Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Teaching empowerment? Gender, sexuality education and the contested pedagogical relations of knowing… Gacoin, Andrée Elizabeth 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2015_may_gacoin_andree.pdf [ 1.1MB ]
JSON: 24-1.0166112.json
JSON-LD: 24-1.0166112-ld.json
RDF/XML (Pretty): 24-1.0166112-rdf.xml
RDF/JSON: 24-1.0166112-rdf.json
Turtle: 24-1.0166112-turtle.txt
N-Triples: 24-1.0166112-rdf-ntriples.txt
Original Record: 24-1.0166112-source.json
Full Text

Full Text

         Teaching empowerment?  Gender, sexuality education and the contested pedagogical relations of knowing and being known with(in) an HIV prevention programme in South Africa   by   Andrée Elizabeth Gacoin  B.A., Rice University, 2002 M.Sc., The London School of Economics and Political Science, 2005     A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF    DOCTOR OF PHILOSOPHY  in  The Faculty of Graduate and Postdoctoral Studies  (Educational Studies)    THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)      April 2015     © Andrée Elizabeth Gacoin, 2015      ii Abstract This dissertation is an exploration of the im/possibilities of knowing and being known with(in) sexuality education. The project was provoked by how sexuality education is framed as a global strategy to prevent the human immunodeficiency virus (HIV) among youth. In particular, the study aimed to problematize how sexuality education is positioned as a site for youth empowerment in relation to gendered identities and relations.   Through feminist and poststructural readings of ethnographic research (Britzman, 2000; Lather, 2007; St. Pierre & Pillow, 2000; Youdell, 2010), this dissertation engages the pedagogical encounters of sexuality education. The pedagogical encounter (doubled through the research encounter) is theorized as a contested site in which educators and learners engage in the messy and always ongoing work of making sense of their lives with(in) place (Ellsworth, 1997, 2005; Massey, 2005).These encounters take form in South Africa, even as the relations explored within them resist a global/local binary of international guidelines, national programmes or local implementation. Within the contours of these encounters, educators from loveLife, a South African non-governmental organization, meet-up with youth in particular moments. Drawing on three opening propositions related to sexuality education as an always political project, this dissertation foregrounds an analytic shift from who youth are and what is known to how understandings of identities and forms of knowledge become coherent within particular pedagogical moments. This shift draws attention to how pedagogical approaches such as loveLife’s are entangled in power-laden understandings of social identities and a perceived (linear) relation to knowledge. In doing so, it destabilizes the claim that youth can be empowered through sexuality education. Within the problematic imperative to “do” sexuality education differently, already present struggles over identities and forms of knowledge point to the  iii necessity of re-articulating what is claimed in and through sexuality education. This dissertation suggests an articulation of sexuality education in which the vulnerability of knowing and being known might become a condition for responsibility to one another and a site for social transformation.     iv Preface  This dissertation is an original and unpublished piece of work, but I could not have done it without the people who have thought with me, and the places that have welcomed me throughout my doctoral studies. The University of the Western Cape (UWC) Faculty Board Research and Ethics Committees provided ethics approval for research in South Africa. The University of British Columbia (UBC) Behavioural Research Ethics Board provided ethics approval for the project, under the title “Teaching empowerment: Grass-roots educators' strategies to address gender equity in the context of the HIV epidemic in South Africa.” The ethics certificate number is H12-00162.      v Table of contents  Abstract .......................................................................................................................................... ii	  Preface ........................................................................................................................................... iv	  Table of contents ........................................................................................................................... v	  Acknowledgements ..................................................................................................................... vii	  Encounters: Gender and the entangled beginnings of this research .......................................... 1	  Chapter one: An introduction ..................................................................................................... 7	  First proposition: A political project ......................................................................................... 13	  Second proposition: A site for social transformation ............................................................... 21	  Third proposition: The problematic “solution” of empowerment discourse ............................ 25	  Invitation: Encounters ............................................................................................................... 28	  Chapter two: Methodology ........................................................................................................ 31	  Research and the localized context ........................................................................................... 32	  The construction of research contexts .................................................................................. 35	  South Africa as the localized context of this study ................................................................ 38	  The contested context of loveLife .......................................................................................... 40	  The gathered stories and selves of an ethnographic research study .......................................... 44	  Doing ethnography ............................................................................................................... 45	  The gathered stories and selves of this researcher ............................................................... 49	  Entering into ethnographic textwork ........................................................................................ 53	  The work of participant observation ..................................................................................... 53	  The work of the interviews .................................................................................................... 57	  The work of educational artifacts ......................................................................................... 60	  On the work of coding ........................................................................................................... 61	  Working the text .................................................................................................................... 63	  The (im)possibilities of doing differently ................................................................................. 65	  Chapter three: Encountering gender ........................................................................................ 67	  An encounter through definition ............................................................................................... 70	  An encounter within the pedagogical approach ........................................................................ 77	  A cultural encounter .................................................................................................................. 84	  Encountering gender equality with(in) words ........................................................................... 90	  The encounters of gender empowerment .................................................................................. 97	   vi Encounters: A pause on gender as target in HIV and AIDS discourses ................................. 101	  Chapter four: Identity and the facilitation of local lives ....................................................... 104	  Who am I ................................................................................................................................. 111	  An encounter with(in) place ................................................................................................ 111	  Global priorities ................................................................................................................. 116	  Mapping the encounter ....................................................................................................... 121	  Contested national subjects ................................................................................................ 126	  Facilitating local lives ............................................................................................................. 131	  Encountering the self within the pedagogical vision of empowerment .................................. 137	  Encounters: A pause on gender as a topic to teach in South Africa ....................................... 141	  Chapter five: Risky forms of knowledge ................................................................................ 146	  Configuring the “facts” ........................................................................................................... 152	  Excessive facts of violence .................................................................................................. 160	  Configuring risky bodies and their pedagogical relation to knowledge ................................. 164	  Excessive bodies ................................................................................................................. 172	  Configuring space and a pedagogical relation to social difference ........................................ 175	  Excessive vulnerabilities within space ................................................................................ 179	  The excessive configurations of knowing empowerment ....................................................... 181	  Chapter six: Re-encountering gender for a feminist politics with(in) sexuality education 187	  Concluding encounters with sexuality education ................................................................... 189	  A political rationality of vulnerability ................................................................................ 189	  Social transformation and a pedagogical relation in the making ...................................... 191	  Productive impossibilities of empowerment discourse ....................................................... 193	  The contributions of this research project ............................................................................... 195	  Research limitations and the limitations of research .............................................................. 198	  The question of future research directions .............................................................................. 201	  A return to the invitation ......................................................................................................... 202	  References .................................................................................................................................. 203	  Appendices ................................................................................................................................. 225	  Appendix A: Individual semi-structured interview protocol .................................................. 225	  Appendix B: Project overview handout .................................................................................. 227	      vii Acknowledgements  My learning in this dissertation has been formed through a pedagogical encounter in which many people and places have guided/provoked/challenged/questioned/supported me  to work at my own limits of knowing and being known.   To the participants of this study Thank you for welcoming me into your work, for sharing your time with me, for gently poking fun at my ingrained habits and bad dance moves, and for trusting me with versions of the worlds you are living. Thank you also to the staff at loveLife who allowed me to conduct this research with the organization.  To places of learning The intellectual communities that I have been welcomed into at the University of British Columbia are formed within the ancestral, traditional and unceded territories of Musqueam First Nation. I am also responsible to the communities that I have encountered through research in South Africa, even as my encounters with(in) those communities have provoked my re-thinking of what it means to be responsible to one another in this globalized world.   To my supervisory committee – Mona Gleason, Lisa Loutzenheiser, James Lees Thank you for your guidance and support through all stages of this project. Mona – your attention to analytic detail and historical depth has been an invaluable contribution to this project, as well as your patience with me as I’ve disappeared into analytic rabbit holes. Lisa – you have continually provoked me to try to think differently, and to encounter learning with(in) the struggles of knowing. Jim (at the University of the Western Cape) - through our conversations you have gently reminded me about the effects of power on educators and youth in South Africa and have guided me to articulate the political stakes of this project. I am also grateful to the professors at UBC who have invited me to engage with educational theories and research methods, and in particular to Claudia Ruitenberg who provided guidance for on theorizing empowerment discourse.   To my family My mother, Mary Holman – thank you for your unwavering support of all of my adventures (as well all your writing advice and pep-talks over skype), no matter how far from home they have taken me. My partner, Gareth Garlick – thank you for believing in me even when the journey of the PhD has seemed without end; I could not have gotten here without your love and your ability to take the piss out of everything (including qualitative research). Meryl and Jack Garlick – thank you for welcoming me to your family and for giving me a home in South Africa. My son, Thomas – your curiosity about the world inspires me to continue working at the limits of who youth are said to be and what they are said to know. My grandfather, John T. Holman – memories of you are entangled in the selves I try to be.      viii To friends To friends who have encouraged me over wine, hikes, cooking nights, bush-taxi rides and 5pm movie dates: Andrea Krüsi, Selina Palm, Jeannie Kerr, Bieke Gills, Karen Attyah, Jacqui Latimer, Karen Martin. To friends like family who are willing to talk about discursive formations of identity and accept sushi in exchange for babysitting: David Butler, Doug MacDougall, Heather Evans. To friends and colleagues at UBC who have asked questions about my work that make me want to take that work further: Isabeau Iqbal, Wendy Hartford, Roselynn Verwoord, Hanae Tsukada, Kal Heer, Paulina Semenec, Sam Stiegler, Rachael Sullivan, Joyce Schneider, Autumn Knowlton, Joel Heng Hartse. Thank you.   I am also grateful for the financial support provided by the Canadian Institutes for Health Research (CIHR) as a Vanier Canada Graduate Scholar, as well as the Department of Educational Studies and the Liu Institute for Global Issues at the University of British Columbia.  1 Encounters: Gender and the entangled beginnings of this research  I begin with a story that goes something like this (told in every space for a particular audience, here for the reader of this dissertation): After I graduated from college I went to Gabon as a United States Peace Corps volunteer. I still remember getting the placement letter and sitting outside my residential college in Houston, Texas, wondering where on earth Gabon was. I was to become a “community health volunteer,” a position that only required three months of on-site training, proficiency in French, and, as I would later recognize it, a strong dose of liberal American idealism. After those three months of training I was dropped off in Lastoursville, a small town of about 8,000 people where the only white people were myself and a group of aging French men who worked for a forestry company. Later, that group of white faces grew by one when the South African man who would become my partner, and the father of my now two year-old son, was temporarily stationed in the region as part of a geological exploration project. He had heard about me before meeting me – the white girl who runs with her dog. In this place I would call home for two years, my stories of belonging and not belonging were already entangled in ongoing “imperial formations” of claims on land, bodies and knowledge (Stoler, 2008).   My job as a community health volunteer started in the hospital where I armed myself with flipchart paper and markers and lectured mothers on how to breastfeed their babies. The vacant, if polite, stares were entangled in what I can now name as a “mode of address” (Ellsworth, 1997), an analytic concept that speaks to who educational subjects are taken to be within multiple relations of power and knowledge. Who did I take these women to be if I, a white American woman with no children, could stand in front of them and tell them what is best for their children? Who did I take myself to be? As our conceptions of self and Other met in this  2 space, what relations of power were already operating through the forms of knowledge I offered them? Faced with my own discomfort, my solution at the time was to turn to the schools. Surely I could teach youth. I had my own experiences of youth behind me, as well as my learned histories of a “good” education. Additionally, I had received training on how to make sexuality education messages “participatory” and “interactive.” I pushed ahead, confident that this was all sexuality education needed.   The Peace Corps provided a “Life Skills” curriculum (Peace Corps, 2001), full of “facts” to learn. In this study I will explore these facts as forms of “configured knowledge” (Ellsworth, 2005): knowledge, formed within multiple relations of power, that has the appearance of being fixed into something “already made” (p. 5). The manual was part of sexuality education as an HIV1 prevention strategy, and one of the topics was “gender,” with a specific focus on gender roles and expectations. I carefully prepared a lesson based on the provided script of gender as “social” and sex as “biological.” Those were the “facts” that I needed to get across, and they operated within a particular framing of “gender empowerment” that linked gender equality with improved health outcomes for both men and women. In other words, the explicit aim of the lesson was to challenge gender stereotypes and promote gender equality in decision-making related to sexual behaviours, such as young women being able to negotiate for condom use. Looking back at the session, my memory has blurred the times and places where these conversations took place – from the Catholic middle school to the porch of my house where a small group of peer educators met in the late afternoon heat of the rainforest. What I do remember clearly is how unprepared I was for the lingering feeling of failure of the sessions. I                                                 1 HIV stands for the Human Immunodeficiency Virus. It is the virus that causes AIDS (Acquired Immunodeficiency Syndrome).  3 sensed, and still believe, that the conversations around gendered identities were important. However, these conversations could not be contained within the neat boxes of the session – clear identities, social roles, equal relationships. There was an “excess” (Orner, Miller, & Ellsworth, 1996), another analytic concept that I will use in this study, which refused to stay within the discursive norms of the “facts.” Here, this excess operated in the space between the curriculum and the complexity that my students and I were living as raced, classed and gendered sexual subjects in particular places and times: talking with my students on the 3km walk from the middle school to town, red dust kicked up by the forestry workers’ Toyota Landcruisers, the girls saying as the older white men drove past: they only stop when they want to sleep with us; teaching the “good” peer educator who nodded as I spoke about abstinence and did not say she already had two young sons; listening to the neighbours who told me to be careful about who I let into my house: (black) men; people will talk. Gender mattered, but how it mattered was outside of what I thought I could “teach” as part of a neatly scripted curriculum.   Across these fragmentary memories, how I made sense of gender was entangled in my own understandings of race, class and sexuality and who was targeted as the subject of sexuality education messages in this town in post-colonial French-speaking Africa. I was labeled as an expert, even as I tried to downplay that expertise through what I considered to be a participatory pedagogical approach. Gender mattered within a specific place and time that my own commitments to particular ways of knowing and being did not, and perhaps could not, allow; a place of encounters, and missed encounters, of multiple relations of power and knowledge (Massey, 2005). These ways of knowing were, and remain, embedded in how I make sense of my own identities; as a white researcher working with black youth, as the mother of a white South African and American son, as an educator who feels responsible to “youth” who have  4 always already escaped my desire to know them. Working with these ways of knowing, and being known, in relation to notions of gender has become a pivotal part of my work.   This story is not the proper subject of this dissertation but I keep returning to it. I have returned to it with my committee when I proposed the study, and with the educators at loveLife, the South African organization I worked with for this study. As I’ve told versions of this story, in those different places and times over the course of my academic studies, I have been struck by a sense of recognition from educators when I’ve spoken about a “something else” within neat scripts of gender, a something that does not fit received notions of gender or messages of gender equality. That something else has been located in the experience of the story, in excess of my carefully prepared academic references and arguments. This is not to equate my and the educators’ experiences, or to romanticize our struggles into some new road map for “what will work” in discourses related to sexuality education as an HIV prevention strategy. Instead I have used that recognition to ask what it might mean to take that something else seriously, and to think outside the “problem” of gender as it circulates within those discourses.  I begin with this story as an always fragmented tale of how I have come to this research, an initial encounter with “gender” as something to script into a sexuality education curriculum and learn about through educational research and practice. I also pause on the story before it settles into the structure of an academic dissertation where the imperative of analytic arguments inevitably solidifies the “data”– words making sense of the world, fragments of experiences, texts I think with and through – into what I claim to know. Here I am reminded of a passage from another story, Barbara Kingsolver’s (1998) novel the Poisonwood Bible:  And you’ll say I did. You’ll say I walked across Africa with my wrists unshackled, and now I am one more soul walking free in a white skin, wearing some thread of stolen goods: cotton, or diamonds, freedom at the very least, prosperity. Some of us know how  5 we came by our fortune, and some of us don’t, but we wear it all the same. There’s only one question worth asking now: how do we aim to live it? (pp. 9-10)   Kingsolver’s question points to a key methodological challenge that I’ve struggled with in this study, as will be explored in Chapter 2: how do I act while simultaneously working with and within the implications of that power-laden action (St. Pierre & Pillow, 2000)? As a researcher, the thread of stolen goods that I carry are formed in and through the positions I have the privilege to occupy (white, American, heterosexual, cisgender) and the knowledge claims that I will make from these positions. For example, the pedagogical approaches that I critique are the ones with which I feel the most comfortable, resonating with years of my own schooling and an ingrained sense of what education “should” be, both in terms of what can/should be taught as well as what constitutes “good” teachers and students. Likewise, the practices of international development that I seek to problematize, such as telling youth what to do, are ones that I have slipped into as I’ve worked with non-governmental organizations. Even as I designed my doctoral project with an explicit commitment to the complexity of power-relations, multiple relations of power haunt the frustration that pervaded my fieldnotes when things did not go as (I had) planned. Beginning from the position that all action is power-laden, part of my inquiry has been to engage my own resistance to particular ways of “knowing” or “doing” as part of the “tangles of implication” (Britzman, 1997) in educational practice and research.  As part of these implications, I cannot, and do not seek to, explain away my position as a white North American researcher working with black and coloured educators and youth in South Africa. The fictive solidity of these identities - black, coloured and white – is a legacy of apartheid’s violent classification of bodies and is maintained through the supposedly neutral descriptive categories used in statistics related to HIV incidence and prevalence (e.g. Human  6 Sciences Research Council, 2014).2 These identities are held separate from the “problem” of gender, even as they are implicit in articulations of who is gender a “problem” for and who offers the “solution.” It is precisely these implications, and their effects, that I argue makes re-articulating what can be thought of as sexuality education as an HIV prevention strategy a crucial political task. Around the world a parade of experts are already telling youth who they are and what they should know. Those same experts then implicitly blame youth (through reports of HIV infection rates, or evidence of unequal gendered power relations) when the “solution” of sexuality education does not live up to its promises. I have been one of those experts, and engaging my positions in this research offers one way to begin to destabilize the supposed neutrality of the “solution” being offered to youth in South Africa through sexuality education.  In this initial encounter with what brought me to this research project, as throughout this dissertation, I am inevitably caught in the tension of arguing that the “solution” of sexuality education is impossible at the same time as I slip into offering solutions to its problems. However, I do not think that this tension can, or ultimately should, be resolved. This tension is part of the political necessity of engaging with the claims made in and through sexuality education in this particular moment. I suggest that it is the impossibility of any “solution” to the dilemmas of knowing and being known that might allow a different articulation of sexuality education: sexuality education as an encounter with(in) contested and power-laden understandings of social identities and their relations, and an invitation towards the possibilities of knowing and being known in ways that are more just for more people.                                                   2 These categories are legacies of apartheid in South Africa, and continue to be used as dominant identity markers, as will be discussed in detail in Chapter 3. I have chosen to use these terms in lower-case in order to mark how they may come to be seen as discrete identities, which carry power-laden effects of those who bear them, while simultaneously resisting an essentialized “Truth” of those identity categories.  7 Chapter one: An introduction  The topic of the debate is teenage pregnancy. There are two groups: the ladies and the guys. The educator stands between the two groups coaxing answers from learners who seem reluctant to participate at first. Everyone speaks in Xhosa and I grasp at meanings from stray English words, watching the learners’ bodies as the debate gains momentum. One group clamours to respond to what the other group said, hands in the air, faces of outrage, peals of laughter, bodies pressed together or hiding behind one another as things seem to get personal. The educator tries to keep the groups quiet, calling on one person to speak at a time. (Fieldnotes, May 2013)   I was observing this debate as part of the fieldwork for my doctoral research project: an ethnographic study that I worked on from 2012 to 2013 in Cape Town, South Africa, with loveLife. loveLife is a non-governmental organization that describes itself as “South Africa’s largest national HIV prevention initiative for young people” (loveLife, 2014a). As I watched the debate, and as I talked with the educator afterwards, I initially engaged this encounter through the three research questions that I posed in framing this study. These research questions were: (1) Within the context of sexuality education in South Africa, how are gender, gender relations and empowerment articulated by educators and youth?; (2) How is meaning making around gendered identities contingent on understandings of other social identities, and in particular race and ethnicity in the context of South Africa?3; and (3) How does the variety of pedagogical approaches to sexuality education in South Africa potentially open up and foreclose expanded notions of gender, gender relations and empowerment? This debate spoke to the first question,                                                 3 In designing the research questions, I foregrounded “race and ethnicity” in order to trouble what I perceived to be a color-blind articulation of “gender empowerment” and explicitly engage with how understandings of “race” continue to underpin notions of identity in a post-apartheid South Africa (Soudien, 2009). While I did not foreground class (in terms of neo-Marxist analysis focused on power relations within a Capitalist system, for example) it is impossible to ignore notions of class when speaking of race or gender in South Africa. In the chapters that follow, I engage “class” in terms of how it is entangled with these other social identity constructions, and the implications for understandings of complex subjectivities.   8 for example, when the learners4 articulated gendered expectations in relation to sexual behaviours – such as a guy expecting sex or having sex to be cool. Turning to the second question, race was not explicitly mentioned in the discussion, but understandings of race were already operating within the implicit focus on cultural expectations for men or women. For example, during the debate one girl said teenage pregnancy should not be terminated because the baby might be the one and only. Her comment pointed to motherhood in South Africa as itself a contested identity where racialized understandings of femininity are entangled in complex and contested economic, social and political processes (Moore, 2013). Drawing from these two readings, the third research question draws attention to how an interactive pedagogical approach (the debate activity) provided a space within which learners explicitly spoke about gendered relations in the context of expectations around teenage pregnancy.   At the same time, within the place of this debate as well as the readings that I could offer, I have come to be troubled by the potential linearity of these questions and its effects. The pedagogical approach was not just about expanded notions of gender (the concern of question 3) but also a constitutive part of their articulation (from question 1). For example, the debate activity addressed learners as already embodying particular gendered subject positions – the guys and the ladies – and then expected them to speak from those positions located in particular (cultural) places. This, in turn, points to a problematic relation to “place” already operating through the questions. Each question can be asked, and answered, differently within a particular place as context (e.g. where it took place). However, when place is conceptualized as a “collection” of “stories-so-far” (Massey, 2005, p. 130), these questions, and my body and                                                 4 Throughout this dissertation I use the term “learners” (instead of “students,” for example) in line with South African educational discourse.     9 identities as the researcher asking them, is part of the constitution of the places that I speak about. For example, the reading that motherhood is a central part of a racialized femininity is located in a place (a black Xhosa speaking township) at the same time as multiple relations of power and knowledge, such as the imperial legacies of the figure of “the subordinate, child bearing African woman” (Reid & Walker, 2005, p. 186), are part of the stories that I as a white North American woman bring to that place. Furthermore, that place was a meeting up of bodies that did not seem to “fit” the research questions: bodies that moved, that clamoured, that hid behind one another. I was confronted with moments of teaching and learning that “never takes place in the absence of bodies, emotions, place, time, sound, image, self-experience, history” (Ellsworth, 2005, p. 55), and which were already exceeding the research questions that I thought I was asking.   