Yan Daudu

A Study of Transgendering Men in

Hausaland West Africa

By

Maarit Sinikangas

A Master Thesis in Cultural Anthropology

Department of Cultural anthropology and Ethnology
UPPSALA UNIVERSITY
May 2004
 Yan Daudu
A Study of Transgendering Men in
Hausaland West Africa

By

Maarit Sinikangas

A Master Thesis in Cultural Anthropology

Department of Cultural anthropology and Ethnology
UPPSALA UNIVERSITY
Supervisor: Sten Hagberg
May 2004
 ABSTRACT

This study deals with yan daudu, transgendering men in Hausaland, West
Africa. The focus is on their role in Hausa society and more precicely in
bori cult which is an old spirit possession cult. Yan daudu have mostly
been mentionned in footnotes and margins and often their position is
defined through different terms of sexuality –yan daudu have been
categorised as homosexuals or transvestites.
In this study yan daudu's sexuality is seen as secondary trait; more
important is their gender variation and productive specialisation. Still this
study considers largely sexuality in order to understand that sexual
categories (for example homo- and heterosexual) are not universally valid
consepts.
The role of yan daudu is studied as a part of the hausa society but
also as an individual choice. This is an attempt to bring the question of self
into antropological discussion. Theorethical base is found in
postructuralism.
Bori cult is studied beside yan daudu. It is one context where yan
daudu live and work. Special focus is put on the claim that the bori cult is a
marginal cult for marginal people, such as yan daudu and karuwai.
Karuwais are women that have often been compared to courtesans and they
are closely associated with yan daudu.

Keywords: yan daudu, bori cult, sexuality, homosexuality, gender, third

gender, marginality, Karuwai,

2
 CONTENTS

INTRODUCTION...................................................................4
Outline of thesis..................................................................7

1. BORI SPIRIT POSSESSION............................................10

Origins of bori....................................................................12
Maguzawa and Islam.........................................................14
Islam and bori....................................................................16

2. QUESTIONS OF MARGINALITY.................................19
Marginality of bori cult.....................................................22

3. YAN DAUDU...................................................................26

4. EARLY ETHNOGRAPHICAL STUDIES

ON SEXUALITY………………………………………..31

5. IS THERE "HOMOSEXUALITY"...................................34
Global aspects of sexuality................................................37

6. SEX AND GENDER DICHOTOMIES............................40

Poststructural theories......................................................44
Third gender?....................................................................47
Yan daudu as third gender................................................50

7. DISCUSSION...................................................................52

CONCLUSIONS...................................................................56

REFERENCES......................................................................59

3
 Introduction

"Where homosexuality was thought to occur in tribal

societies, in the sense of same sex desire coupled with
gender transformation of social role and dress, it has
been seen as the manifestation of something basic,
primitive, biological: a certain kind of essential nature
forcing its way out of the body. We now see how naive
such a view was". Gilbert Herdt (1994:73)

This text of Gilbert Herd is an excellent statement of how in the history of

anthropology the gender-crossing has been interpreted as an expression of
homosexuality. Such a simplifying bias has also been the case in the
writings of Yan daudu (sing. Dan Daudu), that is effeminate men who
have been mentioned in the context of the Hausa Bori cult in the northern
part of the Nigeria and the southern part of Niger.
When I first got interested in yan daudu my main focus was to
study them as homosexuals since that is how they have been categorised in
most of the cases. I found few texts on yan daudu. Most of the authors
mentioned yan daudu in margins or footnotes naming them as
homosexual, transvestites or deviants. The more I learned about yan daudu
the more I felt it was wrong to call them homosexuals and the more
curious I became. Finally, I had to ask what does the term "homosexual"
even mean. What makes yan daudu homosexuals? In fact, are they
homosexuals according the conventional definition of same-sex sexual
relations?
I have three purposes for this paper: first, to deconstruct the terms
homosexual and homosexuality and to show that they are not globally
valid concepts. The term homosexuality is not appropriate for describing
all same-sex relations because of the term’s simplifying bias. Even though
a number of references indicate that many yan daudu have same-sex

4
relations, the role of yan daudu is not explained in terms of homosexuality.
The term has, first of all, too many connotations in English to be a valid
concept (Fry 1986:141). I will show that “homosexuality” is hardly
applicable in all contexts of same-sex relations. Many authors legitimise
the use of the term by claiming that it refers only to physical sexual acts
between two men (or women), but I question if we, as readers, can get rid
of all the connotations the term itself brings, since for us it is loaded with
meanings learned within this cultural context. And when two men are
engaged in sexual intercourse, does that make them “homosexual”? How
to define a homosexual? It can be assumed that same-sex relations are
found all over the world, but to define them they need to be studied in a
local context and defined accordingly. It can be also questionned, wether
sexuality ultimately has such significance for the position of yan daudu.
The second purpose is to show that bori is not a cult of deviants,
but an important part of everyday life for its members. The bori cult is
often interpreted as a form of resistance and of women’s empowerment in
the male-dominant, Islamic Hausa society (for example Masquelier
1993:3, Last 1993:50). It is seen as a possibility for such “deviants” as
“prostitutes” and “homosexuals” to live in a more “understanding”
community and receive appreciation for being what they are (Wall 1988:
167). This view was first presented in the work of I.M. Lewis (1971) who
has divided possession cults into two groups; main morality cults and
peripheral cults. Peripheral cults differ from main morality cults by their
capacity to act upon the common moral rules of society. Because of the
assumed "marginality" of the bori cult, Lewis places it to the group of
peripheral cults. This second purpose is to show that the bori is not only an
empowerment for "marginal people", but it is also an integral part of life
for several people from different backgrounds, not just for "deviants". It is
important to point out that bori fulfills several functions at the same time.
It is a cure and medicine (Wall 1988, Besmer 1983, Echard 1991), cultural
performance (Pittin 1996), tradition (Masquelier 1999) and resistance
movement ( Echard 1991, Masquelier 2001). Above all, it is a way of life.

5
 And finally and most importantly, the third purpose is to study
what is the role of yan daudu and what meaning the gender crossing has. I
will point out that it is not a sign of "homosexuality", which simplifies the
understanding of yan daudu. Yan daudu are mentioned mostly in the
context of bori spirit possession cult, a social and religious institution that
exists in the Hausaland situated in the northern part of Nigeria and the
southern part of Niger. This will be the ethnocraphic context of this paper.
There is no clear "membership" in the cult but the strength of the bori lies
in the collective imagination, in shared beliefs (Masquelier 2001: 9). The
everyday life is lived with the spirits, the other world, which is always
present in the people's life. It is the ritual aspect of bori that has interested
several anthropologists (for example Besmer 1983 Wall 1988,
Onwuejeogwu 1969). Eye catching performances with the musicians,
dancers and singers, ending to the possession itself (which has interpreted
in the name of insanity in the early studies), have directed the interest of
the researchers to the "sacred domain" of the bori (Masquelier 2001: 121).
What makes the situation of bori interesting is its existence next to
the powerful Islam in the Hausaland. Since the islamisation in the 14th
century, the position of bori has been renegotiated several times. Being
considered as a pagan religion it has managed to live side by side with
Islam and has nowadays experienced revivalism as traditional medicine
and folklore. This creates cultural environment for yan daudu where, on
the one hand, there are strict Islamic rules and concepts of gender and, on
the other hand, assumed freedom in the bori cult.
My methodological reflections about the role of yan daudu will
deal with the gender and the position of yan daudu in the Hausa culture.
The methodological background of this work lies in poststructuralism.
After the 1970s, poststructuralists have challenged "…the ideas of fixed
meaning, unified subjectivity and centered theories of power" (Weedon
1999: 100). There are several authors who have used poststructuralism in
order to understand subjectivity and gender. The main poststructuralist
author presented in this work is Henrietta Moore and especially her work
"A Passion for Difference: Essays in Anthropology and Gender" (1994).

6
 In this book she calls for a re-examination of the relationship
between gender and the body in relation to categories of sexual difference.
Even tough the poststructuralists have theorised the body mostly in the
context of feminism and women, it helps to understand the constructions
of sexuality and gender in general. She questions particularly fixed
meanings of bodies and the assumed naturalness of gender categories.
Moore also calls attention to the cultural specifity of the categories sex,
gender and sexuality.
The poststructuralists consider that gender is produced by
discourses as well as individual subjectivity. We learn to be a man or a
woman (or something else) through the discursive practices that guide us.
The subject is a site of different subjectivities constructed by different,
often competing, discourses (Weedon 1999: 104). Still, as will be shown,
this does not mean that individual is totally passive in this process. In the
end of the paper I will call attention to this problem by presenting a term
of Wendy Holloway (1984). She uses the term investment to describe the
choices the individual makes to form his/her gender identity.

Outline of thesis

In the first chapter I will give a general view over the bori cult. The
chapter will deal with different theories on the origins of the cult and show
what kind of meaning the cult has for its members. I will also reflect on the
relations between maguzawa (shortly translated as non-Muslims) and
Muslims, as well as between Islam and bori.
The second chapter will consider the questions of marginality of
spirit possession cults. I will present two different views (the cult in
marginal position versus the cult as integral part of society) by I.M. Lewis
(1971) and M. Lambek (1981). Through their two very different theories I
will study the cultural position of bori cult. The main question of the
chapter will be what kind of meaning bori cult has in Hausa society?

7
 In the third chapter I will present yan daudu whose cultural
environment has been studied in previous chapters. The focus will be on
their productive specialisation and gender variation which has been the
most eye catching element of yan daudu. In this chapter I will also
consider their role in the Hausa society and in bori cult.
The fourth chapter presents some early studies on sexuality. Some
important authors such as Freud and Malinowski will be presented and
their meaning for the later studies reflected. I will also write few words
about the early studies on "homosexuality".
The fifth chapter deals with the concepts of sexuality and
homosexuality. I will present two views that have dominated the
anthropological studies of sexuality, essentialism and constructionism, and
show their impact on anthropological understanding of sexuality. In this
chapter I will also consider the validity of the term homosexuality and to
study whether or not it is universally applicable concept.
In the sixth chapter I will take a closer look on sexual dimorphism
that has been a base for western understanding of sexuality since Darwin's
theory of natural selection. I will study how sexual dimorphism has shown
in different theories (on sex and gender) and finally present some central
poststructuralistic ideas of sex and gender. By poststructuralistic theories I
bring into discussion individual's meaning in creating one's gender. Thus
from gender as cultural model (based on sexual dimorphism) we move
toward theories where gender is (per)formed by individual him/herself.
The seventh chapter is about two different groups of people
(Indian Hijras and North American Berdaches) who are sometimes
categorised as third gender. I will compare these two groups with yan
daudu and analyse if yan daudu can be categorised as third gender.
In the final discussion I will bring forward some methodological
reflections of the role of yan daudu especially in the context of karuwansi.
Their gender variation and productive specialisation will be in focus but I
will also reflect on their motives for taking the role of yan daudu.

8
I will use Holloway's (1984) term investment to discuss why young men
take such position as yan daudu even tough the role is considered as
deviant in Hausa society.

