Ambiguous Pleasures: Sexuality and Middle Class Self-Perceptions in Nairobi

By Rachel Spronk

Among both male and female young urban professionals in Nairobi, sexuality is a key to achieving a ‘modern’ identity. These young men and women see themselves as the avant garde of a new Africa, while they also express the recurring worry of how to combine an ‘African’ identity with the new lifestyles with which they are experimenting. By focusing on public debates and their preoccupations with issues of African heritage, gerontocratic power relations and conventional morality on the one hand, and personal sexual relationships, intimacy and self-perceptions on the other, this study works out the complexities of sexuality and culture in the context of modernity in an African society. It moves beyond an investigation of a health or development perspective of sexuality and instead examines desire, pleasure and eroticism, revealing new insights into the methodology and theory of the study of sexuality within the social sciences. Sexuality serves as a prism for analysing how social developments generate new notions of self in postcolonial Kenya and is a crucial component towards understanding the way people recognize and deal with modern changes in their personal lives.

• Language: English
• Publisher: Berghahn Books
• Release date: May 1, 2012
• ISBN: 9780857454799

Rachel Spronk

Rachel Spronk is Assistant Professor at the Sociology and Anthropology Department at the University of Amsterdam. She has published on intimacy and middle class formation in Kenya, on methodological questions of sexuality research and on the bounds of poststructural approaches to understand how sex(uality) is experienced.

Book preview

CHAPTER I

THE STUDY OF SEXUALITY

This study takes up the challenge set by Malinowksi: to analyse erotic encounters as a ‘fundamental factor in the scientific analysis of society’ (1982 [1929]: I) – in this case, of Nairobi society. I aim to analyse the intertwined discourses on cultural heritage, gender relations and sexuality, the varying directions that young professionals take in their intimate lives, and how these crossing points play out in their personal experiences. Before doing so, however, it is necessary to reflect on the development of research on sexuality in Kenya, and Africa, as this provides the departure point for this study.

Sexuality research in Kenya

The study of sexuality in Kenya shows several interesting shifts in scholarship. The first phase of study took a cultural and socio-economic perspective as a point of departure and analysed institutions of sexuality from the late nineteenth century to the middle of the twentieth century.¹ These studies provide systematic descriptions of initiation, marriage, birth and rites of passage and explore how these cultural institutions relate to the cosmology of a group. From the 1960s onwards, attention shifted to an examination of the transformations of these institutions.² These studies typically describe cultural phenomena, such as the status, roles and powers conferred on men by patriarchal ideologies, elaborating on the division of labour according to gender, reproduction and descent. Sexuality is described as an indispensable aspect of a people’s culture, with an emphasis on the social organization of sexuality. Usually, a chapter in a monograph of a specific ethnic group is confined to sexuality, such as for example ‘Sex life among young people’, a chapter in Kenyatta’s Facing Mount Kenya (1938). It offers an interesting account of the institution of ngweko (fondling): girls and boys were permitted to meet and enjoy ‘lovemaking’ without full sexual intercourse by placing the penis between the thighs of the girl. Kenyatta depicted such practices from an ideal, almost nostalgic, perspective. He detailed the moral rules and obligations pertaining to young people, which he described as being strictly enforced.³ What he left out, however, was information about trespasses or conflicts, which would have elaborated on people’s actual behaviour.

During the second half of the twentieth century, a few anthropological studies emerged in which sexuality was incorporated as a key theme. These studies covered subjects such as prostitution or its various forms, virginity, and female circumcision or same-sex sexuality.⁴ Apart from these studies, research on sexuality in Kenya is limited by the fact that sexual desire and practices are not studied as a distinct and separate theme. Instead, sexuality is studied as a sub-theme of other topics such as circumcision, reproduction, gender, and later, AIDS.

