance in the neighbourhood. It gradually became clear that quite
a few of the mothers in our host families were members of kibuki ritual groups. I remember
well how overwhelmed I felt at the beginning, both by the sheer realism of conversations in
which spirits influenced the European Champions League as well as by the impressive
ferocity and coyness of male and female spirits performing in the trance ritual. Following
this experience, I wrote my MA thesis on a comparison of spirit possession in Tanzania and
Nigeria (a country I knew from earlier travels though I had not witnessed any spirit rituals
before) and focussed particularly on the way Islamic theology rejected and accepted rituals
addressed to jinns in equal measure. For my PhD thesis I wanted to expand my studies of spirit
possession towards Christianity while limiting my scope more tightly to the topic of healing. 1
As an academic category, the term ‘spirit possession’ is generally not applied to trance
rituals in Christian liturgies, but remains limited to Islamic and indigenous rituals.
Anthropologists of Christianity in Africa tend to follow theological terminology and retain
such terms as ‘demons’ and ‘exorcism’. A particular feature of the Marian Faith Healing

1
At the time, in the 1990s and early 2000s, research on social functions and cultural symbolism of spirit
possession were more popular topics.

3
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

Ministry healing services is the presence not only of demons during exorcism, but also of the
Virgin Mary who appears and speaks through a chosen medium. While healing is performed
in the name of Jesus, Mary is the patron saint who is believed to guide the priest, Felicien
Nkwera, and his medium, Sister Dina. The MFHM employs its own blessed water in the healing
services. Similar in Catholic terms to the blessed healing water of Lourdes in France, it is used
to pour over the entire body of the patient, which conforms to traditional healing practices.
It is also taken home to drink and to sprinkle the house as a blessing and as a safeguard against
demonic attacks. During exorcism, it is said that the demons burn fiercely in the coolness of
the water. The MFHM follows an apocalyptic theology in which they have the only access to
an Ark (Safina – another name by which they often call themselves) that will save Tanzania,
and mankind more generally, from the fires of hell. Followers of Nkwera spend much time
praying the rosary which they call a ‘weapon’ in the ‘fight’ against Satan. This militaristic
language is typical of the conservative Fatima movement in global Catholicism. The group
has been summarily excommunicated in Tanzania and I will return to this confrontation
below.
It was my aim to explore these and similar continuities and discontinuities between
religious and cultural spheres in such a multi-religious country as Tanzania without being
stopped by the academic boundaries between Islamic studies, anthropology and mission
history. More particularly, however, I wanted to learn more about the experience of spirits
which were foreign to my own schooling in a rather secular European context and as part of
a liberal Lutheran household in which spirits figure only as figments of the imagination
(Wilkens 2014).
My academic decision was made. Now followed the necessary personal decisions with
my move to Tanzania. In the academic tradition in which I have been trained, ethnographic
fieldwork can last anything from six to 18 months (see below). I planned a stay of nine months
with a return visit a year later. I preferred to stay in the large urban centre of Dar es Salaam
on the Indian Ocean coast so I could have access to a wide variety of religions and
denominations. Also, I grew up in a large city (Munich) and thus felt more at home with that
kind of urban lifestyle. Happily, I was able to reconnect with Anna, a young woman of my age
whom I had met in Germany during a Lutheran youth camp. During the first few months, we
both lived in neighbouring hostels and later we shared a house quite near my research site.
At that point, her husband and young daughter joined her in Dar es Salaam while I was still
unmarried and travelling unaccompanied. Anna grew up in a village near the city of Mbeya
in the southern highlands of Tanzania. A few years after I left, the family moved back to
Mbeya where they are still living now.

[Anna] In 2005, I happened to be a research assistant to my good friend. I found it easy because
of my own background since I went to a [Lutheran] seminary school. At school we had preachers of
different kinds: some were singers, some were teachers and some were fire preachers. I also had a little
experience of spiritual healing at Muhimbili College of Health Sciences. In fact, I learned that there were