In attending to this excess, I am not arguing that I started with the “wrong” research questions. Research questions are a necessary part of making sense within the discursive norms of qualitative research, including a doctoral research study. They are hooks for articulating something that is understandable for the researcher and invite something to be “known” (no matter how partial that knowing is). These questions have guided my analytic focus, even as they have been undone by that focus. Indeed it is the very limitations of these questions, and my attempts to think at those limits, that have given form to a different way of thinking through the arguments of this dissertation. In the rest of this chapter, I lay out three opening propositions on sexuality education as an HIV prevention strategy, propositions that I weave through the chapters and then return to in the concluding chapter. These are (i) sexuality education is always a political project; (ii) sexuality education as a potential site for social transformation necessitates engaging with(in) the complexity of its pedagogical encounters; and (iii) empowerment discourse in sexuality  10 education as an HIV prevention strategy risks shutting down, rather than opening up, the pedagogical encounter as a site for social transformation. As propositions, these statements carry with them an analytic certainty that is in tension with the project I have laid out so far in this introduction. However, rather than an impasse, I argue that this tension is an inevitable part of engaging with sexuality education as an always political project. These propositions explicitly foreground my own commitments to knowledge, and the possibilities of sexuality education that I can articulate with them (as will be discussed in the rest of this chapter). At the same time, the propositions provide a frame within which to engage all claims to knowing and being known as momentary effects of particular configurations of power. In other words, sexuality education is always a “problem:” a site of struggle over who is known and what is knowable. This struggle reaches across, and therefore necessitates attention to, what are often taken to be distinct spatial registers (such as “global” policies and “local” programmes) and temporal imaginaries (such as the injustices of the past and the promises made for the future). As an analytic frame, these propositions are an invitation into struggles over identities and forms of knowledge within the pedagogical encounters of sexuality education.  Before turning to the propositions, I pause on a few of the terms that I am using (specifically: pedagogical encounter, gender, and social transformation). I have come to the term pedagogical encounter by reading the work of Elizabeth Ellsworth (2005) alongside that of Doreen Massey (2005). Ellsworth speaks of pedagogy, and the experience of learning, as an “event in time” (p. 4) that can be engaged from multiple angles. Massey, with an analytic focus on space, describes the coming together of knowledge and understandings of identities within a particular place as “encounters with the apparently familiar but where something continues to trouble, and unexpected lines of thought slowly unwind” (p. 6). Linguistically bringing these two  11 lines of thought together, I use the notion of the pedagogical encounter as a way to momentarily hold analysis in particular time and place (a classroom, for example) at the same time as I engage with how that encounter bears the traces of what cannot be contained within it. This includes traces gestured to in the very notion of “encounter”, such as the colonial encounter or the ethnographic encounter with an exotic Other. In other words, the pedagogical encounter (doubled through the research encounter) is a “simultaneity of stories-so-far” (Massey, 2005, p. 12): stories educators, learners and myself as researcher tell as we engage in the messy and always ongoing work of making sense of our lives within contested relations of power and knowledge. As an always momentary and fragmentary form, the pedagogical encounter cannot offer a “solution” to those relations or fix them into something ultimately knowable. Instead, the encounter is a site within which research may engage with a particular coming together of contested relations amongst power, identity and knowledge that have already escaped the perceived boundaries of that encounter.  As will be explored throughout this dissertation, notions5 of gender as a social identity construction are entangled in those encounters. I understand gender as a becoming (Butler, 2004) whereby people are made understandable as particular gendered identities – such as man, woman, boy, girl – based on norms that operate as the discursive boundaries of those categories. This is what Butler (2004) terms “gender performativity.” The lens of gender performativity works against a view that people can freely choose to embody particular identities, such as by rationally challenging gender stereotypes during a sexuality education lesson. Indeed, dominant norms are made “natural” only by “occluding the ways in which they are performatively established”                                                 5 I use the term “notion” as a way to mark the multiple (and potentially contradictory) understandings held within any term.   12 (Butler, 2004, p. 209). This destabilizes the claim that the “right” pedagogical approach can “solve” the “problem” of unequal gender power relations. Instead, and as will be explored through the encounters of this study, targeting gender identity as a “problem” risks reifying, rather than challenging, understandings of what gender already “is.” As such the question is not just what understandings of gender are articulated in a particular moment (as per my original research questions), but also how the pedagogical approach is implicated in what understandings of gender come to be coherent with(in) sexuality education as an HIV prevention strategy.  Asking how sexuality education might be articulated as a site within which understandings of gender and their relations might be more just for more people is part of a political act committed to social transformation. At first, this might not seem so different from the claim that sexuality education can teach about gendered identities and roles and promote gender equality. However, the shift is from framing teaching about gender roles as a solution to considering how that teaching is a problematic site. Following Butler (2004), a feminist politics committed to the “social transformation of gender relations” (p. 204) is not just about making already knowable gendered subjects (such as particular articulations of “man” and “woman”) and their relations more equitable. It is about working at the limits of the discursive norms of gender, asking how, and at what expense, particular articulations of gendered identities become intelligible. This intelligibility is crucial for having what Butler terms a “livable life” (p. 224): a life that embodies particular norms, such as understandings of masculinity and femininity, which in turn make that life intelligible (and by extension valuable) in specific discursive contexts. Lives that do not fit with these understandings are positioned outside of what is considered “real” in terms of the categories that are needed to make sense of the world. Within the discursive context of sexuality education as a global HIV prevention strategy, this raises disturbing  13 questions related to whose lives are made to count within educational approaches that take gender identity as a universal and knowable target and goal. While the explicit goal is to save lives (HIV prevention), sexuality education risks becoming complicit in sacrificing lives that fall outside the discursive limits of who youth are already said to be as well as what they are already said to know. This is the risk, I argue, that necessitates explicitly engaging with the educational claims and promises of sexuality education as an HIV prevention strategy.  First proposition: A political project  The first proposition guiding this study is that sexuality education is always a political project. This may seem overly obvious. However, as an HIV prevention strategy, the political nature of sexuality education is often masked. In South Africa (Francis, 2010), as well as globally (UNESCO, 2009a), sexuality education is framed as one programmatic strategy for achieving HIV prevention among youth. Drawing on an ideology of “good health for all” (Lupton, 1995, p. 2), itself a profoundly political claim, both public health and educational research have predominantly been concerned with what content should be taught and what is the effectiveness of that content in terms of supposedly neutral health outcomes (Aggleton & Crewe, 2005; Baxen, 2010). However, those outcomes are already entangled in multiple relations of power and knowledge, such as whose knowledge counts as “content,” and what kinds of subjects (defined through their age) are created, and potentially disciplined, in measures of “effectiveness.” Throughout this dissertation, I argue that making these political claims explicit is crucial for problematizing how “gender” has been taken up in sexuality education as an HIV prevention strategy for youth. For example, gendered power relations have been identified as a type of content to include in sexuality education (Bhana & Pattman, 2009). The premise is that messages related to gender equality can be scripted into sexuality education curricula (Senderowitz &  14 Kirby, 2006) and then taught to youth in a range of contexts (e.g. UNESCO, 2009b). The effectiveness of this content can then be measured in terms of (and reduced to) HIV prevention outcomes, such as the increased uptake of pre-defined prevention methods. Problematizing these claims, which are given the illusion of neutrality through a scientific discourse of prevention (Foucault, 1998), turns attention to the “imaginaries of gender, sexuality and relationships” being offered to youth through sexuality education (Lesko, 2010, p. 282). As such, the question is no longer what should be taught, or how effective that content is in terms of pre-defined outcomes, but rather how sexuality education is already entangled in relations of power and knowledge that settle into, and have lived effects for, conceptions of self, Other and their relations. By thinking outside of sexuality education as a “solution” that requires the “active implementation of stable knowledge and rationality” (p. 282), I ask how sexuality education might be articulated in ways that take knowing and being known as ongoing struggles (and will I turn to the educational possibilities within these struggles through the second proposition).   In this dissertation, I foreground the political nature of sexuality education by defining sexuality education (and indeed all education) as a site where what is said to be known, and who knows, is formed through multiple relations of power and knowledge. I purposefully use the term sexuality education rather than sex education. On the one hand, the phrasing references discourses related to HIV prevention in South Africa and globally (Department of Education, 2002; Francis, 2010; Soudien, 2009; UNESCO, 2011). On the other hand, this phrasing explicitly draws attention to education as a technique of power, as well as a site of potential resistance to power relations, related to notions of sexuality. Following Foucault (1998), sexuality has historically been, and continues to be, an “especially dense transfer point for relations of power” (p. 103). For example, colonial discourses deployed particular understandings of sexuality to  15 both mobilize a “local” sexual subject (that could then be targeted with health education messages) and simultaneously mask how those understandings relied upon and reified power-laden constructions of age, race, class and gender (Stoler, 1995; Vaughan, 1991). Those constructions continue today as part of the ongoing imperial formations of international development discourses (Humphreys, Undie, & Dunne, 2008). This includes notions of youth, where youth is understood as a social identity category. As will be discussed in depth in Chapter 4, constructions of age are embedded in a developmental discourse in which adults must guide youth (seen as out of control, irresponsible, irrational) to become proper subjects (Lesko, 2001). A brief discussion of different models of sexuality education can help to ground this starting point.  Models of sexuality education have been based on two underlying, and interlinked, conceptualizations: first, the way in which sexuality is understood (Britzman, 1998) and, second, how the development of youth is conceptualized (Gilbert, 2007). In the North American context, this has led to two dominant models of sexuality education: “the normal” and “the critical” (Britzman, 1998, p. 66). While this analysis cannot be simply extended to the South African context, educational responses to HIV in South Africa are deeply embedded in North American readings of the epidemic (Baxen, 2010; Lees, 2008). Indeed, the mobility of these understandings can be seen as part of the power relations, and their implicit political aims, that move across what are taken to be “local” or “global” contexts.  Within the first model, a “normal” paradigm, sexuality is broadly conceptualized as a “natural” part of human development, and youth consists of a series of biological or psychological stages (Britzman, 1998). As such, the purpose of education is to transmit supposedly neutral information to young people in order to support the development of what is  16 taken to be a normative sexuality. This approach dominated early debates around sex education in the United States (e.g. Carter, 2001), and has been strongly critiqued for relying on eugenic and racist notions of human development (Britzman, 1998; Glaser, 2005; Lesko, 2001). In contrast, and in reaction to the “normal” paradigm, a “critical” paradigm of sexuality education draws on social constructionist views of sexuality and youth, whereby understandings of sexuality and youth emerge as “an effect of discourse, history, culture and the social” (Gilbert, 2007, p. 49). Here, sexuality is conceptualized as an identity, rather than as a biological given. Under this paradigm, sexuality education has the potential to either reproduce normative ideas of sexual identities and behaviours, or provide a space for youth to challenge those constructions and engage in conversations around “who” they are within relations of power (e.g. Campbell & MacPhail, 2002; Fields, 2008; Pattman, 2005).    While these models and their political projects may appear distinct, in terms of whether they propagate a particular construction of “normal” or whether they critically engage with social norms that script what “normal” is seen as, I argue that the HIV and AIDS epidemics have given form to a particular moment in which these models are yoked together, particularly as attention has turned to the “problem” of gender for HIV prevention outcomes. Broadly, HIV discourses deploy a preventative model of sexuality education that in and of itself is not “new” (Britzman, 1998). Historically, debates around sexuality education for youth have coalesced around outbreaks (whether real or perceived) of disease: from concerns around venereal disease in the United States and Canada following World War I (McLaren, 1999) and in Canada in the 1940s (Adams, 1997), to the reading of the threat of disease on racialized bodies in South Africa in the 1930s and 1940s (Jeeves, 2001) and in Colonial East and West Africa between 1890 and 1950  17 (Vaughan, 1991).6 Across these histories and places, the preventative power of sexuality education has been deeply embedded in relations of power and knowledge: who is targeted as the subject of sexuality education; what forms of knowledge they are said to need; and how these identities and forms of knowledge are deployed across space and time. While gendered identities and relations have been at issue across these histories, what is “new” is how the current articulation of sexuality education as prevention operates within a discursive tightening around gender that relies simultaneously on normative and critical models of sexuality education. Drawing on a normative model, there is such a thing as neutral and universal information about “gender” that can be transmitted from teacher to youth. “Cultural relevance,” or “local adaptation,” is a matter of “fit:” making sure that activities designed to transmit a pre-defined body of content are “sensitive to community values” for example (UNESCO, 2009a, p. 19). This is coupled with a critical model by simultaneously positioning sexuality education as a site within which all youth will rationally engage with gendered norms and act for more equitable gender relations. Indeed, the ideologies of participation and challenging power-relations (from a critical model) can be seen as crucial for masking power relations that allow “gender” to be defined and targeted as a problem in the first place. This is an example of how “participation” in development discourse can operate as a form of de-politicization by restricting involvement to a “participatory process” with pre-defined outcomes (Leal, 2007). In other words, words such as “participation” (as well as “empowerment”) are particularly problematic in that they mark a commitment to “transforming development’s relationships” (Cornwall, 2007, p. 475) at the same                                                 6 I am engaging these moments as forms of sexuality education, even if they were not called that per se at the time. For example, educational efforts to address venereal disease in the US in the early 20th century were understood as “social hygiene” (McLaren, 1999).     18 time as they operate as catch-all terms that fail to challenge, and indeed may mask, the ongoing neo-liberal relations of international development. Following Brown (2003), neo-liberalism is a “political rationality” that extends a market rationale into “all aspects of thought and activity” (para. 19). This includes a mode of governmentality of subjects whereby the very “freedom” of the subject to act becomes their regulation, based on the “moralization of the consequences of this freedom” (emphasis original, para. 17). Here, the issue is not just that gender identities are being targeted in the service of HIV prevention outcomes, or that a particular articulation of gender relations are being held up as the “solution” to the dilemmas of HIV prevention. What is also at stake is how those configurations of gendered identities and relations may be seen as constitutive parts of a political rationality in which youth must “act” in particular ways in order for their lives to be seen as valuable.   To problematize this political rationality, I begin by reading the conflation of “normative” and “critical” models of sexuality education, as well as the silencing of their always political nature, as an effect of an analytic focus on “what we know” about sexuality, youth or HIV prevention instead of “how we come to know what we perceive to know” (Baxen & Breidlid, 2004, p. 11). To shift this focus, and in turn attend to the potential effects those ways of knowing and being known have on the subjects they target, I begin by taking sexuality education as a site within which social identity constructions (such as gender, race and class) and their power-laden relations come to be known as such. This departs from both “normal” and the “critical” paradigms of sexuality education by rejecting the view that identity (in terms of how we see ourselves as subjects) comes from an internal True self. Instead, and building from the view that “human beings are made subjects” (Foucault, 1982, p. 777), identity constructions are seen as produced through what is taken as truth. Identities are “discursive fictions” that are momentarily  19 solidified into what are seen as distinct categories in the “materiality of everyday life” (Loutzenheiser, 2005, p. 28). The implication of this conceptualization is that identity can never be a stable target for sexuality education whether the goal is to transmit that identity (as in a “normal” paradigm) or to critically challenge and transform that identity towards a specific goal (as in a “critical” paradigm). It is also impossible to neatly separate “global” goals from “local” implementation. Sexuality education itself is part of the multiple and “larger social rituals of interpellation” (Butler, 1997a, p. 26), or the calling of subjects that creates “taken-for-granted” subject positions in education (Ruitenberg, 2007), that make it possible for gendered subjects to act in recognizable ways. This action always happens within multiple and contested relations of power and knowledge.   I use this reading of identity to problematize (and then try to work at the limits of) two ways that “gender” has taken form in sexuality education as an HIV prevention strategy. Firstly, as discussed above, I question how gender is framed as a teachable topic, such as learning about gender roles and expectations in order to challenge them. Here, my interest is what might happen within sexuality education when there is a shift from identity as a target for the pedagogical encounter to identity as a continual site of struggle within the pedagogical encounter. Secondly, I use this theoretical conception of identity to explore how understandings of social identities are simultaneously irreducible and inseparable within relations of power and knowledge. In line with a critical paradigm of sexuality education, a dominant way of reading the complexity of identity and its relations of power and knowledge has been to focus on “intersectionality,” broadly understood as the “relationships among multiple dimensions and modalities of social relations and subject formations” (McCall, 2005, p. 1771). Feminist work, and particularly black feminist theory, has used intersectionality to respond to critiques that gender identity is considered at the  20 expense of other social identities, including race (e.g. Collins, 2000). However, intersectionality has been critiqued for becoming an additive list of fixed identity constructions that consider only certain identities at the expense of others (Erel, Haritaworn, Rodriguez, & Klesse, 2010). This risk arises, in my view, from the analytic focus on the intersections between identities after they have “become,” rather than as entangled in their very becoming. In other words, “race” is already a constitutive part of how “gender” is being understood within sexuality education as an HIV prevention strategy. By drawing attention to other social identities marked within vague references to “culture” and making something “locally relevant,” I seek to engage always political struggles over knowing and being known that are already happening within sexuality education.   Taking the pedagogical encounter as a site where the entanglements of identity formation within relations of power and knowledge may momentarily be glimpsed, I will argue that these entanglements are precisely what necessitate different articulations of sexuality education as a political project. What is at stake here is not just how sexuality education is “done,” but also how sexuality education conceives the relations of knowing and being known. For example, “race” can be added into a curriculum focused on “gender,” and this may draw attention to the power-laden effects of racism in people’s lives. The risk of taking this addition as a “solution,” however, is that discursive fictions of identity may be reified at the expense of an explicit consideration of how they come to be seen as such. It may also continue to work within, rather than disrupt, a neo-liberal political rationality within which action and responsibility are tethered to an individual (empowered) identity that is discursively removed from the conditions of its formation. The question is what might happen within a political rationality that invited sexuality education  21 to meet, rather than foreclose, power-laden entanglements of knowing and being known already at play within it. This leads me to the second opening proposition of this study.  Second proposition: A site for social transformation   In this study, I work from and develop the view that sexuality education as a potential site for social transformation necessitates engaging with(in) the complexity of its pedagogical encounters. This proposition is grounded in a feminist reading of pedagogy, and how pedagogy is researched, that is concerned with relations of power and knowledge as they take form in and through what come to be seen as educational subjects (such as teacher or learner) or objects (such as the curricula) (Britzman, 1998; Ellsworth, 1992, 2005; Gore, 1992; Lather, 2007; Talburt, 2004b; Youdell, 2006a, 2006b). My thinking has been particularly influenced by feminist readings of, and departures from, a field that is broadly called critical pedagogy. Within this field, pedagogy is conceptualized as the transformation of the consciousness of learners through the project of education (e.g. Apple, 2004; Freire, 2007; Giroux & McLaren, 1986). This conceptualization has been important for destabilizing a view of pedagogy as a scientific and linear process involving the transmission of knowledge from teacher to learner (Bobbitt, 2004; Tyler, 2004) as well as for drawing attention to pedagogy as an explicitly political act. For example, research within this paradigm has highlighted how sexuality education may be engaged in ways that reproduce gendered norms and fail to meet the needs of gendered subjects (e.g. Fine, 1988; Holland, Ramazanoglu, Sharpe, & Thomson, 1998). This approach has also foregrounded youth perspectives on sexuality and their own gendered needs (e.g. Allen, 2008; Muhanguzi, 2011; Pattman, 2005) and made recommendations for how the pedagogical approach can be more “gender-sensitive” (e.g. McNamara, Geary, & Jourdan, 2011) or “gender-transformative”(e.g. Campbell & MacPhail, 2002). However, while critical pedagogy, including  22 feminist articulations of critical pedagogy (e.g. Fine, 1988; hooks, 1994), has explicitly attended to power-relations that can be transformed through education, there are a number of critiques that draw attention to how a critical paradigm may ignore, and itself potentially reify, power relations within the pedagogical encounter (e.g. Ellsworth, 1992; Gore, 1992, 1998; Lather, 2001).  Broadly, feminist critiques of critical pedagogy have highlighted how critical pedagogy relies upon an underlying conceptualization of teachers and learners as rational subjects, and in doing so risks reifying, rather than challenging, power relations between learners as well as between teacher and learner (Ellsworth, 1992; Maher, 2001). Furthermore, particular forms of knowledge are given an implicitly positive value (such as through the notion of being “empowering”). This positioning risks masking how any form of knowledge is always embedded in power relations, such as what learners are said to “need” in terms of knowledge and who has determined these needs (Ellsworth, 1992). In turn, the pedagogical approach (whether critical or otherwise) is always embedded within a particular historical and political context and multiple power relations (Ellsworth, 1992; Gore, 1998). Taken together, these critiques extend what is meant by pedagogy as a political act beyond “what” is taught or “how,” definitional acts that risk reifying understandings of how the world should be, regardless of (or in spite of) how it is for any given person.  In thinking through the contested boundaries of pedagogy itself, I follow Ellsworth (1997) who approaches pedagogy as a “performative act,” an “event in and of itself – not a representation of something else ‘over there’” (p. 154). Brought together with the performativity of identity, this allows a conceptualization of the pedagogical encounter as a space within which social identities are continually being formed within contested relations of power and knowledge. In other words, identities do not simply meet up in the pedagogical encounter (educator, learner,  23 youth, boys, girls) but rather the very “reality” of those identities is “both reproduced and contested” (Butler, 2004, p. 30) in that encounter. Furthermore, this framing engages pedagogy in relation to “knowledge in the making” rather than in relation to “knowledge as a thing already made” (Ellsworth, 2005, p. 1), as will be discussed in detail in Chapter 5. Broadly, sexuality education as an HIV prevention strategy is full of knowledge that has the appearance of being “already made,” such as the “facts” in a curriculum. Working in the service of health-related effects, and the attendant discourse of effectiveness, these “facts” can/should/must be transmitted through the pedagogical encounter. A conceptualization of pedagogy in relation to “knowledge in the making” draws attention to how any “fact” is a particular configuration of knowledge about the world, formed through multiple relations of power. Besides destabilizing the apparent neutrality of those facts, the focus on “knowledge in the making” then engages the pedagogical encounter as a site within which any form of knowledge will encounter, and be encountered by, educators and learners who are actively engaged in making sense of the world.  It is the very place of that encounter that puts multiple stories of self and knowledge (spanning across histories, presents and imagined futures) into relation (Massey, 2005). Working within this encounter is, I will argue throughout this dissertation, a way of beginning to hold sexuality education responsible to the “lived reality” of daily life that is “utterly dispersed, unlocalised, in its repercussions” (Massey, 2005, p. 184). In other words, and beginning from the recognition that any configuration of knowledge carries visions of how the world should be, the pedagogical encounters of sexuality education can be resources for re-imagining received ways of making sense of the world in light of the complexity that people are already living. This re-imagining is not a solution to the dilemmas of sexuality education. It is a refusal to stay within what are currently said to be the possibilities of sexuality education as an HIV prevention  24 strategy, and a move towards a political articulation of sexuality education concerned with how the discursive limits of knowing, and being known, could be more just for more people.  In this articulation of sexuality education as a potential site for social transformation, I am responding to and resisting the critique that poststructural readings of power and the subject destroy the possibility of agency (e.g. Baylies & Bujra, 1995). My starting point is the impossibility of autonomous agency: the premise that a subject could act outside of, even if in response to, the social environment. Autonomous agency is impossible because any identity to overcome (such as gender as a social “role”) is a constitutive part of what makes “self” and “Other” intelligible (Butler, 1997b). At the same time, a subject may always act within understandings of identities in ways that potentially shift the discursive limits of their intelligibility. Butler explains this in terms of a “reiteration:” the assumption of power as an “appropriation” that can work “against the power that made that assumption possible” (p. 13). It is through appropriation that agency “exceeds the power by which it is enabled” (p. 15). Read through this lens, the meeting-up (both literally and discursively) of understandings of social identities and forms of knowledge within the pedagogical encounter is precisely what makes that encounter a dense site for the reiteration of relations of power and their potential appropriation.  This framing of the pedagogical encounter (as a potential site for social transformation that cannot be “known” in advance) is not a “solution” to the dilemmas of identity, even as it is troubled by the inevitable desire for “solution” that any political project invokes. Indeed, my own analytic slide into offering “solutions” is an unresolved tension in this dissertation. This tension, I argue, is symptomatic of how any political project draws coherence from forms of configured knowledge (such as the “facts” of gender equality and inequality) even if the project seeks to work at the boundaries of those forms. The tension is also part of working with the power-laden  25 effects of forms of configured knowledge: the “facts” of sexuality education as an HIV prevention strategy are playing out on the bodies of young people, both in terms of who they are said to be and what they are supposed to know. Given that sexuality education is being widely promoted as a global strategy for HIV prevention, I take it as an urgent political imperative that sexuality education needs to be “done” differently and that this “doing” is an effect of how sexuality education conceptualizes the relation between self, Other and knowledge. One starting point, I argue, lies in taking the very impossibility of empowerment discourse as a pedagogical resource, which brings me to the third proposition of this study.  Third proposition: The problematic “solution” of empowerment discourse  Across this dissertation, I work with the proposition that empowerment discourse in sexuality education as an HIV prevention strategy risks shutting down, rather than opening up, the pedagogical encounter as a potential site for social transformation. I begin by recognizing that empowerment discourse, while serving as a powerful imaginary for positive social change, is itself the site of multiple and contested meanings (for an example in the context of HIV prevention see Baylies & Bujra, 1995). Broadly, as a form of health education, the promotion of sexuality education as a vehicle for empowerment in relation to youth’s sexual identities and behaviours (e.g. Crone, Gibbs, & Willan, 2011; UNAIDS, 2010, 2013a; UNESCO, 2009a) can be seen to draw on two understandings of empowerment: individual empowerment and collective or community empowerment (Laverack, 2004),7 both of which are embedded in western knowledge formations. The first understanding, individual empowerment, can be traced to the field of community psychology in the United States (e.g. Rappaport, 1987; Zimmerman,                                                 7 See Gacoin (2014) for a detailed exploration of public health formulations of empowerment discourse, their implicit educational aims, and the implications for how sexuality education is conceptualized in South Africa.   