9
 2. Bori spirit possession

Spirit possession is a phenomenon that exists all over the world. The
origins of some contemporary possession cults can be dated as early as to
16th century and some are born during the time of colonialism, it therefore
would be misleading to consider spirit possession cults as a relic of an
ancient period (Lewis 1971:97). Many cults are born as a result of social
change. Conversely cults have continuously been changing with the
surrounding society. The capacity to change and modify to current social
situations has kept many cults alive and strong. With the modernisation,
local as well as global elements have been attached to the cults. The most
clearly this can be seen in the spirit pantheons where one can find spirits
named after rock stars, characters from the bible or presidents (Behrend &
Luig 1991: xiii). As Evans-Pritchard interpreted spirits, they are
“refractions of social realities” (Evans-Pritchard 1956: 106)
In the research of spirit possessions there have been three major
approaches; first, the psychoanalytic approach has understood spirit as a
therapy for people in crisis. Second, the sociological approach
concentrates on the “social exclusion and ritual compensation” (Wendl
1999: 120). Third, the feminist approach has been interested in spirit
possession as women’s special way of communication (ibid.: 120). Wendl
criticises previous studies for having concentrated on the functional-
pragmatic dimension of spirit possession. He underlines that since spirit
possession is such a multidimensional phenomenon, it fulfils several
functions simultaneously (ibid.: 120)
The bori is a Hausa cult of possession trance that exists in the
northern part of the Federal Republic of Nigeria and the southern part of
the Republic of Niger in the area called Hausaland (Besmer 1983: 3-5)
The word ‘bori’ itself refers to the act of possession–trance, to
supernatural spirits or mediumship. The etymology of the word, like the
origins of the cult, is uncertain (Last 1991: 56). The number of
participants and the popularity of the cult are impossible to define with
certainty: it varies, and has varied, greatly depending on the political

10
situation (Pittin 1996: 210). The cult varies also geographically and it is
therefore a difficult task to write generally about bori spirit possession
(Pittin 1996: 200). There can be differences between the bori groups even
within the same region but common features can also be found.
Bori is fundamentally a communication between this world and
the other world. It is a set of beliefs in supernatural spirits that can
communicate with the people in terms of good or evil will. The spirits,
always being present in people's lives, can cause several misfortunes and
make the people ill, but they have also the power to cure and bring luck.
(Besmer 1983: 1). The spirits form a spirit pantheon that includes
numerous spirits1 that has their own social and occupational positions in
the spirit world. Every spirit has also its own name and character. These
are known by the participants and this information forms the basis of cult
practices (ibid.: 63).
The possession itself “…is a cultural evaluation of person’s
condition, and means precisely what it says: an invasion of the individual
by a spirit“ (Lewis 1971). People get possessed mostly to search a cure for
diseases caused by spirits. This has made some anthropologists to name
bori as a cult of affliction (Besmer 1983: 21). Involvement in bori is
associated with misfortunes and suffering, but people who are in contact
with the spirits consider themselves lucky (Masquelier 2001: 94).
Affliction can be a mark that the person has angered the spirits, but as well
it can be interpreted that the "victim" is chosen by the spirits to be their
"horse" (Besmer 1983: 22).
The participants of the cult are called yam bori, the children of the
bori. The trance possession taking place in the bori cult the spirits "mount”
the devotees i.e. possess them. The relationship between the spirits and the
devotees is described with this metaphor of a rider and his horse.
Possessed women are called godiya, mare, and men doki, horse (Besmer
1983: 12). In addition there are several other members of the cult in
various positions. The highest authority is Sarkim Bori who is usually a

1
Masquelier estimates that in the Ader region the spirit pantheon contains more than 300 spirits
(Masquelier 2001: 88)

11
man. His authority and tasks vary, but in general he is the one who
promotes the services of the cult and takes care of the gifts and invitations
(Ibid.: 12). The second person in the high position is Magajiyar Bori who
is always a woman. The position of Magajiyar bori is subordinated to
Sarkim Bori. Usually, the tasks of Magajiyar bori are similar, with the
difference that she has a smaller district to control. Magajiyar bori are
generally associated with the karuwansi ("prostitution") since they run
often gidan karuwai, (a house of "prostitutes") (ibid.: 12)
According to Besmer (1983) women participating in bori are
stereotyped as prostitutes. Karuwansi describes the status of an unmarried
woman in an Islamic society. After divorce (or widowhood) a woman who
refuses to remarry takes a new position as a independent woman. Such
women live in the compounds run by Magajiyar bori (Besmer 1983: 13).
In the compounds live also yan daudu.. They are never possessed by spirits
but participate in the cult by dancing and preparing food. The role of yan
daudu will be studied closer in the future chapters. Finally one more
important group of people can be named: the musicians. They are
responsible for the music in the cult and have an important position in
creating atmosphere during the possession.

Origins of bori

The origin of bori cult is much speculated, but there is no certain

information about it. Opinions are commonly divided into two; there are
those who claim it to be of Islam origin and those who consider it as a
relict of ancient animistic religion that existed already before the arrival of
Islam. The oldest assumptions were made by A. J. N. Tremearne (1913)
who claimed that the bori existed already before the arrival of the Islam.
Jacqueline Monfouga-Nicolas dates the development of the bori cult at the
end of the 19th century arguing that it developed as a result of the intrusion
of the Islam into under-privileged minorities and therefore was born as a
form of resistance (Echard 1991: 65). Echard claims that the origins of the
cult are to be found in the 18th century basing this argument on the

12
information obtained from oral traditions. According to Echard, oral
traditions are the only available sources of information that can help dating
the origins of the bori cult. According to him the material shows that the
bori cult developed during the 18th century in the Ader region out of local
elements but being influenced by the migrants. Other data nonetheless
suggest that the cult started during the 19th century among Hausa hunters.
This argument is based on the list of bori leaders2. It shows that most of
the important cult leaders till 20th century belonged to hunter’s lineages.
An important notion is that the hunters have traditionally had the special
knowledge of the physical and Botanical expertise and the current bori
medical techniques are based on that.
Pittin (1996), in his article on Changing symbolic structures and
meanings in contemporary Nigeria, does not define the origins of the bori
cult but mentions that “It is clear that bori has been closely associated with
the kingdom of katsina for centuries..." (Pittin 1996: 202). He sees that the
roots of the bori are in the old spirit-based religion that has lived in
coexistence with Islam for a long time. Lewis (1971) claims that bori
existed already before the rise of Islam with the extend that it did not
include the possession at that time. Possession emerged when women
discovered the spirits men had abandoned (Lewis 1971: 96).
Besmer (1983) contrasts the practice of the bori between the rural
Hausa (Maguzawa) and the urban Muslim Hausa. He contests strongly
Nicolas’(1967) view that bori would be only a women's movement among
the Muslim Hausa, born by the presence of the Islam3. Nicolas sees that
the arrival of the Islam created a new cult where women were able to enjoy
prestige among the male dominant Islam. The spirits of this new cult were
brought from the “old” clan spirits and the “new” Muslim deities (Besmer
1983: 13-149). Besmer claims that evidences testify against this view: bori
cult can also be found among the rural Maguzawa even in those areas
where the influence of the Islam is minor. In fact, Besmer states, the

2
Echard does not name the exact time when the list was collected, but according to my understanding this
was done during Echards fieldwork 1961-1981 in the ader region of Niger.
3
This view is also supported by Masquelier (1996) who states that “...there is evidence that the bori cult
has evolved in conjugation with islam” (Masquelier 1996: 223)

13
practice of the bori is “an integral part of Maguzawa religious culture”
(ibid.: 15). Besmer seems to support the view of bori as a traditional
Maguzawa practice. He writes that “Bori practice is apparently untouched
by Muslim influence even if Allah is placed above an extend pantheon
which includes a few new personalities” (ibid.: 15). Wall considers the
similarity of bori spirit pantheons of the Maguzawa Hausa and Muslim
Hausa to be a clear evidence that bori is a pre-islamic religion. The current
bori is a “...thinly veiled and partially modified paganism existing in an
otherwise Muslim society” (Wall 1988: 166)

Maguzawa and Islam

As written earlier, bori is a phenomenon that varies greatly in time and

space. In the history the position of the bori has changed several times
from the complete denial of bori to “revivalism”: Historical changes of the
position of bori have often been attached to Islam. Since the islamisation
of the Hausa land these two religions have existed side by side (Pittin
1996: 202).
Bori is a cult for both Muslims and non-Muslims, Maguzawa, even
tough it is often associated with the latter group (Last 1991: 56). The
Hausa society is a composition of different groups, but Maguzava are
unique since they are identified by their religion. In Nigeria the term
Maguzawa is often used for all non-Muslims and sometimes it refers to the
“original Hausa” culture. Their current culture is seen as a relict of the
ancient religion before the arrival of the Islam (Last 1993: 269, Besmer
1983:15). As Murray Last (1993) has pointed out in his article History as
religion: De-constructing the magians ‘Maguzawa’ of Nigerian
Hausaland, the label of Maguzawa is a consequence of specific historical
relations of non-muslims who started to trade with the muslims. The word
‘Maguzawa’ comes from the arabic majus which means magician. This
name was given to a group of non-Muslim merchants and mercenaries as a
sign of “protected people”. Muslims were not allowed to deliver products

14
from pagan areas thus their plan was to create a Muslim state in West
Africa and spread their religion and culture. This way they were able to
trade with the protected people and deliver the products all the way to
other Muslim states (Last 1993: 282-283):
In the Hausa area, the Muslim influence and power has been
incontestable. Islam has provided political and social changes affecting
also the identities of the Hausa. (O’Brien 1999: 3-4). Muslims presenting
the visible "power", describes the defencelessness the life of maguzawa.
This shows in two ways; First, Maguzawa have never demanded an
independent state for themselves, but have been willing to migrate in the
occasion of conflicts (ibid: 288). Second, the other power, superior to
Maguzawa, is the other world – the world of the spirits (Last 1993: 289).
The spirits, called iskoki in Hausa, comes from the other world being
invisible but capable to communicate with the people (Wall 1988: 133).
The spirits have both good and bad intentions and thus sacrifices are made
to the spirits to keep them content and favorable. In the case of severe
disease, an individual may have become possessed in the bori cult in order
to reach the contact with the spirit causing the disease. This way the
possessed have the possibility to get cured by the spirits. Often all the
spirits have their special diseases they represent (ibid: 134). Spirits also
have medical knowledge that they can reveal to possessed when they get
familiar to each others (ibid: 134). This describes well the relation between
spirits and humans. Between the living and with the other world there is a
two way relationship; it is possible to live in a constant discussion with the
spirits (Last 1993: 289).
Muslims do not accept the notion of the other world but emphasise
the meaning of the next world (Last 1993: 288). They consider the
worship of the spirits as paganism which prevents access of Maguzawa to
the next world (Wall 1988: 134). Being more concerned about the other
world than their salvation, Maguzawa are accused of being ignorant about
the horrors they will face in the next world. Their religion is seen as a
worship of the evil forces that disobey gods will and thus having
somewhat “black magic powers” (Last 1993: 288). Both Maguzawa and

15
Muslims recognise Allah as supreme God but believes that Allah do not
interfere directly in humans everyday life. Muslims see that Allah has sent
“angels” and “devils” to deal with these things being still under His will.
For Maguzawa these “angels” and “devils” are iskoki or bori spirits and
they can act also independently (Abdalla 1991: 42-43).
Not only the religious differences separate Muslims and
Maguzawa, but also their ways to live their everyday life. While Muslims
live in the walled villages or cities, Maguzawa live in the rural areas
having agriculture as their main source of income. Visitors, as well as
spirits, are welcomed and they are offered a location when Muslims have
more strict rules when it comes to intercourse between people. Also the
clothing differs. Muslims having the obligation to cover completely their
bodies (especially women) have maguzawa no such regulations. Women
are allowed to go out even about bare-breasted. (Last 1993: 288).

Islam and bori

A central question with respect to the relation with Islam is how the bori
cult has survived the pressure of Islam and what is its place in the Islamic
culture. Nicole Echard (1991) sees that bori does not differ from other
possession cults when it comes to their association with the dominant
religious form of the surrounding society (Echard 1991: 67). These two
"groups" have developed as overlapping constituting new sets of practices
and beliefs, rather than two separate paradigms. The bori has been
modified to meet the conditions of Islam in several ways which has
proceed the maintenance of the bori in the Islamic regions (in spite of the
periodical suppression). The relation of the bori and Islam is well seen in
the development of the bori spirit pantheon, especially in its hierarchical
structure (Abdalla 1991: 41). In the arrival of Islam new Muslim spirits
were created and attached to the spirit pantheon. Allah and these new
powerful Muslim spirits were positioned to dominate the spirit world
which reflected the situation of the “real world” (the Muslims taking the
control over political and economic activities) (ibid.: 41).

16
 In the possession-trance the Muslim spirits appear in different
ways than other spirits do: they behave in the way “Muslims do”. These
spirits dress the “Muslim way”, eat the “Muslim food” and do not perform
acts denied for Muslims. Characteristic to these possession rites is also that
they are not presented during the important ritual periods of the Muslims.
This indicates how the spirit pantheon and the ritual performances have
been adjusted to the new political and religious realities of Hausaland
(Abdalla 1991: 42).
The interdependence of Islam and bori is also shown in the travel
narratives of the bori adepts (O’brien 1999: 4) During the trips headed to
Saudi Arabia, bori adepts trade their medical herbals and bori-knowledge.
According to O'Brien these adepts mediate between two cultures gaining
"...symbolic, economic, and social capital" through their illicit trade in
Saudi Arabia. This way these trips work in an empowering way to crease
the appreciation of the bori adepts back in Hausaland. The fact that bori
has relations both to larger Arab world and Hausa culture shows that it
cannot be viewed simply a minor pagan religion. More, bori should be
viewed as a dynamic subculture (O'Brien 1999: 4).
For Muslims, bori represents paganism against Muslim
conventions (Masquelier 2001: 223). It has also been a "statement of
protest and defiance in the face of Islamic hegemony..." (ibid.: 230). The
living side by side has not always been a harmonic co-existence (ibid.:
230). Still, some Muslims take part to bori ceremonies or seek medical
help from bori practitioners. Masquelier notes, that the association of the
local "indigenous elements" with Muslim epistemology has greatly
affected Islam's local practice. There is not only one form of Islam but
many which varies along the context; it should not be reduced to an
essentialist core (ibid.: 226). She concludes: "Neither bori nor Islam
should be perceived as distinct and homogenous entities; they are both part
of a larger set of intersecting, often dissonant, and rarely totally consistent
discourses..." (Masquelier 2001: 47) The devotion to the religion changes,
depends on the individual. Some are unconditional about their faith, some

17
are less, even if they have Islamic status (ibid.: 226). This means that also
Muslims can take part to bori without losing His/Her Muslim identity.

