This art icle was downloaded by: [ Queen Mary, Universit y of London] On: 16 August 2011, At : 11: 13 Publisher: Rout ledge I nform a Lt d Regist ered in England and Wales Regist ered Num ber: 1072954 Regist ered office: Mort im er House, 37- 41 Mort im er St reet , London W1T 3JH, UK Gender, Place & Culture Publicat ion det ails, including inst ruct ions f or aut hors and subscript ion inf ormat ion: ht t p: / / www. t andf online. com/ loi/ cgpc20 Endangered Youth? Youth, gender and sexualities in urban Botswana a CATHY McILWAINE & KAVITA DATTA a a Queen Mary, Universit y of London, UK Available online: 20 Jul 2011 To cite this article: CATHY McILWAINE & KAVITA DATTA (2004): Endangered Yout h? Yout h, gender and sexualit ies in urban Bot swana, Gender, Place & Cult ure, 11: 4, 483-512 To link to this article: ht t p: / / dx. doi. org/ 10. 1080/ 0966369042000307979 PLEASE SCROLL DOWN FOR ARTI CLE Full t erm s and condit ions of use: ht t p: / / www.t andfonline.com / page/ t erm s- andcondit ions This art icle m ay be used for research, t eaching and privat e st udy purposes. Any subst ant ial or syst em at ic reproduct ion, re- dist ribut ion, re- selling, loan, sub- licensing, syst em at ic supply or dist ribut ion in any form t o anyone is expressly forbidden. The publisher does not give any warrant y express or im plied or m ake any represent at ion t hat t he cont ent s will be com plet e or accurat e or up t o dat e. The accuracy of any inst ruct ions, form ulae and drug doses should be independent ly verified wit h prim ary sources. The publisher shall not be liable for any loss, act ions, claim s, proceedings, dem and or cost s or dam ages what soever or howsoever caused arising direct ly or indirect ly in connect ion wit h or arising out of t he use of t his m at erial. Gender, Place and Culture Vol. 11, No. 4, December 2004 Endangered Youth? Youth, gender and sexualities in urban Botswana Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 CATHY McILWAINE AND KAVITA DATTA Queen Mary, University of London, UK ABSTRACT Age is now recognised as a signi®cant social cleavage in research on youth in the South. Using participatory urban appraisal methodologies, this article explores constructions of sexualities among urban youth in Botswana, a country that is currently experiencing an HIV/AIDS epidemic and high levels of teenage pregnancy. We argue that not only are young people sophisticated sexual beings, but that there is a need to adopt more holistic approaches to examining sexualities among them so as to appreciate that constructions of sexualities are multi-faceted, highly diverse and heavily gendered. This appreciation must then be integrated into a multi-sectoral policy approach that moves beyond information provision towards one that addresses changes in gender, cultural and sexual identities. Introduction Age and generation are recognised increasingly as critically important axes of social differentiation within feminist research in the South alongside others such as gender and `race' (Katz & Monk, 1993; Datta,1996; Chant & McIlwaine, 1998). As well as research on elderly people (Clark & Laurie, 2000; Varley & Blasco, 2000), this has led to greater awareness of the experiences of young people, and especially the diversity of youth from a gender perspective (Johnson et al., 1998). Furthermore, attention is turning increasingly to issues of sexualities in relation to young people in the South (Jolly, 2000). In part, this focus is attributable to a growing global youth population and the fact that young people face particular challenges such as poverty, unemployment and lack of access to education (ICRW, 2000). Interlinked with these, and perhaps the most signi®cant challenge facing young women and men today, is that of sex and sexualities. While sexual activity among youth has been identi®ed as a cause for concern for some time due to its social and reproductive health impacts, particularly among women, the impact of the HIV/AIDS crisis on youth has made it even more imperative to understand the intersections between youth, gender and sexualities. Nearly half of all new cases of HIV/AIDS worldwide occur Dr Cathy McIlwaine, Department of Geography, Queen Mary, University of London, Mile End Road, London E1 4NS, UK. Tel: 020 7882 5415; E-mail: c.j.mcilwaine@qmul.ac.uk ISSN 0966±369X print/ISSN 1360±0524 online/04/040483-30 ã 2004 Taylor & Francis Ltd DOI: 10.1080/0966369042000307979 Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 484 C. McIlwaine & K. Datta among young people between the ages of 15±24 years with an estimated 6000 young people contracting the disease every day (ibid.). The gendered nature of the crisis is also clearly evident, with women being particularly vulnerable (Tallis, 2002) and constituting the majority of the 11.8 million 15±25-year-olds living with HIV/AIDS throughout the world (Baylies & Bujra, 2000; ICRW, 2000; Baylies, 2002). In turn, this is primarily linked with unequal gender relations which make young women especially vulnerable to coercive sex and afford them little space to negotiate the nature of sexual relations especially in terms of condom use (Rivers & Aggleton, 1999). While somewhat eclipsed by the HIV/AIDS crisis, the issue of teenage pregnancy also remains a key issue in the realm of health and sexualities among young women in the South. It is estimated that one in six births in the developing world are to young women aged 15±19 years (ICRW, 2000). The impacts of the HIV/AIDS crisis in the South are so severe that the epidemic is now recognised as a key developmental issue in terms of both its causes and effects (Barnett, 2002). This bears particular relevance to Botswana which is now recognized as being at the epicentre of the HIV/AIDS epidemic in Southern Africa and having the highest proportion of youth living with HIV/AIDS in the world (UNDP, 2000). While still in its infancy, a broader developmental perspective towards HIV/AIDS is leading to the adoption of more holistic approaches to combating the disease which combine an epidemiological approach with an appreciation of how sexualities are socially constructed and performed (Kinsman et al., 2000). This has prompted much more interchange between health-orientated research and anthropological studies in relation to sexualities, constructions of masculinities and femininities and HIV/AIDS (Chant & Guttman, 2000; Maharaj, 2001). However, there remain few detailed studies on the importance of sexualities in the South, especially in relation to young people. In light of this, the article has two broad aims: ®rst, to explore the construction of sexualities among young people, and secondly, to examine the awareness of the causes and the effects of HIV/AIDS and teenage pregnancy among young people in Botswana1. We argue that young people in urban Botswana are sophisticated sexual beings, and that the construction of their identities is predicated primarily on sex and sexualities due to the overwhelming importance of HIV/AIDS and teenage pregnancies in their lives. In turn, we highlight how they have extensive knowledge of these phenomena. We also show how these sexualities are highly diverse, in¯uenced variously by traditional cultural mores, but also contemporary development processes. In addition, this challenges the stereotyped notions of a so-called `African sexuality' that assumes a monolithic construction across the continent (Caldwell & Caldwell, 1987). The discussion draws upon empirical research derived from a pilot study with young women and men in the capital city, Gaborone using participatory urban appraisal (PUA) methodologies2. This approach, although ¯awed, ensured the active participation of young people in the research process, while encouraging them to identify concerns which most preoccupied them3. Thus, the article recognises the centrality of sexualities in Botswana as well as the importance of youth as critical actors in the development of their country. Conceptualising Youth, Gender and Sexualities There is a growing consensus that youth, gender and sexualities are all socially Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 485 constructed and negotiated, a process referred to as social constructionism (Giddens, 1992; Nye, 1999). Indeed, it is dif®cult to disentangle the three given that sexualities are a central resource in the construction of youth and gender identities while sexualities are themselves fundamentally shaped by age and gendered power relations (Matikanya, 1993). The ensuing discussion seeks to elaborate upon these intersections. Research on `youth geographies' has emerged in the last two decades and geographers have been at the forefront of examining young people's understandings and sense of place (Aitken, 1994; Skelton & Valentine, 1998; Matthews & Limb, 1999; Holloway & Valentine, 2000). The social constructionist approach to youth has obviously facilitated such research, given its appreciation of the variations in conceptualisations of age and generation over space and time: in short, it acknowledges that context is fundamental in the shaping of youth identities (Ansell, 2002). Thus, while there may be growing agreement on the existence of a `global youth culture', there is also an understanding that youth identities are shaped fundamentally by a range of local factors including family structures, education and labour markets (Valentine, 2000; Punch, 2002). Indeed, identity construction is a process, a performance and provisional, illustrating a contextual and relational positioning rather than ®xed essence (Hall, 1992 cited in Dwyer, 2000; Bondi, 1993 cited in Ansell, 2002). Moreover, the social constructionist approach has enabled the theorisation of age and generation as a social dynamic (West, 1999). Valentine (2000) identi®es two processes as being critical in shaping the construction of youth identities: familialisation and individualisation. On one hand, familialisation conceptualises children and youth as dependants, in need of protection, vulnerable, asexual and irresponsible (see also James et al., 1998). As such, children and youth are rigidly compartmentalised, de®ned in opposition to adults and seen as being in need of adult protection. On the other hand, processes of individualisation, which are derived from conceptualisations of rights and entitlements, allow children and youth to break away from hegemonic and dominant constructions of what it is to be a child or a youth. In so doing, youth are able to `produce their own narrative of self¼ it is a relational processÐin which young people are often strung out between competing de®nitions of their ``identity'' emanating from home, from school and from wider society' (Valentine, 2000, p. 258). Peer group culture emerges as an important force in shaping these alternative identities as it enables children and youth to manage the tensions between conformity and individuality. Transitions from childhood to youth to adulthood are also socially constructed. While research in the North has posited traditionally only two extreme positions, dependence and independence, an in-between position of interdependence appears to be particularly important in the context of the developing world given the vital economic contributions children and youth may be making to the household (Punch, 2002). Research suggests that in the South, young people achieve economic independence earlier than their Northern counterparts while long-term family interdependence is maintained throughout the lifetime (ibid.). However, like youth in the North, the transition from childhood to adulthood is partial, inconsistent and contradictory as young people may engage in some activities which are seen as the preserve of adults (such as employment and sex) while at the same time being dependent on their parents for meals, money and clothing (Valentine, 2000). Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 486 C. McIlwaine & K. Datta The conceptualisation of age as a social process also allows for a greater appreciation of how generation intersects with other social axes such as gender. Yet, while feminist geographers have been instrumental in raising awareness of gender as a signi®cant social cleavage and its intersections with race, ethnicity and class, they have been somewhat slower to appreciate the signi®cance of age (Monk & Katz, 1993). Indeed, some would argue that gender has tended to eclipse age as a social cleavage which is attributable in part to the commonalities and links between women and children/youth (Roche, 1999). Over the last decade, feminist research has particularly focused upon the issue of gender identities and the construction of masculinities and femininities (Laurie et al., 1999, McIlwaine & Datta, 2003). This body of work illustrates the extent to which prevailing gender ideologies in¯uence the construction of masculine and feminine identities through all stages of the life course. Thus, male and female transitions from childhood to youth to adulthood are distinguished by different biological and social markers; young women's transition into adulthood may come later than young men's and, while for some men, youth is a stage of transitory powerless, for women it may be more permanent (Punch, 2002). A ®nal and critical element in the shaping of both youth and gender identities is sexualities. It is within this realm, for example, that the work on the constructions of masculinities has gained widespread acceptance (Datta, 2004). Sexualities can be de®ned broadly as the range of behaviour associated with the ideals, desires, practices and identities linked with sex (Chant & Craske, 2003). Associated with the work of Foucault (1978) on discourses of sexuality as fundamental to the control and regulation of populations, conceptual research has explored how different sexualities are constantly reproduced through discourses played out on bodies (Pringle, 1999). Again, the social constructionist approach has been useful as it has engendered a wider understanding of sexualities such that they are no longer simply equated with the sexual actions of speci®c bodies but rather to understanding the cultural, social and economic contexts