26 2000) and draws on broadly liberal conceptions of power and agency:8 power is a measurable object (either positive or negative) that comes from the individual and will be wielded by an individual over the social environment. The second understanding, collective or community empowerment, is grounded in a view that “health is structured by society” (Stephens, 2008, p. 50). In this view, broadly drawing on neo-Marxist understandings of power and agency,9 power remains an object but it is conceptualized in terms of relations between individuals within a particular social and economic structure. In other words, individual empowerment is inseparable from the collective, and individuals must ultimately act together to “transform the world” (Freire, 2007, p. 48). Both of these framings of empowerment discourse may respond to, and be useful for, what Foucault (1982) frames as particular forms of power-resistance in a given historical moment. For example, learning about laws related to gender equity may support an individual learner to resist “forms of exploitation” (p. 781) in their life. Likewise, engaging groups of learners in explicitly thinking about and challenging specific gendered norms and expectations may be useful for identifying particular “forms of domination” (p. 781) that circulate within collective understandings of gendered identities and their relations. However, and following from the two propositions discussed above, sexuality education is engaged in another form of power-resistance: “forms of subjectivity and submission” (p. 781) where who youth are and what they can/should/must know about their relations to others is already at stake.                                                  8 Liberal here refers to political liberalism (rather than economic liberalism) in which the role of the state is to “secure the freedom of individuals on a formally egalitarian basis” (Brown, 2003, para. 6).  9 Neo-Marxism is a diverse strand of political theory that has extended the focus on political economy in Marxism to include social and psychological factors (Zanetti, 2007).   27 Working within this third struggle, and explicitly engaging its effects, begins from resisting a conceptualization of power as an object. Following Foucault (1998), power “is not an institution, and not a structure” (p. 93), reducible to that which seeks to repress. Power is a productive relation that is everywhere, “exercised from innumerable points” (p. 94). This means that those “forms of exploitation” or “forms of domination” are effects of particular relations of power and knowledge, rather than their source, and that empowerment discourse works at the level of those effects. This reading of power destabilizes the “truth” of empowerment discourse as offering the “solution” to power relations and allows consideration of how empowerment discourse itself is embedded in forms of subjectivity and submission within western knowledge systems. Furthermore, this conceptualization of power destabilizes the implicit educational claim that power can be given by the educator to youth during the pedagogical encounter, including the knowledge of more equitable gender relations. Indeed, if power cannot be “acquired, seized, or shared” (p. 94) then the question of what should be taught to youth about power relations is impossible to begin with.  By working with empowerment discourses within sexuality education, my aim is not to define (or re-define) what empowerment “is.” Any articulation of empowerment discourse is a complex discursive site where different understandings of power and agency play out, as seen in the competing conceptions of individual or community/collective empowerment briefly discussed above. My interest, instead, is to ask what pedagogical possibilities might arise by taking empowerment discourse as an invitation to encounter the “forms of subjectivity and submission” (Foucault, 1982, p. 781) at play within it. In other words, the question is not only what forms of subjectivity are being mobilized within empowerment discourse. It is also how the limits of those forms (and their potential effects) might potentially be glimpsed within the  28 pedagogical encounters of sexuality education. It is at these limits that, I argue, the “solution” of empowerment discourse may become a pedagogical resource: its inevitable failure an effect of the relations of power and knowledge that any “solution” entails, and an invitation for educators and learners to think at the limits of (and potentially shift) discursive norms of knowing and being known that are already being lived.  Invitation: Encounters  I work from and extend these three opening propositions by engaging the pedagogical encounter as a “particular nexus of power and knowledge that gives rise to the field of intelligible things” (Butler, 2004, p. 216). I cannot, and do not seek to, redefine “who” youth are or “what” they know. Likewise, I begin from the recognition that the very words that I have foreground in this introduction – such as “gender” and “empowerment” – are contested concepts in development practice (Cornwall, 2007), including sexuality education as an HIV prevention strategy. One of my tasks in this dissertation is to work against the flattening of meaning that occurs when “gender” and “empowerment” are framed as problems of implementation. For example, “gender mainstreaming” has been taken up in development practice as a programmatic strategy that can guide all aspects of project design, implementation and evaluation.10 Likewise, “empowerment” has been schematized into measurable instruments, such as the “Gender Empowerment Measure” (GEM) that was developed by the United National Development Program (UNDP) to evaluate “whether men and women are able to actively participate in economic and political life and take part in decision-making” (UNDP in Klasen, 2006, p. 257). Beginning from a view that “words make worlds” (Cornwall, 2007, p. 471), the pedagogical                                                 10 See, for example, the work of UN Women, the United Nations Entity for Gender Equality and the Empowerment of Women:, accessed January 22, 2015.   29 encounter offers one site within which to resist this flattening of meaning. These worlds are irreducible to (and yet inseparable from) global discourses, national programmatic priorities and the texture of lived lives. By engaging the pedagogical encounter as a momentary coming together of relations of knowing and being known, I ask what worlds are being made within sexuality education as an HIV prevention strategy. As I trace the opening propositions of this study through the following chapters, the ongoing political question is how those worlds may be lived in ways that are more just for more people.  Chapter 2 explores what it means to study sexuality education as a political project in the context of the HIV and AIDS epidemics. In other words, the opening propositions discussed in this introduction are not only related to what I study, but also how I study it. This methodological chapter engages the “stuck places” of research practice (Lather, 2007) through feminist and poststructural readings of ethnography interested in how identities take form with(in) research in multiple and contested relations of power and knowledge.  Chapters 3, 4 and 5 invite the reader into the pedagogical encounters of this study. Chapter 3 revolves around the question of “teaching gender,” and problematizes the way that gender identities and relations have been taken up as forms of knowledge to teach through sexuality education as an HIV prevention strategy. By destablizing the “solution” offered through notions of gender to the dilemmas of knowing and being known, I engage the pedagogical encounter as a site within which who knows and what can be known take form.  This provides the conceptual grounding for the inquiry of the subsequent chapters: to explore what might happen within sexuality education as an HIV prevention strategy when there is a shift from identity as a target for the pedagogical encounter to identity as a site of struggle within  30 the pedagogical encounter (Chapter 4); and to engage sexuality education’s pedagogical approach as a constitutive part of struggles over knowledge in the making (Chapter 5).  Between Chapters 3 and 4, I pause on practices that define, shape, and may play at the limits of what comes to be understood as gender with(in) particular pedagogical encounters. Alongside the initial story of how I came to this research (prior to Chapter 1), I offer these sections as further encounters with the notion of gender. Each speaks to how gender becomes coherent as a “problem” for sexuality education at the same time as each raises questions that destabilize the coherence of the “solution” being offered through sexuality education.  Finally, in Chapter 6, I re-encounter sexuality education through the political project of this study. Attending to the struggles over knowing and being known within particular pedagogical encounters, I suggest potential ways of articulating sexuality education so that it might, no matter how momentarily, provide a site within which educators and learners mobilize those struggles. It is here, I will argue, that the vulnerability of knowing and being known with(in) sexuality education might become a condition for responsibility to one another and a site for social transformation.    31 Chapter two: Methodology  I first entered into a conversation with methodology as a matter of research practice, or what I did. From the proposal to initial drafts of this dissertation I found this relatively easy to write as a straightforward tale: I worked on this ethnographic research project with loveLife in Cape Town, South Africa, from 2012 to 2013. During this time, I saw myself as conducting fieldwork using the ethnographic methods of participant observation (Hammersley & Atkinson, 2007), formal semi-structured interviews (Holstein & Gubrium, 2004) and artifact analysis (Hatch, 2002). I used these methods to gather what I saw as the data: written fieldnotes, recorded interview transcripts and stacks of educational documents. However, it is precisely when I got to the next “step” of the research practice, analysis, that I became stuck trying to “overlay the linear narrative of methodology on my practice” (St. Pierre, 1997, p. 180). As I sifted through fieldnotes, transcripts and documents I was confronted with “already interpreted data” (St. Pierre, 2011) that was inseparable from the ongoing decisions that I had made within this study to make it “count” as data (Pascale, 2011): from the initial research questions, to where and with whom I worked, to what I literally chose to make visible in words. In other words, my research practice had already enacted, and in doing so masked, what Pascale (2011) calls the “conceptual cut” between processes of formalization (e.g. methods) and processes of interpretation (e.g. analysis), as will be discussed further in this chapter. At the same time, trying to “solve” this cut (by saying that everything is already interpreted, for example) re-centered interpretation as a methodological “will to power” (Lather, 2007). I was stuck with(in) linear research stories that were always (and will always be) gathered in and through the practices of research.  In this chapter, I think about what it has meant to engage the “stuck places” (Lather, 2007) of my research practice through feminist and poststructural readings of ethnography that  32 are interested in “processes of subjectivation in research and representation” (Youdell, 2010, p. 92; see also Britzman, 2000; Lather, 2007; St. Pierre & Pillow, 2000). I take up these stuck places, and their dilemmas, in three ways (and as seen in the three main sections of this chapter). In the first, I explore what happens to the “local context” of an ethnographic research project when relations of power and knowledge are seen to settle in, at the same time as they refuse, a global/local binary. In the second, I ask what ethnography offers for thinking about the stories and selves (including the selves of researcher) told in and through sexuality education as an HIV prevention strategy. Finally, I turn to how the ethnographic “text” might enact some of the dilemmas of knowing and being known through research. Across these discussions, I seek to engage the impossibility of linear research stories (even as I continue to tell them) as an invitation to begin to risk a different kind of relation to the stories and selves of research.  Research and the localized context   This study has taken place during a moment of globalization characterized by an intensity of connections (Sassen, 2007). The conceptual frame of the research – sexuality education as an HIV prevention strategy in South Africa – is constructed through so-called global best practice that can be made “locally-relevant” (e.g. UNESCO, 2009a). The concepts of interest in the project – gender, youth, the “facts” of sexuality education – circulate across the presumed spatial registers of “global” guidance documents and funding priorities, “national” policy and programmes and “local” implementation. As such, one of the key methodological challenges in this project has been how to think about these connections in terms of the “context” of research.  From the start of this project, Tsing’s (2005) conceptualization of “global connections” in ethnographic research has guided how I have engaged and resisted a spatialized global/local binary of research context. Following Tsing, what are considered to be “local cultures” are  33 “shaped and transformed in long histories of regional-to-global networks of power, trade and meaning” (p. 3). At the same time, what is called “global” is not a self-evident “truth or lie” (p. 6). Instead, what is taken to be “global” is only “charged and enacted in the sticky materiality of practical encounters” (p. 1). This conceptualization provides a way to think outside of a hierarchical global/local binary and explore context in terms of “odd connections” rather than “seamless generalizations, inclusive tables or comparative grids” (xi). Indeed, I initially focused on contexts as sites of global connection precisely in reaction to the supposedly seamless generalizations made about gender, youth, and the “facts” of sexuality education as an HIV prevention strategy. The interest, here, was to engage with the spatialized power relations already operating with these identities and forms of knowledge.  This project has taken place in a research moment that Lather (2007) characterizes as a “worldwide audit culture” (p. 2). Take, for example, how research related to sexuality education as an HIV prevention strategy has focused on the effects and effectiveness of discrete interventions (Aggleton & Crewe, 2005). Broadly, this type of research has focused on measuring the success of an intervention in terms of supposedly quantifiable and universal measures of sexual behaviour (e.g. Kirby, Laris, & Rolleri, 2005). Measures lend themselves to discrete characteristics of successful interventions, regardless of “context.” In turn, those characteristics are deployed to make evidence-based “global” recommendations to be implemented in “local” contexts (e.g. UNESCO, 2009a). This type of research (and its uses) draws coherence from, at the same time as it reifies, a global/local binary: there are “local” and “global” subjects, and “global” expertise can be deployed to both identify and solve “local” problems. It also illustrates how dominant epistemological and methodological orientations to researching HIV and AIDS, grounded in “positivist frames of reference that favour conceptions  34 of a medicalized and/or moralized body which operates in a socially detached way” (Baxen, 2010, p. 143), have profoundly limited what counts as educational research in the context of the HIV and AIDS epidemics. While specific methodological approaches may have shifted over the course of the epidemic, such as the move from projection studies to more qualitative work focused on understanding the impact of programmes, the overall focus, following Baxen (2010), has been on “what is known” (and by extension what youth need to know) rather than “how and where knowledge is produced and how we come to [think we] know” (p. 132).  Within the power-laden effects of this research moment, focusing on “odd connections” has the potential to destabilize the coherence of a global/local binary in that it decenters linear or causal explanations (such as the flow of knowledge from one spatial term to the other). It draws attention to power-laden spaces where what is “known” takes form, spaces that refuse neat spatial boundaries, and raises a different set of questions for educational research. For example, what is at issue is not only what effect sexuality education as an HIV prevention strategy has on youth, but also what effects those educational discourses have on who youth are seen to be. Furthermore, when the global/local binary is engaged as a power-laden articulation of the “colonially established framework of homogeneity in the search for answers and solutions to the HIV/AIDS pandemic” (Chilisa, 2005, p. 668), it is a political necessity to engage with what forms of knowledge are being produced and deployed in the name of “helping” youth. At stake here is what is happening to bodies that are erased in and through these dominant epistemological and methodological orientations, the very bodies that are navigating the lived risks of HIV infection and death. For research interested in exploring these struggles, the pedagogical encounter may be engaged as a site within which the “global” and “local” come together in particular articulations of who youth are and what relation to knowledge they are  35 presumed to have. Here, the very incoherence of a global/local binary becomes an opportunity to engage with the materiality of relations of power and knowledge as they play out in people’s lives.  The construction of research contexts   At the same time, it was an “odd connection” during the fieldwork for this research that points to some of the ongoing, and power-laden, troubles of delineating research “contexts.”  I am driving home from a high school outside of central Cape Town and I am wearing one of the loveLife branded t-shirts. I usually take the t-shirt off if I’m not going straight home, but I’ve stopped to get gas. The petrol attendant who comes up to my car asks if he should check the oil and water and then, looking at my t-shirt, asks me what my connection is to loveLife. I am a researcher working on a project with the organization. A minute later he comes back and asks: why is loveLife no longer around? When I ask him what he means, he explains: in the early 2000s loveLife was everywhere but now you don’t see them. I am becoming increasingly uncomfortable as he speaks: loveLife is funded by government, they aren’t doing what they should be doing. I want my daughter to be a part of loveLife. I give what I think are the right answers: I can’t speak for the organization; I will follow-up on what activities might be available. Then I drive away. (Fieldnotes, April 2013)   As I have re-encountered this man and our conversation, both that evening in my fieldnotes and while writing this chapter, I have asked myself why I was so discomforted by this conversation. In many ways, the conversation was a mundane encounter because it took place within the spatial division of social identities that living in my skin in South Africa makes it all too easy to reiterate - rich white woman in a nice car, poor black man working at the petrol station. I could read my discomfort as part of the everyday negotiation of these identities. I could also take my discomfort as part of being asked to speak for an organization I am doing research with, and negotiating perceptions of insider and outsider within research (Dwyer & Buckle, 2009). At the same time, discomfort also operates within, and momentarily points to, what Boler (1999) calls the “inscribed habits of inattention” (p. 17) of research practice. More than unconscious, these inattentions are produced through multiple relations of power and knowledge. For example,  36 while I critique western knowledge systems, my thinking is embedded in those systems. This includes my own commonsense constructions of time, space and relationships between people (Tuhiwai Smith, 2012). As an effect of inscribed habits of inattention, discomfort arose when the man at the petrol station disrupted what I had delineated as the research “context.” I had bounded that context through un-thought acts of research, such as choosing when I chose to embody the researcher role, literally made visible through the t-shirt that I was wearing. In a troubling return to positivist objectivity, I took for granted the privilege of holding the “context” of my life separate from what was studied. While I had carefully troubled conceptual boundaries, I found my own research practices profoundly troubled by boundaries.  In thinking about these boundaries, I have been confronted with what Pascale (2011) critiques as the conceptual cut made between processes of formalization and interpretation in qualitative research. Formalization refers to those processes for systematizing knowledge production: “processes that provide protocols for recognizing relevant phenomenon and transforming them into data” (pp. 16-17). Interpretation, in turn, is about “explaining the significance of evidence” (p. 18), a definition I return to below. Overall, the central premise of Pascale’s argument is that processes of formalization are always already embedded in particular understandings of the world and how research claims to know that world. At that petrol station in Cape Town, I came up against how my own under-theorization of processes of formalization had led to a contradictory understanding of context. In particular, I can read my discomfort at that moment as operating within, and pointing to, a conceptual cut between processes of formalization and processes of interpretation. On the one hand, working with an articulation of global connections, I took context as a site for exploring power relations across what is considered “local” or “global”. On the other hand, I continued to take context as a discrete place  37 that is “unproblematic and naturally occurring” (p. 17) and hold myself separate from that context. In other words, context was a site to study, rather than a “localized context” (Pascale, 2011, p. 17, emphasis original) constructed through research.  At the same time, the very coherence of the cut, and how it took form in my research practice, centers “interpretation” (and attendant notions of understanding) as the goal of qualitative research. Consider, for example, Pascale’s definition of interpretive processes as “explaining the significance of evidence.” (p. 18). Within this aim, the world may be complex/partial/fragmentary but it can ultimately be “known” (Law, 2004). This is a troubling reification of the very ethnographic authority that, as will be discussed below, has conceived self, Other and knowledge through the colonial imperatives of research and its ongoing imperial formations. It also enacts how a researcher’s histories of learning (here, within western knowledge systems) continually and powerfully draw thought into the comfort of its divisions. Here, the methodological task is not to get rid of these terms (such as “methods” or “interpretation”) but rather to engage the impossibilities of thinking outside of them as an ongoing dilemma and provoke attention to relations of power and knowledge settling through them. For example, the encounter at the petrol station, and the “uncomfortable reflexivity” that it prompted (Pillow, 2003), was a momentary enactment of the relations of power and knowledge that give any research “context,” and the practices that seek to study it (which I return to in the following section), coherent. I have used my struggles within this encounter to shape how I speak about the “context” of this research. I purposefully do not provide descriptive geographic or demographic profiles that work in the service of “context” as a bounded place. Instead, I foreground a reading of South Africa and the loveLife programme as “localized contexts” that  38 are given form through, even as they cannot be reduced to, particular configurations of power and knowledge attended to in this research project.  South Africa as the localized context of this study In many ways, South Africa lends itself to an account of global connections. In 2013, South Africa was home to the highest percentage (18%) of the estimated number of people living with HIV in the world (UNAIDS, 2014), and almost a quarter of all new HIV infections are among young heterosexual women aged 15 to 24 (Human Sciences Research Council, 2014). In response, “gender” has been conceptualized as a structural driver for HIV infection in terms of understandings of masculinity and femininity and their unequal relations (Jewkes, 2009), and a key element to address within HIV prevention programmes for South African youth (Bhana & Pattman, 2009). The positioning of gender as a structural driver resonates with international development discourses, within which sexuality education as an HIV prevention response operates. In these discourses (e.g. Hawkes & Buse, 2013; Padian et al., 2011; Parker, Easton, & Klein, 2000), gender relations are framed as a discrete programmatic target necessary to “challenge the conditions in which people live” (Gupta, Parkhurst, Ogden, Aggleton, & Mahal, 2008, p. 766). In South Africa, high rates of HIV infection, and the gendered dimensions of these statistics, have driven educational policy concerned with learning about “relationships, gender, and power” through sexuality education (Francis, 2010, p. 318). At the same time, “local” educational initiatives designed to address these factors, including sexuality education, are entangled in the international flow of “expertise” as well as donor funding. loveLife, for example, has developed its educational programmes drawing on “international” and “local” research (Robbins, 2010), and receives funding from the South African government as well as a range of  39 international partners.11 While these factors are not deterministic of a non-governmental organization’s actions (Choudry & Kapoor, 2013), they do blur the distinction between “local” and “global” priorities and responses and necessitate a conceptualization of the research context that can work across these terms.  At the same time, what drew me to the global connections of HIV and AIDS in South Africa also points to how research is implicated within these connections and what gets called the research “context.” Research related to HIV has been “the major driver of research on sexuality and education in sub-Saharan Africa” (Humphreys et al., 2008, p. 24). This orientation draws coherence from, at the same time as it potentially reifies, understandings of sexuality and gender formed through past colonial and ongoing imperial concerns of reproduction, sexual health, violence and disease. While essentialist and Othering accounts of “African sexuality” in HIV prevention discourses have been strongly critiqued for over twenty years in this context (Humphreys et al., 2008; Patton, 1993a), those accounts haunt an academic citational authority whereby past research legitimizes present research directions and future concerns (including my own path into this research). For example, the universal category of “young women” in the statistics mentioned above erases the “marked racial differences” within these statistics where the highest HIV prevalence rates is among young black Africans (Jewkes, 2009). Within South Africa, the silences around race can be read as a form of resistance to the western pathologizing of African sexualities (Marais, 2012). However, how these bodies are already targeted within the global connections of sexuality education as an HIV prevention strategy is also a critical question. While conducting fieldwork, I was usually the only white body in the room. The loveLife subject                                                 11 A list of current organizations funding loveLife is available at, accessed January 22, 2015.  40 – South African youth – was embodied through black and coloured youth. The aim here is not to say that the “nation” was not complete because white kids were not included. Instead, the very “context” of the nation is itself a dense site for engaging multiple and contested power-relations that operate at and through its boundaries. This includes the “national” programmes that operate within it.  The contested context of loveLife  I was first interested in working with loveLife precisely because of how the organization, through its HIV prevention activities with youth in South Africa, creates the conditions for pedagogical encounters in which multiple discourses related to identity, knowledge and power may be momentarily woven together and engaged through research. In order to understand those conditions, it is helpful to first provide some background on loveLife’s overall aim and how it organizes its activities to achieve this aim.   Founded in 1999, loveLife has a head office, located in Johannesburg, and provincial offices across South Africa. The head office is responsible for the strategic direction of the organization, for running certain services such as the national Call Center, and for the design and development of what loveLife calls its “core programmes.”12 Broadly, the core programmes are “designed to address the individual, structural and social factors that drive the epidemic among South African youth” (loveLife, 2014b). The core programmes are implemented in communities by a national volunteer service corps of youth aged 18 to 25 called “groundBREAKERs.” groundBREAKERs must have finished high school, and are recruited based on their membership                                                 12 A full list of current core programmes is available on the loveLife website (see, accessed January 20, 2015). My research was focused on the two modular educational programmes most explicitly about sexuality education: love4Life and lovingLife and Making my Move.  41 in particular communities that loveLife has targeted for the core programmes. Each groundBREAKER works with loveLife full time for one year, starting with the South African school year in January, and is given a small monthly stipend. loveLife staff educators based in each region, known as “Regional Programme Leaders”, train, support and manage a group of groundBREAKERs within that particular region. Following a 10-day provincial wide training known as “Core 1”, groundBREAKERs are expected to conduct educational activities within their communities. Most groundBREAKERs are based out of schools or community clinics, and activities include implementing structured educational sessions as well as once-off events such as debates and games. Additionally, groundBREAKERs are expected to train, and jointly run activities with, younger volunteer peer educators (ages 12 to 17) known as “Mpintshis.” At the end of the year, groundBREAKERs attend a wilderness camp, known as “Core 2”, to prepare them for exiting the programme and moving onto other opportunities.   Over more than 15 years, loveLife has built up a substantial national presence. In additional to regional offices, there are 22 loveLife Y-Centres, youth centres that “provide educational, recreational and sexual health services” in low-resource areas (loveLife, 2015). groundBREAKERs and Mpintshis work in over 8000 schools across the country.13 The organisation also works in partnership with government and non-governmental organisations within particular areas, for example on activities for special events such as World AIDS Day. Media campaigns, such as radio programming, are part of the daily provision of HIV prevention messages to South African youth led by non-governmental organisations in South Africa.14                                                 13 See, accessed March 30, 2015, for maps representing where loveLife works in the country.   14 The Soul City Institute for Health and Development (, accessed March 30, 2015), is an example of another prominent HIV prevention campaign in South Africa.   42   In the chapters that follow, this reading of the loveLife “context” draws attention to relations of knowing and being known being mobilized within it. For example, in Chapter 4, loveLife’s focus on “marginalized” communities is engaged as a constitutive part of what kind of subject is deployed when identity is framed as a target for HIV prevention. Chapter 5 includes a discussion of how the (female) groundBREAKER body may be disciplined when identity as an educator is conflated with the messages that the educator is expected to deliver. Furthermore, the contested nature of the “context” of loveLife engages research as a constitutive part of (rather than reflection or evaluation of) this pedagogy. In the almost 15 years of loveLife’s history, there have been over 30 research articles published about the programme, speaking both to the high-profile nature of the campaign as well as its contested status. Broadly, there are three debates operating within this literature: what type of content is (or should be) included in loveLife messaging; what is the overall effectiveness of the programme in terms of HIV prevention (such as condom usage) or health outcomes (such as rates of adolescent pregnancy); and who the loveLife youth subject is taken to be. Each of these debates is characterized by binary and polarizing terms. For example, the first debate revolves around a series of controversial billboards15 that are either overly confusing (Hoeken, Swanepoel, Saal, & Jansen, 2009; Jansen & Janssen, 2010; Lubinga, Schulze, Jansen, & Maes, 2010) or a promising trigger for dialogue and youth voice (Etkin, 2013; Harrison, 2010; Jones Royster, 2009; Mitchell, Reid-Walsh, & Pithouse, 2004; Stadler & Hlongwa, 2002). The effectiveness of the programme, as reported in a                                                 15 Examples from a past campaign called “His&Hers” can be viewed on the Communication Initiative website (, accessed January 20, 2015). Examples from the most recent “Nakanjani” campaign are available on loveLife’s website (, accessed January 20, 2015).  