18
 3. Questions of Marginality

The claim that possession is a way for the powerless to express their
problems or misfortunes has long dominated the studies on spirit
possession (Masquelier 2001: 15, Sharp 1991:4). I.M. Lewis (1971)
argued that the most powerless people of the society resist the power by
participating in spirit possession ceremonies. The fact that most of the
participants are women (also applicable to bori) has led to the definition
that the spirit possession is a counterculture of women and social deviants.
In this way the question of gender has had a great impact on
anthropological discussions of spirit possession (Behrend & Luig 1999:
xvii).
Given the fact that people participating in these cults tend to be
members of lower classes or in a socially disapproved position, the cults
have been located outside the mainstream culture (Lewis 1971: 32). Lewis
divides possession cults into two groups; main morality cults and
peripheral cults. This division is based on the evaluation of whether the
considered cult presents (through the action) the general morality rules of
the society or not. According to this definition, the bori would belong to
the latter group, i.e. the peripheral cults. To Lewis there are several
distinguishing marks to define cult as peripheral: First, these cults “play no
direct part in upholding the moral code of the societies”(ibid.: 32).
Second, the evil spirits are usually considered to be from the outside of the
society like from the neighbouring peoples. Third, the targets of these
peripheral spirits are also peripheral, i.e. woman. Fourth, these cults also
“embrace downtrodden categories of men who are subject to strong
discrimination in rigidly stratified societies”(ibid: 32). Many researchers
have categorised assumed “homosexual” men to this group of deviants4.
As the word peripheral indicates, Lewis places the cults “outside”
mainstream culture. He claims that possession cults are women’s
counterculture. Possession cults are a way for the weak and downtrodden

4
According to Besmer (1983) the Hausa bori cult is a cult of deviats since prostitutes,
homosexuals and people from the lower class participate in the cult.

19
(i.e. women) to demand prestige and attention in otherwise masculine main
culture. Likewise according to Abdalla (1991) the Bori-cult gained a large
popularity among women after Islamisation. In the 15th century
islamisation took place in the Hausaland that caused a radical change in
the social status of women. They were excluded from the political and
economic arena. As a way to cope with this situation women turned to bori
possession, a socially sanctioned therapeutic practice (Ibid: 41). The bori
became a “counterculture”.
According to Lewis the spirit possession is, above all, about
ongoing sex-war. It is a way to resist the male supremacy and to gain
prestige and even material gains5. Lewis reflects why men then tolerate
possession in the society and concludes that is because the demands come
from the spirits and not from the women. This way it is possible to agree to
the requests without losing male supremacy. Also he sees that men do not
understand the total meaning the possession has for women: Men consider
it as an illness and cure whereas for women it is a religious drama.

“Consequently, we have a feminist sub-culture, with an ecstatic

religion restricted to women and protected from male attack
through its representation as a therapy for illness” (Lewis 1971:
89)

Lewis compares western Women’s Liberation Workshop to possession

cults in order to see them both as a result of lack of equality. By this Lewis
wants to make a statement of global subordination of women and to show
that women find different ways to resist it. Though, according to Lewis, it
is easier in “our contemporary conditions” to have possibilities for
“directly militant feminist action” (ibid: 87)

5
Lewis gives an example among the pastoral Somali where possession is interpreted as illness
caused by evil sprites (ibid: 75-76). The evil spirits “attack” mostly married “hard-pressed” wives
suffering from the absence of the husband or envy toward other wives of him. Spirits demand trough
wives fine clothing, jewellery etc. that not everyone can afford. Lewis writes: “It will now be clear, I
think, that we are dealing with a widespread strategy employed by women to achieve ends which they
cannot readily obtain more directly. Women are, in effect, making a special virtue of adversity and
affliction, and, often quite literally, capitalizing on their distress” (Lewis 1971: 85)

20
 Lambek (1981) criticises Lewis’ arguments for the marginality of
cults and participants. First of all he claims that spirit possession is not
some counterculture but an integral part of the whole culture. He also
recognises the fact that men are mostly excluded from the cults, but does
not consider it as a sign of women’s marginality (Lambek 1981: 60). To
Lambek, possession is a part of a coherent symbolic system what can be
“readable” as a text (ibid.: 7-8). This is Lambek's theoretical base for his
interpretation of the spirit possession. To him the text is a structure,
separate from behaviour. People can use the structure (i.e. possession) for
their own advantage, as Lewis argued, but it does not define the behaviour.
People can have several models of behaviour, not just one. This idea is
contrary to that of Lewis’ since this denies that women participate to the
possession because of their assumed marginality (behaviour). Why, then,
women are majority in the possession cults? Why does not the both sexes
participate in the same number since the possession has such positive
effects on possessed (like Lewis argued) (ibid: 60)?
Lambek concludes that the symbolic structure of the possession
favours women. This is for two reasons: First, most of the characters
played in the possession are men. When women act these male roles they
are “performed to best effect” (ibid: 61). The women can present the most
explicit aspects of “male performance” of everyday life (i.e. different
notions of power and responsibility in Islamic Mayotte culture). This way
“the process of abstraction and reflection” of behaviour can be carried
further (ibid:61). Were these actors male the result would be more implicit.
Second, the possession requires suitable actors. Lambek states that women
are more acceptable to the role of possessed because possession behaviour
is not appropriate for men. In Mayotte the Islamic law strictly controls the
behaviour of men and their public social role. The “loosing control” in
spirit possession is a controversial act to men’s social role (Lambek 1993:
61).

21
 Marginality of the bori cult

In the discussion above there are two contradictory visions about the
marginality of the possession cults. Lewis views possession as women
empowerment having marginal positions in society. What Lambek has to
offer is a more "positive way" to view the spirit possession. He does not
see it as a result of peoples’ deprivation but an important structure existing
in the society. He does not place spirit possession to a marginal position
but highlights that it is one structure among the others and should be
considered in a larger context. How should the bori cult be viewed? Is the
position of the cult marginal? If so, in what terms it can be viewed as
marginal? This discussion will be started by notion that have been
discussed above. The bori cult have developed aside of Islam that has had
the religious hegemony in the Hausa society. In the terms of Lewis the
bori cult is in the marginal position since it has been considered as
immoral in relation to Islam in Hausa society.
Like in most of the possession cults women's membership in the
bori is considerable. When it comes to the marginality of these women,
many of them are socially unstable, having married and divorced several
husbands. Women also break Islamic norms (such as modesty and
respectful submission to husband) by abandoning their households for
participating the cult. Masquelier shows that among the Mawri, bori is
definitely considered as women's business. Women search help from the
bori healers to ask about pregnancy, babies or to get advises for the
marriage (Masquelier 2001: 92). Still, the situation is not one-sided: both
men and women do have access to membership in the bori. Gender
equality is promoted in the "institutional arena" by three aspects: first, both
man and women can have economic benefits and social status as mediums.
Second, both men and women are expected to follow the same rules of
bori and participate in the performances (by dancing). Third, it is
important for all to "display their capacities as mediums in a public
setting" (ibid:. 95).

22
 How about Lewis' claim that a marginal cult “play(s) no direct part
in upholding the moral code of the societies”(Lewis 1971: 32)? Masquelier
follows the ideas of Lambek. Like him, Masquelier emphasises that spirit
possession is not only a ritual outside the main moral codes but a
constitutive reality of the culture (Masquelier 2001: 123). As she writes:

"…the strength of bori –and the source of its resilience despite

censure and competition –resides precisely in its deep
embeddedness in mundane practices that provide sets of culturally
constituted images and techniques through which mediums and
others are able to share, transform, or contest changing social
realities." (Masquelier 2001: 124)

Bori is a part of everyday life. It is set of practices that do not occur only in
a ritual context but exist constituting the worldview. It should not be put to
a marginal position as a special practice occurring only in the situation of
affliction. The extraordinary that shows in the understanding of the world
of the spirits, is one side (essential side though) of the bori (ibid.: 124).
Was it not to this side the Muslims would not have had any confrontation
with bori. The Muslim elite considers bori as dangerous and hateful
precisely because of these "harmful" pagan spirits.
There have been periods when bori has faced a great suppression,
but in the current situation suppression of the bori is diminished because of
the inner arguments of Muslims (Masquelier 2001: 223). Confrontations of
bori values and Muslim conventions are losing their relevance. In the
recent debate the question has been focused on the nature of Islamic
knowledge and the right way to practice Islam. Two opposite groups in
this debate are the yan izala, a fundamentalist group that aims to purify
Islam, and the mainstream Muslims ('yan tariqa). Bori cult has been left
mostly aside, but used occasionally to lift Muslim profiles in contrast to
the provincial cult adepts. (ibid.: 223).
On the other hand bori has gained a new meaning as "tradition".
Among the Hausa-speaking Mawri villagers the bori has become a help to
preserve a local heritage and to maintain the relations with the spirits in the

23
Islamised environment. For example, women use bori to affirm their
fertility. This is not only for illiterate peasants but also for educated
Muslim elite. Where on the past it was not socially acceptable to show
such backwardness, nowadays it is safe to call attention to one's cultural
roots. The bori presents authentic relict of traditions and is used to confirm
cultural identities. This has also another side: the modern education and
occupations have confirmed the re-evaluation of bori. Villagers with
modern education or occupation get extra prestige by contrasting their
occupational position with the old traditions (Masquelier 1996: 35).
Bori has also gained new recognition as traditional medicine. Bori
rites have been re-evaluated with the increasing respect toward this
medical practice. Even some local authority structures have associated
with the bori functionaries. There are also "western" medical services
available but it does not exclude the use of bori as a medical treatment
(Pittin 1996: 211). According to O'Brien (1999), bori healers in Kano have
such incomes that they have had the possibility to do several pilgrimages
to Mecca. This, according to her, indicates the "continuing popularity and
remunerative importance of the profession". The increasing popularity of
the profession has led some older bori adepts to criticise the younger that
they are in the business only for the money (ibid.: 12). Indeed, the bori has
become a profitable business for some ambitious young men (Masquelier
1996: 36 6)
This chapter has treated the marginality of bori cult. It has shown
that some assumptions about marginality should be corrected. The cult is
not in marginal (especially because of the participants) but an integral part
of some peoples lives. What does this tell about the situation of the
participants? Instead of some assumptions (for example Besmer 1983,
Lewis 1971) the cult has participants from all societal groups. Not all of
the participants are "deviants" or other "low class people" (Masquelier
2001: 93). Besmer have claimed that because of prostitutes, homosexuals

6
From Masquelier (1993): Ritual economies, historical mediations: the poetics and power of Bori
among the Mawri of Niger. PhD dissertation, University of Chicago.

24
and other marginal people, "…bori takes on the appearance of a cult of
deviance" (Besmer 1983: 19).
Since the bori is not only an empowering practice of a small group
of marginal people, it is important to look closer these "deviants" specially
yan daudu who are the main focus of this paper. In the next chapter their
"role" and deviancy will be taken under closer consideration.