From the 1960s onwards, research on sexuality was conducted as a sub-theme primarily from a medical or health perspective. Three fields of sexuality-related research can be distinguished for this period. From the early 1960s onwards, family planning programmes studied sexual behaviour in relation to fertility. The primary interest was in analysing the frequency of the coital act, the use and choice of contraceptives, and to a limited extent spousal decision-making with regard to family size and contraception. This research was largely quantitative in nature and conducted mainly within the fields of demography, population and fertility studies. Even today, these studies play a central role in the production of knowledge about sexuality. From the 1970s, adolescents became a particular focus of study. And from the 1980s onwards, as part of the new focus of interest on public health, the emerging framework of reproductive and sexual health gained popularity. But apart from these demography, population and fertility studies, sexuality has remained a minor research topic. However, from the mid-1980s, the AIDS epidemic changed the climate for sexuality research dramatically. The need to determine levels of AIDS awareness led to the examination of patterns of sexual behaviour and the exploration of ‘risk behaviour’ or ‘transactional sex’ among subgroups of the population such as sex workers, migrants, adolescents, etc. Recently, studies have also looked at sex among discordant couples and the sex lives of infected people.

The impact of AIDS has forced the topic of sexuality onto the broader academic agenda, because sexual practices have become a potential cause of large-scale suffering for individuals, families and communities. This has resulted in a body of scholarship dealing with different social issues in relation to AIDS. Currently, research on sexuality in Kenya tends to be linked to epidemiological, demographic and development studies that are part of AIDS research: essentially, sexual behaviour is studied in relation to HIV infection. For example, the issue of Culture, Health and Sexuality, entitled ‘Recent Research on Sexuality in East Africa’ and consisting of six articles, dealt only with sex and sexuality as they related to HIV/AIDS (Birungi et al. 2009). Most of this literature typically deals with assessing knowledge, behaviour and attitudes regarding sex(uality), HIV and AIDS, violent and/or risky sex, or the age of sexual début of adolescents, the frequency of sexual contacts, and condom use. These quantitative studies have been useful for detecting trends in behaviour in large samples of people or in regions.⁵

Together with the quantitative studies, qualitative studies were conducted to interpret these trends in sexual behaviour.⁶ These interpretative studies have provided much insight into different experiences, practices and justifications of sexual behaviour. However, the literature tends to present only a certain kind of knowledge about sexuality, since sexuality is only studied as a public health problem in a context where the urgency of the AIDS crisis underpins much research. Any approach which views sexuality primarily as a societal problem only highlights certain aspects of sexuality while ignoring others, such as the difference between sexual behaviour and sexual identity, love and intimacy, the desire for affection, or the aspect of pleasure.

Leaving to one side the frequently acknowledged fact that research on sexuality remains limited because sex is a sensitive subject, and hence difficult to study, little of this work has led to the development of an adequate research epistemology for studying sexuality in Kenya, or, for that matter, Africa as a whole. In fact, the terms sex and sexuality are often used interchangeably in sexuality-related research. These patterns reveal a flawed way of thinking about sexuality and AIDS.⁷ Although the scope of sexuality studies has broadened over recent years and much innovative work is now being done, I believe it is necessary to highlight certain trends that continue to impede the understanding of sexuality. The epidemiological or health approaches in the context of AIDS have framed knowledge about sexuality in African societies in a particular way.

Health approaches to sexuality

With epidemiological, demographic and development studies – in short, health studies – the quantification of sex into behavioural frequencies and attitudinal scores has characterized much of the research on sexual behaviour in Africa since the 1980s. While it is important to map patterns of HIV infection among a population, within a region or between regions, there is a need to further interpret this data. Epidemiological studies show patterns but do not explain the causes of these patterns, that is, they cannot explain the motivations of particular groups of people according to their social characteristics such as gender, religion, age, socio-economic situation, etc. Nevertheless, there is a tendency to treat epidemiological data as the end result. In many quantitative studies sex is reduced to an act done either before, during or outside marriage. Further, the term ‘sexuality’ can indicate a range of topics such as reproduction, circumcision, gender, HIV infection and condom use. This instrumental use of the term sexuality indicates a failure to understand different practices and mentalities. Moreover, the methodology of surveys or questionnaires is not always unproblematic with regard to obtaining valid data (see for a discussion Abramson et al. 1995; Hewett et al. 2003). Lastly, the large-scale approach of epidemiological work necessitates a certain amount of generalization and the creation of categories that might not always reflect social realities, such as the category ‘prostitute’ (Pheterson 1990).