43 series of studies commissioned by loveLife (Louw, Peltzer, & Chirinda, 2012; MacPhail, Pettifor, Moyo, & Rees, 2009; Matseke, Peltzer, Mchunu, & Louw, 2012; Mchunu, Peltzer, Tutshana, & Seutlwadi, 2012; Peltzer & Promtussananon, 2003; Peltzer, Ramlagan, Chirinda, Mlambo, & Mchunu, 2012; Pettifor, Kleinschmidt, et al., 2005; Pettifor, Rees, et al., 2005), has been challenged as inadequately defining participation or taking into account the potential influence of other HIV prevention programmes in South Africa (Jewkes, 2006; Parker & Colvin, 2006). Furthermore, the youth subject mobilized through the programme is framed as either a celebration of a new national identity (Bateman, 2007; Etkin, 2013) or a neo-liberal and consumerist projection (Breidlid, 2009; Epstein, 2003, 2007; O'Toole, 2004; Parker, 2006; Prinsloo, 2007).  For this study, instead of getting drawn into the binary terms of these debates and trying to decide the relative truth of one over the other, I engage these research conversations as symptomatic of struggles over identities and forms of knowledge enacted through the loveLife programme. In other words, loveLife as a national programme enacts struggles over who the educational subject is, what can be known through education, and what is the perceived relation between those subjects and forms of knowledge with(in) education. Research cannot offer a “solution” to these dilemmas of knowing and being known through education. As such, the starting point here is not just that this study does not aim to evaluate the loveLife strategy or the effect (in terms of behavioural outcomes) of the organization’s educational messages, aims that are centered around already configured forms of what can/should be known. It is also to start from the position that research that claims to be a neutral evaluation is always already impossible. Knowledge, attitude and behaviour (KAB) studies and intervention impact studies, two of the  44 dominant approaches to educational research in the context of HIV and AIDS (Baxen, 2010), 16 enact the paradigmatic orientations that have limited the scope of educational research in the context of the HIV and AIDS epidemics. Broadening this scope, and attending to how knowledge takes form, engages these studies as themselves formative to the “context” of loveLife. In other words, research is invited through, and in turn invites, a particular gathering of stories and selves. The question is then how research may encounter, and be encountered by, subjects who are always already exceeding the desire to know them in any particular “localized context.”  The gathered stories and selves of an ethnographic research study  Engaging relations of power and knowledge in terms of the “localized context” has raised another methodological dilemma: “how does one act knowing what one does?” (Viswswaran in St. Pierre & Pillow, 2000, p. 1). Early in this study, I conceptualized an ethnographic methodology that takes the “story” of the research not as a linear narrative, but as a site of contradictory stories that implicate both researcher and participants in constructing meaning (Britzman, 2000). My assertion at the time, and one I still make, is that the value of ethnography is the focus that it gives to empirical details. Instead of the “truth” of identity or experience, the empirical here is understood as the power-laden texture of lived lives (Stewart, 2007): the details that are inseparable from, at the same time as they are irreducible to, a particular time and place and the selves that gather there within contested relations of power and knowledge. This conceptualization frames the empirical as simultaneously that “upon which deconstruction does                                                 16 These are two of the three broad trends that Baxen identifies in her review of educational research related to HIV prevention in South Africa. The third trend is studies focused on the impact that HIV and AIDS have on the education system, such as the teacher deaths or the educational needs of students who have lost one or more parents to the epidemics.  45 its work” and that which “survives deconstruction by being that upon which it depends” (Lather, 2007, p. 72), making the task of empirical research  to throw ourselves against the stubborn materiality of others, willing to risk loss, relishing the power of others to constrain our interpretive ‘will to power’, saving us from narcissism and its melancholy through the very otherness that cannot be exhausted by us, the otherness that always exceeds us. (p. 10)  In other words, ethnography may engage the empirical as a resource for exploring the limits of what is said to be known, who claims to know, and who becomes known and knowable within its articulations.   Despite this conceptualization, as I conducted this study (language that itself hints at the problem to come), I found myself drawn into the “doing” of ethnography as a matter of methods. Given the diversity of different “types” of ethnography, what ethnographers “do” has indeed become a dominant definition of what ethnography “is” (Hammersley & Atkinson, 2007): working in an “everyday” context the ethnographer draws on a range of sources, collected over time, in order to provide an in-depth analysis of a particular instance of “making sense of the social world” (p. 4). The risk is that conceptualizing ethnography as what is “done” operates within and can deepen, rather than challenge, the conceptual cut between formalization and interpretation as well as the analytic authority given to interpretation. Indeed, doing ethnography is already a deeply contested space.  Doing ethnography  How ethnography is “done” carries the traces of multiple and contested historical relations of knowing and being known in research. Much of ethnography today seems to define itself in relation to what it is not: the “classic” or “traditional” ethnography of the early twentieth century. Within a positivist epistemological paradigm, “classic” ethnography was taken as an instrument through which objective and scientifically valid “facts” could be known (Lassiter &  46 Campbell, 2010). Culture, in turn, was the gateway to knowledge of people and a site for comparative and generalizable accounts across discrete cultures (e.g. Bateson & Mead, 1942; Mead, 1930, 1961). This framing of ethnography was coherent within, at the same time as it gave a form to, “a master discourse of colonization” (Clair, 2003, p. 3) that shaped views of self, Other, and their perceived relation to knowledge. It was also already a site for potential opposition to these ideas. The German-American anthropologist Franz Boas, for example, strongly critiqued the “anthropological racism” that underpinned ethnography at the time (Clair, 2003).  Following WWII, a “new” ethnography began to interrogate the relation of self to Other in light of shifting social, political and economic relations (Lassiter & Campbell, 2010). This was extended in the 1980s as a series of “turns” in the social sciences that revolved around a critique of a positivist epistemological paradigm and its implications for research (Guba & Lincoln, 2004; Lather, 2007).17 For example, as part of what has been called the linguistic turn, writing is engaged as “part of the invention, not the representation, of cultures” (Clifford, 1986, p. 2). The shift from writing as a “method” was in turn extended by some feminists who drew attention to writing itself as an always political act embedded within broader relations of power and knowledge (Behar, 1995). Another example is methodological work that has re-thought what it means to study “culture” (Eisenhart, 2001). Resisting the cultural evolutionism of anthropologist EB Tylor (Clair, 2003), one dominant approach has been to approach culture in terms of a “web of significance” (Geertz, 2000) to explore through “thick description.” Broadly, these critiques                                                 17 It is outside the scope of this discussion to detail each of these “turns.” As Lather (2007) foregrounds in her discussion of ethnography, deciding which “turn” to feature is a difficult task, and may include: linguistic, structural, critical, deconstructive, rhetorical, cultural, narrative, historical, ethnographic, postmodern, ethical, visual, pragmatic, policy and theological.   47 have been crucial for acknowledging and working with power relations in ethnographic research, both in terms of the historical legacies of studying “other” cultures, as well as the ongoing ethical dilemmas around what it means to know in ethnographic research.  At the same time, the multiplicity of the “turns” in the social sciences, as well as the dissensus between them and the competing understandings of ethnography that they have given rise to, enacts the ongoing complexity of knowing with(in) ethnography (Lather, 2007). The failures of ethnography and ethnographic methods are often taken as a call for “new” methods (Lather, 2007). In critical ethnography, for example, notions of “collaboration” and “participation” have become a type of “method” that seeks to interrupt power relations within research (Fine, 2006; Foley & Valenzulela, 2005). Critical ethnography also engages the subjectivity of the researcher as part of a broader project to “expose relations of power and exploitation” (Foley & Valenzuela, 2005, p. 222). While critical ethnography has challenged power relations within processes of interpretation, the focus on a “better” method to get there (as a process of formalization) risks deepening the conceptual cut between these two processes and propping up the ability to ultimately “know” through interpretation. In other words, there is an implicit assumption that the problems of representation can be solved with the “right” methods (Lather, 2007).  The risks (as well as the seductive pull) of this assumption have worked at the limits of my own “doing” of ethnography. Despite an explicit commitment to attend to processes of subjectivation, as discussed in terms of theoretical commitments in the previous chapter, my research practice was “filled with sacred objects to be recovered, restored and centered” (Britzman, 1997, p. 35) that erupted in frustration in my fieldnotes:  waiting around for things to start; no time tables for work in the school; nothing goes according to plan; sitting at a desk in the regional office; trying to find out who is going  48 where and when; plans change constantly; school holidays; exams starting; exams ending; going on leave; stuck on what “should” happen; not getting enough data (Fieldnotes, January – August, 2013)  It is possible to explain these moments away as subjective reactions to the unstructured nature of ethnographic fieldwork, or the individual anxieties of a doctoral student conducting her first ethnographic project. However, in struggling with the “doing” of ethnography, I’ve re-encountered my actions within the ethnographic field as a “contested and fictive geography” (Britzman, 2000, p. 28). In this field, a white North American cisgender doctoral research student studying “gender” inevitably repeats the histories of ethnography. As Tuhiwai Smith (2012) argues, the colonial histories of ethnography are not a thing of the past. Instead, “research is one of the ways in which the underlying code of imperialism is both regulated and realized” (p. 8). In the context of sexuality education as an HIV prevention strategy, ethnographic research has been positioned as a way to resist hegemonic western categorizations of identity and behaviour (Parker & Ehrhardt, 2001). However, and as will be further explored in Chapter 3, the very notion of being culturally sensitive or “exploring sub-cultural contexts” (Parker & Ehrhardt, 2001, p. 110) operates within and redeploys western hierarchies of knowledge that operate within HIV prevention discourses (Chilisa, 2005), including those constructed through the local/global binary. Here, any approach to ethnography is always already invested in particular ways of knowing (Marker, 2003), including understandings of social identities and the ongoing imperial formations of their relations. The question then becomes not what “method” is best, but how to work within the histories of ethnography “in such a way that the repetition [of those histories] displaces that which enables it” (Lather, 2007, p. 39). This is part of the ongoing political project of sexuality education, as discussed in the opening propositions, and necessitates  49 explicit attention to how the “role” of the researcher is always implicated in (even as it seeks to resist) these relations of power and knowledge.  The gathered stories and selves of this researcher  In proposing this study (and in line with many reports of qualitative research projects that I have read), I first discussed my “role” as researcher in terms of what I would do. I then, separately, engaged my “positionality” in relation to how I saw myself within the research. Initially I saw my “role” as follows. Prior to beginning the study, I had informal conversations with loveLife to define a researcher role for participant observation, one that would complement the organization’s structure, provide opportunities for mutual learning, and give practical help with the day-to-day work of the team. In line with these discussions, I worked in a volunteer position within the loveLife Regional Program Leaders Team of the Western Cape. I began working with the Regional Programme Leaders team in November 2012, as they were getting ready to recruit groundBREAKERs for the following year. Besides spending time getting to know the team and going to end of year events, I helped to interview potential groundBREAKERs. The new programmatic year started in January 2013 with pre-training for the Regional Programme Leaders ahead of the provincial Core 1 groundBREAKER training. I took part as a trainee and received in-depth training on the loveLife programmes and approach to working with young people. I applied this training during Core 1, where I was part of a three-person staff educator team working with a new group of groundBREAKERs. Following Core 1, I worked with the Regional Programme Leaders to identify groundBREAKERs who would be willing to let me work with them over the course of the year as they implemented educational activities. In total, I took part in 33 structured educational sessions that took place at primary schools, high schools and community sites in 7 communities. The majority of my time was spent  50 with two groundBREAKERs, one who identified as a black isiXhosa-speaking man and one as a black isiXhosa-speaking woman, each of whom worked in a black isiXhosa-speaking low income township outside of Cape Town. My role focused on support. The educator and I planned sessions together and, after each session, we talked informally about that day’s activities. During all the educational activities, I followed the lead of the Regional Programme Leader or the groundBREAKER and helped facilitate activities as asked. Over the course of the year I also spent time at the loveLife regional office where I participated in one-off events, the ongoing training of groundBREAKERs and other educational activities as they came up.  While this detailed explanation of “role” may be necessary to give the reader the conceptual grounding for what I “did,” my doing of this role was inextricable from my positionality. The stories of my fieldwork are filled with how others articulated my role as I worked with them, such as how an educator introduced me to a group of learners or a loveLife partner. I was/am/should be:  American Texan white community student tourist one of us Canadian visitor touch-base Regional programme leader good Mpintshi here to observe him her white lady her boss (Fieldnotes, January – August 2013)  These multiple selves are formed within the “tangles of implication” (Britzman, 1997) of this research project. Touch-base, for example, was the nickname given to me by one educator to  51 tease me (and in doing so make explicit) my persistent desire to “know” what was going on. There were also multiple relations of power and knowledge operating through the doubled role/position of white community. At the beginning of my fieldwork, a Regional Programme Leader took me and a French student (who was working with loveLife a part of a volunteer exchange programme) to a school in a black isiXhosa-speaking township outside of Cape Town. As we left the school, walking out of the gates to the car, he looked around and joked that he was making sure he had his white community with him. My position in that white community was drawn through multiple relations of power and knowledge, including the Western knowledge systems that a stream of North American and European foreigners carry with them as they come to “help” with South Africa’s HIV and AIDS response (Lees, 2008). Our gaze on a black community (as a place) was reflected back on the mobility of our white bodies, and the implicit construction of “risk” around our bodies being in that place. The joke marked these relations, at the same time as it may have been an act of resistance to them. The Regional Programme Leaders were often expected to be both tour-guide and baby-sitter to the foreign researchers and volunteers working with loveLife. While I tried to be attentive to these relations, my roles as a researcher were formed within them.   Furthermore, the multiplicity of these selves, including the absence of the selves I did not/could not hear, took form within the encounters of this research project. For example, my understanding of myself as a white woman was mobilized within the relations of power and knowledge of being her white woman: older, richer, English first language speaking, foreign. As a visitor I wielded the power to be here to observe him. As one of us I was at the same time American/Texan/Canadian. The selves refuse any neat separation between who I take myself to be and “processes of subjectivation in research and representation” (Youdell, 2010, p. 92, my  52 emphasis). This is not just to say that my own understandings of who I am influenced the research. Instead, and in line with the argument that I develop in Chapter 4, I articulated and enacted both who I see myself as and who I am taken to be through the research encounter.  Thinking differently about relations of power that settle into conceptions of self is difficult precisely because conceptions of self are crucial to thought and action (Allen, 2006; Taft, 2007). This difficulty, and its potential effects, is explored in Chapter 5 in relation to how educators conceived the pedagogical relation to social difference. Methodologically, I have sought to work within this difficulty by taking an approach to reflexivity that positions the self as a verb rather than a noun (Davies et al., 2004). Here the self of the researcher is inseparable from “complications of constitution – of identity, subjectivity, place, space” (Talburt, 2004b, p. 109) within ethnographic research. One way that I have engaged these complications is through “reflexivities of discomfort” (Pillow, 2003, p. 187) that seek to interrupt the pull towards what is familiar (in terms of who I see myself as and what I claim to know). Throughout this dissertation I have engaged my reactions (such as frustration when things do not go according to the “plan” or a sense of unease in my position) as symptomatic of moments when conceptions of self and attachments to particular forms of knowledge were troubled. For example, moments of frustration, as captured in fieldnotes, mark the power given to stable knowledge (the plan, the objectives, the target) within sexuality education as HIV prevention strategy. At the same time, frustration was a reaction to the impossibility of holding onto that stability. Here, the self of the researcher is always implicated within the “forms of power and meaning” that “can gather themselves into what we think of as stories and selves” (Stewart, 2007, p. 6) with(in) ethnographic research. The text (this dissertation) is then an always partial and contested representation of those gathered stories and selves.   53 Entering into ethnographic textwork The stories and selves of this research are woven in and through what Lather (2007) calls the “textwork” of an ethnographic study. This term foregrounds the strategies of representation (the “work”) that give form to the text, where that text is understood as a multi-layered weaving of the “politics of interpretation, data, [and] analysis” (p. 41). This does not solve the problems of representation but rather foregrounds them. Specifically, in this section, I begin by exploring the “work” that methods perform to make something count as data. I then engage coding as performing its own kind of “work” as it shapes what comes to be the analytic threads of the text. Finally, I articulate an approach to writing that aims to “work” with the text in ways that potentially allow spaces within it for the messiness of how educators, learners and myself as researcher were all making sense of the world with(in) the pedagogical encounters of sexuality education.  The work of participant observation    The first “method” that I engaged was participant observation. As an ethnographic method, participant observation refers to how a researcher actively takes part in “people’s daily lives” (Hammersley & Atkinson, 2007, p. 2). As discussed in relation to my “role,” this is broadly what I did. However, as also explored above, that “doing” cannot be separated from multiple relations of power and knowledge that invited that “doing” in the first place. There is, for example, already multiple silences within the story of my “role.”  As I chat with loveLife staff, one community has repeatedly come up as a good “fit” for my research. It is not too far from Central Cape Town and the groundBREAKER will be working in the schools in English. I follow up with the Regional Programme Leader working in that area and he repeatedly promises to introduce me to the groundBREAKER. Two months pass and I am still waiting. During that time, the Provincial Manager assigns two visiting students (white European women) to work in that specific community for a short-term research project. They soon become known as the Provincial Manager’s spies: the groundBREAKER isn’t at the school; the groundBREAKER isn’t doing what she  54 should be doing. No one seems to know what the students’ research project is about. I decide to ask the Regional Programme Leader about working with a groundBREAKER in a different community. We go to meet her that day. (Fieldnotes, March – April, 2013)   In this encounter, selecting where to work was already entangled in the risks of knowing and being known through research. When the foreign white female bodies became spies, they carried with them multiple relations of power and knowledge: the internal organizational surveillance of loveLife educators through monitoring and evaluation procedures; the external and imperialistic surveillance of “local” sexuality education programmes by “global” actors; the white gaze that disciplines the black body’s movements and actions. Furthermore, what I took to be “logistical concerns” for choosing community sites (such as school time schedules, driving distance from Cape Town and speaking English)18 were already formed through the implicit and explicit rules of western knowledge systems and their hierarchical organization (Tuhiwai Smith, 2012). The use of language, for example, was already a contested site for articulating what is known and who knows. Several senior staff members dismissed my concerns around language, telling me that groundBREAKERs “should” implement in English. The Regional Programme Leaders, in contrast, immediately saw language as a barrier to where I could, or should, work with groundBREAKERs. Following their advice, I approached groundBREAKERs who would supposedly be working in English, and therefore did not work with a translator. At the same time, all language was already passing through multiple and power-laden acts of translation. Despite the centering of English in the loveLife programme (which I take up in relation to notions of gender in Chapter 3), educators and learners spoke in isiXhosa in the vast majority of sessions that I attended. Sometimes the educator translated during the session, or asked that learners speak                                                 18 I did take isiXhosa lessons. These lessons continually reminded me of the vulnerability of learning as well as how little I know.  55 in English for my benefit. Other times I asked the educator to translate after a session and together we re-encountered what had been said in that place and time. These translations were woven into initial impressions and thoughts as I wrote ethnographic fieldnotes (Emerson, Fretz, & Shaw, 1995), and I reflected on them as a part of the process of iterative data analysis, or the ongoing and reflexive process of meaning making during fieldwork (Srivastava & Hopwood, 2009).  While language can be engaged as a limitation for this study (as will be discussed in the conclusion), this limitation may also provoke attention to how all language is always already embedded in power relations (Spivak, 2008). For example, in Chapter 3, I ask how particular forms of power and knowledge settle through the English language as gender has been taken up as something to address through sexuality education as an HIV prevention strategy. The relations of knowing, and being known, entangled in language are a part of the texture of participant observation as a site of “doubt, rather than a confirmation of what exists prior to representation” (Britzman, 2000, p. 32).   I have woven the work of participant observation into a series of vignettes inspired by pedagogical encounters. The fragmentary form of the vignette works against the interpretive authority of a “full” and linear story within traditional ethnographic texts (Britzman, 2000). As evocative scenes, the vignettes are grounded in a “stubborn materiality” (Lather, 2007, p. 10) that is irreducible to a single story, including the inevitable interpretive power of research stories. The “reality” of the vignettes took form through a research encounter with and through a particular time and place. Instead of quotes, I use italics to represent the collection of quotes, translations and remembered moments from these particular encounters. The vignettes provide a way of jumping into the “middle of things” (Tsing, 2005, p. 1) and asking about relations of power and  56 knowledge that both exceed and are constitutive of that “middle.” This was (and continues to be) a middle that is always escaping the desire to know it.  The form of the vignettes is also a part of the ethics of representational choices in ethnographic research. Within the vignettes I use pseudonyms selectively, weighing the importance of the specificity (and always present complexity) of “who” an educator is alongside the risks within a programmatic context where staff performance is constantly tracked and monitored. While this research was explicitly not an evaluation, I have still worried that particular educators could be blamed when educational sessions did not proceed as expected by loveLife (in terms of either process or the stated outcome). As such, and particularly in the context of internal loveLife training events in Chapter 5, I have chosen in certain vignettes to refer to an educator rather than using pseudonyms. My work with learners, in turn, was only within the context of loveLife activities (as per my ethics approval) and those learners did not choose pseudonyms or interact with me outside of the sessions. While learner may seem impersonal, I see assigning pseudonyms to learners as a more risky assumption of identity, particularly given the power of naming in the South African context. For example, in the isiXhosa language, names are drawn from verbs that locate a child within the context and relationships of their birth (Ubuntu Bridge, n.d.) and these relations have been violently disciplined through the imposition of new names, from the Dutch Christian names given to children in a slave school in 1658 to the English names given to black employees by white employers during apartheid. Naming (as well as the absence or refusal of names) is an effect of relations of power and knowledge. I offer the vignettes as a part of taking representation as both an ongoing problem and a resource for thinking through how “substantively, historically, materially [the empirical] came into being” (Sedgwick, 2003, p. 14). They are an invitation into  57 gathered stories and selves, at the same time as they provoke questions pertaining to relations of power and knowledge that refuse to be contained within them.  The work of the interviews   The second “method” that I engaged for this study was individual semi-structured interviews. Specifically, I conducted 9 interviews with loveLife educators (see Appendix A: Individual semi-structured interview protocol) with the stated aim of focusing specifically on meaning-making related to the concepts called out in this study - gender, gender relations, empowerment. Of the 9 interviewees, 6 were Regional Programme Leaders who both manage and implement educational activities in communities, 1 was a head office staff member responsible for designing educational programmes, and 2 were groundBREAKERs working in different community sites. All interviews were conducted in English.19 I also conducted three tape-recorded meetings with past and present senior loveLife staff. These staff agreed to speak on behalf of their work with loveLife and chose to use their own names when I offered to use pseudonyms. All interviewees were recruited using an ethics approved process of informed consent, given a research handout (see Appendix B: Project overview handout), and all chose their own pseudonym. For each interview I invited, but did not receive, feedback on the transcript.  As discussed in relation to participant observation, the data from the method of interview was also produced through its doing. I understand interviews as forming “accounts” of ways of making sense within particular moments and contexts, rather than “reports” of an internal truth or reality (Talmy, 2010). Throughout the interviews, for example, informal conversations in the                                                 19 All of the loveLife staff that I worked with spoke English fluently and conducted much of their work with loveLife in English.  58 office or while driving to a session gave way to a more formal kind of speech. I sat facing the interviewee, a tape-recorder placed between us. I was interviewing them as a “loveLife educator,” and the interviewee and I drew on loveLife discourses as a basis for our shared understandings, even as those understandings were sites of potential resistance to those discourses (as I explore in relation to “gender” in Chapter 3). Broadly, I take the “participant understanding” captured in an interview transcript as a “piece of [participant] and researcher sense-making” (Loutzenheiser, 2007, p. 112) from a particular moment.  In working with the interviews, I first found myself drawn to accounts about the pedagogical encounter. In other words, I engaged the data for what educators think should happen (in terms of a prescribed pedagogical approach, for example) or what they think might arise out of that encounter (such as the goal of youth empowerment). However, within this approach the interviews quickly solidified into interpretations to speak about and potentially for the youth who participated in those encounters (such as what youth need or how to best meet a set of pre-defined needs). This could be explained as a limit of research design (as will be discussed further in Chapter 6). I did not propose to interview learners, and did not have ethics approval for this kind of interview. What I had anticipated was interviewing Mpintshis (the younger volunteer peer educators who work with groundBREAKERs) who were at community sites where I worked for an extended period of time. However, as discussed above, the majority of my time was spent in two community sites and neither groundBREAKER worked with Mpintshis during the school hours when I was with them. Faced with this limitation, I spoke to one of the groundBREAKERs and we planned a series of after school sessions with the Mpintshis. Week after week the sessions were postponed: people forgot, there were after school activities, people were out of town. I then tried to arrange interviews with the Mpintshis outside  59 of these sessions. Again, over the course of several weeks, the interviews were postponed: parents had not yet signed the consent forms, one Mpintshi had her form but it was getting late and it would not be safe for her to walk home after the interview was done. Across these moments I found myself chasing youth voices, a troubling return to the desire for youth voice that I have critiqued within the context of HIV prevention discourses (Gacoin, 2010b).  