4
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

some people who seek healing apart from medical treatment. At the college, I studied health seeking
behaviour, where people seek healing according to the disease and the experiences they have had with
it. In Tanzania, there is a mix of different beliefs such as Christians, Muslims, Hindus and non-believers
who we call pagans. Talking about Christianity, many people here attend services which is different to
Munich where just few believers attend.
I visited Germany two times. Indeed, there is a huge difference between Mbeya City and Munich
in all aspects of life, geographically, socially, economically, and spiritually. The geographical difference
is that Mbeya is hotter than Munich, and Dar es Salaam even hotter, here we don’t have ice.
Economically, Munich is highly developed. For example, infrastructure is very developed and efficient
as in the presence of trains, efficient roads and communication systems. For example, in Munich you
can order goods and they are brought to your home. Here, it is impossible due to the poor post addressing
system. The electricity is stable and it does not go out every so often. Health facilities are more expensive
in Munich than here but most people there have health insurance. Here, not all people have health
insurance. Socially, people here know each other. For example, if someone dies in the neighbourhood,
all people will gather at the family home to join the mourning for the loss. Also, children gather to play
together in any compound where there is space. There are groups of women in most of the streets who
take care of social issues. They invite each other to ceremonies like send-offs,2 weddings and funerals.
They also gather small amounts of money, like one dollar per month, and each month two to five and
even up to fifty women will be given that amount of money to buy home equipment like cooking pots,
blankets, plates, cups or to start a small business. We call it upatu. A good example of that was during
the first and second Covid 19 attacks when almost every family experienced death. But the women
managed to gather together and helped each other. They never gave up, even though it lasted a long
time and affected our economy. These groups were very valuable because the care of Covid 19 patients
is very expensive and when the patient dies, she/he leaves a bankrupt family. So the women’s groups
took charge and handled all the funeral matters and the families gained the advantages.
The first time I visited Munich in 1996, I had the opportunity to join some religious services at a
Lutheran church. Worship is similar in Germany and Tanzania. We both worship God Jehovah, go to
church, have some regulations and procedures to be followed during the services. There is a big
difference according to the type of church, however. The Catholic Huduma za Maombezi in Tanzania
worship God through the Virgin Mary. There are restrictions on the mode of wearing, such as long skirts
for women, the singing is different and there are continuous prayers (the rosary). At the city centre we
found that during service time believers were allowed to pray all together, not only the priest. They also
insisted on God’s presence at the service and they prayed straight to God through Jesus’ name.
Due to the life pattern I passed through, I also happened to find out how the spiritual world
works. I would like to research traditional spiritual healing in Germany. I found that human life is
typically spiritual. If things are done in the spirit, then it happens in the physical. Many things are done
in secret, many people know much about someone’s actual condition, but they can’t say anything due
to the covenants they made with those in power and who control the supernatural powers. If they reveal

2
These are farewell parties for sending off children to boarding school or sending friends back to Germany.

5
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

their secrets, there are hard sanctions given to the betrayers.3 Due to the sanctions imposed on spiritual
expertise, people are likely to lie to someone who knows nothing but who needs their help. Apart from
sanctions there is also bribery where the one witch who doesn’t want his/her secrets to be known is
likely to bribe the seer and the seer if he/she finds that the wrong doer is more powerful than him/her
will accept the bribe or will be afraid of revealing the truth. So, life will go on like that.
I have been working mostly with the people in need as a nurse, as an administrator and so on. I
met people with different experiences where many of their issues had many explanations. For example,
a policeman catches a thief red-handed with vivid evidence and the thief is taken to trial. Then the
police finds that all the things have disappeared together and that the person himself has also
disappeared completely. Also, some people may suffer from a disease that can’t be explained medically,
but when they go to the local doctor, they are healed. In another case, a person’s goat ate a neighbour’s
entire fruit tree which wasn’t able to grow up again. The tree owner was furious and went to look for
the goat’s owner to pay for it. The goat’s owner refused. The next morning, his stomach was swollen and
when he went to seek medical attention, the condition became worse. But when he went back to the
neighbour, the whole neighbourhood gathered and the owner of the fruit tree confessed to have
bewitched owner of the goat. After the meeting the man’s health recovered completely. There are many
examples where people believe that there are hidden powers that make them fall sick. And again, some
people find that maybe there is an argument about something in the family and the next day the
accused wakes up outside while he went to bed in the house. To every situation there is a way to solve
these spiritual matters.
It’s true that in my family we have ancestor spirits just like other African families and there are
family members who served the ancestors long ago. I come from a Royal family. Many people know
stories about our Royal ancestor. Some say that one of our grandparents was called Lwembe and he had
the ability to make live things like cows, chickens, grain, lions etc. For example, when the community
was suffering from famine, he just gathered his followers and bent his head. When he scratched his bald
pate, the grains came out of his head and the people went home with food. Also, when he was young, he
used to go to the bush with other boys to herd the cattle. When they were in the bush and got hungry,
he would take some soil and water and he would make ugali [corn porridge]. With some more soil he
would make a chicken. Then they made a fire and cooked some relish to eat with the ugali. He warned
his friends not to tell anyone about what was done. Sometimes he just made chickens, or he filled a hill
with cattle. When he grew up his father worried that he would take his chiefdom away and people would
follow Lwembe rather than him. So, he planned to kill him. When his commanders approached him, he
turned himself into a lion and was surrounded by other lions. So, he just mixed himself in with the other
lions and they failed to recognize him. His father attempted to kill him several times and Lwembe had
to do something very unique to escape death. This is a true story.