Among the first to study sexuality in relation to AIDS were John Caldwell, Pat Caldwell and Pat Quiggin (1989a, 1989b). They studied sexuality in relation to AIDS in Nigeria, and Africa in general, from an epidemiological perspective from the late 1980s until the turn of the century. Their attempt to study sexuality in greater depth is commendable, although their analyses are debatable.⁸ Their work serves as an example of the dominant trend of sexuality research in Africa. Moreover, their work has been cited many times and has become a standard point of reference. In short, they have produced an ‘African sexuality thesis’, which is used in highly generalized ways to explain higher numbers of HIV infections throughout the continent. They argue that compared to the ‘Eurasian’ system, chastity in the ‘African’ context is little valued; that there is no ‘evidence’ supporting high levels of female pleasure; that the conjugal bond is ‘emotionally weak’; and that sex ‘many times’ is a commercial interaction. They also state that in conjunction with ‘the evidence that [historically] Africans neither placed aspects of sexual behaviour at the centre of their moral and social systems nor sanctified chastity’, African sexuality can be characterized as, ‘above all, permissive’ (Caldwell et al. 1989b). Despite their own caution (see note 5, p.78), it is disheartening to see such huge generalizations being made for a continent as vast as Africa, based only on a few case studies in one particular country and by means of surveys. This type of gross generalization caricatures and devalues complex and diverse emotions and practices. By writing about ‘African’ sexuality in relation to ‘Caucasian’ sexuality, they emphasize differences, rather than possible similarities, such as the relation between poverty or class and sexuality (Parker 2001) or people’s inclination to intimacy (cf. Jankowiak 2008). Their work reveals a narrow interpretation of female sexual pleasure, as well as an ethnocentric and ahistorical analysis of notions regarding morality, sexual behaviour, female sexuality and chastity (for further discussion, see Ahlberg 1994; Heald 1995; Le Blanc et al. 1991).⁹

This work serves as an evocative example of a larger trend in sexuality research in African societies. Its overall perspective presents us with three stereotypes that still impede much research. The first stereotype is the construction of a single entity called ‘Africa’ that does not leave much space for the many variations and diversities that exist on the African continent. The next is that of promiscuous ‘African men’ who have extensive sexual networks and behave irresponsibly. The third stereotype is the construction of the category ‘African woman’, which depicts ‘African women’ as ‘disempowered’ to the point that they ‘often’ have sex for material gain. Related to this latter notion is the idea of a ‘lack’ of female sexual pleasure in ‘the African sexual system’. What is lacking in work that relies on these stereotypes is a sound theorizing of sexuality and a coherent reflection on research epistemology.

When we look at the assets and limitations of epidemiological approaches to sexuality, an interesting contradiction arises. In order to show patterns of behaviour in a population, surveys focus on individual knowledge, attitudes, behaviour and practices. The failure to take into account the causes of these individual factors highlights an inability to consider the social dimensions of sexuality. In response, scholars have been looking at incorporating social aspects of sex in order to explain particular HIV patterns. Thornton, for example, criticizes the standard statistical medical-epidemiological approach of AIDS and sexuality research in Africa and proposes going beyond the individual by focussing on what he calls sexual networks (2008). He rightly points out that statistical aggregates of individual behaviour do not explain why HIV does not progress like most other viral and bacterial epidemics. Instead, Thornton suggests focussing on the issue of the formation of sexual networks, defined as the links between people by way of having sex with more than one steady lifetime partner (serial or simultaneous). However, his important contribution – including wider aspects of how individual shape their lives – has resulted in individual personal motivations and experiences, once again, being overlooked. In order to understand the abstracted tables, figures and mappings, we also need to study people’s personal and intimate considerations.