Instead of becoming stuck in the “failure” to interview youth, this “failure” has invited a different engagement with the interviews. I cannot, and do not aim to, represent what educators or learners think about a particular pedagogical encounter. Instead, in the chapters that follow, the interviews are yet another thread in the collection of stories and selves that take form with(in) the pedagogical encounter. In other words, there is a shift from what is said about the pedagogical encounter to how interview accounts are forms of meaning that may circulate within that encounter. This is another invitation to follow where the pedagogical encounter might go within its already multiple and contested relations of power and knowledge. Drawing upon the interviews also offers opportunities to introduce the educators. These introductions are woven through the chapters (instead of contained in initial “profiles” of participants for example) as a part of resisting the “proper” ethnographic subject who is given voice through their visibility (Talburt, 2004b). Who the educators are seen to be are always partial accounts, formed within my experiences working with them during sessions as well as what these educators chose to share with me when asked a series of background questions (see Appendix B). These articulations of self can be read for the “dilemmas their multiple locations in time and space represent” (Talburt, 2004b, p. 109). In other words, who we all are/were/could be is engaged as an ongoing question within both the pedagogical and research encounters of sexuality education.   60 The work of educational artifacts   The third method that I engaged with for this study was artifact analysis (Hatch, 2002), where artifacts were broadly defined as educational materials that are a part of the loveLife programme. I initially identified two types of artifacts: (i) official loveLife educational materials, such as program guides, training materials, media materials and reports and (ii) educational materials that are created and used by educators to deliver educational activities. I accessed these materials over the course of the study and recorded observations related to these materials (as well as oral permission from educators to use them) in my fieldnotes.   Many of my recorded observations related to these materials are steeped in frustration, particularly around educators not using the “scripts” that loveLife provides (as will be discussed in more detail in Chapters 3 and 4). For example, the love4Life training manual consists of 28 sessions grouped into 12 modules. The module on sexual rights and responsibilities includes two sessions explicitly on gender. Driven by my research questions, I wanted to see how these sessions were “done,” and this desire was a formative part of the pedagogical encounter as discussed in detail in Chapter 3. Working through this frustration as part of my own commitments to particular forms of knowledge has shifted both what I take to be an educational material, as well as how I conceptualize the work of those materials. Following Prior (2008), social science research has increasingly paid attention to the “function” of documents (what it does), rather than just the “content” (what it says). However, “function” has often been defined in terms of what humans do with a document, rather than seeing the documents themselves as active agents. This view underpinned my frustration: I had trouble engaging with the documents outside of what educators “did” with them. Seeing documents as active agents has allowed a conceptualization of the “function” of documents that  61 extends into how they may act as “props, allies, rule-makers, calculators, decision-makers, experts, and illustrators” (p. 828). Specifically, I am interested in how any document (such as a curriculum) is already a part of the pedagogical encounter whether or not it is “used” by the educator. For example, while the loveLife “script” was rarely used, it still operated as an “expert” in the pedagogical encounter, both in terms of how educators drew on its discourses, as well as how my research focus was shaped through it.  This view of the work of documents has, in turn, extended what I consider to be educational materials at play with(in) the pedagogical encounter. For example, I initially took studies related to loveLife’s educational approach as “background” to my study (such as key research debates that I can position this work within). However, and echoing the discussion of loveLife as a localized context, focusing on the work of documents has allowed me to reframe these studies as part of the encounters of sexuality education. Furthermore, this view of documents refuses the very premise that the “right” message (such as gender equality) could ever be scripted into a curriculum and then transmitted in a linear and unmediated relation between teacher and learner. Curricular documents operate within particular imaginaries of the pedagogical relation. They can thus be explored for the relations of power and knowledge that circulate within those always political imaginaries, as well as their limits as documents encounter, and are encountered by, educators and learners. This is the space where documents themselves might exceed their stated intentions.  On the work of coding  Moving from the work of methods into the more explicit work of the text, I have found myself in another “stuck place” of research practice: the space between my analytic need to make data “something [I] can manage” in terms of analysis (Luker, 2008, p. 199) and the  62 reductive move that making something “manageable” inevitably entails. While coding has provided a way to work within this stuck place, I have used it cautiously, attentive to critiques that it can be a positivist practice (St. Pierre, 2011). I took up coding as part of the “wonder of data” where “new connections spark among words, bodies, objects and ideas” (MacLure, 2013, p. 229). I began by using a process of open coding (Saldana, 2013) to develop areas that I called “thoughts to play with”: initial ways of looking at meanings and connections and tensions across the data. Some of these, such as “pedagogies” were in line with my research questions. I used others, such as “knowledge as a non-linguistic event,” to explicitly draw attention to areas that I might have missed within the research questions. In each of these “thoughts” I was interested in connections and tensions across the codes rather than organizing the codes into themes. In the second stage of coding, I used ATLAS.ti v. 7, computer assisted qualitative data analysis (CAQDAS) software,20 as a tool to think differently within the codes. I coded all interview transcripts and fieldnotes using three broad types of coding (Saldana, 2013): attribute coding (such as whether an activity was a “session” or an “event” and where that activity took place, as well as where I wrote up my notes); eclectic coding (using broadly descriptive codes that included sub-types such as “emotion” codes); and simultaneous coding (assigning multiple codes to the same piece of data). I was particularly attentive to simultaneous codes in order to resist a single reading of the data. In the third stage of working with codes, I used multi-coloured pencils to create conceptual mind-maps from the codes. For example, I selected the very broad code “race” and then worked with the AtlasTi software to think with and across fieldnotes, interview transcripts and educational                                                 20 See, accessed January 22, 2015, for a description of ATLAS.ti as well as other available software.   63 artifacts. My focus was on moments of connection as well as tension in relation to that code. I used the colors to visually re-engage with the “thoughts to play with” I initially developed. At the same time, as I grouped codes into those areas by color, I used the visual “mess” (the overlap of colors, lines going in different directions, no clear structure to what color went where) to challenge how I was thinking about the boundaries of those groupings. Finally, I used the conceptual mind-maps to develop broad analytic areas. It was the entanglements of these areas, as seen through the mind-maps, which shifted my focus from the initial research questions to the opening propositions as explored in Chapter 1.   The work of coding enacts the tension between making data “manageable” and the reductive move that managing data inevitably entails. Every code carries with it the risk of becoming the “truth” of the data (St. Pierre, 2001) and a closure of how something has come to count as both “code” and “data.” Furthermore, the organizational pull to patterns is tethered to positivist practices of “causal explanations and universal generalizations” (Pascale, 2011, p. 60). At the same time, rather than an impasse, it was the very process of coding and its risks that continually provoked me to question lines of connection and tension within the codes. Through analysis memos (Luker, 2008) I asked myself questions such as (Lisa Loutzenheiser, Personal communication, October 2013): what do I think I am seeing? What are initial connections and tensions? What might I be missing? In these questions, the codes worked as analytic sparks for problematizing meanings precisely through how they potentially closed those meanings. I then took up those sparks for the work of this text.  Working the text   In the textwork of this study I have engaged with writing as a mode of analysis (Ellsworth, 2005; Richardson, 2004; St. Pierre, 2011). Writing, including ethnographic writing,  64 is a power-laden act both in terms of who is represented and how (Clifford, 1986) as well as who writes and what forms of writing are legible as academic writing (Behar, 1995). This view of writing has been key for destabilizing the supposed neutrality of the text and the transparency of the “facts” presented within it. It has also invited new ways of writing that challenge what counts as (western) academic writing (Richardson, 2004). At the same time, in taking up writing as a mode of analysis, I have specifically focused on the process of writing. Here the interest was how writing could be used to invite excess within the text itself. In other words, the act of writing may be engaged to think about spaces of excess, or that “difference within the story” (Britzman, 2000, p. 38), even as I tell that story.   The stories of this text are told through a strategy of juxtaposition. Drawing from documentary filmmaking, Ellsworth (2005) explains juxtaposition as putting together a diverse range of “source materials” in order to “look at the same event from many different angles and experiences” (p. 13). Here, materials that are “supposed to be separate” are read together, and those “things and ideas that are often seen as being the same” may be read for “volatile spaces of difference” (p. 13). The aim of juxtaposition is to “make something new and different of what we already think we know” (p. 13). Taken to this study, I took the vignettes, as described above, as encounters that were not only to be looked at from different angles, but had also been constituted through multiple forms of power and knowledge. I therefore read each vignette for connections and tensions with other threads of data as well as the broader research literature. I then wrote about the vignettes, juxtaposing those threads of data and research literature within the writing. Through multiple drafts of these chapters I’ve asked questions such as: What else would I expect to see here, and what happens if I do not include what is expected? Are there threads of data that invite seemingly separate readings, and what might happen by putting them together? What  65 might be included here to complicate how I am reading the data? These questions are woven into the structure of the chapters. There is also a purposeful shift between different analytic foci and tones as part of disrupting the comfort (both my own and the readers) found in linear research stories. This includes the sections entitled “Encounters” which mark analytic pauses on the notion of gender that might not seem to “fit” with(in) the pedagogical encounter: a story of how I came to this research project, the “background” of how gender has come to be a target within HIV prevention discourses, and the processes of curricular development that have made gender a “teachable” topic in South Africa. The structure of the text marks these discussions as “outside” of the pedagogical encounter (as they are often seen to be) at the same time as I engage them as inseparable from it and use them to provoke my own thinking about pedagogy and its relations. Overall, I have used writing to invoke and provoke unintended associations and moments of tension within the practices of research. In doing so the text itself has become a repetitive and iterative structure, where the “final” text (the chapters that follow) is simultaneously absence and presence. The (im)possibilities of doing differently  In this chapter, I have explored the “stuck places” of my research practice as part of what has invited me to think about the “doing” of that research differently. While this lays out the path that I take in the chapters that follow, it is also an invitation to readers as they encounter this research. The ethnographic relationship between researcher/author and reader has traditionally been bounded by a hope that a “glimpse of the ‘real’ will guide readers’ thoughts and actions” (Talburt, 2004a, p. 83). This hope carries with it promises of “generalizability” or “transferability” to other contexts, particularly seductive aims in the context of sexuality education as an HIV prevention strategy.   66 My methodological approach has foreclosed, at the same time as it is haunted by, this kind of reading. I challenge the logic of problem/solution even as my arguments inevitably settle into their own analytic solutions. I speak to the problematics of a “cure” in development research (Chilisa & Ntseane, 2010) even as the imperative to “do” differently becomes a methodological cure. As discussed in Chapter 1, I do not seek to explain away this tension. Indeed, it constantly reminds me that sexuality education is already being “done” and that this “doing” has lived effects on the bodies of youth it aims to “help.” My own “doing” in research is a political act that cannot solve its own dilemmas. This is the “doing” from which I invite the reader to think with me, even as I hope that the reader will re-imagine the “real” created through this text in ways that exceed my own claims to know it.    67 Chapter three: Encountering gender   Gustav worked as a groundBREAKER in a black isiXhosa-speaking township outside of Cape Town. Often, before meeting with Grade 6 learners at the primary school, Gustav and I would sit chatting in the office that she shared with another non-governmental organization. One day, some of her office mates invited us to a workshop on “gender” that they had organized for a small group of men and women from the community.  The discussion is about gender roles and the facilitator asks: What are societal expectations related to gender in your community? Gustav responds: if I see a 10 year-old-boy playing with a doll I will think it is strange. He must be gay. (Fieldnotes, April 2013)  As Gustav spoke, I temporarily forgot my own sense of discomfort in this workshop – my position as an outsider to this space of a workshop organized by another NGO, my confusion since most of the conversation up to this point had been in isiXhosa, my personal understandings of community from a childhood in Texas and my travels in and through different communities since. I latched onto Gustav’s words: here was data that responded to my research question asking how educators and learners articulate notions of gender and gender relations. For example, as she conflated playing with a doll with being gay, Gustav drew on an articulation of gender norms (masculinity, femininity) that exist in an “irremediably slippery,” and yet irreducible, relation to notions of sexuality (Sedgwick, 2008, p. 29). Bodies, here a 10 year-old boy, are made intelligible through the “mark of their gender” (Butler, 1990, p. 282). Those bodies are made strange when they work at the boundaries of seemingly mundane norms, such as playing with a doll. Gustav was speaking as a learner in this workshop, but as a groundBREAKER she would herself be expected to guide young learners through lessons related to gender equity. I found myself wondering how this articulation of gendered identities and (Othered) relations would take form in her own classroom, and it was with this question in mind that I later asked Gustav what  68 she had thought about the workshop. She said she enjoyed it, and then added: I love playing devil’s advocate. I asked her if she had been playing devil’s advocate during the workshop. Of course, she said, I don’t believe those things.  One potential way to read this exchange is that, in a troubling return to the promise of objectivity through research, I accepted Gustav’s words as the transparent truth of “belief” and then made them into data (St. Pierre, 2013). I took my own discomfort as something to put aside (Pillow, 2003). Focused on “finding” gender power, I tried to think about power and knowledge outside of the relations of their formation. Indeed, and echoing the very assumptions that this study seeks to challenge, I encountered the notion of gender as something that can be clearly articulated (in a workshop or through a curriculum) and transmitted (from participant to researcher or from educator to learner). Yet, it is precisely because of these slippages that I start here, with one encounter with the notion of gender.  In that workshop, as Gustav and I encountered societal expectations related to gender in your community, we were already entangled in a “meeting-up of histories” (Massey, 2005, p. 4) that stretched across global priorities, national trajectories and local lives. This was a spatial register where a “local” discussion was led by a facilitator from a “national” organization that works across Africa, in line with “global” HIV priorities, to promote the goal of gender equity. Pausing on how Gustav chose to play devil’s advocate disrupts any imaginary of a linear flow across this spatial register, at the same time as it interrupts the logic of information transmission through the pedagogical encounter. As a learner, playing devil’s advocate may have been an act of resistance to the implicit assumption that all members of a community would share the same societal expectations. As a research participant, Gustav was responding to my interest in the workshop and, as she stated that she did not believe those things, she positioned herself as an  69 (implicitly “good”) educator who has “strategies to address gender equity” (See Project Overview Handout, Appendix B). Furthermore, Gustav described her own gender to me as “being a woman and what that means to you as a person,” before adding that she finds it a “difficult question to answer” (Interview, August 8, 2013). She also resisted my hailing of sexual subjectivities (and those found within the encounter of the workshop) when she defined her understanding of her own sexuality in terms of “sexual inclinations.” All of these entanglements foreground an impossible “answer” of gender identities and their relations in any educational encounter. The question then becomes what might happen to notions of gender with(in) sexuality education in excess of neatly scripted answers to transmit from teacher to learner.  In this chapter, I use the notion of encounters with gender to mobilize gender with(in) the inevitable risks of knowing and being known through educational research and practice. Here, I am specifically resisting an analytic approach that would seek to entangle those histories and moves and get back to what gender “is” (in the belief of the educator, for example), even as I recognize the inevitable analytic pull into what can/should be “known” about gender. To work with this tension, I begin from the view that what gender “is” in any moment is “charged and enacted in the sticky materiality of practical encounters” (Tsing, 2005, p. 1). This view foregrounds the focus on gender as something to “teach” through sexuality education as extremely problematic, at the same time as it takes that focus as providing a unique opportunity to work with multiple and contested meanings of the very notion of gender. I invite the reader to momentarily lose the received “truth” of gender by going into the “middle of things” (Tsing, 2005, p. 1). This particular “middle of things” picks up from when I suggested to Lu, a groundBREAKER who described his own gender as a “son of the world” (Interview, August 15, 2013), that we do the love4Life “Gender” session. I had been working for  70 several months with Lu in the black township that he also called home (even as he resisted the identity category of black as part of living in a “democratic South Africa”). I was getting to the end of fieldwork, was feeling anxious that we had not yet engaged with the scripted “Gender” session. Before the class started, Lu and I sat in the reception area of the high school and I asked him how he wanted to run the session with the group of grade 9 learners. He said that we would do the introduction together and then he would have the group create role-plays around gender equality and gender inequality. I was worried (stuck within my own focus on wanting to observe the “Gender” session) because this was not part of the script. At the same time, we did not follow this plan either as we encountered learners and forms of knowledge in this place and time. In this chapter, I engage this particular pedagogical encounter, reading relations of power and knowledge that settled within its contours, at the same time as I use it to encounter and think through (and with) further encounters with “gender” in this research project. Through these encounters, I ask what kind of political project might be available around notions of gender when these notions are seen to be both unsettled, and unsettling, with(in) any articulation of sexuality education.  An encounter through definition   The first activity in the love4Life “Gender” session is entitled “How we learn gender roles.” The broad aims of the activity are to “understand how socialization can trap us into certain roles as men and women” and “challenge some stereotypes about men and women” (loveLife, 2012f, p. 154). Lu began by introducing the topic:  Lu asks: what is gender? One boy says male and female. I continue: are there different ways that boys and girls are brought up? Most of the learners say yes – parents allow boys to go out at night, guys think ladies are weak. Next question: why are there differences? One male learner explains that guys have a penis. No, says another male learner, we are all human. (Fieldnotes, August 2013)    71 This introduction began with a definitional question: what is gender? The response that was given (gender is male and female) might be read as the learner giving the wrong answer. According to the loveLife script, sex refers to “certain physical differences between males and female” and gender refers to “the way we are brought up and socialized” (loveLife, 2012f, p. 157). Following the logic within this sex/gender binary, which is propped up through how sex and gender are discursively constructed in international development discourses,21 the learner should have responded that gender refers to “men” and “women,” or brought in some discussion of social roles. Within a focus on what is being defined, the slippage between sex and gender illustrates the learner’s need for the “correct” information. Furthermore, the slippage could reflect how these terms are often used interchangeably in English and that the educator and learners were English second language speakers. At the same time, I can extend this focus on what is being defined by reading the slippage, and the way it quickly passed unchallenged by both Lu and myself, as an effect of the very instability of the sex/gender binary and its effects. While a sex/gender binary explicitly distances “gender” from biological bodies (sex), the binary remains tethered to a biological truth of bodies through that binary (Butler, 1990). In other words, gender may be a social construction (norms of masculinity and femininity), but the illusion of two fundamentally separate bodies (male and female) is necessary in order to enact those gender roles. Here, the instability of what is being defined is an effect of how understandings of gender remain reliant on underlying constructions of male and female bodies.                                                  21 For example, the World Health Organization defines sex as “biological and physiological characteristics that define men and women” and gender as “socially constructed roles, behaviours, activities, and attributes that a given society considers appropriate for men and women” (see, accessed January 25, 2015)  72 The focus on what is being defined allows engagement with relations of power and knowledge that are circulating within the “answer” given by youth to a question (what is gender?), and thus contribute to problematizing an “answer” that could be held separate from those relations. For example, the “answer” given in this particular encounter drew on understandings of self and Other built on the biological binary of male/female bodies. Far from symmetrical and neutral, binary oppositions prop-up and validate one term through the “subsumption and exclusion” of the other (Sedgwick, 2008, p. 10). This includes the “truth” of biological bodies as male/female and the establishment of masculine/feminine as “expressive attributes” of those bodies (Butler, 1990, p. 283). The risk is thus that relying on a distinction between “sex” and “gender” to teach gender equality, as seen in this encounter as well as in loveLife’s definition of “gender roles” (loveLife, 2012f),22 may actually reify and mask the very power relations that it explicitly aims to address. Working with and engaging the potential effects of this risk necessitates not only analyzing what is taught, but also how that “answer” becomes coherent within the pedagogical encounter.  Each of the “answers” given by learners in this particular encounter revolved around a definitional question: what is gender? This question is built on the assumption that there is a knowable answer, and indeed this is the explicit aim of this activity: making sure that learners understand what gender “is” before the session continues. The issue, however, is that the very move to fix gender into something that “is” has already relied upon, and in doing so redeployed, particular ways of understanding gendered identities within relations of power and knowledge.                                                 22 In a section entitled “Notes for the facilitators: Gender roles,” the following definition is given: “Whilst there are certain physical differences between males and females, the way we behave and the roles we assume as males and females are largely determined by the world we are brought up in” (loveLife, 2012f, p. 157).   73 For example, in the second part of this vignette, I relied upon an underlying binarized construction of gendered identity to ask the question: are there differences in the ways that boys and girls are brought up? Following the love4Life script, the aim of this discussion is to disrupt gender roles as “natural” and challenge “stereotypes that unfairly put people into categories that rob them of their personhood” (p. 157). Yet, the coherence of the question about differences in the ways boys and girls are brought up is reliant upon particular and binarized conceptions of personhood: boys/girls. Learners are thus already being understood in terms of categories based on gender (speaking as boys or girls) as well as age (as youth who are still being brought up). Learners are then expected to take up these positions to speak about how parents treat children differently – boys are allowed to go out at night – as well as articulate particular gendered identities – ladies are weak. In contrast, and in line with their role of “trying to eliminate gender bias” (loveLife, 2012f, p. 156), the educator is positioned as being able to step outside of any bias that they themselves might have. This positioning relies upon, at the same time as it reifies, a view of educators as “deliverers of an uncontested, sanitized and agreed upon body of content” (Baxen, 2010, p. 17). It also props up a view of education as the solution to the “problems” of upbringing, drawing coherence from long histories of sexuality education as a corrective to the racialized and classed (and non-western) family (Stoler, 1995).  Pausing for a moment on the use of English in this encounter can help to further problematize the pedagogical act of definition and multiple relations of power and knowledge circulating within it. The love4Life script, given to each groundBREAKER at the beginning of the year, is published in English. I asked David Harrison, the former CEO of loveLife, about the organization’s decision to use English for the programme materials. Outside of financial considerations (the cost of translation) he also said that early loveLife research showed that  74 English was the “language of aspiration for young people” (Personal communication, July 5, 2013). While the “language of aspiration” points to an equitable future, it is also entangled in contradictory colonial legacies and their ongoing formations. Under apartheid, English was an official language of the apartheid regime. Today, while there are now 11 official languages and fewer English first-language speakers than Zulu, isiXhosa or Afrikaans first language speakers, the governing African National Congress promotes English as the language of the government, and English is widely used in business and the media.23 This is not only to say that young people’s “aspiration” has been shaped by colonial histories or neo-liberal interests. It is also to point to how any “language of aspiration” has already been formed through multiple relations of power and knowledge. For example, the very notion of “gender” has been scripted as something to teach through the western epistemologies of English: from global policy documents to programmatic resources. Here, English gives form to a particular articulation of the very notion of “gender,” constructed through power-laden understandings of identities and their relations (Haraway, 1991). Even if these resources are “translated,” they will still carry the traces of western epistemologies formed in and through the English language. This throws into question the possibility that, as Grace Matlhape, the current CEO of loveLife explained, English could ever be “just the language of the content” (Personal communication, August 7, 2013). Indeed, Grace’s framing of loveLife’s decision to use English is in tension with how she recognized that her own language choices, such as being more comfortable speaking in English than Zulu, could “discourage questions” and “shut young peoples’ voices.”                                                  23 For a list of South Africa’s official languages see, accessed January 22, 2015.  75 In this particular session, as throughout the year, Lu encouraged learners to speak in English, something he described as part of a decision to “practice English.” Partly, this was necessitated by my presence, and as we worked together Lu (who spoke isiXhosa as his first language) repeatedly reminded learners to speak in English, or translated what they said, so that I could understand. As introduced in Chapter 2, this was part of the tension between what “should” happen (I worked with loveLife staff to identify groundBREAKERs who would be implementing in the English language) and the imperative to speak in a particular language that was already entangled in power-laden relations of knowing and being known through language. In this instance, the need to “practice English” was part of our introduction to learners at this school: we had been told that these learners had poor English skills and loveLife should implement in English in order to help build those skills. When I asked Lu about his decisions around language in his pedagogical practice, he said that many learners are “able to understand it,” but that they are “afraid of speaking English around Xhosas” (Interview, August 15, 2013). He explained this as being afraid that “others will laugh at them if they do try to speak [English],” adding “we are not perfect in it” but “we are trying.” In this description, multiple relations of knowing, and being known, momentarily came together through decisions around language in pedagogical practice: the white North American “gender” researcher who speaks English, the positioning of black South African youth as needing to learn English, the “we” who are trying to speak English. In this particular session, these power-laden relations of self and Other were entangled with language as it worked at the limits of what can be defined (gender as sex/gender), who controls the definition (the “correct” answer according to an English script) and the deployment of particular “aspirations” within that definition (here: gender equality). Another day, as we sat going through a session we had just facilitated together, Lu foregrounded the complexity of  76 language when he told me that it really was a shame that I could not understand what learners said in isiXhosa. He added: my translation might sound “nice,” for example, but what they said was “angry” (Fieldnotes, April 2013). Lu’s words marked a limitation of this research (as will be discussed in Chapter 6) at the same time as they foregrounded aspects of language that always escape definition. Language can carry the anger of knowing and being known. It is always formed within multiple histories of power and knowledge, and entangled in the imagined futures for those who can/should speak it. These dilemmas cannot be explained away as a problem of translation or the incomprehension of second-language speakers when defining a term. Instead, if language has the potential to be a “vital clue to where the self loses its boundaries” (Spivak, 2008, p. 202), then a key question is whether the pedagogical approach challenges or reifies linguistic boundaries of self and Other.  