within which they occur (Matikanya, 2003). However, this theoretical research, especially on homosexuality, is primarily Western in orientation (Bell & Valentine 1995). In contrast, research in the South, and more speci®cally in Southern Africa, has tended to adopt a narrower perspective where sexualities are predominantly linked with population, fertility and reproductive health issues; on sex rather than sexualities (Jolly, 2000, 2003; Potts & Marks, 2001). There is a global reluctance to confer youth the same sexual rights that adults enjoy despite the fact that the majority of them are sexually experienced by the age of 20 with premarital sex being common among young people aged 15±19 years (Rivers & Aggleton, 1999). Yet, within the realm of sexualities, youth remain positioned as children with a limited appreciation of the `dif®culties they face in articulating their concerns and in achieving their sexual identities' (West, 1999, p. 526). This can be attributed to negative perceptions of uncontrolled desire among youth, the possibility of homoerotic choices and images of hedonistic and risk taking youth (ibid.). Such perceptions of `unknowledgeable or ill informed adolescents' and `high-risk adolescents' are rife in the literature on youth and HIV/AIDS (Aggleton & Warwick, 1997). Indeed, such is the fear of youth sexualities that in the past, sex education messages have been largely restricted to promulgating the abstinence message, with parents throughout the world also failing to communicate with their children on sexual matters in the belief that a Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 487 denial of information will lead to a delay in sexual initiation (Rivers & Aggleton, 1999). Perhaps ironically, then, it is this con¯ation of sexual activity with adulthood status that often in¯uences young people, especially young men, to engage in sex in order to address intergenerational inequities and to enhance their social status (Rivers & Aggleton, 1999; also Ansell 2003 on Lesotho and Zimbabwe). At the same time, unequal power relations between adults and young people may place youth at particular risk given the incidence of practices such as intergenerational sexual relations. Research illustrates that both young men, but particularly young women, have very little control over sexual relations in such contexts (see below). Other factors affecting the constructions of sexualities among young people include group-based identities or, more speci®cally, peer identities. As Valentine (2000) argues, youth have to position themselves in relation to both adult and peer cultures. Peer group identities are heavily embodied and predicated upon adult notions of heterosexualised gendered identities (ibid.). Given the in¯uence of peers in providing information about sex (in the context of a lack of intergenerational communication about sex) and in shaping sexual identities, it is not surprising that peer education has emerged as one of the most commonly used strategies of HIV/AIDS prevention worldwide. Where research on youth and sexualities has coalesced the principal focus has been on gender and how this structures sexualities among youth, leading some researchers to argue that gender has again eclipsed the signi®cance of age itself in determining the sexual rights of youth (West, 1999). Gender as a marker of youth identities is shaped around ideas of adult heterosexual notions of desirability and morality (Valentine, 2000), and there is increased consensus that gender relations have a signi®cant impact upon sexual and reproductive health behaviours and outcomes (Rivers & Aggleton, 2002). Prevailing gender roles and relations usually mean that young women have less control than young men over sexual relations. For example, men usually retain control of decisions relating to sexual activity in most societies including the frequency and use of contraception. Thus, unequal power relations between men and women are especially critical in understanding unsafe sexual practices (Matikanya, 2003). Age in particular serves as a barrier to the passing of sexual knowledge to young women given the pressures on them to be chaste. Indeed, even where women may possess sexual knowledge, it is often in their best interests to appear to be `innocent' and uninformed. Furthermore, female sexual agency is often denied in that young women are not expected to initiate or actively participate in sex but rather to let it happen to them (Rivers & Aggleton, 1999). The converse is generally true of young men, in that they are both expected to possess greater sexual knowledge and experience (although this is not necessarily true in reality). Pressures on young men to engage in early and multiple sexual relations emanates both from peer groups and older male family members. Research also illustrates that homophobic bullying is often used to push young men into early heterosexual relations (Rivers & Aggleton, 2002). Yet, the importance of including young men in sexual and reproductive health programmes has been particularly highlighted in recent research, which suggests that they are more likely to be willing to discuss sexual matters, participate in sex education and be open to considering alternative sexual practices than older men, as unsafe sexual practices are less entrenched (ibid.). 488 C. McIlwaine & K. Datta Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 The Context in Botswana: youth, gender, sexualities and HIV/AIDS Despite research which suggests that youth is socially constructed, biological age remains the most common proxy measure of stage in the life course as re¯ected in most published statistics on youth. This is evident in the age-based schema identi®ed in the Government of Botswana's National Youth Policy, which de®nes youth as those aged between 12 and 29 years (Ministry of Labour and Home Affairs, 1996a). While we adopted this country-speci®c de®nition of youth, it is not surprising that it was problematic on a number of grounds illustrating the inadequacy of relying upon age delimitations and assumptions about youth in general (Burke, 2000). There is considerable overlap between terms such as child, youth and adult within Botswana. For example, `children' are de®ned as individuals under 18 years of age. At the other end of the spectrum, one can argue that the formal classi®cation of youth until they are 29 years old delays the transition from youth to adulthood, and extends the social control of youth. In turn, this hampers their ability to claim the social, economic and political rights reserved for adults4. This of®cial de®nition of youth is not accepted universally within the country as evidenced by our focus groups discussions which revealed very ¯uid categorisations. Participants referred to themselves as children (as opposed to youth) in some contexts, such as when they were discussing child abuse, and as youth (as opposed to adults) in others. The latter was borne out in our discussions with `outof-school' young men and women who were all unemployed, and despite being aged 25 and 26 years, referred to themselves as youth. This re¯ects a traditional association of adulthood with economic independence (especially among men), the ability to support a family, as well as a host of subjective criteria such as responsibility, appearance and behaviour (see also Ansell, 2003 on Lesotho and Zimbabwe). Therefore, we use this categorisation with care, and with little other option. Notwithstanding debates surrounding the de®nition of youth, youth issues have received little academic attention in Botswana despite the fact that people under the age of 30 constitute 70% of the country's population, with those aged under 15 years making up 40% of the population (Government of Botswana/ UNDP, 1998). Yet, youth remain a high pro®le `issue' in the public realm due both to expectations of young people as future leaders and negative stereotypes of them as exhibiting `high-risk' behaviours among which early and underage sex are emphasised. It is largely within the context of reactionary gendered research on changing household, family and community structures that a limited body of work on youth has grown around two themes. These are ®rst, the importance of age and gender segregation in traditional Tswana communities. Largely anthropological and rural in nature, this research documents the historical importance of age regiments as social institutions that separated young men and women both from each other and their elders (Schapera, 1940). There was a clear demarcation of the responsibilities and duties applicable to each age and gender group. While age regiments (age cohorts) are all but extinct now (especially for urban youth), age and gender remain culturally signi®cant cleavages with the traditional veneration of the elderly (Mompati & Prinsen, 2000). This research clearly illustrates the negative manner in which young women and men continue to be de®ned in terms of qualities or attributes they do not possess such as being unmarried, childless Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 489 and/or without independence and resources. In particular, there is a widely held belief that youth lack access to `knowledge' (distinct from information gained through formal education) which only adults possess and pass on to young people when they are deemed to be on the threshold of adulthood (Burke, 2000). A second, more extensive, body of research exists on gender and sexualities which also encompasses important youth dimensions. This has developed within the context of work on `African sexuality' that is based on perceptions of permissive sexual practices and a lack of shame, particularly among women (Caldwell & Caldwell, 1987). Female sexual agency, in particular, is attributed to the absence of moral and institutional constraints. Although some suggest that the concept of an `African sexuality' has remained intact despite the sexually repressive in¯uence of westernisation (Caldwell et al., 1989), others are beginning to question it (Helle-Valle, 1999). Criticisms lie primarily in the dangers of generalising across different cultures and contexts, as well as making assumptions about predominant heterosexuality5. Research points to the existence of a diverse picture of African sexualities ranging from puritanical to more permissive systems as evidenced by the different positions adopted by Southern African states regarding policies on sexualities. Thus, while post-apartheid South Africa has fairly liberal and progressive formal approaches to gender and sexualities including a pioneering gay rights clause (Cock, 2003), in neighbouring Zimbabwe the parliament passed a new Sexual Offences Bill in 2001 which sought to curb sex work and punish brothel-keepers harshly while also criminalising same-sex relationships (Wojcicki, 2002). Furthermore, while it is often assumed that early sexual experience, as well as a general acceptance of premarital childbearing among women is a common characteristic of Southern Africa, there are no uniform patterns. Recent trends highlight how Lesotho and Zimbabwe have low rates of fertility among adolescent never married women with children, compared with more intermediate rates in South Africa and Namibia and extremely high levels in Botswana (Mturi & Moerane, 2001). The reasons for these differences are complex, but many of them can be traced to variations in traditional cultures and mores and how these have changed in response to westernisation and related processes of globalisation. Finally, the `African sexuality' thesis has been criticised for promoting perceptions of promiscuity which has been particularly harmful to HIV/AIDS prevention campaigns (Ahlberg, 1994). As with research on youth, most work on sexualities and gender in Botswana has focused on anthropological studies in rural areas concentrating on traditional Tswana cultural and social organisations. A range of factors has been identi®ed as in¯uencing the constructions of sexual identities among youth of which early and premarital sexual relations are paramount. It is estimated, for example, that almost half of unmarried young Batswana women aged between 15 and 17 are sexually experienced with 12% having given birth. By the age of 24, 98% are sexually experienced with four of ®ve unmarried women having given birth (Mturi & Moerane, 2001, p. 274). Studies have explained patterns of early and premarital sexual relations and childbearing within the context of changing family structures and especially high levels of female household headship in rural areas (O'Laughlin, 1998). It is often suggested that traditional marriage patterns account partly for premarital sexual relations in that marriage in Tswana society was a process extending over a fairly long period, beginning with the decision to marry and culminating in the Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 490 C. McIlwaine & K. Datta payment of bogadi (bridewealth) when the woman could move into the man's family home (Schapera, 1940; Molokomme, 1996). However, even before the bogadi was paid, a period of living together (ralala) took place and this stage would usually last until a child was born (van Driel, 1996). Therefore, premarital sexual relations were sanctioned culturally, albeit within the broad process of marriage (MacDonald, 1996; Townsend, 1997). This link between premarital sex and marriage was broken in the early part of the 19th century by the labour migration of men to South African mines, which led to more ¯exible and early sexual relations outside the context of marriage (Schapera, 1940; Datta & McIlwaine, 2000)6. While the importance of marriage has declined steadily (in part contributing to signi®cant levels of female headship, much of which is de facto and linked with labour migration)7, the high social status conferred on women with children in Tswana culture, coupled with social opprobrium faced by women who are infertile has continued (Upton, 2001). This has particular implications for teenagers who may, as they get older, come under parental and/or peer pressure to bear children irrespective of their marital status (UNDP, 2000). There is also a popular