I certainly do not want to deny the problems of AIDS and the disadvantaged position of certain groups of women (and men) in Africa as a whole. However, taking into account that AIDS studies are conducted from a particular perspective (which see sex as the problem behind HIV infection), it is important to realize that research on sexuality from this perspective has emphasized only some aspects of sexual relationships, and that sexuality is often studied in relation to other topics. These themes have become the primary source of information on sexuality, which means that sexuality tends to be studied in instrumental terms. This results in a subtle process where the definition of women’s sexuality in African societies is not just complex but fundamentally flawed (Spronk 2005a); and that of men’s sexuality as not just complex but fundamentally coercive (Spronk 2005b). There is virtually no literature on pleasure, the relationship between affection and sex, eroticism, conjugal intimacy, or female pleasure. The difference between this presentation of African sexuality and the literature on Western societies is remarkable. Studies dealing with, for example, erotic pleasure or the meaning of intimacy are almost nonexistent for Africa. However, this is changing, albeit on a small scale (Biaya 2000; Spronk 2011).

Ideally, epidemiological and interpretative studies should complement each other. That this is not often the case is problematic, yet characteristic. Crossing disciplinary boundaries, as Besnier points out, is an arduous but important endeavour. An important step is the ‘appreciation of how scholars in other disciplines couch research questions, what kind of presuppositions underlie these questions, and what counts as evidence as they go about trying to answer them’ (1995: 559). My critical reflections about trends in the development/health research on sexuality should not be interpreted as dismissing the discipline. On the contrary; more effective ways to cooperate need to be explored, such as successful interdisciplinary projects. An interesting approach to the study of sexual patterns where quantitative and qualitative methods are combined is the Relationship History Calendar (Luke et al. 2011). However, there is a tendency to take HIV infection patterns as the end result, thereby excluding or minimalizing the interpretation of contextual factors. This is also due to the fact that research in Africa tends to be monopolized by development-related work that is policy driven and hence dependent on political trends and rapid-assessment approaches. Development projects around the world aimed at promoting population management, disease prevention, and maternal and child health, intentionally and unintentionally shape ideas about what constitutes ‘normal’ sexual practices and, as a result, generate very particular knowledge about sexuality (Adams and Pigg 2005).

The reasons behind these tendencies can be found in what Jansen has called the ‘guises of ethnocentrism’ in development discourse. The belief in and the creation of a systematic difference between the West and the South is one of these major guises. There is also a preference for abstractions over actual social relations, while the South is further perceived as consisting of static and unchanging cultures (Jansen 1989). ‘Africa’ is portrayed as a ‘paradigm of difference’ (Mudimbe 1994). Much of the Western/white discourse on black sexuality still has its roots in the ‘psychological need for projection of Otherness’ (Vaughan 1991: 19).

Sexuality research in the context of AIDS in Africa

When in 2001/2 I explained to Western AIDS or development professionals that in my research I was studying sexuality in the context of modernity in Nairobi, people often responded by saying things like ‘ah yes, condom use is such a problem in these cultures’, or ‘right, AIDS is really killing them off here isn’t it?’ Such remarks reveal two premises. Firstly, that research on sexuality in an African society is often automatically related to AIDS and is not viewed as a possible topic of study in its own right. Secondly, it reveals the widespread assumption that the problem of AIDS – that is, the high HIV infection rates due to sex – stems from the peculiarities of ‘African’ culture. Sometimes people were more explicit and named the problem of ‘African’ cultures. For example, an epidemiologist who was invited to speak about ‘AIDS in Africa’ at the Free University in Amsterdam in 2003 explained that ‘African men like to sow their seeds’ and thus have multi-partnered sex as ‘part of their culture’. Similarly, in 2001 the nongovernmental organization VSO (Voluntary Service Oversees) had the following slogan on a poster which depicted unspecified black people in the background: ‘They believe AIDS originates from witchcraft. We know better. Come and work at VSO to give them a better future.’ These are two random examples of situations one encounters in the context of development work, which point at the necessity for continuous efforts to reflect upon and integrate ethics in the development and research agenda (Nyambedha 2008).¹⁰

