Witchcraft, witch killings and Christianity: The works

of religion and parallel cosmologies in Tanzania

Terje Oestigaard1

Introduction
According to one statistical survey, 60 per cent of the population of Tanzania is
Christian, 36 per cent Muslim and 4 per cent subscribes to another or no reli-
gion. What is more striking about this survey is that 93 per cent of Tanzanians
believe in witchcraft (Christians 94 per cent, Muslims 92 per cent) and tradi-
tional African religious beliefs and practices are prevalent among 62 per cent
of the population (Pew Forum on Religion and Public Life 2010: 34, 64, 178).
Although one should be cautious about these statistics, they may nevertheless
indicate the ubiquity of witchcraft in Tanzania and the prevalence of traditional
beliefs among even Christians and Muslims. The figure of 93 per cent regarding
belief in witchcraft is remarkable, if accurate, and implies that witchcraft beliefs
are far more common in Tanzania than in other countries. Even if the actual
percentage is considerably lower, it would seem that witchcraft is widespread.
The Sukuma group, numbering about five million, mainly live along the
southern shores of Lake Victoria in Tanzania. For more than a century, they
have been integrated into the global world, but in varying degree, and mission-
aries have been spreading the Gospel and Christianising the Sukuma since the
1880s. Traditionally, the society was culturally and cosmologically structured
around the chief and rainmaking. Everything depended on the rain. The chief
was responsible for providing the life-giving rains believed to heal the land, a
power that afforded him legitimacy as ruler. Both chieftainship and rainmak-
ing have now disappeared, while at the same time Christianity is spreading
and both witchcraft and witch killings are flourishing (with more than 500 al-
leged witches killed annually in Tanzania, predominantly among the Sukuma).
Similarly, the role of the ancestors has diminished and tradition is no longer as
important in culture and cosmology. In a religious context, rainmaking is per-
ceived as not working anymore and the role of ancestors is declining. Thus, the
overall questions are: Is the declining role of ancestors enabling both the spread
of Christianity and the increase in witchcraft? How and why are witchcraft and

1. This chapter was presented as a paper at the Satterthwaite Colloquium in Grasmere, the
Lake District, England in 2014, and builds on T. Oestigaard (2014) Religion at Work in
Globalised Traditions. Rainmaking, Witchcraft and Christianity in Tanzania. Newcastle:
Cambridge Scholars Press.

182
 Witchcraft, witch killings and Christianity

Christianity as religious practices seemingly working perfectly together? What

are the consequences when traditions disappear and how are society and cosmol-
ogy being restructured? These questions will be analysed through an approach
that emphasises the work and works of religion.

How religions work beyond structural functionalism

Notwithstanding exegesis, dogma and theology, religions are always concerned
with practical human problems in some way. Our approach is not concerned
with the origins of religions so much as their functions. We relate and explain the
force of religious ideas by reference to the needs of individuals in their everyday
lives, not to society and its forms, Reynolds and Tanner argue. They continue:
We have become convinced by the evidence worldwide that the function of
religions is to respond to human needs, to help people at times of personal crises
(e.g., at funerals), or when they are undergoing a change of status (e.g., at wed-
dings), or generally in relation to everyday strains of normal life. Ours is thus a
functional approach (Reynolds and Tanner 1995:15). Moreover, religion caters
to needs at all levels, but perhaps mostly at the more basic level However
the concept of need is not sufficient to explain what religions appear to be
doing in the world (Reynolds and Tanner 1995:305).
By emphasising that religion functions and works, it is necessary to clarify
my meaning in the term religions work. Concepts such as work and in particu-
lar function have, as part of the criticism of structural functionalism, gained
negative connotations in the history of anthropological thought. In Ritual and
religion in the making of humanity, Roy Rappaport asserted that neither religion
as a whole nor its elements will, in the account offered to them, be reduced to
functional or adaptive terms. He continued: An account of religion framed,
a priori, in terms of adaptation, function or other utilitarian assumptions or
theory would paradoxically, defeat any possibility of discovering whatever
utilitarian significance it might have by transforming the entire inquiry into a
comprehensive tautology (Rappaport 2001:2). This example suffices for me to
make my point: I do not perceive religion in adaptive terms.
Rather, allow me to approach religion from the perspective of the divine
realm (or, at least, how it is perceived and presented by believers), and start with
Christianity and its Jewish antecedents. Christianity works and very much so,
if we are to believe Christians. In Hebrew myth, there is creation from noth-
ing: And God said. Let there be light; and there was light (Gen. 1:3). By the
word alone, God created the cosmos and the world and all within it in only six
days. This is the kind of work I refer to when I say religions work. Importantly,
in Christianity there is a fundamental doctrine that God works in mysterious
ways. Thus, humans may or may not know how and why God works, but that
does not matter. Even if humans have only partial knowledge of this, they have