3
For those readers who do not know these secrets: This is the idiom of witchcraft wherein witches, both male
and female, make pacts with demons in order to control other people. People who live under a spell become ill
or behave in irrational ways. Witchcraft is discovered and disarmed through seers, more commonly called
diviners or traditional healers in anthropological literature.

6
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

This is the history of my family. We opposed ancestor worship because we believe in Christ Jesus.
But things are always complicated. In fact, there is a lot to say about spiritual healing in Tanzania where
many claim to be healing in Jesus’ name, others just lie, and still others use witchcraft in churches and
so on.

3. Methodological interlude
The study of religion as an academic discipline does not have any research methods specific
to religion. Rather it makes use of a wide variety of methods employed in other disciplines,
most notably philology, history, anthropology, sociology or psychology (Stausberg and
Engler 2011). My methodological interest lies mainly in anthropology and so I decided to do
fieldwork for my PhD research with participant observation flanked by interviews and the
analysis of available grey literature.
Doing ethnographic fieldwork in the discipline of anthropology (as opposed to
sociology or European anthropology/folklore studies) generally means conducting research
in a foreign country. The discipline’s global outlook requires intercultural experience. Taking
up residence abroad for several months, or even years, is based on the idea that the
researcher should become acclimatised to the circumstances of local life, to the weather, the
food, the rhythms of the day and the language of everyday communication (phrases and
imagery, but also the morals of what can be said or signalled through gestures and what
cannot). Work on the research project must therefore always be carried out in conjunction
with the overall business of living (Blommaert and Jie 2020). That was why my experience of
sharing a house with Anna was so invaluable to me. Her family taught me about everyday life
in ways I could not have learned in a hostel, as her portrayal of social life in an urban
neighbourhood may demonstrate.
I planned a nine-month stay in Dar es Salaam, the largest city of Tanzania where all
religions are present side-by-side. In this urban context, fieldwork meant visiting churches,
healers and conventions whenever they met. After meetings, I talked to people who had
attended and I made copious notes of what I had seen. I also collected leaflets and posters and
I went to bookstores to buy recommended literature on Islamic and Christian healing of
various denominations. In sociology, this mixed methods approach is known as triangulation.
The primary interest of the project ‘Religious Healing in Tanzania’ was to study a
variety of different forms through which healing was achieved with the help of spirits. This
required a number of clarifications: first, what does healing mean? How does one know what
kinds of illnesses and afflictions there are? And what is the definition of health? Second, the
basic assumption of the project was based on a concept called ‘local religion’ (Grieser 2021:
192) which argued for a closer look at cross-religious and cross-cultural dynamics within a
single locale. In this case, the aim was to observe spiritual healing in various Islamic and
Christian contexts (denominations) and compare it to local variations of traditional medicine.
Before setting out, I therefore read up on medical anthropology – a new field to me as a
scholar of religion – and discussed my research questions with various researchers in the

7
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

field. With a refined set of definitions, I started working on a proposal to apply for travel
funding.

3.1. Gatekeepers
In Dar es Salaam, it soon became clear to me, however, that it would be difficult to keep my
comparative studies evenly balanced. I simply was not able to gain access to Muslim healers.
Being a young white female certainly made it more difficult to communicate with some
healers, but others were friendly enough. I did not find the ‘one’ healer, though, who was
willing to teach me more about his craft. At the same time, I had presented myself at the
anthropology department of the university and had been introduced there to some members
of a Catholic healing ministry specialised in exorcism. Especially one lady was very friendly
to me and invited me to visit a prayer meeting. She acted as my gatekeeper to the Marian
Faith Healing Ministry, which then became my central object of study. As a university
lecturer in philosophy, she understood my academic approach and my interest in research.
At the same time, she was a leading figure in the ministry and could grant me top-down
access, with the approval of the chief priest, to attend the prayer meetings and interview the
patients and believers. With this recommendation, I was welcomed into the group without
reservation. With this top-down approach, however, I was only presented with those stories
about miracle healing that represented the official theology of the group. As I had a lot of
contextual information from other healing groups, from articles in the local press and from
interviews with opponents of the group, I could circumvent this restriction at the Marian
Faith Healing Centre to a certain degree, at least.

3.2. Participant observation

In order to gain insights into the religious practice of spiritual healing groups it was necessary
not only to talk to the people, but also to take part in the prayer meetings and Sunday services
myself. There were many different types of such services, ranging from very tight-knit
neighbourhood groups and church services with a few dozen attendants to huge crusades
(conventions) with hundreds or even thousands of visitors. At the Marian Faith Healing
Centre, healing prayers take place every afternoon in conjunction with the liturgies of
Eucharistic adoration and praying the rosary. They end with the celebration of a word service
or a mass in the evening.
While it may be possible to maintain a certain degree of anonymity in large-scale
services, this is impossible in small groups such as the MFHM. This means that the researcher
must encounter the ‘objects of study’ on a personal and human level, giving information
about his or her own life as much as receiving it. Participant observation is not objective.
Rather, its results derive from the feelings, the impressions and the physical reactions of the
researcher. His or her body is the instrument of observation (Sax 2014). With this premise in
mind, it becomes clear why it is necessary to give some thought to methods of note-taking. It
is helpful to have lists of categories and terms with which to describe architectural spaces or

8
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

physical and emotional sensations. It is also helpful to divide notes into basic descriptions of
spaces or events, in one column, and commenting on that directly, in a second column, in
order to let personal feelings rise to the surface, be recognised and ideally, be reflected upon
at a later time and in a more disinterested frame of mind.