How exactly are studies of sexuality framed by the AIDS paradigm? According to Packard and Epstein (1991), the development of medical research on AIDS in Africa resembles earlier efforts to understand the epidemiology of tuberculosis and syphilis in Africa. In all three cases, early research focussed on the question of why these diseases exhibited different epidemiological patterns in Africa to those exhibited in the West. Early explanations of these differences focussed on the peculiarities of ‘African’ behaviour, while largely excluding the wide range of contextual factors. Packard and Epstein analysed how these initial perceptions shaped the subsequent development of AIDS research, encouraging a premature narrowing of research questions. As early as 1991, the authors warned that, as has happened in the research on tuberculosis and syphilis, this early narrowing down might generate inadequate and inappropriate responses to the AIDS epidemic and limit our understanding of the disease. However, their warning appeared to do little to reconfigure the general scope of AIDS research, probably because of the pressure to act immediately on the imminent crisis. But how did these presumptions originate in the first place?

According to Patton (1992), colonial constructions of black sexuality were revived in the attempt to explain the heterosexual character of AIDS. The idea of ‘Africans’ as being sexually promiscuous by nature became a discursive reality once again.¹¹ In other words, scientific knowledge is not objective or free from the influence of history and hierarchical social relations. The knowledge gained through research on AIDS and sexuality in Africa is far from complete (Arnfred 2004; Heald 2003; Undie and Benaya 2006; see Vaughan 1991 for a general discussion on constructions of race and sexuality in biomedical discourse in Africa).

In her insightful book on the politics of AIDS research in Kenya, Booth (2004) shows how these premises affect research. She analyses in detail how one of the most famous AIDS research projects in Africa was organized, ideologically and practically, and what kind of knowledge on HIV infections, women and men, and sexuality was generated. She highlights what was not asked, what was not thought about, or what was discarded as irrelevant and excluded from the research results. For example, a description of the socio-economic status of the deprived and unattached men, on whom the research focussed, was initially included in the research design. Later, however, this information was ignored when the behaviour of these men was explained.¹² Instead, the concept of ‘African men’ as a self-evident category of sexual agents was used to explain their behaviour. Such ways of thinking about culture figure prominently among Westerners and, ironically, also among Nairobians when discussing people with AIDS outside their own social group (Spronk 1999). In fact, what we may call the AIDS paradigm and the construction of African sexuality are enshrined in the development/health discourse, whose institutions are the main donors for sex research to date.¹³ The perception that geographical difference and promiscuity are central to the problem of AIDS in Africa ‘lies at the heart of the scheme that the World Health Organization [WHO] developed between 1986 and 1988 and that continues to organize international scientific and developmental thinking about the crisis’ (Booth 2004: 50).

From the 1980s onwards, such misconceptions have been seriously challenged by scholars working with computer models to understand HIV patterns, as well as by medical sociologists working in Africa, such as Maxine Ankrah. Nevertheless, the idea that Africans engaging in promiscuous, multiple-partnered sex during their life course is the explanation for HIV infection remains pervasive (see chapter 3 of Epstein 2008, for a stunning account of how scientific research and knowledge production can be severely limited by stereotypes and lack of epistemological reflexivity). During the first decade of the twenty-first century, the concept of ‘concurrent sexual partnerships’, which means the engagement in two or more sexual relationships at the same time, has become the dominant mode of understanding sexuality in Africa. According to its proponents, despite the fact that concurrent relationships are often equated with multiple-partnered relationships, these computer models show that, ironically, people in Europe and the US engage in sexual relations with more partners compared with people in Africa. The difference is that the former engage in serial monogamy, multiple partners successively, and that the latter are more likely to engage with fewer partners but two, and sometimes more, partners at the same time. These concurrent relationships are particular prone to HIV transmission because of HIV’s highly infectious period during the first three months after infection (see Epstein 2008; Thornton 2008). However, there is a huge controversy concerning the validity of the concept, especially regarding its methodological accurateness. The main problem is the lack of detailed estimates on prevalence and characteristic; as a result it is difficult to distinguish between concurrency and serial monogamy and, moreover, there is a tendency to overestimate concurrency levels (Lurie and Rosenthal 2010; Sawers and Stillwaggon 2010). The controversy shows, once again, that there is no shortcut to researching the complexity of sexual relations. Moreover, the idea that some form of risky behaviour peculiar to Africa, rather than, for example, economic or biological factors, is responsible for the spread of HIV, continues to guide research on sexuality and HIV/AIDS in Africa (Sawers and Stillwaggon 2010).