As Lu and I encountered “gender” in this session, the boundaries of the self as a known position were being constructed through pedagogical practices of definition. For example, one learner described gender difference in terms of a biological reality, saying that guys have a penis. In contrast, a second learner seemed to resist this type of biological determinism, even as his statement that no, we are all human draws on humanist conceptions of a “self-contained, authentic subject” (Alcoff, 1988, p. 415). These responses may seem contradictory, but they were both invited through a focus on definition and, in turn, they both rely upon a centering of identity as something known and knowable. As such, they can both be contained by a pedagogical approach that seeks to define identity as something that “is.” The risk of this containment is not only that the pedagogical approach “underplays the extent to which subjectivity in South Africa is a raced, cultured, gendered and classed experience” (Soudien, 2009, p. 41). Additionally, by failing to engage with how conceptions of self are formed through  77 power-laden understandings of the Other, definition may reify the very relations of power that the pedagogical approach aims to challenge. In other words, restricting gender empowerment around a definition of what gender “is” inevitably draws on the power-laden discursive norms (such as the hetero-normative boy or girl) that make gendered identity intelligible at the same time as it holds gendered subjects responsible for overcoming those norms (being empowered).  The perceived necessity of definition within the pedagogical approach is itself a mark of the inherent instability of gender as something that “is.” As seen in this encounter, definition is always a limiting act: an impossible division of self from Other and their formation within entangled relations of power and knowledge. The question then becomes how the pedagogical approach meets and engages the limits that it inevitably repeats.  An encounter within the pedagogical approach  Lu and I held the class outside, sitting on a concrete slab surrounded by overgrown grass. There were long silences, and wanting to boost the energy of the group, I started an agree/disagree activity from the love4Life script. As the facilitator, I read a statement from the script, such as a girl wearing a short skirt is asking to be raped. In response, learners moved to opposite sides of class space depending on whether they agreed or disagreed with the statement.  Agree or disagree? Lu translates for me as the learners speak. While the translations help me follow the session they also feel like an interruption, pausing the discussion as well as the movement of the activity. I keep reading the statements – A girl wearing a short skirt is asking to be raped. A group of boys agree: she is showing off her body and guys are attracted. No, disagrees most of the class, including all of the girls, it is just how she is dressed. I read another statement: It is our culture that men have more than one wife. Most of the boys and one girl agrees. A girl who disagrees explains: you don’t know if those men have an infection. Another statement: A man is head of the household. Most of the boys agree: they pay ilobola24 and work. The girls disagree: women can also work. (Fieldnotes, August 2013)                                                  24 In South Africa, ilobola refers to the practice of a man providing compensation (such as cattle or cash) to the family of the woman he is marrying. It is considered to represent the customary   78  My first reading of gender in this vignette focused on finding gender: exploring the session for how learners understood gendered identities and relations. This has been a dominant way of engaging “gender” in educational research with youth, including in South Africa, and has been crucial for destabilizing individualistic and (supposedly) gender-neutral HIV prevention messages aimed at youth. In this vignette, for example, the comment that boys pay labola and work, and are therefore the head of the household, draws on the discursive construction of a “provider” masculinity that is constituted through co-existing discourses of “tradition” and “modernity” (Hunter, 2010; Jewkes & Morrell, 2010). As part of hegemonic masculinity (Connell, 1987), a dominant and highly valued form of masculinity that is “structured along lines of gendered domination” in South Africa (Jewkes & Morrell, 2010, p. 3), these understandings of gendered identity are in tension with HIV prevention messages that expect young men and women to equally embrace an empowered modern subjectivity. Furthermore, and as seen in the ways that learners make sense of femininity within this vignette, already contested understandings of gendered identities are at play. On the one hand, social norms of femininity and masculinity that position men as active and women as passive (Walker, Reid, & Cornell, 2004), together with a normalization of violence as a part of masculine control of women’s bodies (Jewkes & Morrell, 2010), allows a group of boys to collapse being attracted to girls wearing a short skirt to possessing her body through force. On the other hand, there are moments when girls actively resist these constructions. For example, one girl draws on public health’s “modern” individual (Lupton, 1995), saying that you don’t know if those men have an infection, to speak-back to cultural (and implicitly “traditional”) expectations: men should have more than                                                                                                                                                        negotiation of a marriage (see, accessed January 25, 2015).   79 one wife. In contrast, while it may be said that the girl who agreed with those expectations was “trapped” within them, her agreement can also be explored as an expression of sexual agency within those expectations, even if that agency does not align with HIV prevention goals. Being one of several partners may be a space of negotiation for gendered expectations within a sexual relationship (Jewkes, 2009) or a means to achieve economic aims (Winskell & Enger, 2009). The language of multiple concurrent partners, while framed as a “key driver for HIV infection” (UNAIDS, 2008), may also be appropriated as a woman also having the right to multiple partners. Overall, as moments that highlight contested and power-laden understandings of masculinity and femininity, these readings contribute to problematizing simplistic understandings of youth’s worlds that plague HIV and AIDS prevention work (Bhana & Pattman, 2009).  At the same time, I argue that it is a risky move to unmoor these understandings of masculinity and femininity from the conditions that have enabled them to be articulated. In other words, youth’s understandings of gender are often taken as something for the teacher to address, or the researcher to identify, within the pedagogical encounter. However, the separation of the pedagogical encounter from what is said within it is impossible if youth voice is seen as always embedded within multiple power relations that play out in who speaks, what is expected to be spoken about, and how that speaking occurs (Bragg, 2007; Gacoin, 2010b; Lodge, 2005). In the case of this agree/disagree activity, who spoke was invited through a pedagogical “mode of address” (Ellsworth, 1997): an address through curricula and pedagogies that “invite their users to take up particular positions within relations of knowledge, power and desire” (p. 2). As the educator, I invited youth into this encounter by asking them to embody gendered positions grounded on shared cultural definitions of gender. Those cultural positions, as will be explored  80 further in the next section, were already entangled in contested understandings of self and Other within relations of power. Instead of challenging the effects of those relations, the pedagogical approach deployed them as the basis for participating in the session: there were mutual (agree), even if contested (disagree), understandings of who men and women are said to be. What was spoken about, in turn, was “gender.” As a topic, a particular articulation of gender (gender roles and expectations) was centered through the pedagogical act of definition, as discussed at the beginning of this chapter. How that speaking occurred, in turn, was invited through the pedagogical approach. As I facilitated the activity, I watched as learners physically embodied the gendered binary of the activity: it was us against them; boys against girls. A few learners did break the gendered space, but no one stepped into the physical space in-between the groups that had been instructed to agree or disagree. In other words, the pedagogical use of space worked in the service of particular lines of intelligibility for gendered identities and their relations. Agree and disagree became about the right and wrong answers of gender equality.   When gender equality is a matter of the “right” and “wrong” answers, the “problem” of gender can be addressed through the “solution” of sexuality education. My concern, however, is that the logic of problem/solution masks how the pedagogical encounters of sexuality education as an HIV prevention strategy are themselves “an extremely important terrain of struggle and contestation” (Hall, 1992, p. 272) of representation. Consider for example, another statement from the agree/disagree activity:  I read the statement: A boy wearing tight pants is asking to be raped. Most of the learners disagree. A small group of girls agree: skinny jeans are only worn by guys who are gay. He will be raped by another guy. (Fieldnotes, August 2013)   In this exchange, a particular configuration of the limits of “gender,” and how those limits act on and through the intelligibility of gendered subjects, was momentarily made visible. The implicit  81 address throughout the session had hailed a heterosexual youth subject: boys and girls who were attracted to one another. The conflation of gendered performance (wearing tight pants) with sexual identity (being gay) was a momentary enactment of how the privilege of this heterosexual subject is formed through power-laden understandings of social identities and their relations. Gendered power, particularly masculine privilege, is constructed through what it is not: not feminine, not gay (Sedgwick, 1985). In this particular example, a heterosexual masculine power acts on, and is legitimated through, boys who are perceived to be gay. This is in tension with loveLife’s recognition that “one of the most common forms of verbal violence heard at schools targets staff or students who are perceived as gay” and the explicit educational aim of “creating respect” (loveLife, 2012f, p. 23). “Adding” this message into a curriculum, for example, is an explicit and important act of resistance to these forms of violence. However, the tension between this message and identities articulated within the pedagogical encounter points to the impossibility of calling this message a “solution.” The tension between the “message” and the reification of heterosexual male privilege within that message is a small enactment of how understandings of gender are always already drawing on particular lines of intelligibility in order to be coherent, including the entanglement of gendered identity with other social identities and their power-laden relations. This is further complicated by how social norms are already policed through clothing within educational spaces. Uniforms, for example, were used under the apartheid educational system to inscribe raced and classed articulations of “correct” gender performance (Diko, 2012). While the current South African National Guidelines on School Uniform seeks to ensure that practices related to school uniform do not “impede access to education” or “infringe any constitutional rights of persons” (Department of Basic Education, n.d., p. 1), school uniforms remain a site for the policing of gender norms within multiple  82 relations of power. There is anecdotal evidence, for instance, of school uniform policies (and particularly the requirement that girls wear skirts) being used to discipline gendered identity, including an increase in violence against learners who are (or are perceived to be) non-gender conforming (Kings, 2014).  Focusing on how notions of gender are engaged through the pedagogical approach foregrounds how that approach potentially shuts-down the complexity always already happening within any pedagogical encounter. As the educator in this particular encounter, I found that the structure of the activity pushed me to gather an answer (why learners agree or disagree) and then move onto the next statement. In this move, the “aim” of gender equality potentially reified other inequalities. For example, since the beginning of the HIV and AIDS epidemics, homosexuality has been discursively constructed as a site of “risk” in order to reify heterosexuality as a “safe” identity (Treichler, 1999; Weeks, 2010). Furthermore, the “safety” of this identity has been formed through globalized relations of power and knowledge that make “gender” a problem in heterosexual relations: Africa as a place of “seething sex and rampant death” that requires educational intervention (Patton, 2002, p. xiii). Indeed, standing before this group of learners, my body was a material enactment of the “colonially established framework of homogeneity in the search for answers and solutions to the HIV/AIDS epidemics” (Chilisa, 2005, p. 668). I was a white North American cisgender researcher who wanted to “find” gender and who found my own coherence in the focus on heterosexual relationships. I was using an English script designed to get youth to “understand” and “challenge” gender roles. The gendered subjects being addressed through these already entangled relations of power were all black isiXhosa-speaking learners from a poor township outside of Cape Town. Of course, it could be argued that I should not have been the educator in the first place – that I am an outsider, that it was not appropriate.  83 However, removing my body does not erase the relations of knowing and being known already at play as “gender” is deployed within sexuality education as an HIV prevention strategy. These were the dividing lines of intelligibility already working at the limits of who we were already said to be and what we could supposedly know in this encounter: researcher/participant; north/south; white/black; foreigner/national; homosexuality/heterosexuality. Driven by its own binary logic, a pedagogical approach based on equality/inequality became part of the repetition of the very terms it claimed to challenge. For example, through the binary framing of agree and disagree, leaners were asked to embody particular gender positions in order to then speak about gender power relations and their effects. While learners did discuss gender power relations, as per the aim of the session, how they did so is precisely what destabilizes the claim that understandings of gender can ever simply be transmitted through sexuality education. The pedagogical approach was a constitutive part of what became coherent as gendered identity and relations within that particular moment. It also relied upon, and in doing so reified, understandings of gender identity formed through power-laden understandings of other social identities.  Encountering gender within the pedagogical approach is an encounter with that approach as an always-political act. The point here is not that the pedagogical approach is a (supposedly neutral) tool for transmitting a political project. Rather, the pedagogical approach is itself what Foucault calls a technology of power: a “whole set of instruments, techniques, procedures, levels of application, targets” that “assures an infinitesimal distribution of the power relations” (Foucault, 1995, p. 216). In the encounter discussed in this section, the pedagogical approach mobilized particular relations of power and knowledge and settled them into what can be “known” about gender. Supported by the ideological language of equality, the approach potentially reified  84 understandings of social identities and their relations that have unequal and violent effects in people’s lived lives. This may appear to foreclose a political project concerned with social transformation, and yet I argue that engaging the approach within the pedagogical encounter sets the conditions for beginning to think differently about it. As a technology of power, the pedagogical approach does not act on (supposedly) docile subjects. The approach is part of the texture of the pedagogical encounter at the same time as that encounter cannot be reduced to it. It is within the pedagogical encounter that notions of gender encounter, and are encountered by, educators, learners (and researcher) in ways that may trouble its own conceptual boundaries. This includes the conceptual centering of gender around notions of culture, as seen in the next encounter.  A cultural encounter The coherence of teaching “gender” to youth, as discussed in both the opening encounter of this chapter as well as the agree/disagree activity from the “Gender” script, relies upon culture as a shared conceptual frame. Men and women from the same community can come together to discuss societal expectations related to gender. Girls and boys can (and should) agree or disagree with statements related to gender because they implicitly share the same cultural understandings. Culture is the background to the pedagogical encounter: the shared canvas from which meanings will be made.   My first encounter with another small group of learners in central Cape Town was a small enactment of the “elevated cultural authority” (Treichler, 1999, p. 45) that has been given to gender in HIV and AIDS responses, at the same time as it troubled this authority. An educator and I were meeting with the learners as part of a skills-development program led by another non- 85 governmental organization for “at-risk” youth. The issue, according to the program manager, was the boys:  the boys were drinking; the boys were getting big heads; the boys had mothers who live on the street; the boys needed the life skills that loveLife could provide (Fieldnotes, July 2013)   This introduction to the “boys” was already a particular construction of the boys’ gendered identities. For example, drinking alcohol is part of the gendered construction of masculinity in South Africa. At the same time, the discursive centering around gendered identity, the boys, masks how drinking is simultaneously entangled with the discursive construction of racialized (Jewkes & Morrell, 2010) and classed (Hunter, 2010) identities in South Africa. Furthermore, the coherence of the “solution” of life skills drew on a Piagetian theory of cognitive development, influential in South Africa (Soudien, 2009; VanDyk, 2008), whereby youth need skills to become the right kind of adults. Gender was centered in this account: mothers were the problem, and an implicit threat to the boys becoming the “right” kind of gendered subjects. However, the “solution” of life skills was also constructed through the ongoing imperial formations of “rescue” (of the Other) with western subjectivities and forms of knowledge (Stoler, 1995; Soudien, 2009): heterosexual, white, middle-class. Indeed, the influence of Piaget’s theories in South Africa, and attendant western discourses of humanism, is inseparable from how mass education of the “natives” under colonialism was said to be necessary in order to correctly guide youth through Piagetian developmental stages (Soudien, 2009). In other words, the very notion of “at-risk” boys is inseparable from the racialized and classed landscape of inequality in South Africa. While this discourse is framed as addressing those inequalities, it potentially reifies them by failing to interrogate the relations of power and knowledge that settle into who youth are said to be and what kinds of knowledge they are said to need. Here, the very issue of the “boys” was  86 already a problematic site constructed through multiple histories of knowing and being known as well as power-laden understandings of other social identities.   As the educator and I worked with these learners, this “introduction” did not stay in some hypothetical background space when the session began. For example, in my fieldnotes I found myself writing down all the references to drinking as the educators and I worked with the learners. While this focus was not deterministic, it is a small illustration of how a mode of address is a “structuring of the relationship” between teacher and learner (Ellsworth, 1997, p. 24). That structure is formed through “intimate relations of social and cultural power” that “shapes and misshapes who teachers think students are, and who teachers come to think themselves to be” (p. 6). Here, it was not just that the boys were drinking (and were therefore assumed to be troublemakers, for example). It is that who they were perceived to be as gendered subjects was being deployed through particular relations of power and knowledge. At the same time, as a “projection of particular kinds of relations” (p. 25), the pedagogical mode of address will never be an exact “fit” with conceptions of self and Other operating in a particular moment. This draws attention to potential moments of resistance in any pedagogical encounter: a mode of address only makes sense if its subjects, no matter how momentarily, take up the position being offered through it.  In this particular session, the mode of address to “youth” was inscribed within the aim of enabling learners to explore their “own personal identities” (loveLife, 2012e, p. 11). In tension with this aim, learners responded to the educator’s questions by appropriating the “youth” subject to avoid speaking about the “self”:  This is what influences youth; Youth do things with friends to be cool; Youth see their parents drink; Youth see drinking in the community. (Fieldnotes, July 2013)   87 There are multiple power relations that may have sparked this refusal to speak the “self”: resistance to adults deciding what youth “need”; resistance to how these learners perceived the educators or myself; resistance to how the boys were understood; resistance to the power-laden imperatives of a personal confession (as will be explored further in Chapter 4). Indeed, the very multiplicity of relations of power-resistance is an illustration of how the pedagogical encounter, as the space in which any mode of address and its responses are entangled, is a “social space, formed and informed by the historical conjunctures of power and social and cultural difference” (Ellsworth, 1997, p. 38). In this instance, it was questions of difference that were encountered through the very notion of culture as the session continued.  A learner says that culture influences identity. The educator responds: What do you mean by culture? She adds, speaking to the group of black and isiXhosa first-language speaking learners: I have less culture than you do. One learner gives an example of initiation as an example of culture and the discussion turns to how culture changes over time. Yes, agrees the educator, coloureds used to be expected to get married before having sex. That is also changing. But, she adds, that is religion not culture. (Fieldnotes, July 2013)  The opening statement made by a learner in this vignette, culture influences identity, is coherent with the discursive centering of culture as a shared conceptual frame for individual identity, as discussed in the beginning of this chapter. Furthermore, with the example of initiation, culture is deployed as an explanatory frame for a particular gendered subjectivity. In South Africa, initiation refers to a set of practices used to mark the gendered transition from childhood to adulthood. Framed within the logic of the session, enabling youth to name influences on identity, including culture, is crucial for “dealing with [those influences] efficiently” (loveLife, 2012e, p. 12) and empowering youth to gain “full control” over the future (p. 12).   Yet, it is the very logic of the session that is problematized as this “aim” encounters its subjects within contested relations of knowing and being known. Consider, for example, how the  88 discussion of initiation and change is embedded in a session guided by the logic of HIV prevention. Within HIV and AIDS discourses, initiation has been discursively constructed as a problem because it includes the practice of “traditional” male circumcision. Constructed in a binary with “medical” (modern) male circumcision, an HIV prevention method, “traditional” male circumcision is a risky cultural practice that needs to change (Baylies, 2004; Helleve, Flisher, Onya, Mukoma, & Klepp, 2009). Furthermore, “traditional” male circumcision, as a cultural practice, relies on a conflation between race and culture: it is unruly black bodies that are the problem.   The conflation between race and culture operated within the educator’s statement that I have less culture than you do. The educator, who described herself as coloured and Afrikaans first-language speaking, was addressing black isiXhosa-speaking learners. The coherence of this statement drew on racialized constructions of African sexuality, a powerful colonial formation (Reid & Walker, 2005, p. 186) that has driven both global (Humphreys et al., 2008; Patton, 1993a) as well as South African (Soudien, 2009) understandings of sexuality and the sexual subject. Crucially, in these discourses it is not only the “Other” who is constructed but also the “self.” The educator enacts this construction through difference when she explains her own beliefs as religion rather than culture. Here, her words marked her own understanding of her identity as being brought up Christian at the same time as they distanced her identity from the relations of power that have targeted black culture as a problem. Indeed, separating the self from “culture” is a powerful way to claim the ideal of a non-racialized identity in South Africa and reify a less-than identity for blacks (Bhana & Pattman, 2010), even as it points to this as a deeply problematic move. The privilege of social identity is drawn through the boundaries of what identity is not. Spoken through the coloured body, this rejection of “culture” is an everyday  89 enactment of how hierarchical categories of racialized difference, formed and lived within relations of power, continue to be enacted through the very notion of culture.  Coupled with the rejection of culture, the educator drew on religion to provide an alternative explanatory frame for gendered identity, such as the expectation for women to be chaste (Arnfred, 2004). Indeed, while loveLife is an explicitly secular programme, religion was a pervasive part of the encounters between youth, educators and loveLife messages. Community events often began with a prayer. Invited speakers during events drew on “God” not just as personal faith, but as part of the intelligibility for youth action and protection against HIV. In one session at a school, for example, the educator told learners that he assumed they were all Christian, because good people are Christian (Fieldnotes, November 2012). These encounters with religion are themselves an enactment of the tension between “overcoming” religious identity for HIV prevention and how religious identity has been embedded into understandings of HIV and AIDS (Dilger, Burchardt, & van Dijk, 2010). They are also in tension with loveLife’s vision of a post 1994 “modern” identity that can be known “beyond religion and culture” (loveLife, 2012e, p. 12). While it is widely recognized that culture and religion are two of the dominant discourses for understanding sexuality in South Africa (Arnfred, 2004; Baxen & Breidlid, 2004; Jewkes & Morrell, 2010; Hunter, 2010), exploring these understandings within the pedagogical encounter points to how positioning sexuality education as a site to “overcome” these understandings is already a problematic “solution.” This solution cannot be “fixed” by pointing out that educators will deploy and reify particular cultural subject positions as they teach, and therefore need training, or vilifying religion as an automatic barrier to HIV prevention messages. For example, while the educators I spoke with all identified in some way with religious or spiritual  90 communities, they did so in very different ways: a “higher power” (Makaziwe, Interview, July 15, 2013), “Christianity” (Lu, Interview, August 15, 2013), “ancestors” (Lukanji, Interview, August 12, 2013), “science” (Gustav, August 6, 2013). Read alongside the pedagogical encounter, these articulations of self are a part of the enactment of culture and religion as “highly contentious and contradictory site[s] where the geopolitics of sexuality refuse the stability of cultural, national, gendered and sexual boundaries” (Britzman, 1998, p. 75). It is in the contradictory nature of this site that the claim that messages related to gender equity can be transmitted through sexuality education as long as those messages are “culturally relevant” shatters. Within the pedagogical encounter discussed in this section, educators and learners drew on notions of culture to articulate understandings of self and Other, and these understandings were entangled in other identities, such as religious ones. These understandings worked through, rather that challenged, forms of power that settle into what come to be seen as discrete social identities (such as gender, class or race) and their relations of difference. In other words, the cultural “Truth” of gender was an already impossible starting point for the perceived goal of gender equality.  Encountering gender equality with(in) words  Lu and I had 50 minutes for the “Gender” session. At the end, Lu gave a summary of the key messages for the learners.  Guys have rights, he says, but should also help at home. In my culture, men are the head of the household (I used to think this, he explains to me after the session, but I don’t agree with it now). Girls weren’t allowed to work in the mines, but now they can. There is nothing that women can’t do. There is no specific duty for each and every gender. (That is my role, he later tells me, making sure they know about inequality and equality). (Fieldnotes, August 2013)   In these final words, Lu enacted a particular pedagogical vision of gender empowerment, organized into a linear and rational flow of identities and their forms of knowledge. In this vision,  91 youth should first recognize that there are unequal gendered expectations within their communities – such as men being head of the household. Secondly, youth must understand that these roles can change, as seen in the example that girls can now work in the mines. Finally, youth will accept that there is no specific duty for each and every gender. In this imaginary, the notion of gender is reduced to power relations that circulate through social norms and stereotypes. The problem is power imbalances between gender roles. The solution is education: understand and challenge unequal power relations (within a set amount of time, here 50 minutes). This approach can be seen as a form of critical pedagogy that seeks to transform the consciousness of learners (Apple, 2004). Lu himself embodied this transformation when he told me that he changed his own views because of the loveLife training and defined his own sexuality in terms of a rejection of “male dominance” (Interview, August 15, 2013). Armed with this transformation, he was ready to embody his perceived role as educator – making sure that learners know about inequality and equality.  There are many elements of this account that could be called a success, especially within the discursive framework of the loveLife programme. I worked with Lu for most of the year and, throughout our conversations, he spoke enthusiastically about his role as an educator, as well as the impact that loveLife had on his own life. For example, he told me he was “inspired” by his work as a groundBREAKER, and that, in turn, youth have been “inspired by what I’m doing” (Interview, August 15, 2013). I saw his commitment to “learn from” the learners: listening to feedback from learners after a session, coming up with topics based on issues he saw in his community, trying to make the sessions fun and engaging. In my notes, I remarked upon the sense of rapport that he seemed to have with the learners, and I continue to admire the way that  92 he resisted addressing learners at this school as deviant and unruly subjects (an understanding of youth at this school that I will discuss further in Chapter 4).  Alongside this “success,” and indeed in resistance to the collapse of a particular instance of success as “solution,” there are many parts of this encounter with the discourse of gender empowerment that continue to trouble me. I had suggested this session, and Lu may have been performing the “good” educator in relation to gender and my research interests. What he thought I wanted to know, as well as what I can claim to know, were a constitutive part of this encounter and my readings of it. Furthermore, the stark binary of inequality and equality carried with it implicit judgements of those who failed to step into the side of equality. Read alongside the vignettes discussed previously in this chapter, the pivot of this binary (masculine/feminine) was doubled by a silence around other inequalities that young people might be facing in their lives, such as racism, poverty and homophobia.  As I’ve thought about these silences, I’ve re-encountered gender in the words of the educators as they spoke with me during interviews. I find gender in responses to questions about how they would address issues such as gender and gender relations in their work. Gender should be about “equality” (Gustav, Interview, August 6, 2013), teaching learners that it is a “50/50 situation” (Nkunzi, Interview, June 29, 2013). Gender relations are about the “facts” and making youth aware of “what is happening” within communities (Drifter, Interview, May 28, 2013; Nyla, Interview, May 17, 2013) – teenage pregnancy, rape, stereotypes of men and women. Gender power plays out when learners do not live up to particular expectations, such as taking up the risky position of “independent women” leading to “being abused” or “being excluded” (Makaziwe, Interview, July 15, 2013). Once found, these words can be used to argue for the importance of speaking about gender relations with youth in sexuality education.   