3.3. Narrative interviews

I wanted to know how people made sense of the afflictions touching them and how it fitted
into the world around them (Caplan 1997). For this purpose, I gathered stories from patients
at the Marian Faith Healing Centre in the form of semi-structured narrative interviews. For
me, it was important to get away from concepts of ‘belief’: spirits and demons are not things
you believe in necessarily, they are things you experience and whose actions become
‘sensational’ to you through specific events (Promey 2014). Imagery, tone and posture all
come together in stories to recreate such experiences for the listener. In my study, I often
heard variations of the words ‘I couldn’t believe it!’ Or ‘I didn’t want to believe it!’ But the
ensuing events demonstrated that spirits were indeed involved in their illnesses and/or their
recovery (Wilkens 2018).
Sociologists working in their familiar surroundings sometimes prefer to invite their
interviewees to an office or a separate interviewing room at the university in order to reduce
unnecessary distractions and render the information more ‘objectifiable’. Anthropologists in
the field, however, generally prefer to talk to their interviewees where they are most
comfortable or where the topic of the conversation is most present. Therefore, Anna and I
always spoke to people at some point before, during or after the series of afternoon prayers
which they attended. They were in familiar surroundings and their reasons for being there
in the first place were easily connected to spots in the compound they could point out to me.
Thus, the interviews were often quite lively and performative acts of storytelling which all of
us enjoyed.

4. Reflexivity I: Fieldwork experiences

[Anna] I’m going to explain fieldwork as per my own experience. I have been doing fieldwork for quite
a long time, but its experience depends on the course of action. The truth is that every field has its own
uniqueness. Spiritual healing was a very difficult research topic for me because it needed some evidence
which by its very nature is internal to the patient only.
Speaking about the fieldwork itself, it was easy for me because interviews could be scheduled to
a flexible timetable, and the prayer centre was near our house so we didn’t need much time to travel.
We found kind and welcoming people at the centre who were ready to offer their support. Language
translation wasn’t a problem at all. Everything was fine to me. In my own sociological fieldwork that I
did later in a large private company, I didn’t need an assistant other than an expert on quantitative
methodology.
In the project, I learned a lot about Roman Catholic (Huduma za Maombezi) positions on
spirituality and their mode of healing. To me it was a strange experience since they mostly used water

9
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

to heal and they prayed to the Virgin Mary. I found that they had a very strong stand and were
unshakable against their main church. They are very focused on what they believe. Apart from learning
new things, the research didn’t affect my own spiritual stand. Maybe that was because what we
researched concerned mostly internal afflictions and not physical ones and many people tend to be
affected by visible things rather than the unseen things – including me. So, though I wasn’t affected
personally in my own spirituality, I did find a very unique rule of worship in that institution.

[Katharina] In my case, participant observation meant learning different styles of

prayer from those I had grown up with, including kneeling on hard surfaces for an extended
time (a skill I never mastered), having my head and hair soaked in blessed water on cold
nights and adapting to required styles of clothing and movement. Participation sometimes
means doing things that are fun and exhilarating, and sometimes things that do not agree
with you. This is one of the more complex ethical issues in fieldwork – one that can only be
resolved on a personal and human level (Franke and Maske 2011). With research in religion,
a common question asked by the interlocuters will be whether you are intending to convert.
A very strong missionary attitude can be quite vexing, but in my case my Lutheran
denomination was often taken as an invitation to explain aspects of Catholic ritual in more
detail.4
It was a great help to me to have Anna around as often as possible. She not only helped
with actual translations, but also with navigating the intricacies of polite greetings and
general small talk. When visiting other churches for comparative reasons, such as Catholic
charismatic congregations or Pentecostal churches, I would never have found the places
without her help as a guide on the overcrowded city buses.
Researchers generally return to ‘their field’ for follow-up questions or perhaps for an
entirely new research project based in the same area – the better one knows a group of
people, the more insights one gains into the complexities of their lives. In this regard,
fieldwork in anthropology takes a different approach from that of qualitative sociology.
Sociology is generally not immersive to the same extent and it relies more on the constant
documentation of all observations and conversations (Knoblauch 2003). It is not a ‘lifestyle’
tied into the business of everyday living abroad where observation never ceases.
While I was doing my research, talking to Anna and her friends and taking part in
rituals that were strange and unfamiliar to me, I began to reflect on the question of how I
should translate the various impressions and stories I was gathering into something
‘understandable’ back home in Germany. I was learning to ‘see’ the spirits work and to use
spiritual imagery in everyday conversations – the spirits were becoming real to me.
Reporting on my work back home at my German university, this led me into a new
conundrum: fellow students, members of my grant supervisory board and other colleagues