183
Terje Oestigaard

sufficient to adhere to the cosmic laws given by God. And here God works again.
Jesus was crucified and by taking on human sin, he ensured salvation for all. As
a consequence, if one follows the rules, one may attain eternal life in heaven.
This is a huge promise and a durable work of a God what more can a devotee
wish for? Miracles are another way of illustrating that religion works, and in this
particular case, the Christian God works. Believers truly believe that miracles
take place, that prayers and wishes are suddenly fulfilled, and that sometimes
miracles happen without human requests. These are simply the works of God
originating in and fulfilled through his own plans. However, for sinners, that is,
those who act against God and his will, and this is also a sign of free will (Hertz
1996), God has worked in other ways to prepare a rather unpleasant destiny for
eternity, hell in its various elaborations throughout history.
However, in emphasising the function and work of gods, one must also ad-
dress questions of time and scale. Gods may once have worked in a primordial
era, but may not do so anymore. This does not alter the ontological status of
the gods as existing: in a cosmic perspective, humans are not in a position to
demand that gods work for them (indeed, in Christianity this is seen as heresy).
The Supreme Being among the Sukuma is one such god. It is impossible to reach
him through sacrifices and prayers and he is so remote that he does not interfere
in the fate of individuals. He is not working any longer because he has done his
job. What he created, exists and has never changed, and therefore he has with-
drawn (Cory 1960:15). If there were no such conception of gods, cosmic princi-
ples or divinities working on a grand scale, including the Buddhist conceptions
of karma, there would be no religion as broadly defined. In practice, a kind of
Western atheism would prevail.
Thus, I am not concerned here with the origins or existence as such of re-
ligion. My approach is the believers approach in practice: God, gods and the
ancestors exist, including various malignant forces such as the devil, roaming
ghosts, etc. But they do not merely exist in another sphere: in varying measure,
they influence this world. From this and the believers perspective, what are the
beliefs about how the spirits and divinities of the other world, which are supe-
rior to humans, function and work in this world? How do they intervene and
in which realms do different religions (defined broadly) impact worldly matters?
This approach also emphasises which rituals are performed and how people can
approach their gods, divinities and ancestors with a view to changing daily life
in this world. As I will argue, Christianity and the world of the ancestors, in-
cluding the practice of witchcraft, operate on different premises and affect this
world in substantially different ways. Precisely because of this, they can coexist
and work hand in hand.
A common feature of many African cosmologies is the division of the world
into two distinct realms. One is the visible or the manifest world, which is ob-

184
 Witchcraft, witch killings and Christianity

vious to all, and the other is the invisible or unseen world, which is neverthe-
less as real as the visible world. Indeed, it is more real in that it structures and
defines the premises of this world. The visible world is the everyday world of
living farming, collecting water and fuel, political and economic matters, etc.
The invisible world comprises the ancestors, God and the realm of witchcraft
and the occult, among other spiritual forces. These two realms are intrinsically
linked and the invisible forces largely determine outcomes in the visible world.
The visible world is therefore shaped by a deeper, more real reality, and con-
sequently, it is of utmost importance to control the spiritual and occult forces
that would otherwise harm society (Sanders 2001:169). In many cosmologies,
humans facing problems turn to the ancestors or gods. Religion works and may
have practical consequences here and now.
On the other hand, Christianity, in practice, works quite differently. The
here and now is largely omitted from the realm of this religion. Whether peo-
ple starve or die, plagues haunt society and diseases kill animals, women are
debarred and life-giving rains do not come at the right time, there is a double
perception at work. On one hand, these misfortunes are not believed to be the
Christian Gods work. Although in principle the Christian God can control
these events and processes, since he is believed to be almighty, it is also believed
that such mundane problems are not the concerns of God in his daily and prac-
tical work. Or rather, although mitigating worldly problems may be a concern of
God, usually he does not solve them here and now. Rather, these misfortunes are
placed within the larger cosmological framework in which God works. Thus, on
the other hand, such misfortune is also commonly and in varying degree seen
as Gods penalty for sin. Throughout Christian history, God is believed to have
punished sinners and unbelievers collectively with plagues, droughts and floods.
In other words, it is not for God to solve the problems he has brought down on
sinners and misbelievers. That is for humans to do, cosmologically through obe-
dience and repentance and practically by surviving as best they can.
This does not imply the Christian God does not work (in all senses of the
term, including penalising humans). To the contrary, he does very much work,
with eternal consequences for humanity, if one is to believe Christianity. How-
ever, his main active engagement in the cosmos is after humans die. The grace
of God may send some to heaven for eternity, whereas others may be doomed to
hell. Indeed, this is a very powerful way of having an active god at work, and was
ultimately used by missionaries as the main argument why one should convert
to Christianity.
Apart from eschatology and what may happen after death, the ways in which
the Christian God works as compared to the ancestors point to a crucial aspect
in ones attempts to understand religion and how it works in society. There are
two main differences. First, Christianity mainly works in the other world and