belief among younger women that sexual intercourse during their teenage years enhances their fertility and chances of conceiving in later life (ibid.). Furthermore, birth outside of marriage continues to be a statistical norm and single mothers do not damage their future chances of getting married by virtue of being mothers (Suggs, 1987). Early, and unsafe, sexual initiation may also be rooted in the disappearance of traditional socialisation practices whereby boys and girls were taught responsible sexual behaviour through initiation ceremonies (called bogwera for boys and bojale for girls) (MacDonald, 1996). Taking place three to ®ve years after a child had reached puberty, these ceremonies prepared youth for adulthood. While the period of adolescence among young women (referred to as lekgaribe meaning `young women' or `carefree older girl') was seen as one of sexual initiation and experimentation, such ceremonies also taught responsible sexual behaviour, selecting a partner and marital duties (Suggs, 1987; Meekers & Ahmed, 1999). The demise of these practices is regretted by some on the grounds that it has marked the beginning of a lack of intergenerational communication on sexual matters and a lack of `moral guidance' for the young (UNDP, 2000). At the same time, the fact that sexual intercourse remains symbolic of reaching adulthood explains the relatively high rates of sexual activity among young people. Finally, some researchers suggest that the construction of sexual identities among youth is also in¯uenced by high levels of female headship and weakened social control over the domestic arena, with mothers often having lovers in the absence of children's fathers which leads to young people beginning sexual relations at an early age (Gaisie, 1998, cited in Mturi & Moerane, 2001, p. 272). Linked with these patterns is the predominance of a transactional attitude towards sexualities (Helle-Valle, 1999), also noted throughout sub-Saharan Africa (Caldwell et al., 1989; Baylies, 2002). Partly related to the practice of bridewealth, a key manifestation of this is the phenomenon of `sugar daddies' and, to a lesser extent, `sugar mummies'. In the case of the former, men, who are often well into middle age or older, have sexual relations with much younger women, often in exchange for money, gifts and/or promises of commitment (Leshabari & Kaaya, 1997). Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 491 However, more recently, the intersections between youth, gender and sexualities within Botswana has heightened attributable to two outcomes of high-risk sexual behaviour: HIV/AIDS and teenage pregnancy. Botswana not only has the highest proportion of young people with HIV/AIDS, but also the highest recorded rate of HIV prevalence in the world (UNDP, 2000). Late to acknowledge the HIV/AIDS epidemic, the ®rst of®cial recognition of the disease came in 1985 (MacDonald, 1996). By 1999, adult prevalence was estimated at 35.8% of the population (Kumaranayake & Watts, 2001, p. 453), although the government suggests the ®gure is closer to 28 or 29% (Ministry of Finance and Development Planning, 2000, p. 4). This rate has tripled since 1992, when prevalence was 10% (Poku, 2001, p. 194). Furthermore, differences in HIV/AIDS prevalence rates between urban, peri-urban and rural areas are marginal (MacDonald, 1996). The gender and age dimensions of this epidemic are also clearly evident. By the ®rst quarter of 1996, no less than 68% of reported HIV infection cases were among young women aged 15±29 years (Government of Botswana/UNDP, 1998). Moreover, prevalence rates among young women (34.3%) were signi®cantly higher than among young men (15.8%) (UNDP, 2000; Kumaranayake & Watts, 2001, p. 453). Another particularly affected subsection of the population is children under the age of ®ve, with an estimated 64% of deaths being attributed to AIDS (Ministry of Finance and Development Planning, 2000, p. 4). With such alarmingly high levels, the need to address the situation is critical. Indeed, youth under the age of 15 have been particularly identi®ed as presenting hope for the future (the AIDS-free generation of 2016) given low levels of prevalence (2%) among this age cohort (UNDP, 2000, p. 37). While the perceived reasons for the spread of HIV/AIDS among urban youth are discussed below, there are other factors which contextualise the incredibly high prevalence rates in the country. Until recently, there was widespread denial that HIV/AIDS existed in the country8. This was attributable, in part, to the fact that people refused to believe that certain diseases such as tuberculosis and malaria were related to AIDS (Helle-Valle, 1999), believing instead that AIDS related illnesses were boswagadi (or a disease caused by breaking cultural taboos such as having sex with widows) (MacDonald, 1996). Moreover, the relatively good nutritional status of the population meant that the AIDS epidemic was slower to take hold here than elsewhere on the African continent (interview with representative of the Botswana Network of Aids Services, July 2001). Migration and a good transport system have also facilitated the spread of the disease (MacDonald, 1996; UNDP, 2000). Botswana has been categorised as having one of the most mobile populations in the world. While migration to regional destinations such as South African mines has declined in the postindependence period, rural to rural and rural to urban migration remains important (MacDonald, 1996). Young people are active participants in these migration ¯ows as residential patterns, divisions of labour and the provision of schooling in rural areas necessitates a separation of children, youth and adults across villages, lands and cattle-posts. Moreover, youth also participate actively in rural to urban migration linked with the greater availability of higher levels of education and job opportunities in the towns, as well as being called upon by urban kin to help rear urban-based children (Datta, 1996). The latter especially applies to young women (Datta & McIlwaine, 2000). The separation of young Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 492 C. McIlwaine & K. Datta people from their parents, whether in rural or urban areas, is often cited as one of the reasons for high levels of sexual activity among youth. Arguably, perhaps the most signi®cant factor in the spread of HIV/AIDS is the ways in which sexualities are constructed in Botswana. As the discussion above has illustrated, sexual initiation is intimately related to transitions in life stages. Not only are youth sexualities constructed in a context whereby there is considerable peer pressure to engage in early and premarital sex, having multiple sexual partners is also an important marker of heterosexual masculinities (Datta, 2004). Furthermore, inequitable gender and age relations often mean that young women are at a particular disadvantage in terms of controlling the context in which sex occurs. The prevalence of transactional sex and sexual relations between older men and young women often results in the practice of unsafe sex. Researchers have highlighted the fact that traditionally women are not allowed to deny their husbands the right to have sex, and that men are deemed as being within their right to beat their wives into submitting to having sex (MacDonald, 1996). Notwithstanding the spread of the HIV/AIDS epidemic, there has been little corresponding change in peoples' sexual behaviour (Helle-Valle, 1999; see below). Although somewhat eclipsed by the current HIV/AIDS epidemic, teenage pregnancy has long been identi®ed as a key problem among young people in Botswana (Ministry of Labour and Home Affairs, 1996a). It is estimated that 24% of youth in Botswana are mothers, with health statistics showing that annually 23% of ®rst-time antenatal care clinic attendees are teenagers (Government of Botswana/UNDP, 1998, p. 21)9. High levels of teenage pregnancy are attributable to a range of factors, many of which coincide with those affecting HIV/AIDS. These include, ®rst, a low level of contraceptive use among teenagers despite the existence of a fairly liberal family planning policy with no age restriction on access to contraception services. It is now recognised that the attitudes of health workers towards youth engaging in sex serve as a barrier to access to reproductive programmes and services by teenagers and men of all age groups (Government of Botswana/UNDP, 1998). Secondly, as discussed above, underage and unprotected sexual relations are attributed to either cultural traditions which sanction early sexual initiation or some transformation in these traditions. Health sector workers and researchers have particularly highlighted the health risks that pregnancy poses for young women, such as the contraction of sexually transmitted diseases of which HIV/AIDS is the most recent and deadliest. Moreover, where the cause of death is speci®ed, abortion emerges as one of the top causes of morbidity in the country. Given that access to abortion is all but illegal, it is estimated that most of these deaths (4642 in 1994) are the result of unsafe operations (Government of Botswana/UNDP, 1998)10. Apart from health problems, teenage pregnancy obviously has wider impacts upon the life chances of young women, as it is frequently linked with dropping out of school. Although Botswana can boast of relatively high levels of youth attending school (see below), the vast majority of drop-outs are pregnancy-related; 1993 data indicate that 60% of all secondary school drop-outs (81 and 88%, respectively, of all withdrawals in Forms 4 and 5) among females are caused by pregnancy (Meekers & Ahmed, 1999; UNDP, 2000). Indeed, schools feature quite heavily in discussions of teenage pregnancy, as it is here that many young people have their ®rst sexual encounter11. For example, the National Women in Development Policy highlights the vulnerability of young women to sexual harassment and coercion in schools Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 493 both by their peers and teachers (Ministry of Labour and Home Affairs, 1996b). In turn, high rates of school dropouts obviously have long-term rami®cations for young women in terms of curtailing their education and their subsequent ability to secure formal sector employment. While many young women themselves decide not to return to school after the birth of their child, this is made even more dif®cult by a government policy which requires pregnant students to withdraw from school for a minimum period of a year after which they may resume their education but only in another school (Meekers & Ahmed, 1999). This discussion has illustrated the extent to which research on youth, gender and sexualities suggests fairly generalised assumptions that sexualities and marriage in Botswana are separate social constructions and institutions and that extramarital and early sexual relations are largely condoned by society. This has serious implications for HIV/AIDS prevention and teenage pregnancy reduction initiatives. As with the concept of `African sexuality', however, we argue that the idea of a `Botswana sexuality' is also dangerous and unfounded (Helle-Valle, 1999). Not only is promiscuity, especially among women, condemned publicly in Botswana in both rural and urban areas (thus challenging the stereotype of a permissive society), but constructions of sexual identities are much too diverse to warrant such as rei®ed notion. Instead, sexualities vary among young people according to gender, as well as age and generation which we explore further below. Researching Youth Using Participatory Urban Appraisal (PUA) Methodologies Conducting research with young people is fraught with practical and ethical pitfalls. However, particular research methodologies are more conducive to dealing with some of these dilemmas than others. In attempting to treat young people as social actors, and to respond to recent calls for the increasing participation of young people in society and respect for their opinions (Matthews et al., 1999), participatory research, and more speci®cally participatory urban appraisal (PUA) methods, were used in this research project. PUA is one of a series of participatory research approaches that is most closely aligned with Participatory Rural Appraisal (PRA), and to a lesser extent with Participatory Action Research (PAR). PUA and PRA have their foundations in Paulo Freire's advocacy of popular education as empowerment, as well as applied anthropology, agroecosystem analysis, farming systems research and Rapid Rural Appraisal (RRA). More speci®cally, they refer to `a growing family of approaches and methods to enable local (rural and urban) people to express, enhance, share and analyse their knowledge of life and conditions to plan and to act' (Chambers, 1994a, p. 953). In an attempt to counter top-down data collection practices where the poor and the powerless were often not consulted, PUA and PRA were championed initially by grassroots organisations in the South in the 1980s, and more recently by academic researchers (see Pain & Francis, 2003). The basic principles of PUA include the recognition of `reverse learning' (from research participant to researcher), transferrals of power to allow participants rather than facilitators to determine agendas (often referred to as `handing over the stick'), the encouragement of visual rather than written or verbal accounts of issues, and giving prominence to the voices and perceptions of people themselves (Cornwall & Jewkes, 1995). Thus, as Kesby (2000, p. 424) points out, participants are regarded Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 494 C. McIlwaine & K. Datta as the ones with knowledge and this knowledge is valorised. This ¯exible and iterative approach is conducted mainly through the use of a range of diagramming techniques within groups, such as ¯ow diagrams, listing and rankings, timelines and institutional mapping (see Pain & Francis, 2003; also Chambers, 1994b, 1994c). Participatory methodologies are not without their problems and, indeed, have been subject to considerable criticisms. One of the earliest issues identi®ed was the potential dif®culty in