4
In many cases it can be an advantage to belong to a group of persons – be it because of religious affiliation,
age, or other factors – who is accepted as “not knowing how things work”, as this can invite in-depth
explanations. See further the contribution by Dagrún Ósk Jónsdóttir in this special issue.

10
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

not familiar with spirit possession started commenting on my way of speaking about spirits
as if they were real. When I would say, or write, that a certain kibuki spirit had spoken to me
or made jokes about me (of which I have voice recordings), or when I said that I had been
cursed by a demon at the MFHM, then I was speaking the way people in Tanzania spoke. But
my German interlocutors insisted that I explicitly include a linguistic ‘as if’ marker in every
sentence I uttered. What Germans might accept as social reality in Tanzania, must never
actually appear as ontological reality in Germany.
Now this realisation is nothing new, of course. It is called methodological agnosticism
and has been the prime approach in the academic study of religion for decades (Harvey 2011).
What was new for me, however, coming from a secular upbringing in Western Europe and
from academic training in social constructivism, was the experience of seeing ‘spirits’ in
action, or rather, people in trance, people narrating their own encounters with spirits and
people debating the merits of one spirit healer over another with the respective powers of
their spirit familiars as point of reference (and not the charisma of the healer herself). To me,
this experience of the possibility of an alternative reality is what makes fieldwork so
rewarding. It did not change my own perception of reality and I still do not ‘believe’ that
spirits affect my life, but I certainly learned to pay better attention to the way everyday
events, the senses, emotions and narratives shape and are shaped by different kinds of reality.
I will return to the spirits’ voices in section 6.

5. Reflexivity II: The politics of doing fieldwork

I have described the Marian Faith Healing Ministry as a Catholic prayer group while in
actuality this ascription is highly contentious. The leading priest, Felicien Nkwera, was
ordained in 1968 and served as such for many decades. But his healing ministry led him into
conflict with the Tanzania Episcopal Conference (TEC). Nkwera gave testimony that he had
received visions and messages from the Virgin Mary, Mother of God, giving him her
particular blessing to perform healing rites in her name. In the 1980s, four visionaries were
installed who embodied Mary and other saints during prayer services. Somewhat reminiscent
of trance performances in traditional spirit possession, Mary was believed to inhabit the
visionaries and to speak through them to the gathered audience. While this prophetic
situation was certainly critical for the Church, the final break in the 1990s was precipitated
by conflicts over formal recognition of Nkwera’s performance of exorcism and the correct
way of administering the holy sacrament of the Eucharist. The group’s Marian theology
follows the globally well-established patterns of Lourdes, Fatima and Medjugorje and the
Eucharistic clash fits into the spectrum of ultra-conservative critics of the Holy See. Nkwera
accuses the bishops in Tanzania of being Freemasons fulfilling the wishes of Satan, while he
himself is a true Catholic fighting for the purity of the people, the nation and the Church.
Nkwera was defrocked and his entire community was excommunicated and remains so today.
Nevertheless, the ministry is thriving and people continue to come looking for healing.

11
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

Nkwera welcomed my research for several reasons. He understood its academic

purpose and he was proud of the interest I was taking. The simmering conflict, however, was
very important in situating my work locally. Even though I was simply describing the
supposedly beneficial effects of his healing ministry while taking a neutral position in the
conflict, Nkwera could use my ‘professional’ assessment to further his own cause against the
bishops. This was not a conflict that posed a threat to me personally, neither physically nor
legally (as would have been the case a decade earlier), but it certainly put a particular slant
on my research process.
Nkwera also wanted me to publish my research in both German and English in order
to spread the messages of Mary’s healing word and reach a wider audience of possible future
converts. This, of course, was not my own intention, though I certainly had no problem with
the wish of reaching a wide readership. The question I had to reflect on, however, was how
far these political and missionary intentions of the people I was studying would have an
adverse effect on my own academic work. I decided that I needed to be open about these
expectations and positionings. Reflecting on the political sensitivity of my work thus became
important during the analysis of my data and, later, at presentations of my work (as again
now, in this paper). My interviewees, as I mentioned above, were chosen for me by a
gatekeeper who certainly was aware of the issues at stake. When I write about healing
successes at the Marian Faith Healing Centre these successes (‘miracles’) are certainly true
on a personal level, or more precisely on the level of the narrative identities I analysed
through the specific autobiographical data I collected. They do not, however, paint a balanced
picture of all members or casual visitors of the Centre, some of whom went away disappointed
to search for healing elsewhere.