185
Terje Oestigaard

the hereafter, whereas ancestors work in the here and now in society. Thus,
the religions resolve different problems in different realms. Christianity may
promise an eternal life in heaven, but cannot (or at least does not) procure the
life-giving rains upon which people are utterly dependent for life and well-being
or solve any other acute problems here and now. The ancestors can resolve the
latter problem. This brings us to the second difference: the ancestors are active
and can be contacted and requested to solve current problems. Thus, the ances-
tors can be activated, or, in other words, humans may manipulate the divine world
for their own betterment. This is also in the interests of the ancestors, since the
descendants are their heirs and family. An approach such as this is not possible
in Christianity. There is no way humans can manipulate God to secure health
and wealth through ritual and sacrifice. One may pray, but the outcomes are
highly uncertain. Miracles are believed to take place in Christianity, but where,
when and why is a mystery. Even Christians acknowledge that miracles to solve
problems in this world happen rarely. The rule of the game is that humans have
to solve their problems themselves. The ways and means by which humans en-
gage in the world and solve their problems are, however, judged after death, and
include severe penalties for eternity.
This has created a space for different religious practices to operate on different
scales by different means. Wisjen and Tanner state that the Sukuma see religion
in terms of the options it provides rather than the obligations it creates. They
have periodic problems and their religious practices tend to be periodic (Wijsen
and Tanner 2000:30). Thus, conversion to Christianity is not a straightforward
displacement of one religion by another. Instead, it is a process of syncretism or
parallel cosmologies. The different religious systems offer different possibilities
and solutions to various problems. The benefits of Christianity are also largely
its main shortcoming it does not solve the most acute problems here and
now, the problems that have to be resolved. Consequently, religious solutions
are sought elsewhere. This highlights the premises on which the Sukuma world
works. Causation in their thinking is animate rather than inanimate. An event,
particularly an unfortunate one, has to be caused by someone or something, liv-
ing or dead, with malevolent intentions toward the sufferer. There are no pure
accidents in Usukuma (Wijsen and Tanner 2000:47). As will be discussed, this
is the realm in which belief in witchcraft flourishes.
Although it was held that one could achieve much the same results by pro-
pitiating the ancestors, there was a sense that in the modern world it is better to
use the modern way and not approach the ancestors. They are connected with
the past and tradition, but witchcraft is connected to the challenges of today and
tomorrow. Thus, witchcraft has to some extent replaced the role of ancestors and
is perceived as a way of being modern.

186
 Witchcraft, witch killings and Christianity

Witchcraft and witch killings

According to the 2002 Tanzania Witchcraft Act (Cap. 18 [R.E.2002]), witch-
craft is defined as sorcery, enchantment, bewitching, the use of instrument of
witchcraft, the purported exercise of any occult power and purported possession
of any occult knowledge (Tanzania Human Rights Report 2011: 33). All acts
of witchcraft are punishable if murders are committed, and in January 2009
Prime Minister Misengo Pinda stated that those caught in the act of murdering
albinos should be killed on the spot (Bryceson et al. 2010: 374).
A phenomenon peculiar to the Sukuma in Tanzania has been the intensity
of witch killing, although such killings also occur in other regions. Since the
1960s, witch killings have increased significantly in Sukumaland (Abrahams
1994:15). According to Tanner in the 1950s, old men could still remember that
before the Germans occupied the country witchcraft was rare (Tanner 1956a:
443). Similarly, in the 1960s witchcraft incidents were rare (Per Brandstrm
personal com.). Still, the problem of witchcraft was acknowledged by the British
and the territorys high court produced a table of murder statistics in mid-1944
(Mesaki 1993: 64).

Table 1: Murder cases in Sukumaland 19351943

Year Murder cases Witch-related
1935 35 5
1936 20 4
1937 39 4
1938 40 7
1939 26 3
1940 41 5
1941 30 3
1942 35 4
1943 56 11
Source: Mesaki 1993.

There are, of course, uncertainties about the actual number of witch killings and
many killings may not have been reported to the authorities. In any case, even if
the actual number was two or five or even ten times higher, it would nevertheless
have been modest compared to the rate from the 1970s onwards. In short, there
seems to have been a dramatic increase in murders associated with witchcraft.
Importantly, statistics in Tanzania are highly unreliable, and there are serious
methodological obstacles regarding witch killings. Apart from the fact that the
statistics are not published and are largely based on second- or third-hand ob-
servations (for example, anonymous police sources largely impossible to check),
it is also unclear how witch killings are documented, or what is categorised as

187
Terje Oestigaard

a witch killing. Witchcraft may also be used as a local term for violent death
and police officers often use it loosely (Stroeken 2010: 199). Thus, the actual
numbers may be higher or lower than the figures presented here.
The killing of suspected witches became gradually more noticeable in the
1960s and notoriously so by the mid-1970s. By the end of the 1980s, the situ-
ation seemed to be out of control and the government instituted the Mongela
Commission on witchcraft in 1988 to investigate the phenomenon. The com-
missions conclusions were alarming (Mesaki 2009: 72-3) Between 1970 and
1984, 3,693 suspected witches were killed 1,407 men and 2,286 women (Me-
saki 1993: 98). Here it is important to add a comment on the statistics, since the
number of killed witches is 3,693 whereas the number of witch-related cases was
3,333. Mesaki explains: It should be noted that the number of cases does not
necessarily match the number of those killed because a single case can lead to
several deaths while in other cases no murders occur (Mesaki 1993: 99).

Table 2: Witch killings in Tanzania 19701984

Year Cases Men Women
1970 106 31 84
1971 181 50 133
1972 133 49 106
1973 308 135 192
1974 216 74 156
1975 242 87 168
1976 219 68 175
1977 189 94 131
1978 164 41 172
1979 225 81 163
1980 220 116 119
1981 294 199 158
1982 221 97 151
1983 324 144 212
1984 291 141 163
Total 3,333 1,407 2,286

Source: Mesaki 1993.