25
 4.Yan daudu

Yan daudu (sing. Dan Daudu) are a strictly distinctive social category of
males, who have adapted feminine mannerism, speech and dress in Hausa
society. They have been mentioned mostly in footnotes and margins in
texts concerning the bori cult and nearly always it is the status of deviant
that has been addressed. Yan daudu have access to religious life but they
do not participate in rituals as possessed persons. In bori performances yan
daudu dance womenlike and donate money to cult-adepts, specially when
the spirit Dan Galadima appears (Besmer 1983: 18). The term dan daudu
comes from Dan Galadima, a son of Galadima in the spirit pantheon. Dan
Galadima is considered as a loose living, handsome man who is popular
among women (Greenberg 1946: 42). It is not known, however, why yan
daudu are associated with him (Besmer 1983: 18).
Mostly yan daudu have been categorised as homosexuals,
transsexuals or transvestites without considering their specific role any
further. Some of the yan daudu do have same-sex sexual relations with
other men but, as I will point out later, partaking in same-sex actions is
neither necessary nor a sufficient criteria for the status of Dan Daudu.
Moreover, their status is to be defined through their work and the most
visible feature, gender crossing. These two key features of Yan daudu's
role could be named as productive specialisation and gender variation
(Roscoe 1993: 332).
The productive specialisation indicates that more than simply
viewing Yan daudu as homosexuals, they should be considered as a socio-
occupational group (Kleis & Abdullahi 1983). They work and live in the
areas where prostitution exists, mediating between the clients and the
prostitutes, managing relations with authorities and recruiting new
prostitutes (ibid.: 45). In the context of karuwanci, yan daudu participate
also to work that are considered to be women's domain in the Islamic
sphere. They, for example, prepare and sell food, mainly luxury snacs such
as fried potatoes, eggs and chicken (Wall 1988: 153)

26
In the Hausaland karuwanci is an old pre-Islamic institution that has
increased as the consequence of urbanisation, greater geographical
mobility and the rise in marriage expenses (followed by prolonged
bachelorhood) (Kleis & Abdullahi 1983: 51). Karuwais live and work in
gidan karuwai, the community of karuwai which are usually placed
outside the birni the traditional set of compounds. The space outside birni
is called waje that means stranger quarters. Waje is inhabited by migrants
from the birni and non-Muslims form the central and southern parts of the
Nigeria. In waje the social pressure of Islam is minor and thus the
atmosphere is more tolerable - also toward karuwanci (ibid.: 43). Many of
the karuwais are divorced women who have refused to move back to their
childhood home. In Islamic Hausa society, Karuwanci provides a
temporary income, often more substantial than that earned in other
professions typical for women (for example preparing food or potteries). It
is mostly temporary, since several karuwais remarry and leave the
karuwanci (Dunbar 1991: 76).
Even tough karuwanci is common it is not a profession of much
appreciation. It is disapproved by the Islamic morality norms and it
violates several norms of Hausa womanhood (ibid.: 41). For example,
karuwais live outside the seclusion which is not normal for Muslim Hausa
women. Karuwansi offers possibility for women to move into spheres not
allowed for married women, and have broader social contacts than
otherwise. For example, unlike married women, they participate in public
dances or to drink beer with men. Thus their social sphere is vaster than
that of other women in Hausa society (Kleis & Abdulahi 1983: 42)
Another feature of the role of yan daudu is the gender variation.
Their highly visible way of dressing and behaving "women like" has led a
lot of reseachers to define them as homosexuals or transvestites. Gender
crossing in the case of yan daudu means that "biological men" have
adapted mannerism of woman; they wear wrappers (Hausa female
garment), speak in a "high-pitched falsetto voice" and move "swaying".
(Kleis & Abdullahi 1984: 44). Yan daudu also dance publicly with
karuwais. Public dancing is addressed commonly for women. Ames writes

27
about dancing men: "Some bori dancers are men, although female adepts
far outnumber them, but they too are viewed as social deviants" (Ames
1982: 128).
Yan Daudu consider themselves as women, rejecting totally their
male identity according to Kleis & Abdullahi (1984). Yet, according to
Gaudio, yan daudu sees their feminine behaviour, dancing and joking as
play, wasa, or crazyness, iskanci. The womanlike behaviour is not serious
and thus they keep their male identity (Gaudio 1998: 119). What they all
agree on is that commonly yan daudu are considered as men in Hausa
society (Gaudio 1998: 119, Kleis & Abdullahi 1984: 49).
Being a dan daudu does not exclude the possibility of getting
married and having family. In the Hausa society marriage and the first
child complete the social status of a man (Wall 1988: 57). They also spend
time in women's company. Normally it is not allowed for a man to be
alone with women, apart from his wife. Yan daudu are freely with
karuwais without limitations. Their relation is not sexual but includes very
intimate conversations and erotic lore. Young yan daudu are socialised by
karuwais and later vice versa (Kleis & Abdullahi 1984: 46).
Given the strict separation of the spheres of men and women (in
Islamic society) researchers have come to interpret the existence of such
category as yan daudu in the context of the bori cult. As written earlier, the
cult has been seen to provide a more tolerable environment for such
"deviants" as "homosexuals". Since yan daudu deviate from the traditional
role of the man (i.e. dressing, behaviour and productivity) they are
considered as deviants among Hausa. They face periodically
disapprovement from religious leaders (other than bori) and authorities
that consider yan daudu as immoral. This encourages "...the abusive
treatment, including arrest, extortion, and physical violence, that 'yan
daudu often face at the hands of police and young hooligans" (Gaudio
1998: 122). On the other hand attitudes toward yan daudu can be good
natured joking. Their visible feminist mannerism and clothing can become
a subject of humored jesting. (Besmer 1983: 19).

28
The statement made about the deviancy of yan daudu requires some
clarifying. As criticised earlier, being a yan daudu, is not a simply question
of sexuality. Nor is their deviancy a reason for their participation to bori
cult. O'Brien (1999) underlines that the reason is the possibility to dance to
popular garaya music. Deviancy has been considered an unifying indicator
between the bori adepts, karuwai and yan daudu. They all share the
ambivalent relationship to a Hausa political and economic order but it is
not enough to consider them as united group sharing social marginality
(i.e. a group of deviants). Solidarity is not self-evident: Well-educated
karuwai would never participate in the bori cult organised by people from
the lowerclass. Moreover, relationships between karuwai and yan daudu
are due to "overlapping social worlds and bonds of friendship" (O'Brien
1999: 10).
Several authors have categorised yan daudu as homosexuals. One
of these texts that will be studied closer is Rudolph Gaudios (1998) "Male
Lesbians and Other Queer Notions in Hausa". What makes his article
special is that it concerns mainly the sexuality of yan daudu. His article
contains valuable information about the lives of yan daudu but, as I will
point out, he has done the research from perspective that transpires the
whole studies. Gaudio underlines that yan daudu are occupational category
but concentrates to study them as "gays" or "homosexuals". With these
terms he refers to men "who are conscious of themselves as men who have
sex with men, and who considers themselves to be socially […] distinct
from men who do not have this kind of sex (Gaudio 1998: 117).
Gaudio starts his article by telling about his own sexual orientation
and how he assumed that being gay would help him to communicate and
get contacts with the "local gays", i.e. yan daudu. He had also heard that
Hausa society has a reputation for homosexual activity. Later he notices
that such a view was naïve since the social world in Nigeria is different
from the American one (Gaudio 1998: 118). During his research he
becomes very aware of the cultural differences but he does not question
the "homosexuality" of yan daudu at any point. To the contrary he
criticises that researchers have refused to see the homosexuality of yan

29
daudu (ibid.: 119). According to him most of the yan daudu are
homosexuals but they want to hide it from the society since
heterosexuality is the norm.
Gaudio characterises the assumed "homosexuality" of yan daudu
as "transgenderal" and "gender-defined" based on the "heterosexual"
features of their "homosexual" relationships. Yan daudu have often male
partners to whom they refer as miji, husband. This relationship follows the
heterosexual norms in grammatical and semantic structure (names), sexual
intercourses (the masculine one is the insertive partner) and in the way
partners treat each other. For example the "male partner" is expected to
give yan daudu presents such as money and clothing (ibid.: 120-121).
In particular, Gaudio accuses Kleis and Abdullahi (1983) for not
dealing with the sexuality of yan daudu. They are one of the few who were
studied particularly yan daudu. Kleis and Abdullahi's argument is that the
terms homosexual or transvestite are not applicable in Hausa society but
do not explain their claim any further.
My attempt in the future chapters is to deconstruct the terms
homosexual and homosexuality in order to scrutinise the text of Rudolf
Gaudio and others who have used the term homosexual to describe the
sexuality and the role of yan daudu.

30
 5. Early ethnographical studies on sexuality

Studies on sexuality are flourishing at present, but that has not been always
the case. In her book The geography of perversion (1996), Rudy E. Bleys
writes that ethnographic narratives have only recently become sources for
historical research on non-Western sexuality. According to her,
naturalising sexuality has prevented intellectual historians from seeing
diversity in sexuality – the historical dimension of sexual categories and
culture. The discourse mostly concentrated on the moral aspects of
sexuality (Bleys 1996: 3). Beth Maina Ahlberg (1994) suggests that
anthropologists’ lack of interest on sexuality studies is due to the historical
context of anthropology. In the Victorian era, when anthropology was still
a young discipline, sexuality was a taboo subject, surrounded by silence.
Additionally the fact that anthropology was mainly used to research on
African social systems to colonial rule, reduced the interest in sexuality
(Ahlberg 1994:224-225).
Freud occupies an important position in the work of many authors.
It is often said that Freud started the research on sexuality, but the French
historian and philosopher Michel Foucault claims otherwise (Sarup 1993:
71). According to Foucault, the problem of sexuality was widely
considered in medicine and psychiatry of the nineteenth century. John
Beattie sees that Freud’s contribution to other disciplines (than psychology
and psychiatry) has been remarkable but mostly indirect, like his ”visit” to
anthropology. His efforts to recover the origins of culture in his work
”Totem and taboo” (1918) did not convince anthropologists (Beattie 1964:
28), but he showed the primacy of symbolic elements in human thought.
Freud refocused the discussion from natural (biological) sexuality to
sexuality’s psychic dimensions (Bleys 1996: 5). Foucault asserts that the
role of psychoanalysis was liberating. “Psychoanalysis was established in
opposition to a certain kind of psychiatry, the psychiatry of degeneracy,
eugenics and heredity” (Sarup 1993: 71).
Psychoanalysis had a great impact on the anthropological studies
on sexuality. One of the first anthropologists, interested widely in sexuality

31
was Bronislaw Malinowski who became interested in psychoanalysis after
it was recommended to him by Dr. C. G. Seligman (Malinowski 1929:
325). Malinowski used psychoanalysis for his studies of sexuality in the
Trobriand Islands (Ombolo 1990: 18). In ”Sex and repression in Savage
society” (1927) Malinowski tested whether psychoanalysis could be
adapted to study on so called primitive societies. To Malinowski,
psychoanalysis was mainly ”a theory of the influence of the family life on
the human mind” (Malinowski 1927: 2). He studied the Trobriand
islanders’ family closely to show that the general attitude towards
sexuality and the type of relations between parents and children have
effected on the different types of complexes of family. ”Complex of
family” is a term that Malinowski used to describing a family’s structured
feelings and typical attitudes toward kinship and sexuality. His interest lay
in researching the kind of effects the complex of family could have on
some general habits, social organisation, myths and legends etc. Because
Malinowski showed that families are constituted and variate in time and
space, he also concluded that the complex of family vary.
Malinowski was also interested in the œdipus complex, which
Freudian analysts believed to be universal.7 He demonstrated that the
complex of œdipus was not universal after all – it was only associated with
the Western, patriarchal family model (Ombolo 1990: 25, Malinowski
1927). This he concluded by analysing the complex of family in Trobriand
Islands. Malinowski brought up the distinctive features from the
patriarchal family where the œdipus complex could be applied. One of the
distinctive features was the fact that in matrilineal systems, children were
sent to live with the mother’s brother, who was considered as children’s
provider. The biological father’s role was to be a friend, unlike in Western
society, where it was supposed to be characterised by power and authority.
Because of the taboo concerning the relationship between sister and
brother, the uncle had to keep distance between him and his sister, but he
was the one with authoritative power. Malinowski therefore denied that

7
The complex of œdipus is a freudian concept with two components: A love for the mother and a hatred
for the father

32
psychoanalytical generalisations could be adapted to every society. Still,
he did not totally abandon psychoanalytical methods. As he wrote: ”…And
my results showed beyond all doubt how even a theory which has, in light
of investigation, to be partly rejected can stimulate and inspire”
(Malinowski 1929: 325).
In ethnocraphical texts and travel narrations in the 19th and 20th
century, same sex relations were often mentioned only briefly and
evaluated along European standards of morality (Bleys 1996). In
anthropology, they were concerned not to exist in tribal societies. A lot of
different explanations were given to support this view - for example easily
available women (Bleys 1996: 166). From Freudian basis, some
anthropologists considered the “lack of homosexuality” as a sign of
balanced personal development. In psychoanalysis, homosexuality was
seen as a neurosis (ibid: 5). According to Murray and Roscoe (1998)
anthropologists denied the excistence of tribal "homosexuality" even when
they were faced with the clear evidence (Murray & Roscoe 1998: XIII,
Bleys 1996: 166). In the West, the “primitive man” was considered to be
close to nature, and tasks of the “primitives” were mostly considered as
beins reproductive-centric. Thus "homosexuality" was seen as an
impossible or unlike feature in tribal societies. Because the black Africans
were seen as the “most primitive” they were supposed as “the most
heterosexual”.
When "homosexuality" was identified, it was often explained as
being an Arabic import. The linguistic ”proofs” were given to confirm the
view that "homosexuality" did not exist in Africa before the Arabs
introduced it. Terms for homosexuality were traced to Arabics, and
indigenous African languages were announced to be ”free” from the terms
(Bleys 1996: 167-168). Generally speaking it can be said that
ethnographical texts (in the 19th and early 20th century) have nearly always
been either negative or positive regarding the same sex relations, hardly
ever neutral. The majority of ethnographers have presented these relations
as a negative feature in tribal societies (ibid: 167-168).