6. Reflexivity III: Theorising spirits in the study of religion

In section 5 I discussed the process of learning how to see the work of spirits and hearing
their voices during the period of fieldwork. Things get really complicated, however, when we
start theorising these voices in anthropology and the study of religion. Who is it exactly
whose voices are being heard? Who is the ‘author’ of an utterance spoken during a spirit
possession trance? Is it the dancer’s voice in an alternate state of consciousness, the spirit’s
voice, or the voice of the community? Who is listening to these utterings? Where do the
subconscious, the theatrical and the spiritual meet and when are they incommensurable? I
have discussed some of these questions within a framework of narrative analysis that leans
on performance theories and aesthetics of religion (Wilkens 2020). They are crucial questions
fieldworkers have to grapple with both at the moment of entering the field as well as at the
moment of leaving the field and re-entering the framework of academic theorising.

6.1. The multiplicity of spirit voices

In spirit possession, people under trance speak differently from the way they normally do.
Psychologists and psychotherapists speak of dissociations and personality disorders, usually

12
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

with reference to spontaneous and unwanted episodes of dissociation. But in Tanzania, and
in many other parts of the world, dissociation is trained and wanted. Though it might be a
symptom of psychological distress, it is not a symptom of a personality disorder. As a scholar
of religion, I was more interested in the attribution of the voices to spirit personalities rather
than to split personalities, but coming from a culture in which spirit possession is not
recognised, the medical framework always influenced my perception of the situations I was
witnessing. My analytical work has since focussed on the ways of translating spirit possession
into psychological terms and on the ways in which psychotherapy can accommodate spirits
as long as they remain on a level of narrativity rather than ritual (Wilkens 2020).
Not only psychologists are interested in voices, but also social anthropologists. The
question is then on whose authority the voice is speaking. If a spirit is speaking, it holds
higher authority than either the human host or the human priest leading the ritual (Lewis et
al. 1991). As such, the spirit can order humans to do things they would otherwise not do. The
human host could make no such demands, especially not on her own behalf (e. g. ‘take better
care of this person’). It is, however, unreasonable to assume that all patients at all times are
faking spirit voices in order to gain spiritual authority. Is the voice of the spirit then that of
the community? Or of the priest? It is a conundrum in social theory that demonstrates the
shortcomings of theoretical individualism and remains to be solved even after decades of
research on spirit possession (Quack 2014).

6.2. Healing and believing

The origin and authority of the many voices speaking in spirit possession confuse and
intrigue not only European observers, but also lay observers in Tanzania. As I mentioned
above, ‘belief’ in spirits comes second to the experience of spirits – hearing their voices at
night or through a medium, feeling their touch as a shiver or a rustling, dreaming of them,
watching their performative actions in trance rituals. The social context influences people as
to whether these sensations are taken seriously and whether they are attributed to spirits.
People who are afflicted with chronic pain, anxiety, depression and traumatic life events
often look for explanations of what is going on. Often, they put their trust in their close
friends and family. In Tanzania, these quite often suggest spirits as the cause of weird
sensations surrounding the affliction and send the patient to a ritual expert. In Germany,
such sensations will generally be remarked on to a lesser extent and while ritual and spiritual
explanations certainly exist, the majority of people would attribute them more to
psychosomatic effects rather than to the malevolent intervention of spirit beings. Alternative
therapeutics and vernacular rituals represent minority positions, not majority positions as
in Tanzania.
Spirit healing in Tanzania (and other places in the world) has no creed and no regular
Sunday service. Rather, it is situational, contextual and demand-oriented. Participation thus
does not depend on a belief in the spirits themselves, but rather in the efficacy of the healing
process attributed to them. When I spoke with people in Tanzania, many of my interlocutors

13
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

were critical of superstitious beliefs in spirits while accepting the rituals as part of everyday
life – a good and helpful part. Some saw demonic actions everywhere at work and were
convinced believers in the necessity of exorcism. Some medical students saw demons at work
in psychiatric patients under their care, but could not express this in medical terms. Some
were very critical of spirits until they were afflicted themselves and became convinced that
spirits were the root cause. And some mounted awareness campaigns to reduce the suffering
caused by exorcisms and witch hunts arising from the false belief in superhuman powers.
For the researcher, the question of belief is suspended during the collection and
analysis of data (methodological agnosticism). It is important, however, to remember that
belief is not a switch that has an ‘on’ and an ‘off’ position. Certainly, belief can be
propositional resting on a system of symbolic structures called theology, with which one
might agree, or not. But belief is just as much a question of interpreting sensations, trust,
childhood memories and storytelling. Belief-as-sensation is gradual and can be trained, a
process Tanya Luhrmann calls absorption (Luhrmann 2020). Scholars of religion should
therefore not bracket belief categorically, but rather pay attention to those details of
narrative, performance, sensation and argumentation that give clues as to how reality is
perceived and interpreted and how belief is created (Koch and Wilkens 2020). The treatment
of individual patient’s traumas, the ritual structures of trance initiation groups, and a given
society’s perception of spirits all contribute to the dynamic development of belief and
scepticism among practitioners and academics alike (van den Port 2005).