A related misconception has plagued much AIDS research. ‘Culture’, of course, has been held responsible where ‘cultural practices’, such as ‘wife inheritance’ in western Kenya, were seen as contributing to HIV infection. The over-representation of the practice of wife inheritance in the literature of the early 1990s is remarkable, since in fact it occurs in only a few of the many ethnic groups in Kenya, and within these groups it is practiced by a minority (Nyanzi et al. 2008). But because of the attention it received from Western researchers, it came to be seen as a typical ‘Kenyan’ practice responsible for the high rates of HIV infection, hence misrepresenting the nature of the epidemic. The same is true of the ‘sugar daddy’ stereotype (Luke 2005). Another effect of the ‘AIDS paradigm’ is the defensive reaction from people from Africa against such discriminating attitudes, as we have seen from Patrick’s account. The political correctness of development workers and scholars and the defensive attitudes of people from Africa (who, ironically, all invoke ‘culture’), have sometimes resulted in a stalemate, where the subject is avoided altogether.¹⁴

The current development/health discourse has thus framed sexuality research in African societies to a large extent. The representation of women’s and men’s sexuality in medical discourses on AIDS in Africa demonstrates fixed certain assumptions about people’s social and presumed exclusively heterosexual identity (Epprecht 2008), their risk status and their responsibility in AIDS prevention (Hunter 2010; Silberschmidt 2001), and their sexual lives. As a result, AIDS campaigns have missed the point by focussing solely on the incidence of particular attitudes and practices, instead of studying the various reasons why people engage in sex. When one reads the literature on Western societies, it becomes clear that people engage in sex for many reasons. To name a few: the pursuit of pleasure; a desire for intimacy; an expression of love and/or affection; erotic expression, definition of a gendered sense of self; procreation; domination; money; or any combination of these reasons. From the eight reasons listed here, only the last four are described in the literature on sexuality in Africa and only recently is the component of pleasure being taken up. We can conclude that for Africa, sexuality has long been understood in terms which do not allow for personal and erotic specification.

I have discussed major trends in the research on sexuality and its impressive expansion. After more than three decades of AIDS intervention, this is a good moment to reflect on research, policy and implementation, as many new initiatives testify. This is not to say that there have not been studies that challenge these trends through different approaches. Within the social sciences a growing literature is looking at sexual conduct in its social context, and many proposals within the development/health discourse are reflecting on more inclusive approaches to sexuality (such as providing ‘sexual counselling’ within AIDS projects, or including notions of ‘sexual pleasure’ in project outlines). The establishment of the Africa Regional Sexuality Resource Centre (ARSRC) in 2003 is a great initiative, aimed at giving visibility, depth and legitimacy to the field of sexuality. In short, there are two major challenges to research about sexuality in Africa. Instead of focussing on frequency of sex in the context of HIV infection, the focus should be readjusted to the sexual act in relation to intimacy, eroticism, psycho-social aspects, socio-economic status, etc. Second, a more adequate research epistemology is urgently needed; without adequate theoretical grounding and methodological sophistication, research will remain flawed. The challenge therefore is to reformulate sexuality research, taking into account how, on one hand, society organizes experience and structures desire and motivation, and how, on the other hand, livelihoods or lifestyles frame people’s sexual behaviour. To understand how this interaction takes place requires not only the mapping of explicit gendered power relations; it also requires the exploration of people’s motivations, justifications, and agency. This study is therefore positioned within current debates in the social sciences about the social and cultural analysis of sexuality.