Mwanza and Shinyanga regions accounted for no fewer than 2,246 of these
witch killings, and in the following period, from 1985-88, a further 826 were
killed in Sukumaland, for a total of 3,072 in this area between 1970 and 1988.
The statistics on the killings from 197084 were broken down on a regional
basis by Mesaki (1993: 99).
Thus, the killing of witches has mainly been a Sukuma phenomenon and
among the 2,246 Sukuma killings before 1984, 1,869 were women and 377

188
 Witchcraft, witch killings and Christianity

Table 3: Regional distribution of witch killings 19701984

Region Cases Men Women
Iringa 132 120 16
Kagera 82 77 70
Kigoma 17 10 7
Mara 3 2 1
Mbeya 13 13 -
Morogoro 1 1 -
Mtwara 1 6 -
Mwanza 1,098 269 927
Pwani 2 1 1
Ruvuma 4 3 1
Singida 731 675 185
Shinyanga 1,022 108 942
Tabora 227 122 133
Total 3,333 1,407 2,286
Source: Mesaki 1993.

were men aged over 15 years, a ratio of 5 to 1. Of the women victims, 62 per
cent were 41 and older, that is, post-menopausal. On a monthly basis, some
10-12 people were killed for suspected witchcraft among the Sukuma (Wijsen
and Tanner 2002: 135), with an average of 160 deaths a year in Mwanza and
Shinyanga regions. For Tanzania as a whole, there were 246 killings per year on
average.
From 1984 to 1993 there are no reliable figures for killings of alleged witches.
A newspaper reported in 1998 that 325 people had been killed in Shinyanga
region from 1996 to 1998: 133 in 1996, 102 in 1997, and 90 between January
and October 1998. Another survey conducted by TAMWA (Tanzania Media
Womens Association) revealed that between 1993 and 1998 in Mwanza region
alone, 318 elderly people had been killed. According to a police official in Shin-
yanga region, however, when other innocents who are killed in remote villages
are taken into account, murder would be seen to occur more or less daily in
Shinyanga and Mwanza regions (Mesaki 2009:73), although not all these deaths
are related to witchcraft. A leaked Ministry of Home Affairs report indicated
that 5,000 people had been killed between 1994 and 1998 (Duff 2005), al-
though not all these deaths were necessarily related to witchcraft.
Prime Minister Pinda told parliament in late January 2009 that 2,866 el-
derly people accused of being witches had been murdered in the past five years,
for an average of 573 a year (Banda 2009). The police in Mwanza reported in
February 2009 that more than 2,585 old women had been killed in eight (of 21)
mainland regions of Tanzania over the previous five years (Tanzania Human
Rights Report 2009: 21).

189
Terje Oestigaard

Table 4: Witch killings in 8 regions 20042008

Region Deaths in the five years prior to February 2009 Average deaths per year/region
Mwanza 698 140
Shinyanga 522 105
Tabora 508 102
Iringa 256 52
Mbeya 192 39
Kagera 186 37
Singida 120 24
Rukuwa 103 21
Total (eight districts) 2,585 517
Source: Tanzania Human Rights Report 2009.

Based on these numbers, every third day an old woman accused of being
a witch was killed in Mwanza region, and every fourth day in Shinyanga. The
actual number of killings of suspected witches was believed to be higher, be-
cause not all such killings are reported to the police. As one respondent from
Shinyanga put it: Its risky to inquire of the police about a relative killed due to
witchcraft because you will be regarded as an accomplice. And they might
end up taking your life as well (Tanzania Human Rights Report 2009: 21).
The number of witch killings may total more than 1,000 a year. In a news-
paper interview in 2005, Simeon Mesaki asserted that in Shinyanga region a
minimum of 300 witch killings take place annually, and the total was most
likely the same in Mwanza. According to an official who wanted to remain
anonymous: The government figures are very low, not accurate. I know a much
higher number, and even that is not the full situation (Duff 2005).
Official statistics and statements suggest that in the period 197084 an av-
erage of 246 people were killed as witches, whereas today almost 600 people
are killed annually. In each year from 1970 to 1988, an average of 160 persons
were murdered in Mwanza and Shinyanga region, whereas this number has now
increased to 245. In Tabora region, the annual average has increased from 17
killings to 102. However, according to the Tanzania Human Rights Report
2010, in that year about 50 people were killed in Tanzania because of witchcraft
accusations. Although this number is provisional, compared to previous years
there has been a significant decrease (Tanzania Human Rights Report 2010:57).
In 2011, however, the national figures were higher according to the Tanzania
Human Rights Report:
Between 2005 and 2011, about 3,000 people were lynched to death by fearful
neighbours who believed them to be witches. This suggests that an average of
500 elderly people, old women in particular [are] killed [on] suspicion of be-
ing witches annually all over Tanzania. For instance, 242 people were killed due
to witchcraft beliefs in Shinyanga alone from January 2010 to June 2011 [P]

190
 Witchcraft, witch killings and Christianity

olice statistics [also show] the growing trend of killings due to witchcraft beliefs
from 579 in 2010 to 642 in 2011 (Tanzania Human Rights Report 2011:34).

The discrepancy between the numbers given in the 2010 report and the 2011
report was explained by new data provided by the police in a letter. In the Tan-
zania Human Rights Report 2012 it is reported that 630 persons were killed as
alleged witches (Tanzania Human Rights Report 2012: 31). In the Tanzania
Human Rights Report 2013 it is written that witchcraft related killings claimed
the lives of 765 people 505 women and 260 men (Tanzania Human Rights
Report 2013: 36). In 2013, a new form of abuse was encountered, with many
people being buried alive for various alleged misdemeanours, often including
suspicions of witchcraft. In sum, despite gaps in official statistics and the short-
comings of other surveys, the conclusion appears to be unequivocal: the killing
of accused witches has increased dramatically.
As a comparison, in the whole of Europe there were about 90,000 prosecu-
tions of witches and some 45,000 executions during the 300 years the main
witch craze lasted (Levack 2006:23). If the rate of witch killings in Tanzania
is 500 per year or more, then there will be more witch killings in Tanzania in
only a century than there were during Europes much longer witch craze. The
estimates may be too high, but even if only 150 witches are killed each year in
Tanzania, this is equal to the rate of executions throughout Europe. Indeed, the
witch craze seems to be spiralling out of control, as evidenced by, inter alia, the
killing of albinos and the even more recent trend to kill people for their wealth.