7. Conclusion: Research and friendship

I think it is best to do research with a friend rather than with a person you do not know personally. It is
also important to note that what is best is not just being a friend but living together. That gave us
enough time to discuss so many questions of research and of our personal lives. Living far apart it ’s not
easy to meet the target. To me as a Christian, and as a mother, with other family matters to cope with,
if my friend is close to me, I can do better for both sides. I can’t even distinguish personal and research.

Blending the personal and the academic gives anthropologists and scholars of religion unique
insights into other cultures and religions. My friend and assistant Anna and I have attempted
to shed some light on this way of gathering, contextualising and theorising data from worlds
that stand apart from one’s own. This type of immersive and cross-cultural approach to
research opens up vistas to current theories on aesthetics, affect and embodiment, but it also
emphasises the need to reflect on the humanity and ethics of the fieldwork encounter.
Though this may be quite challenging, it certainly is also very rewarding. In any way, it goes
far beyond the circumscribed project standing at the centre of a PhD dissertation.
Research of this type, however, continues to rest on the basic assumption that a Euro-
American researcher travels to Africa (or someplace similarly exotic) in order to ‘collect’ data
which then belongs to him/her within the framework of the research project. Hopefully, this
attitude will change as more and more universities in Africa, Europe and the US (and other

14
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

places) institutionalise decolonial methodologies through research funding, professorships

and conferences. What remains for the future is the normalising of ethnographic fieldwork
being conducted by Africans (and others) in Europe, not just among diaspora populations,
but with regard to every sphere of life. This should not just be an attempt to ‘provincialize
Europe’ and Western academic methodologies, rather it should actually help balance out the
amount of research being conducted on all continents alike.

Recommended Reading
Though both my personal and my academic biography are quite different from his, it is
perhaps William Sax’s (2014) approach to participant observation on spirit possession that
has most deeply influenced my own perceptions of the translation process between life and
academe as presented in this article. There are, however, also many other noteworthy
collections on fieldwork experiences in a variety of locations and concerning a variety of
topics. Borneman and Hammoudi (2009) and Faubion and Marcus (2009) may serve as
examples. There is very little literature on fieldwork and participant observation in the study
of religion, but the short summary by Graham Harvey (2011) provides excellent insights into
similar spirit-oriented research topics as I have discussed here.
In anthropology, interviews are an essential tool for obtaining different kinds of
information – expert knowledge on a specific subject, personal narratives or the type of
arguments in a conflict. But techniques of how to conduct and later analyse an interview are
seldom discussed on a methodological level. I therefore turned to sociology and the great
work being done there on biographical interviews and other types of narrativity (Lucius-
Hoene and Deppermann 2000, 2004). Brembourg (2011) also provides an invaluable overview
of sociological interviewing techniques specifically for the study of religion.

Bibliography
Behrend, Heike, and Ute Luig, eds. 1999. Spirit Possession. Modernity and Power in Africa. Oxford: James
Currey.
Blommaert, Jan, and Dong Jie. 2020. Ethnographic Fieldwork: A Beginner’s Guide. 2nd edition. Bristol, UK,
Blue Ridge Summit, PA: Multilingual Matters.
Borneman, John, and Abdellah Hammoudi, eds. 2009. Being There: The Fieldwork Encounter and the Making
of Truth. Berkeley: University of California Press.
Bremborg, Anna Davidsson. 2011. “Interviewing.” In The Routledge Handbook of Research Methods in the
Study of Religion, edited by Michael Stausberg and Steven Engler, 310–22. London, New York:
Routledge.
Brown, Candy G., ed. 2011. Global Pentecostal and Charismatic Healing. Oxford: Oxford University Press.
Caplan, Pat. 1997. African Voices, African Lives. Personal Narratives from a Swahili Village. London and New
York: Routledge.
Faubion, James D, and George E Marcus, eds. 2009. Fieldwork Is Not What It Used to Be: Learning
Anthropology's Method in a Time of Transition. Cornell paperbacks. Ithaca: Cornell University Press.