93 At the same time, the educators and I continually lost what we could articulate as gender with(in) our words. As described at the beginning of this chapter, Gustav told me that answering the question “how do you define your gender” was difficult. Makaziwe, another Regional Programme Leader with loveLife, said that the very notion of gender needed “unpacking,” noting that most conversations about gender happened within jokes rather than a particular session (Interview, July 15, 2013). In the interview, she used the language of “individual” to resist how categories of gender and sexuality “cluster” people (including herself) and how these clusters are “excluding” some people “to society.” Here, the words of gender, doubled through its entanglements with sexualized identities, are always already a limiting act, drawn through who is included and who is excluded within its boundaries. These words were in excess of both the interview and the pedagogical encounter, even as they worked at its limits. In thinking about these words within the discursive space of the interview, I pause on two in order to re-encounter gender with(in) their spaces.  Varni, a Regional Programme Leader, first defined gender as “male and female” and as something that can be “taught” (Interview, June 10, 2013). However, as she continued to speak, she said that it is a “broad topic if you really go into [it].” At that point in the interview, trying to understand what she meant by “broad,” I asked her about an image from the love4Life manual that represents gender as a spectrum. The image, which has the potential to disrupt binarized understandings of gender, was a new addition to the curriculum in 2013, and was part of a loveLife session that had been revised by a North American woman introduced to me as the “gender expert” (Fieldnotes, January 2013). In response, Varni said that she had not seen the image. It is possible that she did intend to read this session before using it, although the interview took place several months into the “implementation” of the programme. It is also possible that  94 she had disregarded the new session as irrelevant, resisting the foreign “expert” and/or process of top-down curricular change. At the same time, as she continued to speak, she invoked gender as a “complex topic” and said that the curriculum only takes young people through a “very small part of gender.” She also pointed to the complexity of gender when she defined her gender as female, but then said that this definition was something that she had grown up with: “you’re either male or you’re female.” Throughout the interview, I found myself chasing this “complex topic,” wanting Varni to articulate exactly what she meant. In doing so, I slid into my desire for knowledge and definition and potentially shut down the very complexity I was seeking.  The difficulties of defining gender, and how these difficulties arise within particular attachments to self and knowledge, were foregrounded in the interview with Nkunzi, another Regional Programme Leader. I had been speaking to Nkunzi for about thirty minutes and had reached the end of my interview protocol. Thinking that the interview was pretty much over, I turned the question that is asked in expectation of closure: do you have any questions for me? At most, I thought he might ask me something about the process of the interview or my role as a researcher. Instead, he laughed and asked: “what does gender mean to you?” (Interview, June 29, 2013). In asking this question, Nkunzi assumed the power that I had held up to this point in the interview: the researcher finding out how participants give meaning to gender. He asked me to explicitly name my own ways of making sense of the world. In responding, I lost my own conceptualizations of gender with(in) my words.  What does gender mean to me? That’s a good question. Um, I think for me gender… is a very um…it really depends on the different definitions that people give. So, for me it’s how we understand ourselves as male or female within a particular community. So I find it hard to define how I define gender in you know the community where I grew up is going to be very different from maybe how you define gender within your community, but I think all of that is gender. So that would be my – my understanding of it. (Interview with Nkunzi, May 27, 2013)   95 I initially responded to my words with surprise and a vague feeling of horror. In this response, I had tethered gender to the fixed (and biological) categories of male and female. I drew on notions of “my” and “your” community, and in doing so reified “place” as a site where gender “is.” This slippage, and its tensions with the theoretical framework of this study, enacts some of the texture of continually shifting relations of power and knowledge in and through educational research. The “answer” was situated in a particular interview space, and drew on loveLife’s discourses around gender as a shared basis of understanding for the conversation. Alongside a stumbling explanation of finding gender “hard to define” (in line with my theoretical commitments), I positioned gender as something that could be defined as something that “is” within a community (as per the loveLife script). This is a small enactment of how any “answer” takes form in an already contradictory and power-laden space.   In attending to the power-relations of that space, I have also been confronted with what I find to be a more difficult implication: my “answer” drew on what I thought Nkunzi could hear or what he wanted to hear. Throughout the interview, Nkunzi spoke about gendered identity in terms of cultural expectations. For example, he spoke about going to the bush (a reference to initiation) as a key part of becoming “a man” within the black isiXhosa-speaking community that he grew up in. In tension with the reading that I provide of the pedagogical encounter in the previous section, I took these articulations of “culture” as part of his truth about gender and the basis for what he would be able to understand about identity. Drawing on the language of “particular communities,” I re-centered my own position as the expert who can speak outside of these communities and ultimately package knowledge into something that “they” can understand. As Nkunzi and I sat across from one another, the “answer” gave voice to power-laden binaries of who knows and what is known: black/white; South African/American. It also risked silencing the  96 complexity within notions of gender that were already circulating with(in) the space of the interview. For example, when I asked Nkunzi how he defined his gender, he said he understood it in terms of the word he had chosen for his pseudonym, which means “bull” in isiXhosa. Through the research act of choosing a pseudonym, Nkunzi marked a way of understanding identity that exceeded the concerns of research confidentiality. He named himself by drawing on an understanding of identity, within the isiXhosa language, where names are a way to identify the subject within their relations (Ubuntu Bridge, n.d.). This naming exceeded the conceptual separation of gender from social identities at the same time as it exceeded the binary of man/woman that settle through the western epistemologies of the English language.   In re-encountering gender within these interviews I have come to ask how the difficulty of finding gender might be part of the condition for beginning to think differently about relations of power and knowledge within notions of gender. Following Spivak (2008), the pedagogical encounter (and here a research encounter through the interview) is an opportunity to “share with you what I have learned about knowing, that these are the limitations of what I undertake, looking to others to teach me” (p. 22). Across these encounters with(in) words, it was the moments we (learners, educators, researcher) failed to “speak” gender that notions of gender may have escaped their definitional weight. This encounter is always an incredibly murky space. Rights, culture and equality are themselves contested terms (Cornwall, 2007). Yet, sexuality education relies on these words in the promise of gender empowerment. These words haunt an educator’s summary of inequality and equality in a particular pedagogical encounter, as well as my own violent closure around what gender “is.” I am not arguing that we need to get rid of these words, but I am asking what it might mean to take them as part of the limits of what comes to be articulated as gender within sexuality education rather than a description of gender’s truths.   97 The encounters of gender empowerment   I am attending an event in one of the oldest buildings in Cape Town, now used as a museum dedicated to the history of slavery in South Africa. Organized by a corporate partner, the event has brought together youth (later described to me as street children), “motivational” and “guest” speakers and educators from loveLife. The speakers, two women, start the event by sharing personal stories of sexual and domestic abuse, and describe how they overcame that abuse to reach their dreams. Two groundBREAKERs then stand up to do something “interactive,” an impromptu role-play showing a boy pressuring a girl to have sex. The message of the role-play is you have to say “no.” (Fieldnotes, November 2012)   It may seem overly simplistic to return to a discussion of “ saying no” as the enactment of gender empowerment. While saying “no” has a long history within sexuality education discourses (e.g. Carter, 2001), saying “no” has also been analyzed as a complex discursive space and critiqued for placing responsibility for sexual refusal on the individual, and particularly women, in spite of power relations within this space (Kitzinger & Frith, 1999). This would seem to foreclose “no” as a strategy to address the “social” issue of gender power. Yet, within the loveLife imaginary of gender empowerment, “the ability to say no – and be respected for it – is at the heart of equal relationships” (loveLife, 2013b, p. 40). This lends coherence to how, during this event in Central Cape Town, saying “no” was the “solution” to unequal gender power relations in a sexual relationship.  Gender empowerment as saying “no” is a comforting tale: gender power settles into the verbal enactment of rational knowledge that has been transmitted through the pedagogical encounter. At the same time, it is precisely the effects of that power that speak against the simplicity of this tale. The imperative to say “no” in this particular pedagogical encounter was in tension with the stories of sexual and domestic abuse that had preceded the role play. Saying no was also an individualistic and atemporal act, and, as such, a silencing of multiple histories and presents coming together in the place of the pedagogical encounter, a museum dedicated to the  98 histories of slavery in South Africa. Slavery is an extreme example of how the ability to say “no” can be taken away. Its relations of economic, political and social dispossession haunted the inequalities faced by street children whose ability to say “no” is not only about gendered power but also about lived relations of risk on those streets. It is in these tensions that “no” begins to shatter, no longer able to tether action to a fixed gendered identity or knowledge as a concrete solution to relations of power.  By engaging “no” alongside the encounters of this chapter, I am suggesting that part of the discursive staying power of “saying no” in sexuality education is an effect of how the pedagogical relations of knowing and being known are conceptualized. The desire to “empower” subjects with knowledge is a powerful conceptual imaginary in education, and saying “no” lends itself to this imaginary. In other words, no matter how complex the individual is said to be, or how contested knowledge is, saying “no” is the verbal mark of a subject who is able to rationally resist relations of power based on the education that they have received. However, and as explored throughout this chapter, that mark is formed through westernized understandings of subjects and their forms of knowledge. As such, it is always already an articulation of a particular, and power-laden, imaginary of who youth are and what relation to knowledge they are perceived to have. This includes imaginaries circulating through gender empowerment: boys and girls who will, through the “facts” of sexuality education, rationally work towards relations of gender equality.  Throughout this chapter, I have argued that it is necessary, but not enough, to talk about complexity within notions of gender in sexuality education. While gendered meanings may be complex, the risk is that the analytic frame (the notion of gender) is somehow propped-up as a neutral and universal concept. This might not seem like a “new” argument. The term “gender”  99 has sparked decades of feminist debates, including: the linguistic limits of the term (e.g. Haraway, 1991); the distinction, or lack thereof, between sex and gender (e.g. Butler, 1990; Rubin, 2011); and what power relations are deployed and reified when gender is a “problem” for international development initiatives (e.g. Cornwall, Harrison, & Whitehead, 2007; Saunders, 2004). However, and what I argue is “new,” sexuality education as a global HIV prevention strategy has the potential to both reify as well as challenge the power relations within these articulations of gender. As seen in the encounters with gender explored in this chapter, taking up “gender” as something to teach relies upon and reifies particular understandings of gendered identity, particularly the heteronormative boy and girl, as formed through globalized relations of power and knowledge. How culture already operates within these understandings is in tension with calls to make sexuality education “culturally relevant.” One key effect of centering notions of gender on culture is to mask the inseparable, and yet irreducible, relation between social identities as well as the privileges that play out in their discursive boundaries. The pedagogical approach, in turn, cannot be held separate from these relations. The pedagogical approach is itself a technology of power and as such it gives form to the political rationality (the vision of the world and its relations) guiding sexuality education as an HIV prevention strategy.  At the same time, following notions of gender into the pedagogical encounter speaks to the impossibility of ever simply “transmitting” them. Notions of gender are unsettled, and unsettling, within those encounters. They operate within relations of difference, even as they may be re-articulated in ways that challenge what that difference is said to be. For example, educators and learners are already encountering gender power relations through notions of culture (rather than as a product of culture) and this could be engaged as a resource for re-imagining how power takes form in and through multiple relations of social difference. Likewise, educators are already  100 struggling with the “small part” of gender explicitly being engaged in any pedagogical approach (as in the research interview). Those struggles invite an articulation of sexuality education that exceeds what can/is/should be known about gender identities and their relations.  Destabilizing the premise that sexuality education could be the “solution” for relations of power and knowledge necessitates explicit engagement with how sexuality education as an HIV prevention strategy conceptualizes pedagogical relations of knowing and being known. I take this up in the following chapters through two shifts that aim to work at the limits of how sexuality education as an HIV prevention strategy is/can be articulated. First of all, there is the question of what might happen if (no matter how momentarily) the logic of sexuality education is unmoored from gender as something that “is.” I take up this question in Chapter 4, asking what it might mean to shift from identity as a target to identity as a site of struggle. Secondly, there is the question of what might happen to sexuality education in excess of any claim of knowledge, and knowable subjects, within the pedagogical approach. This is the focus of Chapter 5: exploring a conceptual shift for sexuality education from knowledge as something already made to knowledge as something always in the making. Across the chapters, I am interested in what might happen to all social identities and their entangled relations in the pedagogical space between agree and disagree.     101 Encounters: A pause on gender as target in HIV and AIDS discourses  It is the 1980s.   Women are represented as “vessels” and “vectors” in HIV and AIDS discourse (Patton, 1993b). The purity of the victim is positioned against the immorality of the perpetrator: the unknowing wife, the prostitute, the mother who passes an infection to her unborn baby. “Good” and “bad” settle around the racialized and classed body: white, heterosexual, middle class; prostitute, bad mother, African woman (Kitzinger, 1990). Scientific and medical authorities enact and reinforce particular understandings of gender (Treichler, 1999), promising safety through chastity, fidelity and marriage (Wilton & Aggleton, 1991). Sexuality education medicalizes and fragments the gendered body (Holland, Ramazanoglu, Scott, & Thomson, 1994).  It is the 1990s and  there is a “heterosexual” epidemic of HIV and AIDS in Africa. Dominant international discourses posit that this epidemic is driven by “exotic sexual practices and polygamy” that deviate from “proper” heterosexuality (Patton, 1993a, p. 130). “African heterosexuality” is “unsafe” and needs “rapid reorganization into bourgeois families” (p. 132). As a part of these families, “culturally defined gender roles” have been “identified as points of concern for Africa” (Brown, Sorrell, & Raffaelli, 2005, p. 586). This is the “women’s epidemic” according to Peter Piot, then the head of UNAIDS, the Joint United Nations Programme on HIV and AIDS (in Baylies, 2000). African women are said to need “education and empowerment” (Brown, Sorrel & Raffaeli, 2005, p. 586).   It is World AIDS Day 2004.   The theme is “Women, Girls, HIV and AIDS”, which aims to draw global attention to the role of gender inequality in driving the epidemic. The UNAIDS (2004) update report argues that women and girls are increasingly affected by HIV and AIDS in all parts of the world. The situation is even more “disturbing when viewed up close” (p. 7). In South Africa, “young women (aged 15-24 years) are three to six times more likely to be infected than young men” (p. 7). Two years later, a 2006 report by the Southern African Development Community warns that a “lethal cocktail” of “biomedical, political, economic and cultural forces shapes the gendered dynamics of the HIV epidemic in South Africa” (in Gilbert & Selikow, 2011, p. 325). Traditional ABC (abstain, be faithful, use a condom) messages are inappropriate. Educational initiatives need to focus on “enabling environments” and women’s rights (Baylies, 2000).  It is 2014 and   102 “gender power” is widely accepted as a “structural issue” that will “directly determine the social vulnerability of groups and individuals (Parker, Easton, et al., 2000, p. S23).25 Epidemics around the world vary, but “gender inequalities and biological differences still make women and girls especially vulnerable to the epidemic” (Gay et al., 2010, p. 17). A website,, funded by the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR) and the Open Society Foundations26, presents themes based on “a comprehensive review of data from HIV/AIDS interventions for women and girls in nearly 100 countries.” An Interagency Youth Working Group (IYWG), funded by U.S. Agency for International Development, exists to “provide global technical leadership to advance the reproductive health and HIV/AIDS outcomes of young people aged 10 to 24 in developing countries” (Interagency Youth Working Group, 2014). The 2013 IYWG toolkit, “It’s about more than just sex,” provides over 30 educational materials that were developed by “internationally recognized global health or academic organizations” (FHI 360, 2013, p. 5). While many of these materials were developed in the United States, they are “culturally adaptable and appropriate for low-resource settings” (p. 5).    Across these named moments targeting gender, notions of gender splinter under the weight of received meanings and imagined futures. I can weave these moments into points in time in order to trace a named, if always incomplete, trajectory. That trajectory is one in which the notion of gender has solidified into a global problem conflating bodies and identities. Each point can be seen as a site in which AIDS, as an “epidemic of signification” (Treichler, 1999, p. 1), has defined not only the “problem” of gender, but also what is meant by “gender” as a problem. These are moments that claim to offer gender as something separate from other social identities, at the same time as gender is only coherent through its entanglements with these other identities.  Reading these as points in time may be one way to disrupt the linearity of the account. Global discourses are not free floating in some undefined out-there. They work in global space as a “meeting-up of histories” (Massey, 2005, p. 4). Each point in this particular narrative reaches across time and space, working within and masking the power relations that make the points noticeable. The construction of gender as a “problem” reaches into imperial memories where                                                 25 See also: Gupta, Parkhurst, Ogden, Aggleton & Mahal (2008); Padian et al. (2011)  26 See; accessed January 22, 2015.   103 “Africa” was targeted with a medical discourse that was “heavily gendered and sexualized in its language” (Vaughan, 1991, p. 21). “Global” guidance on “local” educational initiatives extends through colonial encounters and takes “shape in relation to the ambiguities that give modernity its character” (Soudien, 2009, p. 25). These points in time are also produced through my research interests, shaped through my critical trajectories and attachments to knowledge. The points themselves are part of the “terrain of struggle and contestation” that is the HIV and AIDS epidemics (Hall, 1992, p. 272). Each of these points is a potential provocation to the notion of gender, put into the place of the pedagogical encounter, where “unexpected lines of thought slowly unwind” (Massey, 2005, p. 6). These points refuse knowledge about gender as something already made before it comes to the pedagogical encounter, just as they refuse knowledge about gender as something already made within that encounter. What might their undoing provoke within sexuality education?    104 Chapter four: Identity and the facilitation of local lives  Lu and I were meeting with a primary school class of 12 and 13 year old learners for the first time. Before the session started, he told me the topic for the day would be “Who am I.”  We all sit perched on desks and plastic chairs that have been pushed to the sides of the room. Lu gives the instructions: step into the center of the circle and say your name, birth date, where you come from and what makes you special. A few learners volunteer right away: a girl describes living in a safe house because her mother left her; another girl speaks about how her parents had died. When no one else volunteers, Lu shares stories from his own life. Then he begins to call on the learners one by one. After the session, Lu tells me he is disappointed that more learners did not participate. It can be hard to speak out in a group setting, I say. He responds: maybe it was because a lady was there. (Fieldnotes, March 2013)   On that day in a black isiXhosa-speaking township outside of Cape Town, I was part of a pedagogical encounter in which learners were asked to embody their “reputation of being one of the highest risk groups for HIV/AIDS in the world” (Leclerc‐Madlala, 2002, p. 23). As a national HIV prevention initiative, loveLife aims to reduce HIV prevalence among youth, currently estimated at 7.1% across the country (Human Sciences Research Council, 2014). The stories that were shared within the room are part of the texture of risk in youth’s lives. For example, the girl who spoke about her parents having died may be one of the 16.9% of South African children ages 18 or younger who has lost a parent to the disease (Human Sciences Research Council, 2014). Others shared experiences of male violence, pointing to norms of masculinity in South Africa that center around the ability of men to control women and other men (Jewkes & Morrell, 2010), as discussed in Chapter 3. These statistics and norms, and the social issues that they foreground, give impetus to an urgent response for youth within HIV prevention discourses. In this session, as youth stepped into the circle, each youth body and identity was simultaneously the problem (through his or her vulnerability) as well as the solution (in the educational promise of a different future). Once they were shared, the stories of “Who am I” were supposed to help  105 youth discover their individual “purpose” and have “full control” over their futures (loveLife, 2012e, p. 12). This is an articulation of individual empowerment where “power-from-within,” accessed through education, will allow the individual to exercise “power-over” the social environment (Laverack, 2004). Sexuality education was the vehicle for this empowerment, drawing coherence from international framings of sexuality education as a universal “evidence-informed” response for youth (UNESCO, 2009a), as well as national discourses within which education in South Africa is understood as a “tool for empowerment and the basis of a new kind of citizen” (Hammett & Steaeheli, 2013, p. 327).  Yet, and at the same time, the embodiment of “Who am I” in this session refused to stay in place. For example, when I spoke about it being hard to share in a group setting, I was reading the learner’s silence as a potential resistance to what I saw as a Foucaultian (1998) confessional space: the directions to share something special about yourself becoming an incitement to speak the “very constitutions of the self” (Lupton, 1995, p. 11). However, even the learners who did share stories were potentially resisting the incitement to speak the “self” as separate from the multiple power relations giving form to those selves. Instead of the envisioned celebratory and individualistic stories of what makes you special, speaking the “self” was a momentary articulation of power relations intertwined in the formation of those selves. Furthermore, the incitement to speech was turned back on my own body when Lu resisted my understanding of the silences, saying maybe it was because a lady was there. In this statement, who I was in this session was an embodiment of multiple histories of unequal power relations defining self and Other in South Africa. There is the white lady whose domesticity, woven into a privileged form of gendered heterosexuality, was constructed through the deviance of the black Other under apartheid (Glaser, 2005). There is the middle or upper-class white lady who continues to exercise  106 power over the racialized bodies that populate her household: gardeners, maids, nannies. There is the foreign lady (consultant, NGO worker, researcher) who is constructed as an expert about youth through the mobile formations of her western epistemological knowledge systems.  As Lu invoked this lady in the pedagogical encounter, he pointed to an “I” that cannot be fixed into clearly defined identity categories (such as gender, sexuality, race, class) or understood outside of its Others. In this chapter, I begin with this encounter as an invitation to begin to think about identity within the pedagogical encounter as a place of “loose ends and missing links” (Massey, 2005, p. 12) entangled in multiple relations of power and knowledge, including research. These loose ends and missing links always refuse the fixing of identity, and that is why I start by foregrounding them: to trouble identity, including but not limited to gendered identities, as something to know through sexuality education. My interest here is what kind of political project might be possible with(in) sexuality education when any instance of who the subject is, including youth, is foregrounded as a problematic identity that is encountered by educators and learners. I engage the “throwntogetherness” of a place (Massey, 2005, p. 140), here a classroom that holds both a pedagogical and a research encounter, to mobilize understandings of who youth are within the “unavoidable challenge of negotiating a here-and-now” that is always “drawing on a history and a geography of thens and theres” (p. 140). This, I will argue, contributes to destabilizing who the subject of gender empowerment discourse is said to be within sexuality education, at the same time as it engages struggles over identities and their relations as a site of possibility for beginning to articulate sexuality education differently.  Shifting the analytic focus from identity as target to identity as a site of struggle, resists a conceptualization of pedagogy as a linear and fully rational, relation among educators, learners, and knowledge (as introduced in Chapter 1). For example, framed within pedagogy as a linear  107 relation, the analytic focus in reading the classroom encounter described above would be on what, either in terms of the method (what worked or did not work?) or the knowledge that was supposed to be transmitted (what did the educator teach? what did youth learn?). This focus may be operationally useful, particularly for programs like loveLife that seek to “scale-up” educational initiatives for youth and measure the “effect” of those initiatives in line with donor priorities. However, focusing on what ignores, and in doing so potentially reifies, multiple and constitutive relations of power and knowledge within the pedagogical encounter itself. The shift, then, is to focus on how particular understandings of identity take form within that encounter.  In order to attend to how understandings of identity take form, Doreen Massey’s (2005) conceptualization of place extends thinking about identities within the pedagogical encounter in relation to the places of those encounters. While I explicitly attended to pedagogical spaces during my fieldwork, notions of “place” took me by surprise as I worked with the vignettes drawn through the experiences of this study. Place exceeded detailed accounts of “location” or “background.” In the interviews, for example, educators answered questions about gendered identities and expectations through notions of place – a family’s kitchen, a boy going “to the bush,” 27 a particular classroom. Those particular classrooms were located in communities, but the lived relations of those places refused to be neatly bounded and contained. I also found multiple places within my fieldnotes, and my own relation to those places was entangled in the privilege I had to move between them and engage them as a “place” for research (as discussed in Chapter 2).                                                 27 Nkunzi used this phrase (going “to the bush”) to refer to an initiation ceremony that marks the transition from “boy” to “man” in the community where he grew up in the Eastern Cape of South Africa (Interview, June 29, 2013). In other communities, depending on the physical landscape, this may be called going “to the mountain.”   108 As part of the shift from place as an “object of study” to place as a “framework for analysis” (Larsen & Beech, 2014, p. 195), Massey (2005) provides a conceptualization of space and place that engages these geographic concepts as interrelated and constitutive sites that are a key part of the performativity of identity. Responding to dominant understandings of globalization, she argues that understandings of space often rely on and re-tell stories of modernity with an “essentialist billiard-ball view of place” (Massey, 2005, p. 68): space as a vague surface made up of discrete places. Those places may collide (think of colonial encounters) or be increasingly interconnected (think of globalization) but they ultimately are discrete representations of geographic location and the people who live there. Refusing the underlying binary distinction between place (as local or fixed) and space (as something “out there” or global), Massey shifts to a reading of space as a “simultaneity of stories-so-far,” produced within multiple relations of power. Place, in turn, is a “collection of those stories” (130). In other words, instead of “points on a map,” place itself is an event where what gets called “here” is an “intertwining of the histories in which the spatiality of those histories (their then as well as their here) is inescapably entangled” (p. 139). Taken to the pedagogical encounter, this view mobilizes “place” as a constitutive part of the stories of selves that are collected within it: stories of the “problem” of youth and HIV prevention, stories of “who” youth are, stories of the “solution” of empowerment through education. As such, place offers a way to attend to the lived effects of power while simultaneously resisting the closure of identities around, or in terms of, those effects. Pedagogical places are sites where “forms of power and meaning…gather into what we think of as stories and selves” (Stewart, 2007, p. 6). That gathering is always incomplete and unstable. Once collected, stories and selves are always undoing themselves, reaching across their entangled histories, contested presents and imagined futures.   