33
 6. Is there "homosexuality"?

In anthropological studies of sexuality, two major views have been

dominant: essentialism and constructionism (Vance 1998: 160). In social
science, essentialism is a pejorative word that has rarely been defined in
academic discussion but used more and more in a rhetorical sense to
“dismiss positions with which one does not agree” (Petersen 1998: 3). This
has caused the common polarisation of essentialism and constructionism.
Constructivism is easier to define but should not be left totally
unproblemised. In studies on sexuality, social constructionism has:
…challenged the “natural” status of many domains, presenting the
possibility of a truly social inquiry as well as suggesting that human
actions have been and continue to be subject to historical forces and, thus,
to change” (Vance 19988)
In the studies of sexuality, constructivist writings have assumed
“homosexuality’s” inter-cultural nature, emphasising the variation of its
“cultural forms”. This view has been a cornerstone for those cultural
constructivists who have rejected the transhistorical and transcultural
definitions of gender and sexuality and highlighted the variation of their
different expressions. Contrary to constructivism, essentialists emphases
that human behavior is natural and there can be found an essence of human
(often sexuality is understood biological and fixed which will be soon
discussed). Constructionism contests that the claimed “natural” is
production of human action and history instead of biology. Different
cultural variations of sexuality can be found throughout time and space.
Even though constructivism has managed to break through the
essentialist ideas of sexuality as a purely biological drive and show the
variety of different expressions of sexuality in different cultures, it is
strange that it has not “questioned the existence of universal categories like
homosexual and heterosexual” as Vance states (Vance 1998: 163). This
particularly concerns reinterpretations of history made to support the view

8
First published in Homosexuality, Which Homosexuality? A. Van Kooten Nierkerk & T. van Der Meed
(eds). Amsterdam: An Dekker

34
that homosexuality has existed throughout the time. Vance names this
historising sexuality (ibid.: 160). Creating gay and lesbian history was
mostly a project of early constructivism. Nowadays there are several
views on studies of sexuality within the constructivism9; Vance stresses
that constructivists have different opinions how much finally constructed.
How much is due to culture and socialisation? How much is “essential”?
Common agreement is that physically identical sexual acts can be
given different subjective and social meanings since the act itself does not
carry any meaning. Some constructivists go further assuming that the
individual’s desire is socially constructed, and the most radical
constructionists claim that even sexual impulse is constructed by history
and culture and thus there is no biological lust (ibid.: 164). Many assume,
nevertheless, that there exist certain sexual behavior or physical relations
understood as sexual, that vary contextually (ibid.: 165). This supports my
understanding of sexuality: extreme constructionism can hardly answer to
all questions and, further more, I do not find it meaningful to try to prove
that sexuality is throughout constructed. Is it even possible to define
sexuality entirely?
Grosz (1995) lists three forms of essentialism, whereof biologism
is the most recognised form (in Petersen 1998: 5). Biologism considers the
social and cultural factors as effects of biological factors. It has been used
to explain, among other things, the “essence” of men and women and,
what is interesting for my study, “homosexual orientation”. The second
form of essentialism is naturalism. In naturalism certain attributes attached
to women and men are considered natural without any biological
explanation. For instance, women are seen as naturally caring. Grosz
considers Freudian psychoanalysis as such essentialism since it ignores the
influence of the subject’s social position and emphasises the genital
morphology. 10 The third form of essentialism, universalism, is also based

9
I take the liberty to write ”inside the constructionism” even tough it is notable that constructivism does
not form any school, but theory.
10
Freud explained the difference between sexes by psychological developement that was always attached
to the same pattern: when the child notice the bodily difference (i.e. penis) between his/her parents (i.e.
man and women) it makes it possible to create the own sexual identity. Freud, Sigmund (1961):
Introduction à la Psychoanalyse. Payot: La Petite Bibliothéque Payot

35
on biological assumptions, such as commonness of women and men. It
differs, however, from biologism in fact that it is conceived in social
terms. Universalism comprehends the idea of a united group of women (as
well as men) that is ahistorical and cross-cultural (Ibid.: 1998: 5-6). The
same way "homosexuals" have often been understood as one category.
In current research scholars often limit the concept of
"homosexuality" to describe only same-sex intercourse in order to
“undress” it from western connotations of homosexuality. This would
present the common constructionist view on sexuality. Fewer have
questioned the use of the term itself, as above commented. Gilbert Herd
(1995) suggests that the term homosexuality should be put in quotation
marks because there is not such a unitary entity as "homosexuality". More
like it, he states, there exist “several species” of the same-sex relations that
can be classified as four kinds of practice; age-structured, gender-
structured, role- or class-structured and egalitarian-structured
"homosexualities" (Herdt 1995: 75). While this view still assumes
particular forms of sexuality, Peter Fry (1986) argues that homosexuality is
not a valid term for scientific research because of its multiple meanings
(Fry 1986: 141). This notion includes the idea that we can never define the
term in a united way. He certainly has a point there since the term has so
many connotations depending on the context.
Michel Foucault (1978) calls attention to time and context
specifity of "homosexuality". It can be said that the category called
"homosexuality" was born in the 19th century when the "world of
perversion" was discovered (Foucault 1978: 40). Before this, till the end of
the 18th century, sexuality was controlled by three major codes; canonical
law, civil law and the Christian pastoral. They all defined what was
"normal" and approved and what was condamned. The norm was the
sexual relation of a husband and wife, everything else such as adultery,
rape and sodomy were considered as equally condemned. Thus the same-
sex relations were seen as a category of forbidden acts. They were "against
the nature", but more importantly, against the law. When the category
"unnatural" was discovered, the "homosexuality" was taken apart as its

36
own category and "homosexuals" were considered as a sub-race. The
members of this special category were viewed sexually peripheral, deviant
and even sick (ibid.: 40). The sexual orientation of "homosexual" was
always present and it was the cause to all his actions. He became ..." a
personage, a past, a case history, and a childhood, in addition to being a
type of life, a life form, and a morphology, with an indiscreet anatomy and
possibly a mysterious physiology" (Foucault 1978: 43). The categorisation
of homosexuality was also constituted in psychiatry, psychology and
medicine that confirmed the view of homosexuality as something
permanent, a quality of a person, contrary to previous comprehension of
sodomy which was totally a temporary aberration. The history of the
category called "homosexuality" shows that it is not a universal,
essentialist category but a specific term born to describe the people
suppressed to modal and medical criteria.

Global aspects of sexuality

The problem of the use of the term homosexuality is at the same time
ethical and theoretical. Ethical in the sense of whether or not we have the
right to define some people as homosexuals if they themselves do not
consider themselves as such and theoretical in a sense that, because of the
term's politiced history, whether or not we ever use it as a neutral concept
having only one meaning? Thus in the end the question is, whether the
concept can be used at all?
Murray and Roscoe (1998) defend the use of the term in the book
Boy-wives and female husbands: studies in African homosexualities. The
purpose of their approach is to show that homosexuality is a global
phenomenon that has been suppressed in anthropological and sociological
research due to West’s negative attitude toward homosexuality. Even
though the study is declared constructionist it could be viewed as
essentialist; even if it is trying to show cultural particularity of different
“homosexualities” it still assumes them as based on the same phenomena.

37
As Herdt underlines, it is important to interpret the phenomena rather than
consider them as literal realities. When one claims such entity such as
homosexuality one has to take under precise research the cultural factors
(Herdt 1993: 24). Murray and Roscoe argue that when scholars state that
homosexuality is not an African phenomenon it has real social
consequences since “they stigmatise those who engage in homosexual
behavior and those who are grappling with gay identities” (ibid.: XXII).
Still, we cannot assume that all who have same-sex relations consider
themselves as "homosexuals", or engage precisely in "homosexual
behavior". On the other hand it is understandable that those engaged in
modern gay ideology understand the situation differently.
Within globalisation, the ideas of "homosexuality" and gay-
identity have spread across the world and new groups of people have
expressed their will to live like western "homosexuals"11. This means that
they want to live according to their sexual desire and have long-term
relationships with another man (Altman 2001: 8) 12. The major factor that
has spread the modern gay-ideology is the “language of HIV/AIDS
control, surveillance, and education” (Ibid.: 8). Aids has opened new
channels to offer western ways to see sexuality -offering a model of gay-
life and identity it has developed a new gay consciousness. Another factor
has been gay-organisations which spread the gay-ideology. Thus for
homosexuals who are a part of "the global homosexual consciousness", the
research done about “homosexuality” has great significance.
To be correct it is necessary to highlight that "homosexuality" can
never be adopted in one way. More, the western picture of
"homosexuality" is used as a resource to create new, local versions of it.
And thus there is not only one gay-identity but several (Altman 2001: 2).
Even though Western "homosexuality" has new meanings, it still has the
base in certain ideology: the ideology of acceptance (politics), special
sexuality and implications to subculture.
11
Also Herdt considers homosexuality as a modern phenomenon (Herd 1997: 39)
12
Altmans research on Asian “homosexualities” can offer new and intresting perspectivs to
“homosexuality”, globalization and modernity. It shows how the “homosexual identity” has a
great meaning in modern negotiation over lifestyles and identities.

38
 What comes to the western influences of the "gay world", it is
worth to notice that it goes also vice versa. Western gay world can be
influenced by other cultures. Indeed, anthropologists have had a significant
role for the people in creating alternative sexual identities.
Since the anthropology has a long history with the theories of
different sexualitites (anthropologies have done research about "third
genders", "transsexualism" and "transgendering" both in "other cultures"
and in Europe and Americas) anthropological data and theory have
become a material for popular culture, and especially to gay and lesbian
cultures (Weston 1993 in Moore 1999:159).

39
 7. Sex and Gender dichotomies

Now that it has been demonstrated that homosexuality is not a valid,

universal concept of a specific cultural category of sexual behaviour, there
is a need to look closer on how the ideas of sexuality and its categories
have been formed in the research of sexuality. In the very heart of this lie
the western understanding of "male" and "female", i.e. the sexual
dimorphism. In the previous chapter it was explained how the
understanding of sexual dimorphism is nowadays a form of essentialism -
biologism. Now we will take a closer look to the "naturalness" of this
dimorphism and reflect how it has influenced on the understanding of sex,
gender and sexuality.
The studies of sexuality are best understood when the terms sex
and gender are contextualised to the western tradition. First, in the western
culture sexuality has long understood as a consequence of the dimorphism
of the (biological) sexes, as will be pointed out below. This means that
sexuality has been seen as a part of the "biological bias" even though
cultural constructivists have shown the diversity of sexuality among
different cultures. As I will highlight in this chapter, the evolutionary
theory and the dimorphism created by evolutionary thinking have had a
great importance on the research of divergent sex and gender roles and in
this way, on sexuality.
Secondly, sexuality has been an important aspect in the research of
variable forms of sex- and gender crossing. Often gender-crossing have
been interpreted as a sign of "homosexuality" or transsexuality. This rises
also a question of the connection of sex and gender which is specially
important when one does not seem to be fitting in either of the assumed
sexes or genders. Individuals such as Indian Hijras and Berdaches fall
between these dualistic categories. Even though there are several examples
from different cultures that speak against to the dimorphism of sex and
gender, it is still strongly present in different studies. The dichotomous
pairs man/ woman, masculine/ feminine are being in the back ground when
interpreting gender crossing.