15
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

Franke, Edith, and Verena Maske. 2011. “Teilnehmende Beobachtung als Verfahren der
Religionsforschung. Der Verein Muslimische Jugend in Deutschland e.V.” In Religionen erforschen:
Kulturwissenschaftliche Methoden in der Religionswissenschaft, edited by Stefan Kurth and Karsten
Lehmann, 105–34. Wiesbaden: VS Verlag für Sozialwissenschaften.
Grieser, Alexandra. 2021. “‘European History of Religion’ Revisited: Modelling a Pluralist Approach.”
In ‘Religion in in Culture – Culture in Religion’: Burkhard Gladigow's Contribution to Shifting Paradigms in
the Study of Religion, edited by Christoph Auffarth, Alexandra Grieser and Anne Koch, 191-220.
Tübingen: Tübingen University Press.
Harvey, Graham. 2011. “Field Research: Participant Observation.” In The Routledge Handbook of Research
Methods in the Study of Religion, edited by Michael Stausberg and Steven Engler, 217–44. London,
New York: Routledge.
Knoblauch, Hubert. 2003. Qualitative Religionsforschung. Religionsethnographie in der eigenen Gesellschaft.
Paderborn u.a. Schöningh UTB.
Koch, Anne, and Katharina Wilkens, eds. 2020. The Bloomsbury Handbook of the Cultural and Cognitive
Aesthetics of Religion. London: Bloomsbury Academic.
Larsen, Kjersti. 2008. Where Humans and Spirits Meet: The Politics of Rituals and Identified Spirits in Zanzibar.
New York: Berghahn Books.
Lewis, I. M., Ahmed al-Safi, and Sayyid Hurreiz, eds. 1991. Women's Medicine. The Zar-Bori Cult in Africa
and Beyond. Edinburgh: Edinburgh University Press.
Lucius-Hoene, Gabriele, and Arnulf Deppermann. 2000. “Narrative Identity Empiricized: A Dialogical
and Positioning Approach to Autobiographical Research Interviews.” Narrative Inquiry 10 (1): 199–
222. doi:10.1075/ni.10.1.15luc.
———. 2004. Rekonstruktion Narrativer Identität. Ein Arbeitsbuch zur Analyse Narrativer Interviews.
Wiesbaden: Verlag für Sozialwissenschaften.
Luhrmann, Tanya M. 2020. “Absorption”. In The Bloomsbury Handbook of the Cultural and Cognitive
Aesthetics of Religion, edited by Anne Koch and Katharina Wilkens, 85-96. London: Bloomsbury
Academic.
Nisula, Tapio. 1999. Everyday Spirits and Medical Interventions. Ethnographic and Historical Notes on
Therapeutic Conventions in Zanzibar Town. Saarijärvi: Gummerus Kirjapaino Oy.
Promey, Sally M., ed. 2014. Sensational religion: Sensory Cultures in Material Practice. New Haven: Yale
University Press.
Quack, Johannes. 2014. “Porous Dividuals? Complying to Healing Temple (Balaji) and a Psychiatric
Out-Patient Department.” In Asymmetrical Conversations: Contestations, Circumventions, and the
Blurring of Therapeutic Boundaries, edited by Harish Naraindas, Johannes Quack, and William S. Sax,
56–84. Epistemologies of Healing 14. New York, NY: Berghahn Books.
Sax, William. 2014. “The Multiple Worlds of Ethnographic Fieldwork: A Personal Account.” Zeitschrift
für Ethnologie 139: 7–22.
Stausberg, Michael, and Steven Engler. 2011. “Introduction: Research Methods in the Study of
Religion/s.” In The Routledge Handbook of Research Methods in the Study of Religion, edited by Michael
Stausberg and Steven Engler, 3–20. London, New York: Routledge.
Thompson, Katrina D. 2017. Popobawa: Tanzanian Talk, Global Misreadings. Bloomington and
Indianapolis: Indiana University Press.
van de Port, Mattijs. 2005. “Circling Around the Really Real. Spirit Possession Ceremonies and the
Search for Authenticity in Bahian Candomblé.” Ethos 33 (2): 149-79.

16
 Marburg Journal of Religion Vol. 25 No. 1 (2024) – Article 6

Wilkens, Katharina. 2011. Holy Water and Evil Spirits: Religious Healing in East Africa. Berlin: Lit.
———. 2014. “Inkorporierte Imagination. Geistertänze und Exorzismus in Ostafrika.” In Religion -
Imagination - Ästhetik. Vorstellungs- und Sinneswelten in Religion und Kultur, edited by Lucia Traut and
Annette Wilke, 107-30. Göttingen: Vandenhoeck & Ruprecht.
———. 2018. “‘Instant Miracles Are Rare, but It Happened to Me’. Faith-Healing in Urban Tanzania.”
Numen 65 (2-3): 207–31.
———. 2020. “Narrating Spirit Possession.” In Narrative Cultures and the Aesthetics of Religion, edited by
Dirk Johannsen, Anja Kirsch, and Jens Kreinath: Brill.

17