reaching consensus and ignoring power relations within focus groups on grounds of gender, age and ethnicity (Guijt & Shah, 1998; Cornwall, 2000). Indeed, some suggest there is a possibility of social inequalities being actively reinforced through the PUA research process (Mohan, 1999). This relates to another key criticism that the methodology tends to overstate the extent to which genuine participation is achieved and that the name may hide what are essentially top-down consultations with marginalised groups (Cooke & Kothari, 2001). In reality, much participatory appraisal is brief and lacks the depth facilitated through other qualitative methods (Pain & Francis, 2003). This is partly attributable to the failure to problematise the concept of participation itself, especially the extent to which it can lead to empowerment, and the potential of the methodology to reduce a host of social inequalities, as is so often promised (Cooke & Kothari, 2001; Cornwall, 2003). Moreover, some have argued that participatory appraisal, and especially that associated with Robert Chambers, has been substantially under-theorised in relation to politics and power (Kapoor, 2002). Finally, there are also constraints in analysing PUA data. Despite the intention to give voice and agency to research participants, analysis often highlights the views of the researchers over the participants (Moser & McIlwaine, 2004), with researchers maintaining their position of power (Pain & Francis, 2003). While bearing these caveats in mind, we suggest that PUA has certain advantages over other traditional social science methods when working with young people (although we do not advocate that it supplants these). These include the potential to get young people involved and interested in the research process more easily, thus giving them greater voice and agency which is especially important in a context, both in Botswana and elsewhere, where young peoples' views are often disregarded as unimportant (Young & Barrett, 2001a, 2001b; Winton, 2003). PUA attempts to address, at least in a limited way, unequal power relations between researcher and researched which are especially marked between adults and young people (Morrow & Richards, 1996; Robson, 2001). This also deals partly with some of the ethical dilemmas related with power relations (Johnson et al., 1998). The ¯exibility of PUA allows for the exploration of the differential abilities of young people, giving those who may be less articulate a chance to participate (Matthews, 2001; but see below). Participatory approaches have also been increasingly used for working on issues of sexual and reproductive rights, and especially on HIV/AIDS (Kesby, 2000). Similar advantages have been highlighted for using this approach such as creating a space for people's voices, generating dialogue, and facilitating learning and action (Cornwall & Welbourn, 2000). The current project was fully aware of the limitations of PUA, and we should stress that the process we deployed was akin to `shallow participation' (Cornwall & Jewkes, 1995; Kesby, 2000). As a pilot study, we did not have the time or resources to follow the research through with the young people in a sustainable way. Instead, we focused on `its ability to facilitate in practice participants' own Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 495 deconstruction/reconstruction of the categories and meanings that structure their lives' (Kesby, 2000, p. 432). Thus, bearing in mind the drawbacks as well as advantages of PUA, research conducted in Gaborone in July 2001 used participatory diagramming with a total of 29 focus groups in Gaborone, 27 of which were undertaken in schools and two carried out through a youth community drama non-governmental organisation (NGO), the latter to elicit viewpoints of out-of-school youth. The decision to work primarily through schools re¯ects quite high school enrolment rates in Botswana, even if 17% of the population aged between seven and 13 are not enrolled in school (mainly disabled, very poor and those living in remote areas) (Burke, 2000, p. 206). Nine of the school focus group discussions were carried out in a state senior secondary school, 12 in a private senior secondary school and six in a state community junior secondary school12. Among the 29 focus groups conducted in Gaborone, 10 were with male-only groups, six with female-only groups and 13 with mixed gender groups. In total, 154 participants were included in the focus groups; 70 (45%) of these were young women, and 84 (55%) were young men. In terms of ethnicity, the majority were Tswana, including virtually all the participants in the state schools, and the majority in the private school. Only in the latter were a minority of Asian and European youth included in the groups13. The combination of community and private schools also meant that the socio-economic status of participants varied from those with fairly meagre resources, to the children of the wealthy and powerful (although the latter were in the minority). The majority of the focus group members were aged between 15 and 18 years, with a few participants aged between 12 and 14, and between 21 and 2714. The focus groups were conducted by the two authors during discussion or `prep' lessons within the schools. Recognising that `informed consent' is reduced when young people are in a classroom context (Otaala, 1998), we were always careful to explain who we were, the nature of our research and what we planned to do. However, although we were keen to emphasise our role as researchers, being introduced by teachers tended to associate us with those in authority in the school thus creating unequal power relations from the outset (see also Ansell, 2001). This was compounded by using English as the main form of communication rather than Setswana. Yet, while working in Setswana may have been preferable, all the young people were ¯uent in English, given that all the discussions were held in English-medium schools. In addition, we worked only with those groups who were willing. In a couple of instances, discussions were started and then abandoned when it was obvious that young people were not interested or lacked the concentration to learn the basic tools of PUA15. Other problems also arose in some groups where young men dominated discussions in mixed gender groups while others experienced dif®culties in reaching consensus (ibid.). In some cases the quieter, more reticent members of groups had dif®culties in expressing their views. However, we tried to ensure that these people also had the opportunity to participate, and in some cases we broke groups into smaller units (pairs, for example), to allow this to take place. It is also worth noting that most young people spoke in the third person when discussing many issues, especially those linked with sexualities. While this has advantages in terms of allowing people to talk more freely, especially about issues which are generally thought to be taboo, it also means that information on personal experiences is lost. In terms of speci®c diagrams, we focused on listing and rankings (making lists of the issues that most affect young people, and then ranking their importance Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 496 C. McIlwaine & K. Datta usually through the use of an `onion' diagram) (see Moser & McIlwaine, 1999), causal ¯ow diagrams (identifying an important issue and mapping the causes and effects), and timelines (abstract graphs that assess changes over time in, for example, gender relations). Overall, the focus group discussions were viewed positively by the young people and in numerous cases, they became fully engaged with the issues reporting that the tools and exercises were `good fun' (see also Manyau, 1998). Also in line with the ethos of the methodology, although we could not conduct any follow-up work ourselves, we returned copies of all the focus group tools prepared by the young people to be discussed in follow-up classes run by teachers (and by youth leaders in the case of the NGO). Finally, although PUA can be conducted as a methodology in its own right, it is being used increasingly in combination with other methods such as quantitative questionnaire surveys, but particularly with in-depth interviews (Moser & McIlwaine, 1999). This acts as a form of triangulation, corroborating or not the ®ndings from the PUA discussions. In the current study, the PUA was combined with six in-depth interviews with representatives from the key organisations working with youth in the country16. Deconstructing Sexualities Among Young Men and Women in Gaborone, Botswana oung people's identity construction in Gaborone was mediated strongly by sexualities, and in particular by their perceptions of HIV/AIDS and teenage pregnancy and how these intersected with other sexual practices. Contrary to what is often believed or perceived, young people were very complex sexual beings. In our empirical research, sex and sexualities emerged as the most important set of factors affecting the lives of young people17. While the total number (which extended to 46) and nature of issues identi®ed were diverse, more than half (56%) were related to sex and sexualities. Of these, the two most prominent preoccupations were HIV/AIDS and teenage pregnancies, both mentioned in around two-thirds of all focus groups. The most serious concern was the HIV/AIDS crisis, with 19 of the 29 focus groups raising it as a key issue, regardless of the gender of the group. Also, when young people were asked to rank the most important problems in their lives, HIV/AIDS again emerged as their most important priority, although the mixed-gender and all-male groups were slightly more likely to prioritise it. Broadly similar patterns were obtained in relation to teenage pregnancy, with 18 of the 29 groups identifying it as important. Perhaps not surprisingly, the all-female and mixed-gender groups were most likely to discuss and rank this, with only three of the 10 all-male groups mentioning it (compared with four of the six all-female groups). Other concerns related to sexualities also dominated discussions, especially sexual abuse and rape of young people, premarital and underage sex, and sexually transmitted diseases, all of which were seen as interrelated. Interestingly, all discussion of sex and sexualities referred to heterosexuality. This perhaps re¯ects the taboo surrounding homosexuality in Botswana, as well as the fear in seeming to deviate from hegemonic heterosexual gender identities. Endangered Youth 497 Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Awareness of the Causes of HIV/AIDS and Teenage Pregnancy Among Young Men and Women The dominant preoccupation with HIV/AIDS and teenage pregnancies among young people in Gaborone re¯ects wider societal concerns, with negative attitudes towards underage and unprotected sex underlying both. Most young people displayed high levels of knowledge about the reasons for and consequences of HIV/AIDS and teenage pregnancies. They also appreciated that incidence was shaped by a range of interrelated social, cultural and economic factors. In terms of the speci®c reasons for contracting HIV/AIDS, young people were well versed in public health messages emphasising that the disease was spread by using unsterilised needles, sharing razors and through contaminated blood transfusion (see Fig. 1). All the groups that discussed HIV/AIDS made some reference to unprotected sex as a major cause. This was usually linked with what was usually referred to as `promiscuity' and/or `early sexual encounters' (see Fig. 1). Only a minority propagated inaccurate stereotypes about contraction of HIV. For instance, some young people identi®ed prostitution as the cause of HIV/AIDS18, and in one case a young man suggested that HIV could be contracted through kissing. Overall, most young people were aware of public health messages about the dangers of HIV/AIDS, especially the importance of condoms as prevention. Many groups referred to the widely propagated safe sex campaign message, sometimes referred to as the ABC of AIDS prevention: `Abstain, Be Faithful, Condomise'. Unprotected sex, multiple partners and sex at an early age also underlie discussions of teenage pregnancies (see Fig. 2). These were linked with a further range of themes that contributed to people engaging in potentially dangerous sexual activity that could lead to both HIV/AIDS and teenage pregnancies. In a somewhat contradictory way in light of the obvious absorption of public health messages, the most commonly cited cause of HIV/AIDS and teenage pregnancy Figure 1. Causal ¯ow diagram of HIV/AIDS drawn by a group of 5 young men and 1 young woman (aged between 15 and 18) at a state community junior school. Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 498 C. McIlwaine & K. Datta was lack of information about sex and contraception (see Fig. 2). This was often related to cultural factors in terms of the taboo surrounding sex, as one 15year=old young woman, Parline, noted: `In African society, there is no discussion about sex', with a second young woman, Bakang, going on to say that: `It's rude to talk to adults or parents about sex'. In a similar vein, a 16-year-old young man, Kagiso, argued that because parents were `too cultural and traditional, [they] can't talk to a child about teenage related issues, i.e. sex, relationships so they [youth] rather hang out with the wrong people because they are listened to'. This was echoed further by a young man, Phineas, who said that: `Our parents are not open to us about sex issues, they are too reluctant to talk about sex issues'. This lack of communication contributed not only to lack of information, but also to widespread mistrust between parents and children19. At its extreme, according to one all-female group, this could lead to a number of different behaviours, attitudes and psychosexual tendencies, including sexual and substance abuse, a lack of esteem, depression, anger and violence, and suicidal tendencies. Moreover, this also highlights how young people look to their parents for guidance and legitimation on sexual matters, yet they feel their parents are refusing to accept their sexualities (see also West, 1999). As well as parents' roles, this lack of information on matters relating to sex and sexualities is linked partly with the cultural demise of initiation ceremonies and other societal changes brought about by modernisation processes. Schools and public health campaigns have emerged as the major sources of information about sex, together with peer groups. Yet, young people were critical about the quality of information provided. Although one young man in a mixed gender group Figure 2. Causal ¯ow diagram of teenage pregnancy drawn by a group of 4 young women and one young man (aged between 16 and 17) at a state senior secondary school. Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 499 asserted that: `You must be from Mars if you do not know about contraception', others complained about the lack of instruction on `the proper use of condoms' and easier access to contraception. One 20-year-old young man, Stephen, from an out-of-school youth group, argued that: `Condoms and sexual booklets must be delivered every week to our doorsteps'20. Furthermore, although the Gaborone youth were fairly well-informed about sexual health, the majority felt that society as a whole lacked access to information, especially in rural areas. For example, referring to knowledge about AIDS, one 17-year-old young man, Ratanang, pointed out that: `Some people are still not aware, especially among herdsmen in the villages'. Associated with wider cultural practices, as well as the fairly rapid economic growth experienced in Botswana in recent decades, transactional sex also emerged in several discussions as a reason for HIV/AIDS and teenage pregnancy. Although some referred to the role of prostitution in the spread of HIV, discussions of transactional sex mainly revolved around how some young women had sexual relations with older and usually rich men in exchange for gifts and money as part of the `sugar daddy' phenomenon. Although this was linked with the notion of `economic hardship' and the need for survival (see Fig. 2), all the groups that identi®ed this practice noted that these relationships also gave girls access to non-essential consumer items such as mobile phones and Nike shoes, which many felt necessary to cement their identity as young people who were part of a global culture (see also Kambou et al., 1999; Baylies, 2002)21. One young man, Prince, pointed out: `After school, you will see all these luxury cars picking up girls, girls like nice cars, cell phones'. This was corroborated by a group of female respondents, one of whom asserted that: `Young people go out with people old enough to be their parents. They do it for the money, cell phones and things'. While, in itself, the phenomenon of `sugar daddies' does not automatically lead to HIV or teenage pregnancies, the unequal power relations inherent in these relationships makes it dif®cult for young women to negotiate with older men about the use of contraception (Kambou et al., 1999; Jewkes et al., 2001)22. As noted in research elsewhere, many older men will often refuse to use condoms in the belief that sleeping with virgins and young women will cleanse their blood of HIV (UNDP, 2000). Although not reported in the schools in which we worked, it has been noted elsewhere that in some cases female students are coerced into `granting sexual favours' to male teachers in return for decent grades (ibid., p. 30). Together with intergenerational and transactional sex, substance abuse referring to both drugs and alcohol, and sexual abuse were also seen to contribute to the practice of unsafe sex and promiscuity (Figs 1 and 2). Identi®ed as one of the key problems facing youth in the National Youth Policy (Ministry of Labour and Home Affairs, 1996a), many young people felt that drugsÐwhich they classi®ed as cigarettes, alcohol, marijuana (often referred to as dagga) and glueÐwere all on the increase, especially in the previous 10 years (see also Ministry of Labour and Home Affairs, 1996a). In turn, young people reported that both genders were less likely to control themselves sexually, or to use contraception, when under the in¯uence of drugs or alcohol. Even more frequently discussed was how substance abuse was linked with sexual abuse and rape (Figs 1 and 2). As one young 20-yearold woman, Sinah, from an out-of-school group, noted: `Alcohol abuse leads to sexual abuse which leads to teenage pregnanciesÐthey are all linked'. Sexual abuse and rape were identi®ed repeatedly by respondents as key factors contributing to the spread of HIV/AIDS and teenage pregnancies; in all but one Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 500 C. McIlwaine & K. Datta group discussing the former and all but three discussing the latter (Fig. 1; see also Varga, 1997; Wood et al., 1998). Asymmetrical power relations driven by age and gender inequalities meant that young women were often forced into having sex through the use of violence, acknowledging that women of all ages are potentially at danger. As one young woman reported, men who rape do not use condoms, and so the chances of contracting HIV and/or becoming pregnant were heightened. Some participants put forward alarming reasons for the incidence of rape, with one all-male group identifying it as linked with temptation in that `boys get turned on because these girls they wear mini-skirts', or what another male-dominated group identi®ed as `dress-code'. While other groups attributed rape to the in¯uence of alcohol and drugs, the majority of participants related it to prevailing constructions of hegemonic masculinities. Always discussing the issue in the third person, one group of ®ve male participants said that young men in general felt the need to assert themselves in relationships, especially when women would not consent to sex. One of them, Mpotsang, stated: `If a boy proposes to a girl that he loves her and she turns away then he might get angry and rape her'. While all the male and female participants condemned rape as unacceptable female participants, in particular, tended to view it as a hazard of growing up in Botswana and an integral part of gender relations in what one young woman referred to as `man-centred society' (see below). Fifteen of the groups noted that rape was on the increase, a trend corroborated nationally. A recent report estimates that nationally three of ®ve Tswana women of all ages have been the victims of some form of sexual or physical abuse, with the incidence of rape rising by 11% between 1997 and 1998 (UNDP, 2000, p. 29). More than half (58%) of all victims of rape nationally were between the ages of 11 and 20 years, of whom 41% were under the age of 16 (Emang Basadi Women's Association, 1999, p. 11). The young age of rape victims is also related to the more `ef®cient' transmission of HIV due to the high risk of damage to genital mucosa (UNDP, 2000). The constructions of sexualities among young people and their awareness of HIV and teenage pregnancy were in¯uenced heavily by peer pressure, which was another major factor discussed by young people for engaging in early and unprotected sexual liaisons. Twenty-four groups identi®ed peer pressure as linked with `trying to ®t in'. While the participants reported that peer pressure had different manifestations according to gender, it invariably involved pressure to have sexual relations. For young women, there was great pressure to have a boyfriend with whom they had to have sexual relations if they were to `keep them' (their boyfriends) and in the words of one young woman, Edem, `to prove my love for him' (see also Jewkes et al., 2001). For young men, on the other hand, they were under pressure to prove their sexual prowess in terms of having as many sexual partners as possible. This was very much part of conforming to constructions of hegemonic masculinities in relation to traditional Tswana culture where having multiple sexual partners is important, and where sex remains the main way of getting to know a woman (Datta, 2004; also Kambou et al., 1999 on Zambia). As alluded to at various points thus far, underlying the constructions of sexualities among young people and attitudes towards sex was the nature of gender roles and relations. Historically, the constructions of masculinities and femininities have been linked with strongly patriarchal cultural traditions which have been passed down through the generations (Emang Basadi Women's Association, 1999). As stated in the National Youth Policy, young women are often Endangered Youth 501 restricted from participating fully in development due to various socio-cultural practices (Ministry of Labour and Home Affairs, 1996a, p. 9). Among the youth in Gaborone, the nature of gender identities was viewed as predominantly patriarchal. One young 17-year-old woman, Ntoba, felt strongly that: Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Discrimination of women is strong in African culture¼ when women act in certain ways, men tell them to calm down¼ most women are discriminated against, both in the tribes in rural areas and even here in class. As well as blaming traditional Tswana culture that is strongly predicated on the subordination of women, Ntoba also felt that women were socialised into accepting sexist behaviour: `We grow up to think women are inferior without thinking why'. Another woman, Tshepang, pointed out the rami®cations of gender discrimination in that: `Young women lose their self-esteem'. This construction of passive femininities was echoed among all the youth, together with a construction of masculinities according men greater power and status in society. A young woman, Kehumile, complained: `It's the male mentalityÐthey think they're superior to us'. Of the 15 groups who explored the issue of gender roles, 11 came up with traditional, and often stereotyped views, such as linking women's roles with reproductive activities such as mothering and housework, and men's with breadwinning and protection. However, young men (in all-male and mixed-gender groups) were much more likely to identify stereotyped roles than women. For instance, one all-male group identi®ed women's roles as `nurturing children', `satisfying the sexual needs of her husband' and `domestic chores', and men's as `breadwinning', `satisfying the sexual needs of his wife', `handy work and maintenance' and `security of the family'. In contrast, an all-female group was keen to stress that women and men shared breadwinning and decision-making in the household and that women were just as likely to be pilots and truck drivers as men. Eight of the 15 groups who explored the issue of gender roles went on to look at how women's status had changed over time (with most looking at the last 20 years). All eight concluded that women's position in society had improved and identi®ed the government and the NGO sector as being responsible for promoting this change. Interestingly, two of these groups drew time-lines of both men's and women's rights and felt that as women's situation had improved, men's rights had declined. Despite these shifts, most young people, especially women, felt that gender identities were characterised by deep-seated inequalities that provided the foundation for the construction of sexualities, as well as widespread discrimination in other ®elds. One all-female group felt that gender oppression was also reinforced by the use of force or what they called `abuse of women', both physically and sexually (see also Datta, 2004). One ®nal, more prosaic issue that was reported by participants as in¯uencing early sexual relations was linked to the dearth of leisure facilities in Gaborone (and elsewhere in the country)23. As one young man, Kitso, in a mixed gender group argued: `Recreational facilities must be provided because otherwise the young have nothing to do, they have idle minds', which leads to experimentation. Although this experimentation was linked with alcohol and drug abuse, one allmale group suggested it was associated most commonly with unprotected sex, HIV/AIDS and teenage pregnancy. This group blamed the government for failing to allocate funds for recreation or to implement their National Development Plans, 502 C. McIlwaine & K. Datta as well as blaming parents who they felt were not concerned about what their children were doing. This broad view was echoed by the Chief Executive Of®cer of the National Youth Council who pointed out that: `Recreation for young people is very vital but we don't really have a society where young people can play in an organised environment'. Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Awareness of the Effects of HIV/AIDS and Teenage Pregnancy Among Young Men and Women Young people were not only aware of the causes of HIV/AIDS and the reasons for early and unprotected sexual relations that led to teenage pregnancy, they were very knowledgeable about the consequences of both phenomena. This knowledge contributes to constructions of sexualities among young people. In relation to HIV/AIDS, discussions tended to identify both the biomedical and social effects for individuals, as well as wider social and economic consequences. Seven of the 19 groups who directly identi®ed HIV/AIDS as a serious issue all mentioned death as the ultimate outcome, with one group suggesting that these deaths were caused by people committing suicide to avoid the pain of an AIDS-related death. For those living with AIDS, all the groups were well aware of the stigma associated with the disease. Figure 1 highlights the nature of this social opprobrium through references to speci®c terms such as `shame and disgrace', `isolations of victims', and `reduces the number of friends'. Six of the groups mentioned how this stigma led many sufferers to resort to drug and alcohol abuse as a way of coping. Other groups mentioned how families were also affected, with family breakdown sometimes related to living with someone with HIV/AIDS. At a societal level, young people were well aware of the effects of HIV/AIDS on the economy. Not only did HIV/AIDS bring poverty to individuals and their families due to the `loss of breadwinners' (see Fig. 1), but HIV/AIDS also had wider rami®cations for the economy in terms of productivity of the workforce, as well as the availability of a labour force. One all-male group also pointed out that AIDS `distorts the government budget', while another mixed gender group highlighted the importance of `population decrease'. The rami®cations of teenage pregnancy were also perceived at a number of levels from the individual to the societal. For individuals the most commonly identi®ed consequence was young women having to drop out of school, which was related to dif®culties in ®nding jobs due to lack of educational quali®cations (see above). Teenage pregnancy was also strongly associated with HIV/AIDS and sexually transmitted diseases because of the lack of condom use (see Fig. 2). As with HIV/AIDS, many groups identi®ed a high degree of stigma, which some argued could lead to the girl's family rejecting her, or even to girls committing suicide (see Fig. 2). While many of the consequences were linked with the birth of babies, abortion and abandonment of babies was also mentioned frequently. Abortion was perceived as a key issue. One young woman, Marga, argued that: `If they knew about the dangers of abortion, they would not fall pregnant'. Others were still more dramatic with one young man arguing that teenage pregnancy could lead to death as: `Their bodies are not mature, they die during labour', with another young woman suggesting that teenage pregnancy led to infertility. Unlike traditionally tolerant attitudes towards early sexual relations and childbirth, teenage pregnancy among these urban youth was viewed negatively. For young urban women, this re¯ects not only how cultural practices have changed, but how Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 503 their identities and life trajectories relate to ®nishing their education and securing good jobs. Young men and women in Gaborone were therefore keenly aware of HIV/ AIDS and teenage pregnancy. Indeed, both the construction of their identities and sexualities were in¯uenced heavily by these two phenomena, with teenage pregnancy a particular concern among women. While several factors were related to Tswana cultural traditions, such as transactional and intergenerational nature of sexual relations encompassed within the `sugar-daddy' phenomenon, others were linked with contemporary urban trends and public health campaigns, such as the awareness of safe sex, as well as broadly negative attitudes towards teenage pregnancy and underage sexual relations. Therefore, as suggested at the outset, rather than conceptualising sexualities among young people as monolithic, they are diverse, dynamic and in¯uenced by a host of social, cultural and economic circumstances. Yet, despite this dynamism, constructions of sexualities remained heavily imbued by gender inequalities. Masculinities were strongly predicated on the exercise of male power and men's rights to demand sex, by violent means if necessary, while women's femininities emphasised their nurturing role and passive acceptance of sex which in some cases was used to access material possessions. Overall, what was striking was how young people were agents of their own sexualities in terms of knowledge and awareness. Yet at the same time, they still required their parents' recognition of their sexualities. Indeed, this re¯ects Valentine's (2000) distinction between familialisation and individualisation where young people require both dependence and independence from parents with respect to sexualities. Interventions and Policy Implications for Addressing HIV/AIDS and Teenage Pregnancy In line with our aim to treat young people as legitimate social actors, as well as the evidence that they were active and knowledgeable about sex and sexualities, we asked them to suggest possible interventions to address the HIV/AIDS crisis and prevalence of teenage pregnancies. Two main approaches were identi®ed. The ®rst, less prevalent approach laid greater emphasis on addressing the symptoms of AIDS and teenage pregnancies. This focus tended to be advocated among young men, although by no means did all young men express these types of views. For example, one all-male group had a heated debate with one participant, Opelo, arguing that: `There should be a selective culling of people with AIDS or they should be isolated from society', re¯ecting the extent to which AIDS remains misunderstood and plagued by taboo. His friend opposed this by arguing that this was against the human rights of people with AIDS. In relation to teenage pregnancies, young men were also more likely to suggest that teenage mothers should be excluded permanently from schools (a view not expressed by any young women). The second most commonly suggested solution focused on prevention, such as improved access to information and contraception and especially condom use, as well as the need to develop recreational facilities to prevent young people turning to sex as a leisure pursuit. These views were shared most commonly by young women, who tended to consider the root causes and consequences of the problems before suggesting solutions. For example, one young woman, Resego, pointed out in relation to teenage pregnancy that: `More sex concepts should be Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 504 C. McIlwaine & K. Datta taught in schools, not only causes but also social, economic and health effects'. Having said this, young men were also keen to stress the importance of prevention. However, while young women usually accepted sex as an inevitable part of growing up and in the construction of sexualities, only the men mentioned the importance of abstinence, or what one young man called, `keeping yourself pure' as a solution to AIDS. More pragmatically, another young man, Simon, also advocated abstinence on the grounds that: `Condoms burst, or girls who have used contraception still fall pregnant'. As noted earlier, a key component of this prevention approach shared by young women and men was the role of parents. Repeatedly, young people suggested that parents should take a much more proactive role in sex education, or as one young woman stated: `Parents should be con®dent when talking to their children about sex', while another suggested that `parents should learn better parenting skills'. As with young people's awareness of HIV/AIDS and teenage pregnancy, on the whole they recognised that prevention required holistic interventions that extended beyond a narrow, biomedical health perspective (see Butcher & Welbourn, 2001; Tallis, 2002). One group of three young women and two young men identi®ed four broad areas of intervention to address AIDS as well as the institutions or groups that should be responsible for implementing projects in these ®elds. These included `education' that was the remit of schools, the government, and the media; `breaking cultural misconceptions', that should be the responsibility of schools, the government and respectable members of the community; `funding' which should come from the government and the international community; and `addressing promiscuity' identi®ed as the responsibility of youth. Another group of six young women identi®ed a similar initiative to address teenage pregnancy that would involve sex education to be provided by schools, access to contraceptives to be provided by the government, recreational centres developed as part of community projects, and peer counselling to be carried out by youth volunteers. Albeit belatedly, this recognition of a holistic perspective has been shared by national government policy with reference to HIV/AIDS. Since the late 1990s government policy has shifted from an under-resourced focus on health, towards one whereby HIV/AIDS initiatives have been mainstreamed into all government ministries in a multi-sectoral way led personally by the President (UNDP, undated)24. Youth have also emerged as an important focus of government initiatives following the identi®cation of AIDS and unemployment as the two major challenges facing youth in Botswana in their National Plan of Action for Youth (Ministry of Finance and Development Planning, 2000). Speci®c preventative measures aimed at youth include the introduction of HIV/AIDS education to primary and secondary school curricula and youth programmes which explore young people's knowledge of preventive measures and their attitudes towards sexual practices. This increasingly coordinated strategy towards HIV/AIDS has also been echoed within the NGO sector. A youth NGO sector has been growing in recent years and engaging in a range of activities with HIV/AIDS prevention, adolescent sex and reproductive health programmes forming an integral part of most programmes (interview with Deputy Executive Director, Botswana National Youth Council, July 2001)25. The gender NGO sector has now recognised the importance of age and especially the `girl-child' in relation to reproductive health and HIV/AIDS prevention (interview with Emang Basadi Women's Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 505 Organisation, July 2001). In addition, most NGOs within the health sector, which is coordinated by the Botswana Network of AIDS Service Organisations (BONASO), focus on sexualities as a way of understanding HIV/AIDS and teenage pregnancies (interview with a representative from BONASO, July 2001). Much less attention has been directed towards teenage pregnancy. Although understandable in that the developmental impact is less than that of HIV/AIDS, there are serious rami®cations of teenage pregnancies for young women in particular (see earlier). Indeed, recent policy documents, such as Vision 2016, have called for a review of the policy of exclusion from school of young women once they fall pregnant. It has been suggested that young mothers should be able to resume schooling as soon as possible after con®nement and at the same school that they left (BOCONGO, 2000). Yet, the tendency of both policies towards HIV/AIDS and teenage pregnancy, especially among the government, has been to focus on information provision. However, as noted by the young people above, a multi-pronged strategy towards both phenomena is required with provision of information being only one element. Therefore, the youth were aware that information alone will not prevent HIV/AIDS or teenage pregnancies and that absorbing information from public health campaigns is not the same as changing actual behaviour (see also Welbourn, 1999; West, 1999). Unfortunately, we have no speci®c evidence of this beyond recognition of these issues as young people tended to talk in the third person about sex and HIV/AIDS, and few discussed their own personal experiences. Furthermore, without a longitudinal analysis, we had no way of knowing the actual effects of public health campaigns in terms of behaviour change. However, what is more certain is the need to recognise a holistic approach to preventing HIV/AIDS and teenage pregnancy that extends beyond the biomedical and health approach to incorporate an understanding of constructions of sexual identities among young people in particular, and especially the gendered dimensions of sexualities. While still in its infancy in Botswana, there is a need to further develop policies and programmes that embrace communication skills, group work and role play (West, 1999). This has been attempted within the NGO sector, for instance, by the Reetsanang community-based drama organisation targeting out-of-school youth; they regularly stage shows aimed at raising HIV/AIDS awareness, as well as developing plays and stories for radios shows (interview with representative of Reetsanang, July 2001). Yet there is much greater scope to develop these types of activities, possibly along the lines of the wellknown `Stepping Stones' project initially developed by ActionAid in Uganda. This is a training package using participatory techniques within communities that focuses on a host of issues such as gender and generational inequalities, communication and relationship skills, with HIV/AIDS being the underlying guiding principle. It has been adapted for a range of countries in the South, and has been shown to be very effective in addressing the behaviour and cultural changes required to deal with transforming sexual practices (Welbourn, 1999; Butcher & Welbourn, 2001). Conclusion This article has explored the importance of the constructions of sexualities among young people in Gaborone, Botswana in relation to the key issues Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 506 C. McIlwaine & K. Datta of teenage pregnancy, as well as the more recent HIV/AIDS epidemic. In highlighting the need to recognise the agency of young people, our discussions with youth raise a number of conceptual and policy-related issues. Conceptually, the discussions have highlighted the need to deconstruct the notion of a `Botswana sexuality' and to recognise that youth sexualities are diverse, ¯uid and heavily gendered, being in¯uenced by the dominant Tswana culture, as well as a host of other social, cultural and economic circumstances associated with public health campaigns. In other words, youth sexualities should be conceptualised as holistic and deeply embedded in society, rather than as a range of reproductive health responses. Underlying this is the need to recognise young peoples' sexual identities as equally important and sophisticated as adults, especially by their parents. This also resonates for the formulation of policy. From a policy perspective, young people have demonstrated how they have received and understood public health messages. While this is heartening in itself, the key and much more dif®cult challenge is how such messages can be translated into practice. For this to occur, sexualities need to be conceived as a holistic construct rather than as a public health issue as constructions of sexualities are affected by a diverse range of social, economic and political factors. In turn, this requires a need to re-engage with the debate on youth agency and empowerment. In a dominant Tswana culture which values obedience and discipline in young people, youth have to be given a chance to speak, to be seen as active agents of change. Critical to this is the need to address gender disparities which are placing young women in particular at risk from unsafe sexual practices and violent sexual relations. There is a concomitant need to focus on young men as gendered beings who need to be convinced of the positive role they can play in transforming constructions of sexualities and challenging hegemonic masculinities. Young people must therefore be given spaces within which to construct alternative sexual practices. While there have been some attempts to develop these spaces through working with young people in partnerships similar to the Stepping Stones approach, there remains considerable scope to extend these much further. Only by engaging with young people as legitimate social actors and recognising the diversity of their sexual identities will further progress in addressing the HIV/AIDS and teenage pregnancy crisis in Botswana be made. Acknowledgements We would like to thank the Nuf®eld Foundation for funding the research project entitled `Youth, gender and violence in Southern Africa', award number SGS/ 00496/G, on which this article is based, as well as the Government of Botswana for granting us permission to conduct the study. We are also extremely grateful to the staff and students of Gaborone Secondary School, Maru-a-Pula Secondary School, Bokamoso Community Junior Secondary, Marang Community Junior Secondary as well as the Reetsanang Association of Community Drama Groups. Finally, we would like to thank Sylvia Chant for her helpful comments on an earlier draft of this article, as well as the three anonymous referees together with Linda Peake for their constructive criticisms. Endangered Youth 507 Notes 1. 