40
 The ideas of sexual dimorphism are strongly rooted in Darwin's
theory of natural selection which assumed that natural selection gives a
certain function to male and female sexuality: the reproduction (Herdt
1993: 25). "By sexual dimorphism is typically meant a phylogenetically
inherited structure of two types of human and sexual nature, male and
female, present in all human groups" (Herdt 1993: 25). This thinking has
reached the point where it is considered as a base for all actions of "man"
and "woman". Being universal, existing in all places and times, it is the
"principle of sexual dimorphism" [italics added] (ibid.: 26). After the 19th
century Darwinism, this view was confirmed by the science of sexology. It
stated that heterosexuality is normal and necessary form of sexuality
because of its reproductive nature (ibid.: 28) and that two sexes "male" and
"female" form an universal structure that maintains the law of
reproductivity. "Homosexuality", which did not belong to this category,
became the counter pole for "normality" as we have already seen in the
previous chapter.
"Male" and "female" became objects for medical definitions: the
clinical and sexological practices were standardised in order to define all
the human beings either "male" or "female" at birth. Individuals who did
not fit to either of these categories (i.e. whose "biological sex" were
uncertain) were considered as abnormal (ibid.: 30). Soon this dimorphism
was adapted also in the research of social sciences. It followed the ideas
that sexology had produced. The distinctions between nature and nurture,
biology and society, and heredity and environment were formulated in the
paradigms of essentialism versus constructionism that have been reduced
to mean biologism versus culture. This analogy is, however, misleading
(ibid.: 30). This has also been the case in the research of sex and gender.
The concept of gender came to existence meaning "learned cultural
elements" whereas sex referred to "biological elements" (Herdt 1993: 30).
But as will be showed this dichotomy has not been unproblematic.
The distinction of sex and gender was created in the 1970's in
social sciences but soon it spread to all humanities. First, gender was used
to describe the "social sex” i.e. to explain the cultural ideas of “man” and

41
“women” and how individuals are socialised to become such (Moore 1999:
151 [1988]. This kind of anthropology was referred to as women's
anthropology and it focused mainly on women's life explaining the
universal subordination of women. The most known theorists of gender at
the time were Michelle Rosaldo and Sherry Ortner, whose main
contribution was based on the idea of the existence of major dichotomies
considered to be universal.
Rosaldo (1974) presented the cross-cultural public sphere –
domestic sphere dichotomy as the cornerstone of female oppression. The
domestic sphere was linked to childcare, thus to women, and it was given
lower status than the public one. The subordination gap depended on the
level of differentiation between the two spheres (Public vs. domestic).
Ortner theorised gender in the context of nature-culture dichotomous pair
in her well known and much commented article “Is female to male as
nature is to culture?” (1974). She based her binary opposition on Lévi-
Strauss’ idea that the nature -culture dichotomy has global proportions,
and continued by developing that idea. She claimed that representations of
woman are close to nature when, accordingly, men are close to culture.
This means that female subordination is situated in “cultural explications
of biological difference” (Pine 1996: 254). These binary oppositions were
based on a self-evident assumption that women’s biology (reproduction)
kept women apart from men and gave them lower status (Pine 1996: 21).
In the 1980s the critical focus of the studies shifted and was
renamed the anthropology of gender (Moore 1999: 152). The anthropology
of gender studied gender relations as a structuring principle of men and
women in relation to each other. The gender was still seen as cultural
marker of the sex. What was important for anthropological theory, gender
became a tool that helped to understand cultural variation of categories of
man and woman. A remarkable work of this period is Ortner’s and
Whitehead’s (1981) Sexual meanings - The cultural construction of gender
and sexuality where the authors consider gender, sexuality and
reproduction as symbols that can be analysed and interpreted only in the
context of the culture in which they exist (Pine 1996: 257). They state that,

42
in order to define terms such as ”male”, ”female”, ”reproduction” etc.
(terms that belong to the study of gender and sexuality), one has to take the
whole cultural context into consideration, “see the larger context of
interrelated meanings” (Ortner & Whitehead 1981: 2). These terms are
symbols that should be related to other cultural meanings and symbols on
one hand, and to the forms of social life and experience on the other
(Ortner & Whitehead 1981: 2).
Their work has been much appreciated, as it shows that gender and
sexuality cannot be detached from the cultural context; they challenge
essentialist assumptions of sexuality and gender. Nevertheless, the
weakness of this view is that it assumes only one variable gender system;
later work in anthropology has shown that there can be several
contextually and biographically diverse gender systems in one culture
(Sanday and Goodenough 1990; Strathern: 1987; Moore 1994: 559). We
will return to this topic in the next chapter.
Until the end of the 1980's the categories of sex and gender were
considered to follow directly the biological categorisation and “what was
acquired through socialisation was really no more than a cultural gloss”
(Moore 1994: 54). In the 1970s the radical message was that gender
existed whereas in the 1980's the question was whether or not sex exists.
Thus universal dichotomies were abandoned “as reflections of dominant
Western discourse, historically situated and socially and culturally
specific” (Pine 1996: 255). The assumptions of universal subordination
and binary oppositions were to be used only in western contexts. Moreover
it was said that these theories never managed to make a profane distinction
between sex and gender, because of the assumptions that gender is a
reflection of basic sex differences (Pine 1996, Moore 1994a: 816).

43
 Poststructuralist theories

Poststructuralist theories have considered largely the problematic of

gender, subject and body and have questioned fixed meanings given to
these concepts. Henrietta Moore is one of the theorists who have taken up
poststructuralist theories in order to understand gender and subjectivity.
Moore (1994) questions the concept “biological sex”, that has been
regarded as “natural” opposition pair to gender and agrees with
Yanagisakos and Colliers’ claim that both sex and gender are socially
constructed (Moore 1994a: 816, Yanagisako & Collier 1987). According
to Moore, Yanagisako and Collier (1987) claim that “Bodies,
psychological processes and body parts have no meaning outside of
socially constructed understanding of them. Sexual intercourse and human
reproduction are not just physiological processes, they are also social
activities” (Moore 1994a: 816). Thus if we understand that also ‘sex’ is
socially constructed, biological sex loses its validity as a “base” for the
gender constructs. This notion has led several poststructuralists to theorise
gender as free floating artifice that is built through discourses.
The path breaking author for many poststructuralists has been a
French philosopher and historian Michel Foucault who questioned the idea
that sexuality is something constitutive and, that human beings are sexed
and sexual by nature. He saw that sexuality, as well as gender, is a product
of social discourses (and powerful institutions) that make “sexuality” and
“homosexuality” to look like they were “axiomatic truths”. Foucault
(1978) writes in his book The History of Sexuality that “…sexuality is not
a natural reality but the product of a system of discourses and practices
which form part of the intensifying surveillance and control of the
individual” (Sarup 1993: 72). Also gender difference is constituted
through discourses. "[…]The meanings ascribed to bodies are culturally
produced, plural and ever changing" (Weedon 1999: 102) which means
that there is no essential core of gender inside person. The subjectivities of
feminine and masculine are negotiated within different discourses in the
middle of several power relations.

44
 To insist, for example, that the meanings of 'woman' are neither
singular nor stable and are effects of power, including class and
racial power, is to suggest that what it means to be a woman varies.
"This is not, however, to suggest that particular, socially and
culturally produced forms of gendered subjectivity do not contain
elements which may be shared by women across classes, races and
cultures. Nor is it to suggest that particular forms of embodied,
gendered subjectivity are somehow inauthentic and not lived as real.
Ideas of true femininity and masculinity are replaced by competing
discursive constructions of gender." (Weedon 1999: 102)

The disciplinary production of gender creates a false image of stabilised

gender. This means that we experience gender as something coherent since
the gender is applied (learned) through these discourses (Butler 1999: 172-
173). People incorporate13 the gender from the discourses and carry out it
in his/her action gestures and desire. This creates an "…effect of an
internal core of substance" (ibid.: 173). Still, it is important to mark that
this happens "on the surface of the body" (ibid.: 173). This is what Butler
calls the performative character of gender. Since the gender is formed
through different discourses, it is separate from biological sex and free to
be “played” or “performed”. Butler theorises that discourses form an
idealised gender which we wish to incorporate as well as possible. This is
an idealisation of gender coherence. To put it differently, things we do to
perform the gender (i.e. acts, gestures, words) form a sense of a substance,
but is really actions on the surface. The essence or identity that these acts
are meant to present, are "…fabrications, manufactured and sustained
through corporeal signs and other discursive means" (Butler 1999: 173).
The single gender system approach (that was theorised is the
1970s and 1980s) goes further towards a static picture of an individual and
culture; gender is a cultural model in which individuals must be socialised.
This view gives little room for “failure, resistance and change”(Moore
1994: 55-56). Those who do not fit in the assumed “gender system” are
13
Term by Foucault (1979) in Discipline and Punish: the Birth of the Prison. New York: Vintage.

45
considered cultural variation. The poststructuralist view that Moore
presents assumes that there are several discourses on gender, where the
subject takes different positions, instead of one gender system (ibid.:: 55).
We cannot state that an individual is one subject. More likely, the
individuals are “multiply constituted subjects who take up multiple subject
positions” within different discourses (ibid.:55). Thus a subject is actually
a location of several subjectivities. The man and woman categories are
produced through several discourses and an individual understands his/her
gendered subject depending on the discourse. To put it differently, there
are several femininities and masculinities, as Connel (1987) argues (In
Moore 1994: 59). This view minimise the meaning of the "physical sex"
and highlights the "...individual and social factors, such as occupational
preference, behavior and temperament, religious experiences and so forth"
(Roscoe 1993: 342).
What is important to notice is that these categories can be
oppositional and mutually exclusive, which means that an individual can
live out several contradictory positions. The individual self does not
experience this as disorder but constitutes the sense of self by three factors:
the physical fact of having a body, the experience of identity, and the
historical continuity of subject positions. As mentioned above, for Butler
the self is construed by discourses and acted out in gestures, which give
the sense of "permanence", the self. In my opinion, the poststructuralist
theorisation of identity and gender is suitable for explaining the gender-
crossing that many texts see as a direct consequence of sexuality. Men
who behave or dress like women are considered to do so in order to show
their “homosexuality” or “transsexuality”. According to Moore the
construction of gender is not attached to biology and this way to
reproduction and is free to be "played" as Butler reflects.

46
 Third gender

Persons who diverge from the traditional idea of sexual dimorphism are
sometimes categorised as third gender. This gender category includes
those people who do not fit into the dimorphic categories of two genders
i.e. man and woman. Herdt (1993) starts his definition of the third sex by
naming what the third sex is not. Notable is his strong opposition that the
members of this category are not "homosexuals" (Herdt 1993: 47).
As already Foucault pointed out, "homosexuals" became a sexual
category in the 19th century which was later confirmed by the medical
science of sexology. (Foucault 1978: 43). "Homosexuality" was for long
studied as a such category assuming that the sexual orientation of those
committing to same sex relations is an adequate principle to name them as
"homosexuals". However, same-sex desires cannot be interpreted only in
the name of sexual acts. They should be linked to their "social
classification and expression by age, gender class and egalitarian modes of
social ideas and relationships" (Herdt 1993: 47).
The persons categorised as the third gender constitute a social
person and cultural reality. Thus they should be viewed from their social
position, not their "sexual orientation" (ibid.: 47). Also, the application of
the sexual dimorphism or the heterosexual/homosexual duality is part of
the western culture and should not be automatically considered as valid in
other cultures. This concerns also the term bisexuality that follows the
sexual dimorphism and this way does not manage to get rid of the western
tradition (ibid.: 49). To conclude these thoughts briefly: "There is no
absolute link between sexual orientation and....a third gender" (ibid.: 50).
The best illustration of the third gender is their social role and
position in society. The role itself is culturally constructed and includes
different social tasks and certain statuses. It is important that the social
structure maintains the role and the category of third gender. This way it is
given a stronger position in longer period (Herdt 1993: 60). This category
could be described also as an ontological entity. It is a distinctive subject
where people identify to. Like Moore stated, an individual constitutes the

47
sense of self by three factors: the physical fact of having body, the
experience of identity, and the historical continuity of subject positions.
Famous examples of such ontological third sex/gender entities
are, for example, Indian Hijras and North American Berdaches. Berdaches
are "womanlike" men who have been documented in approximately 150
North American societies.14 Until recently they have been studied only
marginally, but during the last 20 years more interest have been shown
toward this group of people (Roscoe 1993: 330). Many of the researchers
have translated the Berdache as a "homosexual", "transsexual",
"hermaphrodite" or "sodomite" (ibid.: 332).
Like in the case of yan daudu, the roles of berdaches is to be
defined in other than sexual terms. There are three key features that define
their roles. These are productive specialisation, supernatural sanction and
gender variation. Like yan daudu, also berdaches have their special
"profession". Many of the male berdaches work with domestic work and
crafts when females hunt and are in leading positions. Supernatural
sactions refer to berdaches as mediums of supernatural powers. As we
have noticed before, yan daudu have no such powers. Their role is totally
mundane, without any participation in possession itself. The third feature,
gender variation, has appeared to be the less certain marker of berdache
status. Only some of berdache cross-dresses or it might be occasional.
What comes to the information about their sexual relations, some had
same-sex relations, but not all (Roscoe 1994: 332-335).
Thus berdaches are a distinctive group of men and women who
are specialised in their own society. In what terms can berdaches be
defined as third gender? Definitely one term is that their status is
considered as a separate category, different from "man" and "women". The
name berdache is the sign of the category. As as described earlier, they
have specialised occupational practices that includes specific symbols,
such as "distinct color of feathers" and separate life-cycle rites from others.
(Men and women have their own rites, when berdaches form the "third