Albino killings and the mining industry

Albinos have traditionally been stigmatised and discriminated against in society
and there are numerous rumours that albino children have been killed at birth
as mercy killings. In agro-pastoral communities, albino babies have also been
placed in the cattle kraal gateway and trampled to death, but those who survived
were allowed to live. Births have been reported as stillbirths and albinos have
often been called zeruzeru, which is believed to refer to zero, an archaic term
for ghost-like creatures, probably referring to the lack of pigment and denial
of personhood (Bryceson, Jnsson and Herrington 2010: 3678). In the past,
giving birth to an albino was seen as a bad omen. The whole family was stig-
matised. Other villagers were not allowed to marry the albino, but people were
also reluctant to marry any other member of that family for fear that their own
children would be albinos. Albinos were stigmatised in the same way as people
with leprosy. Even a visit by an albino was seen as a bad omen and people refused
to eat and share food with them.
Today, however, this has in principle changed, and people marry members of
families with albinos. Even so, the lives of albinos have gone from bad to worse.

191
Terje Oestigaard

When albinos are born, men often accuse their wives of unfaithfulness with the
devil or white people, and there are also widespread beliefs that children born
with albinism are punishment for an ancestors wrongdoing or for having sex
during menstruation (Ackley 2010: 44). The sexual connotations of albinism
have taken new form in recent years, the most disturbing being the belief that
sex with an albino may cure HIV/AIDS (Baker, Lund, Taylor and Nyathi 2010:
177). Albino women are even raped because of this belief (Alum, Gomez and
Ruis 2009: 89).
Albino killings in Tanzania began in 2006, but there were also earlier re-
ports. From 2000 to 2009, the number of albinos killed was 68 (Tanzania Hu-
man Rights Report 2011: 36). However, from 2007 to 2009 alone 59 albinos
were killed (six in 2007, 37 in 2008 and 16 in 2009) and nine were mutilated. In
2010 and 2011, there were no reports of albinos being killed (Tanzania Human
Rights Report 2010: 589), whereas one albino was killed in 2012.
The killing of albinos among the Sukuma is a recent phenomenon and
seemed to come from nowhere (Bryceson, Jnsson and Herrington 2010: 368,
379 fn. 10). In the 1980s and 1990s, there were rumours that bald people would
bring prosperity and as a consequence such people were reported decapitated
and their heads used in magic potions. Thus, apart from witch killings, there
are indications that people were killed and used for various purposes among the
Sukuma, although these practices would also be part of the world of ancestors.
Still, the killing of albinos is new. Why albinos and why as part of witchcraft?
According to the Tanzanian government, there are about 1.5 million artisa-
nal miners out of a total population of some 40 million, and Mwanza and Shin-
yanga have seen more mineral rushes than other regions (Bryceson, Jnsson and
Herrington 2010: 379). The work is dangerous and the miners have no medical
insurance, so many of them believe they need powerful witchcraft protection.
Moreover, if they have money, there is always the danger it might be stolen. They
also believe they need good luck to find the minerals. Thus, there is a double
need for witchcraft: protection and good fortune.
It is impossible to know which parts of an albinos body will be used, but of-
ten it is the legs from below the knees. It is believed that one cannot start work-
ing in the mines without first going to a traditional healer, and success or failure
depends on the effectiveness of the medicines. And even if a man has no place
or mine where he can start digging, if he has consulted a traditional healer he
can start searching anywhere, even outside the mines, and achieve success and
riches. The miner does not pay the healer in advance. However, after the miner
has struck gold or diamonds, he will return to the traditional healer, who will
instruct him on how to use his finds. In many cases, the traditional healer will
claim the first find as payment for providing good luck, and the next find of gold
or diamonds will belong to the miner. Consequently, many traditional healers