2. 3. Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. At the last census, Botswana had a population of 1.6 million people, which was divided equally between rural and urban settlements (Central Statistics Of®ce 2001). Research was conducted in Gaborone, Botswana for a three-week period in July 2001. Although this project set out to look at the relationship between youth, gender and violence in South Africa and Botswana, the inductive nature of the research facilitated by the use of PUA methodologies meant that young people were able to identify and prioritise issues that they felt were most important in their lives. Thus, while violence was explored with youth in South Africa, and identi®ed by youth in Botswana, sexualities and related issues (such as teenage pregnancy, HIV/AIDS) were consistently prioritised. Hence, the focus on sexualities here. In Botswana, passing from one stage to another of a lifespan has not traditionally been measured by age. For example, Suggs' (1987) research among the Bakgatla revealed that women's lifespan was divided into four stages (mosetsana, lekgaribe, mosadi and mosadi mogolo) which corresponded roughly with the Western categories of girl, adolescent, woman and old woman. In turn, the passage from one stage to the next was measured by competence in decision-making, motherhood, the ability to provide for the household and the establishment of managerial household independence and not biological years. In challenging these assumptions, some interesting research is emerging within the ®eld of masculinities (Morrell, 1998, p. 2001), on issues such as same-sex relationships in South Africa (Louw, 2001; Phillips, 2004), and on masculinity, sexuality and HIV in the South African gold mines (Campbell, 2001; Moodie, 2001). Other researchers have noted similar attitudes in neighbouring Southern African countries (Jewkes et al., 2001; Mturi & Moerane, 2001). The institution of marriage has been undermined further by the workings of the pastoral economy based on cattle-raising; some suggest that for men, buying cattle serves as insurance for old age rather than having a wife and children (Timaeus & Graham 1989). Prior to this, there was a widespread belief that AIDS was a disease which primarily affected homosexuals, and as it was (mistakenly) assumed that there were no homosexuals in Botswana, there was no HIV/AIDS (MacDonald, 1996). Here, we are con¯ating the term `teenager' with youth. Among the proposed strategies and actions to come out of a review of women and health, is recognition of the need to prioritise abortion as a major public health problem and develop policies and programme that address is properly (Government of Botswana/UNDP, 1998). Particular spaces, such as schools, have been identi®ed as key sites in which gendered and sexual identities are understood and constructed (see West, 1999; Ansell, 2002). The education system in Botswana is based on seven years of primary school, three years of junior secondary school, and two years of senior secondary school. There are over 650 primary schools in the country, 200 community junior secondary schools and less than 30 senior secondary schools (Burke, 2000: 206). Youth have been the major bene®ciaries of educational developments that have taken place since independence in 1966. They are more literate and have more advanced levels of education than their predecessors (Government of Botswana/UNDP, 1998). In the 1990s, it was estimated that 85% of youth under 15 years were in school compared to 70% in the 1970s and 1980s and a mere 40% in the 1960s (Women's Affairs Department, 1994; Meekers & Ahmed, 1999). Most citizens of Botswana are members of the Tswana ethic group. Other minor ethnic groups include the Bakalanga, Basarwa and Baherero as well as a smaller number of citizens of Asian and European origin (Ministry of Finance and Development Planning, 1997). Information on rural±urban origin of young people was not collected in the focus group discussions. With hindsight, it would have been useful to see if any variations existed according to this axis. This was less of a problem in Botswana than in other research conducted by the authors in 2001 in Cape Town, South Africa. Here, we attempted to use PUA focus groups with street children with little success. This was mainly because most had little education, and many were constantly under the in¯uence of drugs (Datta & McIlwaine, 2001). Interviews were conducted with representatives of the National Youth Council, Botswana Network of AIDS Services, Botswana National Youth Centre, Emang Basadi Women's Association, Childline Botswana and Reetsanang Association of Community Drama Groups. This information was elicited using `listings' where participants were asked to list issues affecting them. A total of 29 listings were carried out. Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 508 C. McIlwaine & K. Datta 18. While prostitution itself is not a cause of HIV/AIDS contraction, it can facilitate the spread of the disease through the practice of unprotected sex. 19. This is also related to low levels of contraceptive use among youth who are scared that their parents may come across their contraceptives (Meekers & Ahmed, 1999). 20. This lack of information on sex education was also echoed by the Chief Executive Of®cer of the National Youth Council in an interview with the authors in which she argued that negative attitudes by both parents and health workers led to young people being uninformed, as well as being scared to visit clinics and health centres to obtain contraception. Indeed, the provision of a `youth friendly' clinic was one of the National Youth Council's key programmes, and which was proving quite popular, serving around 30±50 young people a day. 21. There is a clear distinction in most African cultures between transactional sex (or survival/ informal sex work as it is sometimes referred to) and commercial sex work. While commercial sex is commonly associated with colonisation and westernisation, transactional sex is associated with economic need and poverty, the giving of services other than sex (e.g. domestic services) and the functional nature of the practice (i.e. money is sent home, used to pay school fees, becoming head of a household). For these reasons, transactional sex is socially sanctioned (Wojcicki, 2002). 22. This was re¯ected in an interview with a representative of the Botswana Network of AIDS services (BONASO) in which he questioned: `How can someone of 18 negotiate with someone at 30 when it comes to sex?'. 23. This concern was also identi®ed in an interview with a representative of the National Youth Centre, who said their main priority was building a football pitch and a gym. 24. This includes annual sentinel surveillances initiated in 1992 under the auspices of the Ministry of Health (through the National AIDS Control Programme), data collected by all health institutions who are now obliged to report all cases of HIV/AIDS to the AIDS/STD Unit, the approval of a National Blood Transfusion Policy, community mobilisation to provide home-based care and the formation of a National Orphan Policy. Moreover, several high-level committees have been established to including the Parliamentary Committee on HIV/AIDS and the National AIDS Coordination Agency which seeks to co-ordinate the multi-sectoral response to the disease (Ministry of Finance and Development Planning, 2000). 25. NGOs in Botswana are divided into nine sectors including the Youth, Reproductive Health and HIV/AIDS and Women and Gender NGOs. They all fall under the umbrella organisation, the Botswana Council of Non-Governmental Organisations (BOCONGO). References Aggleton, Peter & Warwick, I. (1997) Young people, sexuality, Hiv and Aids education, in: L. Sherr (Ed.) Aids and Adolescents, pp. 24±36 (Amsterdam, Harwood Academic Press). Ahlberg, B. (1994) Is there a distinct African sexuality? A critical response to Caldwell et al., Africa, 64, pp. 220±242. Aitken, S. (1994) Children's Geographies (Washington DC, Association of American Geographers). Ansell, Nicola (2001) Producing knowledge about `Third World Women': the politics of ®eldwork in a Zimbabwean secondary school, Ethics, Place and Environment, 4, pp. 101±116. Ansell, Nicola (2002) `Of course we must be equal but¼': imagining gendered futures in two rural southern African secondary schools, Geoforum, 33, pp. 179±194. Ansell, Nicola (2003) Secondary schooling and rural youth transitions in Lesotho and Zimbabwe, Paper presented at a research symposium on Uncertain Transitions: Youth in a Comparative Perspective, University of Edinburgh, 27 June 2003. Barnett, Tony (2002) The social and economic impact of Hiv/Aids on development, in: Vandana Desai & Robert B. Potter (Eds) The Companion to Development Studies, pp. 391±395 (London, Arnold). Baylies, Carolyn & Bujra, Janet, with the Gender and AIDS Group (2000) Aids, Sexuality and Gender in Africa: collective strategies and struggles in Tanzania and Zambia (London, Routledge). Baylies, Carolyn (2002) Hiv/Aids and older women in Zambia: concern for self, worry over daughters, towers of strength, Third World Quarterly, 23, pp. 351±375. Bell, David & Valentine, Gill (Eds) (1995) Mapping Desire: geographies of sexualities (London, Routledge). Botswana Council of Non-Governmental Organisations (BOCONGO) (2000) Bocongo News, 2. Burke, Charlanne (2000) They cut Segametsi into parts: ritual murder, youth, and the politics of knowledge in Botswana, Anthropological Quarterly, 73, pp. 204±214. Butcher, Kate & Welbourn, Alice (2001) Danger and opportunity: responding to Hiv with vision, Gender and Development, 9, 51±61. Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 509 Caldwell, John & Caldwell, Pat (1987) The cultural context of high fertility in sub-Saharan Africa, Population and Development Review, 13, pp. 409±437. Caldwell, John, Caldwell, Pat & Quiggin, Pat (1989) The social context of Aids in sub-Saharan Africa, Population and Development Review, 15, pp. 185±229. Campbell, Catherine (2001) Going underground and going after women': masculinity and Hiv transmission amongst black workers on the gold mines, in: Robert Morrell (Ed.) Changing Men in Southern Africa, pp. 275±286 (London, Zed/ Pietermaritzburg, University of Natal Press). Central Statistics Of®ce (2001) The National Population Census (Gaborone, Government Printer). Chambers, Robert (1994a) The origins and practice of Participatory Rural Appraisal, World Development, 22, pp. 953±969. Chambers, Robert (1994b) Participatory Rural Appraisal (Pra): analysis of experience, World Development, 22, pp. 1253±1268. Chambers, Robert (1994c) Participatory Rural Appraisal (Pra): challenges, potentials and paradigm, World Development, 22, pp. 1437±1454. Chant, Sylvia & Gutmann, Matthew (2000) Mainstreaming Men into Gender and Development: debates, re¯ections and experiences (Oxford, Oxfam). Chant, Sylvia & Mcilwaine, Cathy (1998) Three Generations, Two Genders, One World: women and men in a changing century (London, Zed Books). Chant, Sylvia & Craske, Nikki (2003) Gender in Latin America (London, Latin American Bureau/ New Brunswick, New Jersey, Rutgers University Press). Clarke, Fiona & Laurie, Nina (2000) Gender, age and exclusion: a challenge to community organisations in Lima, Peru, Gender and Development, 8, pp. 80±88. Cock, Jacklyn (2003) Engendering gay and lesbian rights: the equality clause in the South African Constitution, Women's Studies International Forum, 26, pp. 35±45. Cooke, Bill & Kothari, Uma (2001) Participation: the new tyranny (London, Zed Books). Cornwall, Andrea & Jewkes, Rachel (1995) What is participatory research?, Social Science and Medicine, 41, pp. 1667±1676. Cornwall, Andrea & Welbourn, Alice (2000) From reproduction to rights: participatory approaches to sexual and reproductive rights, Pla Notes, 37, 14±112. Cornwall, Andrea (2000) Making a difference? Gender and Participatory Development, Ids Discussion Paper 378 (Brighton, Institute of Development Studies). Cornwall, Andrea (2003) Whose voices? Whose choices? Re¯ections on gender and participatory development, World Development, 31, 1325±1342. Datta, Kavita & Mcilwaine, Cathy (2000) `Empowered leaders?'