14
Also berdaches, who are "biological" woman, dressing as man, are reported. However they are less
in number, as are also studies concerning thems

48
group" of initiated.) In some groups, the existence of this category is also
confirmed "…in the same context in which male and female gender
categories were defined" (Roscoe 1994: 339).
Another example can be taken among the hijras of India. They
differ yan daudu and berdache, since they go through emasculation i.e.
their genitals are removed. Their occupational role includes "performing at
the birth of a male child and at marriages and as servants of the goddess at
her temple" (Nanda 1994: 373). They also work as prostitutes. The word
hijra is a masculine noun that is translated often as "eunuch" or
"hermaphrodite". These both terms refers to their impotence in their sexual
relations with women (ibid: 380).
Hijras have adapted feminine mannerism and dressing and many
of them identify themselves as women. Still they clearly show that their
feminine gender is a "play". Contrary to the traditional role of women
hijras bring out their sexuality aggressively and exaggerate their dress and
manners. They also participate to occasions that are uncommon for
women. They, for example, dance publicly and smoke cigarettes (ibid.:
328).
As it is with yan daudu and berdaches, the signals of gender-role
vary among the hijras. Not all of them adopt female names or wear
women's clothes. Some might dress as men and grow a beard. Some hijras
can also put off their genital operation in order to have more time to
consider it through. The communities of hijras allow several gender roles
and do not necessarily demand a solid gender identity from their members.
Even tough many of hijras are socialised as men in their childhood they
still take this alternative gender role as hijras. According to Nanda (1994)
there are two explanations for this. First, many of them have had
ambivalent gender socialisation in their childhood. Many of the hijras
have been supported to bring out both feminine and masculine behavior in
their families. Most of the hijras have joined hijra communities in their
early adulthood. Second, most of the hijras enjoy being sexual receptors
for men. This has led them to define themselves as "…effeminate man, no
longer capable of having

49
 Yan daudu as third gender

What is common to both hijras and berdache is their religious role in the
society. Hijras participate in blessing the newborn and act as servants of
Mother Goddes and berdaches are believed to possess supernatural forces.
Yan daudu lack the supernatural aspect. The most obvious indicators of
yan daudu status are economic and transgendering attributes. Yan daudu
have a position in the context of karuwanci as recruiters of new karuwais
and as mediators between clients and karuwais.
In which terms yan daudu could be interpreted as a third gender
category? As berdache and hijras, are yan daudu a distinctive category of
people who have their special role? Still it was impossible to find any
evidence that Hausa would refer it especially as a gender group. Yan
daudu themselves find them as men who behave like women but do not
share any special marks that separate them from two other genders of
Hausa society (i.e. man and women). They hold to gender signs that are
commonly shared with the difference that they blend these signs. Also the
social structure of the Hausa society seems to support only two gender
categories, i.e. man and women These categories are confirmed by the
strict Islamic rules (for example women's seclusion) and brought out in
everyday life. Boys are raised as boys and girls as girls. The socialisation
of yan daudu starts when the other yan daudu recruits a new member for
the occupation. By this it can be assumed that yan daudu do not form a
category of third gender.
Finally, is it important to categorise yan daudu by their gender? Or
can the whole concept be used at all? As it was presented in the previous
chapter, gender can be understood as a set of subject positions that are
taken in different discourses. This way it is possible to see the subject as a
person who acts according to his/her own interests. Traditionally
anthropologists have had the tendency to privilege culture against personal
interests (Weston 1993, Graham 1998 in Moore 1999: 158). The context of
karuwanci and bori cult create discourses for yan daudu where they carry
out their role. Yan daudu are not forced to this role "by their nature" but

50
chooses it according to their interest. In the next discussion it will be
looked closer how this poststructuralist understanding of gender can be
applied in the case of yan daudu.

51
 7. DISCUSSION

The post-structuralist understanding of the subject is useful especially

when explaining the situation of the Yan daudu. Many authors have stated
that the yan daudu are “homosexuals” treated as deviants not fitting either
of the gender categories (for example Wall 1988; 167, Besmer 1983: 19).
A closer look at the yan daudu role shows that gender crossing has
explanation other than just being an expression of “homosexuality”. As
different discourses produce different subject positions, we can see that
yan daudu can negotiate gender, depending on the context. Same-sex
relations are a totally irrelevant indicator for example in the context of
karuwanci. On the one hand yan daudu are considered as (biological) male
and thus by karuwais they are seen to present the male power (Kleis &
Abdullahi 1983). On the other hand yan daudu present ”feminine men”:
The femininity of yan daudu is understood as a “play” (According to
Gaudio 1998) and thus it is given a different meaning. For yan daudu,
femininity presents a kind of equality in relation to women since they work
among them. This seems to be a quite contradictory situation. Why do yan
daudu take these different subject positions in the institution of
karuwanci? And why taking these positions is important?
Kleis and Abdullahi assume that yan daudu are acceptable to both
prostitutes (women) and the male establishment because of the gender
crossing. The male establishment appreciates yan daudus’ position as a
representative of male power and because of yan daudu's assumed sexual
ambivalence they are not a (sexual) threat to prostitutes (i.e. potential sex
partners) (Kleis & Abdullahi 1983: 45). Or rather, they do not break the
moral rule of separation. In the Hausa community the distinction between
man and woman is strict and it is seen nearly in every aspect of life.
Women are isolated from the men’s sphere both symbolically and
concretely. Men having more public roles, women are isolated into
seclusion inside the household, according to the Islamic rule 15(Kleis &

15
(this does occure in all house holds -It is more common in upper class or among those who afford
to have their wife at home)

52
Abdullahi 1983: 39, Wall 1988, Gaudio 1996, Besmer 1983). According to
Barbara Callaway (1987) the strict separation is of Islamic origin. Any
woman who is seen with another man than her husband is considered to
have a sexual relationship with this man. Because of the separation
between men and women in society, they are not assumed to meet for any
other than sexual reasons (Callaway 1987: 44). The only women who can
cross the boundaries between men and women are karuwais.
Karuwanci has been a way for women to gain economical
independence in society -normally wives are much dependent on their
husband’s incomes, excepting some minor economical activities to have
incomes such as selling food (Coles & Mack: 1991: 19, Callaway 1987:
76). Those who are engaged in karuwanci can also take social liberties
such as drinking beer with the men or otherwise showing her
independence of male authors - thus they are a complete contrast to a
secluded and sexually confined wife (Callaway 1987: 43-44, Kleis &
Abdullahi 1983: 50). Even though many authors describe the karuwanci
nearly as a positive and appreciated phenomenon, they are socially in very
low ranked category. They are deviants, deviating from the common
female role (Kleis & Abdullahi 1983: 43)
Kleis and Abdullahi (1983) assume that all the strict rules are
supposed to keep women subordinated to men and exclude them from
men’s dominant spheres (Kleis And Abdullahi 1983: 40). Kleis and
Abdullahi consider the karuwansi as a potential threat to masculine
hegemony. The only institution for diminishing this threat and conserve
the masculine power are yan daudu. They are the ones who are in the
position to “control” karuwanci: they mediate between the karuwais and
the clients and manage relations with authorities. The role of the yan
daudu appears to be professional (Ibid.: 44). For his thesis Abdullahi
(1984) interviewed several Yan daudu and 57% of them admitted of being
yan daudu for economic reasons. Second biggest group (22%) named the
influence of friends and associatios. Many of them were from the
countryside and the came to the town for Koran studies or to look for a job
(Abdullahi 1983:53) As mentioned this profession is considered as deviant

53
in Hausa society. What motivates them to commit themselves to the role
and profession as discussed above?
Wendy Holloway (1984) has introduced the term investment to
explain why an individual sometimes takes subject positions in completely
contradictory gender discourses. According to her, individuals invest in
particular subject positions on the basis of emotional commitment and
personal interest. Personal interest is related to the relative power that the
individual assumes in the subject position (Holloway 1984: 238). The
expectations are sufficient to waken the interests – the position does not
necessarily provide such a reward. Investment becomes essential in order
to understand the role of yan daudu. What kind of assumptions can be
made about the investment and yan daudu?
First, there is an economic reward. As mentioned above those
who commit themselves to the role of yan daudu are often teenage boys
moving from the countryside to urban centers. Since most of them do not
have any social contacts or any incomes there, it is tempting for them to
take offered jobs in the context of karuwansi (Kleis & Abdullahi 1983:53).
Also, in later age, the role of yan daudu makes it possible to earn an
income. Thus we can assume that part of the reason men taking the role of
yan daudu is economical.
Second, even though the “profession” is seen as marginal and
deviant, yan daudu enjoy certain power within the institution of
karuwansi. Even if they have an ambivalent gender status, they still
present the masculine hegemony in the karuwansi. They are not
considered as women but ambivalent men and thus they are given the
power status of men (he is the “patron” of prostitutes). Yan daudu are
acceptable to both prostitutes (women) and the male establishment because
of the gender crossing. The male establishment appreciate yan daudu’s
position as a representative of male power and because of yan daudu’s
assumed sexual ambivalence they are not a (sexual) threat to prostitutes
(i.e. potential sex partners).
Third, what resides in the power is prestige. This prestige is
enjoyed mostly inside the institution, but according to Abdullahi (1984),

54
they are occasionally given even some administrative responsibility (Kleis
& Abdullahi 1983:47). As Pittin has pointed out, a religious context may
"permit or encourage the reflection of alternative positions or interests
within the community" (Pittin 1996: 213)
The fourth reward for yan daudu is attached to their sexual
relations. Even if we cannot state that all yan daudu have same-sex
relations, their role enables them to meet potential (sex)partners. Rudolf
Gaudio (1996) emphasises the meaning of same-sex relations for yan
daudu. He considers yan daudu as a gay community and interprets the role
to be primarily a “gender-play” based on their “homosexuality” (Gaudio
1996: 119ff). To my mind this view seems to simplify the understanding
of yan daudu, overlooking economic, professional and prestigious rewards
of yan daudu. But since Gaudio seems to present quite a convincing
description of same-sex relations (that was the main focus of his research
in Hausa society) it is possible to assume that yan daudu can also meet
potential (male) sex-partners in the “shelter” the role offers. Abdullahi
(1993) mentions that 7% of the yan daudu he interviewed attributed
becoming yan daudu to their nature. Unfortunately Abdullahi does not
give any further explanations if this "nature" implies to their sexuality
(Kleis & Abdullahi 1984: 53) .
Finally, yan daudu invest in relations with other participants, as
O'Brien (1999) has emphasised. In the context of bori cult, karuwai and
yan daudu create bonds of friendship which is not because of shared
marginality but overlapping social worlds (O'brien 1999: 10).

55
 8. Conclusions

Inspired by Henrietta Moore’s book, A Passion for Difference (1994), I

decided to approach the role of yan daudu by focusing on gender identity,
subjectivity and their relation to society. Anthropologists have avoided
questions of identity and subjectivity by emphasising that gender systems
are culturally constructed - individuals have been seen as a part of the
collective. (Moore 1994: 55). The problem is situated between the
individual and the culture. Traditionally, an individual subject is
considered in anthropology subordinated to culture, but in reality
individuals are not “singular entities which require a cultural imprint”
(ibid.: 54). The relation between an individual and his society is surely full
of resistance and change, not only submission to entities (culture). It is
important to see the individual as an active actor in the dynamics of
society.
The post-structuralist understanding of gender highlights the
dynamics and change of the subject and society. This is particularly
relevant to gender relations. It allows us to gain a more complex idea of
gender and move further from essentialist assumptions of gender as a
consequence of biological categories of man and woman. There are not
just two genders, but a multiple set of gender discourses that overlap with
the discourses of age, sex, class and kinship. Gender identity is not a
simple entity, which is a result of direct socialisation, but a complex set of
different identities that the subject takes out in different gender discourses.
Gender discourses, as subject positions, are multiple and often
contradictory. It is therefore possible to take into account the resistance
and change in gender identity and why individuals take positions that can
look contradictory to their identity. For instance why yan daudu take a
position that is not appreciated in society or why he dresses as a “woman”
if he is a “biological” man. I have shown that the gender-identity is
attached to other spheres such as production and politics and thus only the
sexuality cannot define one’s identity.