192
 Witchcraft, witch killings and Christianity

are exceedingly rich, at least by local standards. If a miner does not return to the
healer with his share, it is believed that all of his haul will magically disappear
and the miner will be left with nothing.
Some believe the origin of albino killings has its roots in macabre Nigerian
movies, which are popular among the miners when they have time off. One po-
lice officer allegedly commented that an influx of Nigerian movies, which play
up witchcraft, might have something to do with it, along with rising food prices
that were making people more desperate. According to other Tanzanian police-
men, the value of a killed albino, including all four limbs, genitals, ears, nose
and tongue, may amount to USD 75,000 (Ackley 2010). Others believe albinos
are killed because they are not useful and will not be missed in society, while
another explanation is the analogy between the rare occurrence of gold and the
rareness of albino charms. How this practice began will most likely remain hid-
den, but it is undoubtedly propagated by the waganga (medical practitioners).
Even though nobody really knows why and how albino charms work, the belief
that they do keeps such beliefs and practices alive (Bryceson, Jnsson and Her-
rington 2010: 3689, 371).
The use of body parts in medicines is based on the assumption that it is pos-
sible to appropriate another persons life-force literally through the consumption
of that person. It is also believed that the acquired life-power is much greater
if the body parts are removed while the victim is still alive (Vincent 2008:43).
This has other implications: The albino fetish has become the most expensive
because it is perceived as harnessing spirits that are far more powerful than any
plant or animal charm that waganga could otherwise offer (Bryceson, Jnsson
and Herrington 2010: 371). As a consequence, there has also been a transition
from healers to dealers (Bryceson, Jnsson and Herrington 2010: 364).
Although it is often claimed that the use of body parts in witchcraft is recent,
this is only partly correct. In Tanzania, among the Sukuma, for instance, there
were practices in the past involving such uses. These were, however, very limited.
Recent ritual innovation has led to an increase in such practices and to their in-
corporation into spheres where previously there were no such traditions. In this
sense, current witchcraft practices build on and continue older worldviews and
practices, which are reinterpreted and more widely adopted.
Chief Charles Kafipa of Bukumbi chiefdom explained past traditions of us-
ing body parts in medicines and how they worked (personal communication).
The killing of albinos and use of their body parts in medicines is not part of tra-
dition and, indeed, the chief argued strongly that it was against tradition. More-
over, according to him, those who kill albinos and use this type of medicine will
not be successful in their endeavours: killing people for medicine is murder and
has no ritual effects, and such killings are seen by all as truly horrible.
Still, according to the chief, the bodies of chiefs and albinos are very impor-

193
Terje Oestigaard

tant and powerful traditional medicines if they are properly handled by quali-
fied healers. Albino bodies are only effective as medicines if their owners die
naturally. The extraordinary power of such medicines derives from the belief
that albinos are composed of something different on account of their complex-
ion: it was believed there is something missing or something extra in their com-
position. Traditionally, albinos were buried beneath the interior mud floors of
their homes, with only the nearest family at the funeral. Thus, a healer could
only get body parts with the consent of family members. This practice was both
legal and had superior ritual potential. There were also beliefs that albinos are
somehow alive after death, and it is this that gives the medicines their special
power. Today, with cement floors in modern houses, albinos are buried in out-
side graves. This has resulted in the looting of albino body parts. Even where the
albino died naturally, such stolen parts are ineffective as medicine, because they
were taken without family consent. Thus, the agreement of relatives is necessary
if the medicines are to be effective, and the other practices now escalating do not
even work as ritual, the chief said.

Religions at work
The poignant question then becomes why both witchcraft and Christianity are
on the increase in Africa? What kinds of solutions do the different religious
systems offer, and how do they differ, since Christians also use witchcraft? The
short answer is simple: witchcraft is believed to be more efficient than Christi-
anity. More importantly, witchcraft is believed to work in this world, solving
mundane and practical problems here and now and creating health, wealth and
prosperity. It is generally agreed that the Christian God, although omnipotent,
does not bother about these trivial matters. His ambitions for humanity are
more fundamental, namely the ultimate destiny of humans in heaven or hell.
Still, for poor people without food and water or facing other miseries, the mun-
dane is far from trivial. Consequently, the problems of the mundane have to be
solved in one way or other. The Christian God is utterly silent in providing help
or solutions and uninvolved in daily life and the miseries of common people.
Thus, the parallel religious cosmologies work perfectly well together: one solves
the problems of this world and the other the problems of the next.
One may therefore say that Christianity has once again created a devil, al-
though not intentionally. By strongly opposing ancestral tradition for more than
a century, the church has largely undermined this tradition, but has also created
a cultural and religious vacuum. But no such vacuums can persist. The miser-
ies and problems the ancestors were believed to resolve continue, and may have
become more acute with increased poverty. Christianity as a religion is basically
not intended to resolve current miseries such as these, but witchcraft can do so
perfectly here and now, it is believed.

194
 Witchcraft, witch killings and Christianity

Regarding how God works within Pentecostal Christianity, or at least in

the preaching, I draw on two complementary examples. First, there were the
two American Pentecostal missionaries I met struggling to convert people. They
were frustrated by what they called the lack of self-sacrifice among the Sukuma.
One of the missionaries had refused to bless a congregant because he had not
paid his weekly dues to the church. According to the missionary, converts had
to follow the same principles as they themselves did and make sacrifices to God.
The missionaries had paid their air-tickets to Tanzania with their own money,
and each week they donated money to the church. These missionaries stated
that the person denied the blessing had once given money at the altar, and they
concluded he had money, or had to find it somewhere and prioritise God! As
they put it, what good are excuses that they had to support their family and not
the church when they face God on Doomsday?
This example highlights the emphasis on godly matters and their interaction
with worldly concerns. From these missionaries perspective, the world hereafter
is the sole reference point: what happens after death and the consequences of
being judged on Doomsday. However, worldly matters do matter to the people.
Children need school uniforms and fees have to be paid, the family needs food
on the table and medical bills have to be met. In the missionaries overall cosmic
framework, these are mere excuses.
Second, the charismatic founder of the successful Pentecostal International
Central Gospel Church in Accra, Rev. Mensa Otabil, gave some illuminating
insights into the ways God is believed to work in one of his many fire-and-
brimstone sermons. While blaming the sad state of affairs on African politi-
cians, he said: But you cant claim anyones money by faith its illegal. If you
want to have money, there is only one way: work I prayed to God to prosper,
but we have to change economic structures and social structures. If I dont have
that opportunity, I can pray all I want and Ill still be poor You can pray all
you want, but it wont help unless we start looking at the structures of our
nations (Van Dijk 2004: 1789). In other words, you can pray all you want in
church for food and money, it will not help: this is not the way God works. He
is not uninterested in his children and human suffering, but he has a bigger and
more important issue to deal with, the devil. Or something else.
A further example of how Christianity is generally believed to work can
be seen in the old woman I met during the drought of 2011 in Sukumaland,
one of the three worst droughts she had experienced in her whole life. She ar-
gued strenuously in favour of rainmaking rainmaking worked and provided
prosperity to the community. However, the tradition had been lost, which she
regretted, and now all had become Christians. During the drought of 2011, they
went to the church and prayed to God for rain and food, but as she said, nothing
happened and they left the church even hungrier. The Christian God does not