. Perspectives on women heading households in Latin America and Southern Africa, Gender and Development, 8, pp. 40±49. Datta, Kavita & Mcilwaine, Cathy (2001) Final Report to the Nuf®eld Foundation on the project `Youth, gender and violence in Southern Africa' (London, Department of Geography, Queen Mary, University of London). Datta, Kavita (1996) Women owners, tenants and sharers in Botswana, in: Anne Schlyter (Ed.) A Place to Live: gender research in Africa, pp. 112±126 (Uppsala, Nordic Africa Institute). Datta, Kavita (2004) A coming of age? From Wid to Gad to `add-men-and-stir' in urban Botswana, Journal of Southern African Studies, 40 . Dwyer, Claire (2000) Negotiating diasporic identities: young British South Asian Muslim women, Women's Studies International Forum, 23, pp. 475±486. Emang Basadi Women's Association (1999) The Women's Manifesto: a summary of Botswana women's issues and demands, 2nd edn (Gaborone, Lentswe La Lesedi). Foucault, Michel (1978) The History of Sexuality, vol. I. An introduction (London, Allen Lane). Giddens, Anthony (1992) The Transformation of Intimacy: sexuality, love and eroticism in modern societies (Cambridge, Polity Press). Government of Botswana/UNDP (1998) National Gender Programme Framework (Gaborone, Ta Publications Botswana). Guijit, Irene & Shah, Meera Kaul (Eds) The Myth of Community: gender issues in participatory development, pp. 131±140 (London, Intermediate Technology). Helle-Valle, Jo (1999) Sexual mores, promiscuity and `prostitution' in Botswana, Ethnos, 64, pp. 372± 396. Holland, Jeremy & Blackburn, James (Eds) (1998) Whose Voice? Participatory research and policy change (London, Intermediate Technology). Holloway, Sarah & Valentine, Gill (Eds) (2000) Children's Geographies (London, Routledge). International Center for Research on Women (ICRW) (2000) The Critical Role of Youth in Global Development, Issue Brief, December 2001 (Washington DC, Icrw). Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 510 C. McIlwaine & K. Datta James, A., Jenks, C. & Prout, A. (1998) Theorising Childhood (Cambridge, Polity Press). Jewkes, Rachel, Vundule, Caesar, Maforah, Fidelia & Jordaan, Esme (2001) Relationship dynamics and teenage pregnancy in South Africa, Social Science and Medicine, 52, pp. 733±744. Johnson, Victoria, Ivan-Smith, Edda, Gordon, Gill, Pridmore, Pat & Scott, Patta (Eds) (1998) Stepping Forward: children and young people's participation in the development process (London, Intermediate Technology). Jolly, Susie (2000) `Queering' development: exploring the links between same-sex sexualities, gender and development, Gender and Development, 8, pp. 78±88. Jolly, Susie (2003) Gender myths and feminist fables: repositioning gender in development policy and practice, Paper presented for the International Workshop Development myths around sex and sexualities in the South, Ids July 2003. Kambou, Sarah Degnan, Shah, Meera Kaul & Nkhama, Gladys (1999) For a pencil: sex and adolescence in peri-urban Lusaka, in: Irene Guijt & Meera Kaul Shah (Eds) The Myth of Community: gender issues in participatory development, pp. 110±120 (London, Intermediate Technology). Kapoor, Ilan (2002) The devil's in the theory: a critical assessment of Robert Chambers' work on participatory development, Third World Quarterly, 23, pp. 101±117. Katz, Cindi & Monk, Janice (Eds) (1993) Full Circles: geographies of women over the life cycle (London, Routledge). Kesby, Mike (2000) Participatory diagramming: deploying qualitative methods through an action research epistemology, Area, 32, pp. 423±435. Kinsman, John, Nyanzi, Stella & Pool, Robert (2000) Socializing in¯uences and the value of sex: the experience of adolescent school girls in rural Masaka, Uganda, Culture, Health and Sexuality, 2, pp. 151±166. Kumaranayake, Lilani & Watts, Charlotte (2001) Resource allocation and priority setting of HIV/AIDS interventions: addressing the generalised epidemic in Sub-Saharan Africa, Journal of International Development, 13, pp. 451±466. Laurie, Nina, Dwyer, Claire, Holloway, Sarah & Smith, Fiona (Eds) (1999) Geographies of New Femininities (Harlow, Longman). Leshabari, M.T. & Kaaya, S.F. (1997) Bridging the information gap: sexual maturity and reproductive health problems among youth in Tanzania, Health Transition Review, 7, pp. 29±44. Louw, Ronald (2001) Mkhumbane and new traditions of (un)African same-sex weddings, in: Robert Morrell (Ed.) Changing Men in Southern Africa, pp. 287±296 (London, Zed/Pietermaritzburg, University of Natal Press). Macdonald, David (1996) Notes on the socio-economic and cultural factors in¯uencing the transmission of Hiv in Botswana, Social Science Medicine, 42, pp. 1325±1333. Maharaj, Pranitha (2001) Male attitudes to family planning in the era of HIV/AIDS: evidence from KwaZulu-Natal, South Africa, Journal of Southern African Studies, 27, pp. 245±257. Manyau, Charity (1998) Situational analysis with children living in dif®cult circumstances in Zimbabwe using participatory rural appraisal, in: Johnson, Victoria, Ivan-Smith, Edda, Gordon, Gill, Pridmore, Pat & Scott, Patta (Eds) Stepping Forward: children and young people's participation in the development process, pp. 69±75 (London, Intermediate Technology). Matikanya, Richard (2003) Long-haul drivers, masculinities and the HIV/AIDS epidemic. PhD thesis in preparation, Department of Geography, Queen Mary, University of London. Matthews, Hugh & Limb, Melanie (1999) De®ning an agenda for the geography of children: review and prospect, Progress in Human Geography, 23, pp. 61±90. Matthews, Hugh (2001) Power games and moral territories: ethical dilemmas when working with children and young people, Ethics, Place and Environment, 4, pp. 117±178. Matthews, Hugh, Limb, Melanie & Taylor, Mark (1999) Young people's participation and representation in society, Geoforum, 30, pp. 135±144. Mcilwaine, Cathy & Datta, Kavita (2003) From feminizing to engendering development, Gender, Place and Culture, 10, pp. 345±358. Meekers, Dominique & Ahmed, Ghyasuddin (1999) Pregnancy-related school drop-outs in Botswana, Population Studies, 53, pp. 195±209. Ministry of Finance and Development Planning (1997) National Development Plan 8 (Gaborone, Government Printer). Ministry of Finance and Development Planning (2000) Mid-Term Review of National Development Plan 8 (Gaborone, Government Printer). Ministry of Labour and Home Affairs (1996a) National Youth Policy (Gaborone, Government Printer). Ministry of Labour and Home Affairs (1996b) National Women in Development Policy (Gaborone, Government Printer). Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 Endangered Youth 511 Mohan, Giles (1999) Not so distant, not so strange: the personal and the political in participatory research, Ethics, Place and Environment, 2, pp. 41±54. Molokomme, Ataliah (1996) State intervention in the family: a case-study of the child maintenance law in Botswana, in: Rajni Palriwala & Carla Risseeuw (Eds) Shifting Circles of Support: contextualising gender and kinship in South Asia and Sub-Saharan Africa, pp. 270±301 (New Delhi, Sage/London, Thousand Oaks). Mompati, Tlamelo & Prinsen, Gerard (2000) Ethnicity and participatory development methods in Botswana: some participants are to be seen and not heard, Development in Practice, 10, pp. 625±637. Monk, Janice & Katz, Cindi (1993) When in the world are women? In: Cindi Katz & Janice Monk (Eds) Full Circles: geographies of women over the life cycle, pp. 1±26 (London, Routledge). Moodie, T. Dunbar (2001) Black migrant labourers and the vicissitudes of male desire, in: Robert Morrell (Ed.) Changing Men in Southern Africa, pp. 297±316 (London, Zed/Pietermaritzburg, University of Natal Press). Morrell, Robert (2001) The times of change: men and masculinity in South Africa, in: Robert Morrell (Ed.) Changing Men in Southern Africa, pp. 3±37 (London, Zed Books/Pietermaritzburg, University of Natal Press). Morrell, Robert (Ed.) (1998) Of boys and men: masculinity and gender in Southern African studies, Journal of Southern African Studies, 24, pp. 605±630. Morrow, Virginia & Richards, Martin (1996) The ethics of research with children: an overview, Children and Society, 10, pp. 90±105. Moser, Caroline & Mcilwaine, Cathy (1999) Participatory urban appraisal and its application for research on violence, Environment and Urbanisation, 11, pp. 203±226. Moser, Caroline & Mcilwaine, Cathy (2004) Encounters with Violence in Latin America: urban poor perceptions from Colombia and Guatemala (London, Routledge). Mturi, Akim & Moerane, William (2001) Premarital childbearing among adolescents in Lesotho, Journal of Southern African Studies, 27, pp. 259±275. Nye, Robert (1999) Introduction: on why history is so important to an understanding of human sexuality, in: Robert Nye (Ed.) Sexuality, pp. 3±15 (Oxford, Oxford University Press). O'Laughlin, Bridget (1998) Missing men? The debate over rural poverty and women-headed households in Southern Africa, Journal of Peasant Studies, 25, pp. 1±48. Otaala, Barnabas (1998) Children's participation for research and programming in education, health and community development: selected experiences in Africa, in: Johnson, Victoria, Ivan-Smith, Edda, Gordon, Gill, Pridmore, Pat & Scott, Patta (Eds) Stepping Forward: children and young people's participation in the development process, pp. 135±142 (London, Intermediate Technology). Pain, Rachel & Francis, Peter (2003) Re¯ections on participatory research, Area, 35, pp. 46±54. Phillips, Oliver (2004) The invisible presence of homosexuality: implications for HIV/AIDS and rights in Southern Africa, in: Kalipeni, Ezekiel, Craddock, Susan, Oppong, Joseph & Ghosh, Jayati (Eds) HIV and AIDS in Africa: beyond epidemiology, pp. 155±166 (Oxford, Blackwell Publishing). Poku, Nana K. (2001) Africa's Aids crisis in context: `how the poor are dying', Third World Quarterly, 22, pp. 191±204. Potts, Deborah & Marks, Shula (2001) Fertility in Southern Africa: the quiet revolution, Journal of Southern African Studies, 27, pp. 189±205. Pringle, Rosemary (1999) Sex/sexuality, in: Linda McDowell & Joanne P. Sharp (Eds) A Feminist Glossary of Human Geography, pp. 248±251 (London, Arnold). Punch, Samantha (2002) Youth transitions and interdependent adult±child relationships in rural Bolivia, Journal of Rural Studies, 18, pp. 123±133. Rivers, Kim & Aggleton, Peter (1999) Adolescent sexuality, gender and the Hiv epidemic, Undp, Hiv and Development Programme, available at: áwww.undp.org/hiv/publications/gender/adolesce.htmñ. Rivers, Kim & Aggleton, Peter (2002) Working With Young Men to Promote Sexual and Reproductive Health (London, Department for International Development, Safe Passages to Adulthood Programme). Robson, Elsbeth (2001) Interviews worth the tears? Exploring dilemmas of research with young carers in Zimbabwe, Ethics, Place and Environment, 4, pp. 135±142. Roche, Jeremy (1999) Children: rights, participation and citizenship, Childhood, 6, pp. 475±493. Schapera, Isaac (1940) Married Life in an African Tribe (London, Faber and Faber). Skelton, Tracy & Valentine, Gill (Eds) (1998) Cool Places: geographies of youth cultures (London, Routledge). Suggs, David (1987) Female status and role transition in the Tswana life cycle, Ethnology, 26, pp. 107± 120. Downloaded by [Queen Mary, University of London] at 11:13 16 August 2011 512 C. McIlwaine & K. Datta Tallis, Vicci (2002) Gender and HIV/AIDS: overview report, Bridge Report (Brighton Institute of Development Studies, University of Sussex). Timaeus, Ian & Graham, Wendy (1989) Labour circulation, marriage, and fertility in Southern Africa, in: Ron Lesthaeghe (Ed.) Reproduction and Social Organization in Sub-Saharan Africa, pp. 365±400 (California, University of California Press). Townsend, Nicholas (1997) Men, migration and households in Botswana: an exploration of connections over time and space, Journal of Southern African Studies, 23, pp. 405±420. United Nations Development Planning (UNDP) (2000) Botswana Human Development Report 2000: towards an Aids-free generation (Gaborone, Petadco Printing House). United Nations Development Planning (UNDP) (undated) Moving Upstream and Engaging Governments in Strategic Policy Advocacy: the case of UNDP Botswana's HIV & AIDS Programme. Available at: www.undp.orgñ. Upton, Rebecca (2001) `Infertility makes you invisible': gender, health and the negotiation of fertility in Northern Botswana, Journal of Southern African Studies, 27, pp. 349±362. Valentine, Gill (2000) Exploring children and young people's narratives of identity, Geoforum, 31, pp. 257±267. Van Driel, Francien (1996) MarriageÐfrom rule to rarity? Changing gender relations in Botswana, in: Rajni Palriwala & Carla Risseeuw (Eds) Shifting Circles of Support: contextualising gender and kinship in South Asia and Sub-Saharan Africa, pp. 51±78 (New Delhi, Sage; London, Thousand Oaks). Varga, Christine (1997) Sexual decision-making and negotiation in the midst of AIDS: youth in KwaZulu/Natal, South Africa, Health Transition Review, 3, pp. 45±67. Varley, Ann & Blasco, Maribel (2000) Intact or in tatters? Family care of older women and men in urban Mexico, Gender and Development, 8, pp. 47±55. Welbourn, Alice (1999) Gender, participation and Hiv: a positive for change, in: Irene Guijt & Meera Kaul Shah (Eds) The Myth of Community: gender issues in participatory development, pp. 131±140 (London, Intermediate Technology). West, Jackie (1999) (Not) talking about sex: youth, gender and sexuality, Sociological Review, 47, pp. 524±547. Winton, Ailsa (2003) Youth, social capital and social exclusion: examining the well-being of the young urban poor in Guatemala City. PhD thesis, Department of Geography, Queen Mary, University of London. Wojcicki, J.M. (2002) Commercial sex work or ukuphanda? Sex-for-money exchange in Soweto and Hammanskraal area, South Africa, Culture, Medicine and Psychiatry, 26, pp. 339±370. Women's Affairs Department (1994) Report on the National Seminar on the Incorporation of Gender into the Sadc Programme of Action and Preparations of Regional and World Conferences on Women (Gaborone, Government Printer). Wood, Katherine, Maforah, Fidelia & Jewkes, Rachel (1998) `He forced me to love him': putting violence on adolescent sexual health agendas, Social Science and Medicine, 47, pp. 233±242. Young, Lorraine & Barrett, Hazel (2001a) Adapting visual methods: action research with Kampala street children, Area, 33, pp. 141±152. Young, Lorraine & Barrett, Hazel (2001b) Ethics and participation: re¯ections on research with street children, Ethics, Place and Environment, 4, pp. 130±134.