The dynamics of sexuality: the focus of this study

Despite Bronislaw Malinowski’s call ‘[T]hat which means supreme happiness to the individual [the erotic phase] must be made a fundamental factor in the scientific treatment of human society’ (Malinowski 1982 [1929]: 6), it is interesting to note that the study of sexuality has a tendency to ignore the erotic pleasures of sexuality, in favour of studying gender, identity, kinship, or reproduction. In this part I would like to address the question why anthropologists tend to ignore, in the words of E.E. Evans-Pritchard, ‘the flesh and blood’ in their writings on sexuality (1974: 9).

Anthropology and sexuality

As Andrew and Harriet Lyons point out in their book on the history of anthropology and sexuality, it is curious that sexuality has rarely been, and is still hardly a dominant theme in ethnographic research despite strong interest in the topic on the part of some of anthropology’s founding practitioners and their descendents (2004: 2). In the 1920s and 1930s, anthropological studies emphasized the cultural variability of sexuality. Malinowski’s work on the Trobriand Islands and Mead’s work on Samoa showed that the sexual lives of primitive societies were quite variable and both scholars used these social facts as a mirror for their own societies to highlight the cultural nature of marital sexuality (Malinowski 1982 [1929]) and adolescent sexuality (Mead 1950). Malinowski provided an interesting research programme into sexuality as one of the key sites of social life whereby to study culture. Unfortunately, his legacy lay dormant until much later. It is generally understood that during the Great Depression and the two decades following the Second World War there was a relative silence among anthropologists concerning sex (Lyons and Lyons 2004; Tuzin 1995). The structural-functionalist orientation introduced an approach which looked at sexuality in relation to social structures. Moreover, the topic was decentralized in the sense that sex and gender were reconceptualized by the more disembodied terms of marriage, family and social structure. E.E. Evans-Pritchard’s work on Zande homosexual and heterosexual eroticism in the 1920s was not published until 1974, and he suggested that his generation of anthropologists may have ‘lost the flesh and blood’ in their writings about African societies (1974: 9).

In the post-war years the influence of Levi-Strauss redirected anthropology’s attention towards rituals and cosmology. The study of sexuality came to be limited to sexual symbolism, that is, how symbols of sex and the body referred to plants, colours, and social groups. In the 1950s and 1960s a differentiation between British and US anthropology took place. The theoretical direction in anthropology in Britain had for some time ruled out any consideration of individual motivation and bodily processes, whereas US anthropology was more open to such considerations. Nevertheless, it was notable that even the neo-Freudian culture and personality movement in the United States produced few ethnographic descriptions of adult sexual behaviour. The culture and personality studies approach focussed on the daily practices of socialization, hence on the social institutions of sexuality such as initiation rituals, marriage, etc. (Benedict 1934).

Since the 1960s, sexuality has reemerged as a focus in anthropological theorizing because of pioneering work on sexuality through feminist and gay and lesbian studies. So far, the study of sexuality had been the domain of sexology and other biomedical disciplines, apart from the above mentioned studies, whereas the new approaches are generally recognized as studying the social construction of sexuality. Feminist efforts focussed on a critical review of theories which used reproduction to link gender and sexuality, thereby explaining the inevitability and naturalness of women’s subordination (see Vance 1991: 876). This theoretical reexamination separated sexuality and gender into two different concepts. A second impetus arose from issues that emerged through the examination of male homosexuality in nineteenth-century Europe and North America. The emerging gay and lesbian scholarship distinguished homosexual behaviour and homosexual identity and found the latter to be a recent historical development (Katz 1976; Weeks 1977). As such, this body of work introduced the concept of the relation between sexuality and identity. These two approaches – studying the politics of sexuality regarding gender and identity – have since remained the dominant themes in sexuality