195
Terje Oestigaard

solve problems here and now and does not provide food on the table when it is
really needed.
From this perspective, Christianity, missionaries and evangelisation have
been highly successful in eradicating ancestral tradition as religious practice, but
have largely failed to replace the role of the ancestors with the Christian God:
these are basically different entities or realms working at different scales within
and among humans. God belongs to the other world, and so do the ancestors,
but whereas the ancestors, properly propitiated, help the living in their daily
lives, God does not intervene in this world among his children. Christianity has
therefore created one cosmology, but also left a whole cosmology ripe for rein-
vention. It has replaced the role of the ancestors in the world beyond with God,
but not fully displaced the role of ancestors in this world and their abilities to
solve problems for the living. Christianity has diminished that role in this world,
but not the logic of how the ancestors work. And that logic of causation has
proved more durable than the power of the ancestors themselves: this is point of
departure for witchcraft. It provides the means to manipulate or approach the
forces of the otherworld for the betterment of this world.
Whereas Christianity is immune from criticism about what happens in the
otherworldly realm (and on this side), this is where witchcraft is superior. Witch-
craft can, or at least is believed to, resolve any kind of problem in this world: it
is the source of wealth and success and can deprive others of the same riches and
resources and ultimately cause their deaths. The outcomes are possible to meas-
ure and adjudge in a world where poverty increases while some become exceed-
ingly rich. The precise ways in which witchcraft works are impossible to know,
but the mere fact that someone, somehow has gained immense wealth whereas
others suffer more and even die, is all the evidence needed that witchcraft works,
and is at work.
Accidents and misfortunes still happen and this is where witchcraft as logic
and religious system has its supreme force. Whereas in the ancestral cult such
misfortune could be explained in terms of the ancestors not solving the prob-
lems, a failing that over time could challenge the rationale of such beliefs,
in witchcraft there are always others to blame. Somebody else has employed
stronger and more effective medicines. Misery and evil, or the success of others
in becoming wealthy and powerful, becomes the ultimate evidence of why a per-
sons own medicines and witchcraft did not work. This circular, self-referential
evidence is complete, and the only way to break out of this vicious circle is by
employing even stronger and more dangerous medicines, ultimately by using
human body parts. The very logic of how witchcraft works is also the source of
its increase. The consequences of witchcraft can only be combated by employing
more witchcraft.

196
 Witchcraft, witch killings and Christianity

Conclusion
The burden of evidence in Christianity, in which it is strictly impossible to prove
the existence or effectiveness in this world of the Christian God, favours the
spread of witchcraft. Witchcraft also works in mysterious, albeit more mysti-
cal and magical, ways. Although there are many unknowns, witchcraft is still
more tangible both as regards how and why the rituals work and the outcomes
of the rituals. The healer makes medicines, using whatever ingredients, which
may include human body parts; he propitiates and incorporates the ancestors,
the other worldly realm well known in the communitys culture; and the client
physically employs the medicines either by ingestion, or by applying it to the
body or sprinkling it. The materiality of witchcraft and its operation on exist-
ing cultural premises and horizons of understanding strengthens the beliefs.
Everything is here and now from the healers and his medicines to the partaker
and his problems. And it continues the logic of causation on which the ancestral
tradition was based: it is possible to actively interfere with the spirits to improve
conditions in this world.
Explaining why witchcraft increases, based on a religious logic that religions
work although nobody can know for sure how, why and when, seems somehow
straightforward. People seek refuge in the magical world given the premise that
religion works. This is a world in which changing the premises and the outcomes
of daily life and affairs is believed to be possible. The church may also promise
this, but very rarely are the outcomes explicitly evident. The church does not
promise a one-to-one relationship between prayer and good health, successful
work, prosperous marriage or excellent exam results. Witchcraft does or is be-
lieved to do so. Whether this happens is another question, but then within this
logic there are sufficient ad-hoc explanations legitimising the cosmological sys-
tem and its effectiveness, even when it fails. Explaining why witches are being
killed, and at a seemingly accelerating rate, is more difficult, because an increase
in witchcraft does not necessarily imply intensified witch killing.
In a cross-cultural perspective, human sacrifice is at the pinnacle of any sac-
rificial system, but it is difficult to pinpoint the exact reasons for, or cultural
patterns behind, the prevalence among the Sukuma of human killing within a
ritual framework. Rainmaking rituals have historically included human sacri-
fice, as for instance in ancient Egyptian, Mayan and Aztec civilisations. Howev-
er, sacrifices carried out in this manner require a sacred person to consecrate the
offerings. The sacrificial victim incarnates the society as a whole and therefore
human sacrifices can be made only for the benefit of the collectivity (e.g. Valeri
1985:49). Strictly speaking, the killing of albinos for personal wealth in the
mining industry is at odds with current sacrificial theory. Since the empirical
data are not wrong and these theories are based on other data, it would appear

197
Terje Oestigaard

that something peculiar and particular is going on in Tanzania. From a purely

logical perspective, human sacrifice is understandable given certain premises
regarding how gods and the cosmos are perceived to work. From a human per-
spective though, it is harder to understand how anyone can believe in such prac-
tices, let alone carry them out.