56
This work has been based on the texts written about Hausa culture and yan
daudu. During the writing process I have faced questions that would
definitely require fieldwork among them. The ethnographic data found
about yan daudu has been minimum and, as criticised through this paper,
often only in footnotes. Valuable material has been Gaudios and Kleis' &
Abdullahi's articles that have focused mainly on yan daudu. These articles
have lightened up the mystics around them. In other texts many writers
have considered yan daudu as something bizarre. In a way there has been
"double otherness". By categorising yan daudu as homosexuals and
deviants, anthropologists have considered them as "other" of the "other
culture".
Other ethnographical data presented in this paper have focused on
the bori spirit possession. Bori is dynamic set of beliefs that is also
important way of life for several people. It brings the extraordinary to
everyday life and everyday life to extraordinary. Anthropologists have had
different views on bori: it has been considered as healing, tradition and
resistance toward Islam and colonial powers. In this paper I wanted to
emphasise the complexity of bori, that at the same it can have several
functions. The most essential of them depends on the time and place, the
context.
For yan daudu, bori has given the environment where they work,
live and enjoy themselves. Karuwais organise often bori ceremonies where
yan daudu dance and prepare food. Yan daudu also mediate between
karuwai and their clients. The importance of their occupational role
became the most important feature of yan daudu in this paper. Even tough
the sexuality was discussed quite largely it was done in order to show that
our interpretations of transgendering are often made from a certain point of
view.
Transgendering people are found in all over the world. Their
position is sometimes institutionalised (third gender) sometimes reduce to
deviancy. There are some who remove their organs (or replace them more
correctly) and some who uses gestures or other signs to bring out their
gender. The variation is endless. Yan daudu are one group of men who

57
dress themselves women like and represent an institutionalised form of
transgendering people.

58
 REFERENCES

Abdalla, Ismail H.,1991, Spirit Possession as Therapy: Bori Among Non-

Muslims in Nigeria. Lewis, I.M. Al-Safi, Ahmed & Hurreiz, Sayyid
(eds), Women’s medicine: The Zar-bori Cult in Africa and Beyond.
Edinburgh: Edinburgh U.P.

Abdullahi, S.A, 1984, A sociological analysis of the institution of ‘dan daudu’ in

Hausa society. Thesis. Kano: Bayero University

Ahlberg, B.M., 1994, Is there a distinct African sexuality? A critical response to

Caldwell. Africa: Vol. 64, No. 2:220-242

Altman, Dennis, 2001, Global Gaze/Global Gays. Intersections and Essays.

Hawley, John (ed), Postcolonial and Queer Theories. Westport:
Greenwood Press

Ames, David W., 1982, Context of Dance in Zazzau and the Impact of Islamic
reform. In African Religious Groups and Beliefs. Ottenberg, Simon
(ed.), Delhi: Archana Publications

Andersson, Ulrika, 2002, Working with spirits among Muslim Hausa in Nigeria.
A Study of Bori in Jos. Working Papers in Cultural Anthropology, No.
11. Department of Cultural Anthropology and Ethnology. Uppsala
University

Beattie, John, 1964, Other cultures. Aims, methods and achievements in social
anthropology. London: Routledge and Kegan Paul LtdBehrend, Heike
& Luig, Ute, 1999, Introduction. Heike Behrend & Ute Luig (eds),
Spirit Possession ,Power and Modernity in Africa. London: James
Currey.

59
Besmer, Fremont, E, 1983, Horses, Musicians & Gods: The Hausa Cult of
Possession-Trance. Massachusetts: Bergin & Garvey Publisher.

Bleys, Rudi C., 1996, The Geography of Perversion. Male-to-Male Sexual

Behaviour outside the West and the Ethnographic Imagination, 1750-
1918. Cassel. London

Butler, Judith, 1999, Gender Trouble: Feminism and the Subversion of

Identity. New York: Routledge

Callaway, Barbara, 1987, Muslim Hausa Women in Nigeria. New York:

Syracuse University Pess
Connel, Robert, 1987, Gender and Power. Cambridge: Polity Press

Dunbar, Roberta Ann, 1991, Islamic Values, the State, and “the Development of
Women”: The case of Niger. Coles, Catherine & Mack, Beverly (ed.),
Hausa Women in the twentieth Century. Wisconsin & London: The
University of Wisconsin Press

Echard, Nicole, 1991, Gender relationships and Religion: Women in the Hausa
Bori of Ader, Niger. Coles, Catherine & Mack, Beverly (ed.), Hausa
Women in the twentieth Century. Wisconsin & London: The University
of Wisconsin Press

Evans-Pritchard, E. E , 1956, Nuer Religion. New York : Oxford University

Press

Foucault, Michel, 1978, The History of Sexuality, Volyme 1. An introduction.

London: Peguin books Ltd.

60
Fry, Peter, 1985, Male Homosexuality and Spirit Possession in Brazil.
Blackwood, E. (ed.), The Many Faces of Homosexuality:
Anthropological Approaches to Homosexual Behaviour. New York:
Harrington Park Press

Gaudio, Rudolf P, 1998, Male lesbians and other queer notions in Hausa.
Murray, Stephen O. & Roscoe, Will (eds), Boy-wives and Female-
husbands: Studies of African homosexualities. New York: St. Martin’s
Press.

Graham, M, 1998, Follow the Yellow Brick Road: An Anthropological Outing

in Queer Space. Ethnos 63(1): 102-32

Greenberg, J.H, 1946, The influence of Islam on a Sudanese Religion. New

York: J.J. Augustin.

Grosz, E, 1995, Space, Time and Perversion. New York: Routledge

Herdt, Gilbert, 1994, Introduction: Third sexes and Third Genders. Herdt,
Gilbert (ed.), Third Sex, Third Gender: Beyond Sexual Dimorphism in
Culture and History. New York: Zone Books

Herdt, Gilbert & Bower, Andrew, 1995, Bisexuality. Toward a Comparative

Theory of Identities and Culture. Parker, Richard & Gagnon, John
(eds), Cenceiving Sexuality: approaches to sex research in a
postmodern world. London: Routledge

Holloway, Wendy, 1984, Gender difference and the production of subjectivity.

Julian Henriques, Wendy Holloway, Cathy Urwing, Conze Venn &
Valerie Walkerdine (eds), Changing the Subject: Psychology, Social
Regulation and Subjectivity. London: Methuen

Kleis & Abdullahi, S.A, 1983, Masculine Power and Gender Ambiquity in Urban
Hausa Society. African Urban Studies, No 16, p.39-53

61
Lambek, Michael, 1981, Human Spirits. A Cultural Account of Trance in
Mayotte. Cambridge: Cambridge University Press.

Lambek, Michael, 1993, Knowledge and practice in Mayotte local discourses of

Islam, sorcery, and spirit possession. Toronto : University of Toronto
Press

Last, Murray, 1991, Spirit possession as therapy: Bori among non-muslims in

Nigeria. Lewis, I. M. (ed), Women’s medicine. Oxford: The Alden Press
Ltd.

Last, Murray, 1993, History as Religion: De-constructing the Magians

‘Maguzawa’ of Nigerian Hausaland. CHRETIEN Jean-Pierre Karthala
(ed.), L’invention religieuse en Afrique. Histoire et religion en Afrique
noire. Paris: Éditions Karhala et ACCT.

Lewis, I.M, 1971, Ecstatic religion. An anthropological Study of Spirit

Possession and Shamanism. Harmondsworth: Penquin books Ltd.
Masquelier, Adeline, 1999, The Invention of Anti-Tradition. Dodo Spirits in
Soutrern Niger. Behdend, Heike & Luig, Ute (eds), Spirit Possession.
Modernity & Power in Africa. Oxford: James Currey Ltd

Masquelier, Adeline, 1993, Narratives of Power, Images of Wealth: The Ritual

Economy of Bobi in the Market. Comaroff, Jean & Comaroff, John
(eds), Modernity and its malcontents: ritual and power in postcolonial
Africa. London: The university of Chicago Press Ltd.

Masquelier, Adeline, 2001, Prayer Has Spoiled Everything.Possession,

Power,and Identity in an Islamic Town Of Niger. Durham & London:
Duke University Press.

Moore, Henrietta, 1999, Whatever Happened to Women and Men? Gender and
other Crises in Anthropology. Moore, Henrietta (ed.), Anthropological
theory today. Cambridge and Oxford: Polity Press

62
Moore, Henrietta, 1994a, Understanding sex and gender. Tim Ingold (ed.),
Companion encyclopedia of anthropology: humanity, culture and social
life. London & New York:. Routledge

Moore, Henrietta, 1994, A Passion for Difference: Essays in Anthropology and

gender. Cambridge and Oxford: Polity Press

Murray, Stephen O & Roscoe, Will, 1998, Preface. “All very confusing”
Murray, Stephen O & Roscoe (eds), Boy-wiwes and female husbands:
Studies in African homosexualities. New York: Will. St. Martin’s Press

Nanda, Serena, 1994, Hijras: An Alternative Sex and Gender Role in India.
Herdt, Gilbert (ed), Third Sex, Third Gender: Beyond Sexual
Dimorphism in Culture and History. New York: Zone Books

Nicolas, Jacqueline, 1967, “Les Juments des Dieux”: Rites de possession et

condition féminine en pays Hausa. Etudes Nigériennes, No.21.
Niger:IFAN-CNRS
O'Brien, Susan, 1999, “Pilgrimage, Power, and Identity: the Role of the Hajj in
the Lives of Nigerian Hausa Bori Adepts.” Africa Today 46, 3-4
(December, 1999).

Ombolo, Jean-Pierre, 1990, Sexe et société en Afrique noire. L’anthropologie

sexuelle beti: essai analytique, critique et comparatif. L’Harmattan, Paris

Onwuejeogwu, Michael, 1969, The Cult of the “Bori” Spirits Among the Hausa.
New York: Anchor Books.

Ortner, Sherry B. & Whitehead, Harriet, 1981, Introduction: Accounting for

sexual meanings. Ortner, Sherry & whitehead, Harried (eds), Sexual
meaning. The cultural construction of gender and sexuality. Cambridge:
Cambridge university press.

Petersen, Alan, 1998, Unmasking the masculine. ‘Men’ and ‘Identity’ in a

Sceptical Age. London: Sage publications ltd

63
Pine, F, 1996, Gender. Barnard, A and Spencer, J (eds) Encyclopedia of Social
and Cultural Anthropology. New York: Routledge, 253-262.

Pittin, Renée, 1996, Possession and Dispossession: Changing Symbolic

structures and Meanings in Contemporary Nigeria. Parkin, D. Caplan, L.
& Fisher, H. (eds), The Politics of cultural performance. Oxford:
Berghahn Books

Rosaldo, M, 1974, Woman, Culture and Society: A Theorethical Overview.

Rosaldo, M. and Lamphere, L. (eds), Woman, Culture and Society.
Standford: Stanford University Press.

Roscoe, Will, 1993, How to become a Berdache: Toward a Unified Analysis of

Gender Diversity. Herdt, Gilbert (ed), Third Sex, Third Gender: Beyond
Sexual Dimorphism in Culture and History.New York: Zone Books.

Sanday, Peggy R. & Goodenough, Routh (eds), 1990, Beyond the Second Sex:
New Directions in the Anthropology of Gender. Philadelphia:
University of Pennsylvania Press

Sarup, Madan, 1993, An introductory guide to post-structuralism and post-

modernism. Hertfordshire: Harvester Wheatsheaf.

Sharp, Lesley A, 1991, The power of Possession in Northwest Madagascar.

Contesting Colonial & National Hegemonies. Behrend, Heike & Luig,
Ute (eds), Spirit Possession. Modernity & Power in Africa. Oxford:
James Currey Ltd.

Strathern, Marilyn, 1987, Understanding Inequality. Cambridge: Cambridge

University Press

Tremearne, A. J. N, 1913, Hausa Superstitions and Customs: An Introduction to

the Folk-lore and the Folk. London : John Bale, Sons & Danielsson

64
Vance, Carole, 1998, Social Construction Theory: Problems in the History of
Sexuality. Nardi, Peter & Schneider, Beth (eds), Social perspectives in
lesbial and gay studies: a reader. London: Routledge

Wall, Lewis L, 1988, Hausa Medicine. Illness and Well-being in a West African
Culture. Durham and London: Duke University Press.

Weedon, Chris, 1999, Feminism, Theory and the politics of difference. Oxford
and Massachusetts: Blackwell Publisher Ltd.

Wendl, Tobias, 1999, Spirit Possession, Slavery and Ritual Consciousness. A

Case Study of the Tchamba Cult among the Mina in Togo. Heike
Behrend & Ute Luig (eds), Spirit Possession, Power and Modernity in
Africa. London: James Currey. p. 111-123.

Weston, K, 1993, Lesbian/Gay Studies in the House of Anthropology. Annual

review of Anthropology 22: 339-67.

Yanagisako, S. & Collier, J, 1987, Toward a unified analysis of gender and

kinship. Collier, J & Yanagisako, S (eds), Gender and kinship: Essays
Toward a Unified Analysis. Stanford: Stanford University Press

65