References
Abrahams, R. (1987) Sungusungu: Village Vigilante Groups in Tanzania. African
Affairs 86(343): 17996.
Ackley, C. (2010) The Fetishization of Albinos in Tanzania. University of Chicago,
p. 44. Available at http://www.underthesamesun.com/sites/default/files/The%20
Fetishization% 20of%20Albinos%20in20Tanzania.pdf (Accessed on 2 June 2015).
Alum, A., M. Gomez and E. Ruis (2009) Hocus pocus, witchcraft, and murder: The
plight of Tanzanian albinos. International Team Report 2009 Tanzania. Available
at http://www.underthesamesun.com/node/7 (Accessed on 2 June 2015).
Baker, C., P. Lund, J. Taylor and R. Nyathi (2010) The myths surrounding people
with albinism in South Africa and Zimbabwe. Journal of African Cultural Studies
22(2): 169-81.
Banda, S. (2009) Tanzania: PM cries over albino killings. Africa News, 1 February.
Available at Africahttp://www.africanews.com/site/Tanzania_PM_cries_over_
albino_killings/list_messages/22930 (Accessed 25 May 2011).
Bryceson, D.F., J.B Jnsson and R. Herrington (2010) Miners magic: Artisanal
mining, the albino fetish and murder in Tanzania. Journal of Modern African
Studies 48(3): 35382.
Cory, H. (1960) Religious beliefs and practices of the Sukuma-Nyamwezi tribal
group. Tanzania Notes and Records 54: 1426.
Duff, O. (2005) Tanzania suffers rise of witchcraft hysteria. The Independent, 28
November. Available at http://www.independent.co.uk/news/world/africa/tanzania-
suffers-rise-of-witchcraft-hysteria-517157.html (Accessed 25 May 2011).
Hertz, R. (1996) Sin and expiation in primitive societies. Occasional Papers No. 2.
British Centre for Durkheimian Studies, Oxford.
Levack, B.P. (2006) The Witch-Hunt in Early Modern Europe. Third edition. London:
Longman.
Mesaki, S. (1993) Witchcraft and witch-killings in Tanzania. PhD thesis, University of
Minnesota.
Mesaki, S. (2009) The Tragedy of Ageing: Witch Killings and Poor Governance
among Sukuma. In Haram, L. and B. Yamba (eds) Dealing with Uncertainty in
Contemporary African Lives. Uppsala: Nordic Africa Institute, pp. 4290.
Pew Forum on Religion and Public Life (2010) Tolerance and Christianity in Sub-
Saharan Africa. Washington DC: Pew Research Center.

198
 Witchcraft, witch killings and Christianity

Rappaport, R.A. (2001) Ritual and religion in the making of humanity. Cambridge:
Cambridge University Press.
Reynolds, V. and R. Tanner (1995) The Social Ecology of Religion. Oxford: Oxford
University Press.
Sanders, T. (2001) Save our skins. Structural adjustment, morality and the occult
in Tanzania. In Moore, H. and T. Sanders (eds) Magical Interpretation, Material
Realities: Modernity, Witchcraft and the Occult in Postcolonial Africa. London:
Routledge, pp. 16083.
Stroeken, K. (2010) Moral Power. The Magic of Witchcraft. Oxford: Berghahn.
Tanner, R.E.S. (1956) The Sorcerer in Northern Sukumaland, Tanganyika.
Southwestern Journal of Anthropology 12(4): 43743.
Tanzania Human Rights Report 2010. Dar es Salaam: Legal and Human Rights.
Tanzania Human Rights Report 2011. Dar es Salaam: Legal and Human Rights.
Tanzania Human Rights Report 2012. Dar es Salaam: Legal and Human Rights.
Tanzania Human Rights Report 2013. Dar es Salaam: Legal and Human Rights.
Valeri, V. (1985) Kingship and sacrifice. Ritual and Society in Ancient Hawaii. Chicago:
University Press of Chicago.
Van Dijk, R. (2004) Beyond the rivers of Ethiopia: Pentecostal pan-Africanism and
Ghanaian identities in the transnational domain. In van Binsbergenand, W. and
R. van Dijk (eds) Situating Globality. African Agency in the Appropriation of Global
Culture. Leiden: Brill, 16389.
Vincent, L. (2008) New Magic for New Times: Muti Murder in Democratic South
Africa. Tribes and Tribals, Special Volume No. 2: 4353.
Wijsen, F. and R. Tanner (2000) Seeking a good life. Religion and society in Usukuma,
Tanzania. Nairobi: Paulines Publications Africa.
Wijsen, F. and R. Tanner (2002) I am just a Sukuma. Globalization and Identity
Construction in Northwest Tanzania. Amsterdam: Rodopi.

199