109 In this chapter, I engage these stories for what I am calling the emplaced weight of performativity. I am using this concept to think about the lived weight of identity and its effects on selves and bodies, while thinking “besides” (Sedgwick, 2003) the binary terms of a discursive/material division that I have found restrictive. This binary arose within the Social Sciences as a part of, and in response to, the “linguistic turn” that broadly focuses on how language and discourse is constitutive of what comes to be seen as “reality.” In the 1990s, for example, poststructural readings of the subject and power were critiqued for failing to engage with the material effects that power and knowledge have on particular lives (e.g. McNay, 1992; Sawicki, 1991). While one response has been to turn away from the linguistic and re-ground materiality within the principles of physical science,28 Massey’s theorisation of place provides an analytic language through which to engage the materiality that I argue is always already within discursive understandings of identity. Here, discursive constructions of identity have lived effects, both in terms of what is understandable as “oppression” (such as unequal relations between what are taken as known subject positions – man/woman; black/white) as well as norms that determine whose life can even count as livable (Butler, 2004). It is this discursive violence that then gives rise to physical, social or emotional violence within the places where the self gathers. Engaging the self with(in) place is not a “solution” to the dilemmas of identity but rather one way to explicitly engage with multiple relations of power and knowledge and their lived effects.  The rest of this chapter engages identity as a site of struggle through three sections. In the first section, “Who am I,” I work across multiple discursive constructions of “youth” that are                                                 28 See, for example, a recent strand of feminist thinking broadly called the “new materialisms” (e.g. Alaimo & Hekman, 2008).  110 often held separate both in terms of their spatial register (global, national or local) as well as their boundaries as a “target” for the pedagogical encounter. Juxtaposing and working across these registers, I analytically engage place as a site to think within co-constitutive understandings of identities and their relations that have always already escaped the pedagogical desire to know them.  In the second section, Facilitating Local Lives, I engage identity as a site of struggle within loveLife’s pedagogical approach. Being able to know personal identity is central to loveLife’s approach to working with young people. For example, the Love4Life Faciliator’s Manual, focused on “healthy sexuality” (loveLife, 2012f, p. 4), and the lovingLife and Making my Move Facilitator’s Manual (loveLife, 2012e), which aims to “educate young people on a range of positive life choices” (p. 3), both start with sessions entitled “Who am I.” The premise, as stated in the lovingLife manual, is that knowing personal identity is “the most important element to success and a positive, bright future” (loveLife, 2012e, p. 11). In this section, I ask both what political rationalities are at play within this approach (in terms of imagined relations between self, Other and knowledge), as well as what happens to sexuality education when this approach is seen to foreclose the very complexity of social identities it claims to liberate.  Finally, I extend this discussion by exploring struggles over identity with the pedagogical aim of “teaching empowerment.” Here, I ask what happens to “teaching empowerment” if there is no longer a knowable youth subject to target with power (to “overcome” their social identities). In doing so, I explore ways that this impossibility might be a resource for imaging social identities and their relations differently.   111 Who am I  An encounter with(in) place  We drive by stand-alone concrete houses and neatly kept yards enclosed by outer walls. There is no one on the streets. Trevor describes the school as the best in the area. The learners are very disciplined. When we get to the school we wait in the lobby for the teacher to come get us. The lobby is filled with pictures from special events and community activities.   During the session, Trevor asks the group of grade 7 learners: how many people are from (a neighbouring community)? He later tells me that this neighbouring community is violent with a lot of gangs. In response to his question, about a third of the learners, including the teacher and Trevor raise their hands. What negative things have you heard about the community? The learners fidget. No one answers. The teacher responds for them: boys are told they will become gangsters; girls will be taxi queens29 or prostitutes.   Trevor sums up the lesson: each person is in charge of his or her own life. No one should be determined by what other people say they are going to be. (Fieldnotes, April 2013)   In this vignette, Trevor, a groundBREAKER working in a coloured suburb outside of Cape Town, and I were driving to a school for the first time. The description that I provide could be approached as realist ethnographic detail (Britzman, 1998; Youdell, 2010): the detail serves to create a sense of place for what is taken as the analytically “interesting part.” For example, focusing on gendered identities and relations, the construction of boys as gangsters may reflect a construction of South African masculinity as violent (Bermudes & Da Cruz, 2004; Leclerc-Madlala, Simbayi, & Cloete, 2009; Walker et al., 2004). Youth gangtersism is also entangled in class inequalities that have become the “problem” of racialized and gendered bodies in South Africa: the white “duck tail” gangs of the1950s (Glaser, 2005); the black “tsoti” who threatens order within historical (Glaser, 2000) and contemporary imaginaries;30 the “well-established”                                                 29 “Taxi queen” refers to a young woman who has transactional sexual relations with (older) male taxi drivers, for example in exchange for a ride to school.  30 For example, the 2005 film “Tsotsi” followed the life of a young gang leader in Johannesburg, and received critical acclaim in South Africa as well as international film festivals. See, accessed January 22, 2015.   112 gangs in the Cape Flats of the Western Cape, where coloured (and mixed-race) peoples were moved during apartheid.31 Understandings of femininty are likewise already complex within this place. The positioning of girls as prostitutes is coherent within long-standing and binary constructions of women within HIV and AIDS discourses as either innocent or deviant (Patton, 1993b). This gendered binary is simultaneously an effect and a potential reification of imperial discursive legacies in which “prostitute” and “African” are collapsed as deviant subjects (Kitzinger, 1990). Understandings of taxi queen, in turn, both extend deviant constructions of female sexuality (Strebel, Shefer, Potieter, Wagner, & Shabalala, 2013), and, through the transactional nature of this position located within particular social inequalities, may resist passive constructions of female sexuality in the South African context (Bhana & Anderson, 2013; Jewkes, 2009; Jewkes & Morrell, 2010). Across these readings, already multiple understandings of social identities are meeting within a “risk environment” that may increase susceptibility to HIV infection (Barnett & Whiteside, 2002). What these readings leave unexplored, however, is how that “risk environment” plays a constitutive role in the very intelligibility of identities at play in this pedagogical encounter.  When place is engaged as part of the performativity of identity, these initial readings of how power relations settle in and through understandings of identity are extended. For example, Trevor draws on the physical landscape, the school being the best in the area, as he describes learners as disciplined. Spoken in relation to a coloured community, this disciplined identity can be read against constructions of coloured identity in South Africa as “often associated with poverty, immorality and sexual promiscuity” (Bhana & Anderson, 2013, p. 551). This                                                 31 See, for example, recent media coverage of gangs in the Cape Flats:, accessed January 22, 2015.   113 construction of “coloured” identity draws on its particular historical legacies, a label imposed by whites during apartheid, as well as negative discursive constructions of this racialized identity as “mixed,” not white or black (Adhikari in Bhana & Anderson, 2013). Disciplined, in contrast, may be used to evoke a middle-class identity that has been historically white (Soudien, 2009). As we move into the space of the school, this disciplined youth identity is given public recognition through photos of special events and achievements. Here, disciplined is part of the enactment of the emplaced weight of performativity: place as a physical and discursive site through which youth identities are always already entangled in racialized and classed markers refusing temporal and spatial boundaries and creating particular registers of intelligibility.  Furthermore, as we all meet in the session, place does not fade into the background. In contrast to the privileged place of the middle-class youth subject, Trevor draws on the discursive power of “gangsterism” to construct a neighbouring community as deviant. In turn, his question – who is from that community – collapses the “risk” of place with the “risk” of subjects who are from that place. This binary of risk/safety relies on the coherence of community as a bounded geographic place, at the same time as it inscribes particular constructions of identity within those places. This collapse is further enacted in the following question – what negative things have you heard about that community? By simultaneously addressing those who are from that community as well as those who are not, this question has the potential to speak back to those negative things (from those within the community) as well as reify and Other those negative things (from those outside of the community). The teacher takes up this question to invoke the “risk” of becoming within these places: boys will become gangsters, girls will be taxi queens or prostitutes. Within this encounter, social identities were not only meeting up with “risk” in a place (the classroom, for example). Instead place was a site within which the “risk” of social  114 identities are inextricably intertwined: a particular articulation of race, gender and class that settles in and through place and its imagined boundaries. Social difference plays out at these boundaries: a community full of “promise” versus a community full of “risk” and “risky” subjects.  Crucially, across these moves, who youth are said to be has already been discursively collapsed with where they are from and this is in direct tension with the “message” of the session: each person is in charge of his or her own life and no one is determined by what other people say. The coherence of this message relies upon, and redeploys, a neo-liberal conception of a free-floating subject. This subject is central to the loveLife imaginary: from the multiple educators who framed their task as getting youth to discover themselves as “individuals” to senior staff members who told me that the core of programme was about “self-awareness” (Grace Matlhape, Personal Communication, August 7, 2013) and valuing young people’s “own sense of themselves” (Scott Burnett, Personal communication, October 3, 2013). The explicit claim, within this imaginary, is that knowing the self will enable the self to act. At the same time, there is an implicit claim operating here related to responsibility for that action. A key element of neo-liberalism as a political rationality (as introduced in Chapter 1) is how it “equates moral responsibility with rational action” (Brown, 2003, para. 15). In other words, a sense of personal identity is key for action and this action (here: being in charge of his or her own life) serves as the moral compass of identity. This places the consequences of acting, or not acting, fully on the individual “no matter how severe the constraints on this action” (para. 15). Read within this rationality, being in charge of his or her own life is not a liberating possibility, but rather a problematic shifting of responsibility for power-laden social relations onto the shoulders of learners.   115 Engaging the “message” of this particular session as a part of a neo-liberal political rationality further sets who youth are said to be in motion across multiple relations of power and knowledge. Within loveLife’s pedagogical approach, the discursive centering of the individual is already in tension with how the organization goes about working with youth. As introduced in Chapter 2, loveLife positions itself as providing “nationwide community-level outreach” (loveLife, 2014a) but specifically targets “marginalized” peoples and communities (loveLife, 2012a; Robbins, 2010). In South Africa, the geographic landscape of community is one that continues to be drawn through the racialized dividing practices of apartheid (Hunter, 2010). Under apartheid, the South African population was divided into racial categories based primarily on physical appearance (such as “white”, “black” and “coloured”) and access to urban spaces was determined by racial classification. While apartheid laws were repealed in 1991, community membership draws on these historically racialized boundaries at the same time as the boundaries are re-drawn through severe economic inequality that restricts access to housing and movement between communities. As the “context” for the loveLife programme, learners are targeted based on their membership within a particular community and groundBREAKERs are recruited from, and expected to work within, those communities. Here, the point is not that a message should be adapted for a particular place (such as addressing issues of gangsterism) but rather that the “I” being addressed is already a contested site for relations of power that give form to particular understandings of the gendered, raced, classed and sexualized body. Furthermore, I argue that reducing moral responsibility to the ability of this “I” to act is a profoundly limiting move, not only in terms of the lived barriers to action, but also in terms of what it means to be responsible to lives lived within relations of power and knowledge. Following Butler (2001), it is the co-constitutive and intimate relation between conceptions of self and Other that grounds moral  116 responsibility. In opposition to a political rationality of neo-liberalism, responsibility in this view lies in the “limits of self understanding” as the “predicament of the human community itself” (p. 37). In other words, it is through understandings of self and Other that all subjects are already intimately connected and bear ethical responsibility for the effects of power mobilized through these understandings.  In making this argument, my concern is that the focus on identity as something to know (who youth are) is both a refusal of the complexity of youth identities, as well as a closure of responsibility for how understandings of self and Other settle with(in) place. Sexuality education as an HIV prevention strategy is filled with stories of who youth are and what they need: imperialist stories concerned with the deviant African subject, development narratives diagnosing “local” problems, international “best practice” guides that promise the solution. As long as moral responsibility is equated with individual action, the tellers of these tales (such as “expert” policy makers, programme developers and researchers) will not be held to account for relations of power and knowledge deployed in the name of “helping” youth. In the pedagogical encounter discussed in this section, the learners’ silence may have marked a small act of resistance to who they were said to be, including their position as “local” subjects targeted with sexuality education to address a “global” problem.  Global priorities   In 2013, as part of the African Union Summit, UNAIDS released an Update report (UNAIDS, 2013b) with a cover that features a black child running across an open field of grass. The young boy, wearing a faded t-shirt and jeans, runs towards the viewer, into strands of grass that blur in the foreground, as his arms and legs are captured in motion. He looks down at the ground, biting his tongue in concentration. As part of a story entitled “How Africa turned AIDS  117 around,” his healthy, lively body can be seen as a symbol of a successful “future for the AIDS response” (p. 16). A key part of sustaining that response, according to the accompanying “call to action,” requires “empowering young people, especially young girls, by ensuring they have access to sexuality education” (p. 16). The boy is carrying a backpack, the imagined weight of school-books carrying the world’s educational hopes and promises.  Within global discourses, sexuality education explicitly emerged as a global strategy for HIV prevention for youth with the 2009 publication of the International Technical Guidance on Sexuality Education by the United Nations Educational, Scientific and Cultural Organization (UNESCO, 2009a).32 Located within a broader shift in HIV prevention responses from individual behaviour change strategies to interventions aimed at socio-cultural and structural influences on behaviour (Altman, 1995; Gagnon & Parker, 1995; Gupta et al., 2008; UNAIDS, 2001), the document aims to “assist education, health and other relevant authorities in the development and implementation of school-based sexuality education” (p. 3). The guidance document provides a rationale for sexuality education, reviews the evidence base for the impact of sexuality education, and then provides concrete characteristics for successful sexuality education.  One reading of the document is that it responds to the critique that many educational responses to HIV and AIDS have put the individual in a “social vacuum” (e.g. Boler & Aggleton, 2005; Parker, Barbosa, & Aggleton, 2000). The guidelines are premised on the view that the “sexual development of a person is a process that comprises physical, psychological, emotional, social and cultural dimensions” (p. 5). Building on this broad, but still individualistic, view of                                                 32 Educational responses have been a key part of global HIV prevention initiatives since the disease was first identified in the United States in the 1980s.   118 sexuality, the guidelines bring together individual attributes (such as decision making skills) with social factors (such as gendered roles and power relations). At the same time, the document continues to rely upon particular constructions of the individual in relation to those “social factors.” Specifically, the document mobilizes two dominant conceptualizations of youth identity within relations of power and knowledge.   Firstly, the promotion of sexuality education as a global strategy for HIV prevention uses age to give coherence to a supposedly universal category of youth. For example, the Guidelines state that all youth need “adequate preparation for their sexual lives” as they approach adulthood (p. 2). This preparation consists of “correct knowledge” as well as “positive values and attitudes” (p. 2) that must be received “before [youth] become sexually active” (p. 3). Reading age as an invented category rather than a natural property of youth (Lesko, 2001) draws attention to how these claims are embedded in a developmental discourse in which adults must guide youth (seen as out of control, irresponsible, irrational) to become proper subjects. The focus on “age-appropriate” information relies on discursive constructions of childhood “innocence” and carries implicit assumptions about the kinds of knowledge youth need, potentially restricting what topics are considered appropriate (Bhana, 2008). For example, the discursive collapse of “age” with the “need” for particular forms of knowledge played out in this research project when the educator team in the Western Cape made an informal decision that the lovingLife programme (on positive life choices) was appropriate for primary school learners while the love4Life programme (on healthy sexuality) was appropriate for high school learners. In turn, the presumed universality of age obscures how “youth” is also always constructed through gendered, raced and classed relations of power and knowledge. The point here is not only to problematize how “age” creates an undifferentiated category of who youth are in HIV and AIDS discourse that is then used to  119 determine what they need (Boler & Aggleton, 2005). This is an important critique for exploring relations of power and knowledge outside of the pedagogical encounter.   At the same time, age-related assumptions about youth are an ongoing and potentially contested relation within the pedagogical encounter. For instance, Gustav (who was introduced in Chapter 3) spoke about age, and the claim that learners are “too young” for particular kinds of information, as “intertwined” with “HIV, gender and all these sexuality things” (Interview, August 6, 2013). Giving the example of “15 year-old” dating a “22 year-old guy,” she explained that age was coupled with gender in a relation of “respect” that forecloses a young woman questioning a guy who “proposes a certain thing.” She then countered the claim that learners are “too young” for certain kinds of information by aligning herself with youth (she was 22 at the time of the interview) and drawing attention to knowledge already circulating with the pedagogical encounter: “the things that they say when there’s no adults in the room would shock you.” Instead of taking this up as an alarmist warning (e.g. there are out-of-control youth who need adult intervention), my interest is how “the things they say” are always instances of actively making sense of the world. The risk here is that the power relations circulating in age-related assumptions about youth work to shut-down, rather than engage with, the complex ways that youth are already understanding their identities and their worlds within sexuality education as an HIV prevention strategy.   The second conceptualization of youth identity operating with(in) these global HIV prevention discourses is inseparable from understandings of place: place has already been engaged as a constitutive part of who youth are. In other words, it is not only that “global” guidelines will be implemented in “local” contexts, but also that those “local” contexts are constitutive of how youth should/can/must be addressed through education. Consider, for  120 example, the statement that in addition to being “age-appropriate,” all sexuality education should be “culturally relevant” (UNESCO, 2009a, p. 2). This statement is coherent as a response to the critique that many HIV prevention initiatives have drawn on universalizing and western rather than “local” understandings of sexuality (Boyce et al., 2007; Parker, Barbosa, et al., 2000). However, and extending the discussion of “culture” in Chapter 3, understandings of social difference that operate in and through being “culturally relevant” are tethered to bodies through place. Western tropes of “African sexuality” have brought together place (Africa) within “static and ahistorical” understandings of “culture” in order to construct the “African sexual subject as quintessentially other” (Reid & Walker, 2005, p. 186). Rather than a departure from its imperial histories (Patton, 1990, 2002), the claim of being “culturally relevant” thus extends those histories in new formations of what is “global” knowledge (the Guidelines) and how that knowledge can be extended to already known “local” subjects (youth). As argued in Chapter 3, understandings of identities, and their relations of social difference, are formed through notions of culture. Rather than taking this pedagogical encounter up as a potential site within which to re-think those relations of social difference, the risk is that global imperatives to be “culturally relevant” conflate culture and place and in doing so reify already power-laden articulations of social difference.  As the boy runs across the cover of the UNAIDS document, he runs into a gaze entangled in imperial legacies. He is running across a historical field where black youth, and the specter of their potentially threatening adult selves, are a problem (Carton, 2006; Vaughan, 1991). He is running through a place, a grassy field, which mobilizes his blackness as a part of a rural/urban binary that works to reify understandings of the “modern” west as developed and “traditional” Africa as underdeveloped (Patton, 2002). He is running within a present where a global agency  121 tells African youth that their behaviours and their identities are crucial for the “continent’s future” (UNAIDS, 2013b, p. 15). He is running in the shadow of his teenage self whose masculinity will potentially be a “problem,” his identities and lived relations silenced in the gaze of the “expert.” He runs in and through my research, in my own closures around “youth” even as I struggle to hold the term open. He is entangled in who he is, constructed through notions of age and place that appear to be so commonsensical that they often remain unchallenged. To problematize those relations I follow him into the imagined classroom. Here, I ask how struggles over identity are encountered with(in) place, and how sexuality education might engage those struggles in articulating a project of social transformation.  Mapping the encounter  The groundBREAKERs are presenting maps that they have drawn of their communities. A young man presents his community, a large town in an agricultural district in the Western Cape. His nick-name is Mitchell’s Plain, which is an urban coloured community outside of central Cape Town. Someone explains to me: it is because he looks like he comes from there.   On the map, the groundBREAKER has drawn white, black and coloured areas. He explains his map: Each area has its own sports stadium. All the poor people live in the black and coloured areas. The hospital is in the white area. Another groundBREAKER, a black female, asks: why are people in your town racist? He responds: it isn’t racist, that is how it is.   Lukanji, the Regional Programme Leader leading the training session, steps in: racism is a reality. She adds, speaking of loveLife’s work in schools: I would be happy to get a school in a white area. They don’t usually need us because they have resources and are living their own lives. (Fieldnotes, February 2013)   The mapping activity in this vignette, drawn from the provincial training event for new groundBREAKERs, was part of a session on “community mobilization” (loveLife, 2013a). Broadly, community mobilization is part of a participatory model of international development  122 that aims to resist top-down, donor-driven, approaches to development.33 groundBREAKERs are expected to use community mapping as a way of getting an “inside view” of communities, “identifying resources and community infrastructure that may be beneficial to our programmes” and “establishing baseline information about a community” (loveLife, 2012c, p. 59). The use of community mapping resonates with participatory health education theories and methodologies as they have been taken up in HIV prevention work with youth (e.g. UNAIDS, 1999). The educator (here the groundBREAKER) conducts a “hands-on” activity with community members to create a community map. While the map may be combined with other “community mobilization” research techniques (such as demographic surveys or identifying government and leadership structures), the key focus of the map is on the physical spaces of the community: “major infrastructure,” “prominent landmarks” and “other features” of the landscape (loveLife, 2012c, p. 59). The underlying premise of the map, and the discourse of community mobilization it relies upon, is coherent within dominant understandings of collective or community empowerment:community members “must all work together to solve problems and bring change for the better” (loveLife, 2012c, p. 48). In this training event, the groundBREAKERs worked on their own (each being from different communities) and created maps that showed the layout of particular communities and the services within it.  These maps, however, were not only a representation of the physical spaces or services (as per the stated intentions of the activity). Through the map, the groundBREAKER evoked and engaged contested relations of self and Other within and through the physical space of the community. This exceeded both the map’s focus on “infrastructure” as well as a demographic                                                 33 See, for example, toolkits for “community action” developed by the International AIDS Alliance (, accessed January 21, 2015).   123 profile of community characteristics (such as a numerical tally of those who are “African,” “coloured,” “Indian,” “black” or “Other”) (loveLife, 2012c, p. 52). It was through the places of the map that race and class, for example, were brought together: poor people live in the black and coloured areas. Infrastructure was entangled within these constructions through the privilege of access: the hospital is in the white area. Here, the map can be read as a momentary representation of how apartheid brought “race, class and space” together in unique ways that continue in policies, such as housing, that have “worked through apartheid’s geography, rather than providing a basis for its transformation” (Hunter, 2010, p. 110). At the same time, the map moved beyond representation when two groundBREAKERs used the places within it to momentarily work at the boundaries of what is and is not said about race in the context of their lived lives. One groundBREAKER seemed to take the places in the map as an exception for how things are supposed to be in a democratic South Africa, asking why people in this town were racist. In contrast, the groundBREAKER presenting the map used the place of the map to re-inscribe race in the fabric of how it is. These statements were a small enactment of co-existing and contradictory ways of living the notion of race in South Africa: transcendence or being trapped within the effects of racialized identity categories (and this will be explored further as an effect of simultaneous discourses shaping the national subject in the following section). Understandings of race operated through lived relations (racism) that play out in who people are said to be and what they are said to know.  The educator in this session, Lukanji, was a Regional Programme Leader who described herself as black and isiXhosa first-language speaking. As she joined the conversation, she took-up the threads of this already contradictory encounter as a way to talk about race in relation to the loveLife programme. This was the only time I explicitly observed this conversation  124 happening and, while I cannot say it did not happen in other places or times, it struck me as unusual precisely because of the colour-blind discourse of purpose and opportunity that loveLife uses throughout its educational materials. Specifically, Lukanji used notions of place, in terms of the communities that loveLife works in, to bring attention to how the loveLife subject (the youth who needs loveLife) is constructed in opposition to white people who do not because they have resources and are living their own lives. In this very political statement, Lukanji seemed to acknowledge that there is a division between youth subjects (white versus Others) that is potentially reified within loveLife’s work. Indeed the invocation of this as a material “fact” (an explanation of how things are) works as a reification of the limits of whose “lives” are deemed as needing loveLife and what categories of social difference (here: race and class) play out in the privilege of subjects being able to live their own lives. Her statement also echoes her understanding of her own privilege in terms of her gender and sexual identities: “normal” in relation to how others come to “be defined as not normal” (Interview, August 12, 2013). Located within these already multiple relations of power and knowledge, Lukanji’s statement that she would be happy to work in a white area refuses a simplistic reading of problems of implementation (where white youth just need to be added in). Echoing the discussion in Chapter 3, the key point here is that constructions of “whiteness” are deployed, and simultaneously masked, in aspirational discourses of youth potential and are already at play in any articulation of who that youth subject is.  In this particular pedagogical encounter, engaging notions of identity through the map exceeded the map’s constructions of place, even as those places were taken up as a way to think about identities and power relations that have always already refused to remain in place. This gestures towards how articulations of identity, including loveLife’s very question of “Who am I,”  125 could potentially be engaged as sites of struggle within a map even as the “tool” of the map inevitably flattens their contours. In working with this reading of the pedagogical encounter, I am interested in the political potential of this map to spark conversation around how apartheid normalized “identity in racial terms” (Soudien, 2009, p. 44) and potentially resist the normalizing power of those identities. Silencing race, in vague references to community infrastructure for example, does not erase it. Rather, silence risks reifying race as something constantly present but never engaged as something to think about differently within the pedagogical encounter (Hammett & Steaeheli, 2013).  Within this specific encounter, the “place” where this mapping activity occurred was itself a site of privilege: a beach-front upper-middle class town on the Western Coast of South Africa. The white privilege of this space, seen in vacationing families and foreign tourists strolling on the beach, was disrupted by over a hundred and fifty youth bodies, only one of whom was white. Perhaps this was an embodiment of the empowered youth subject of the loveLife imaginary. At the same time, for the groundBREAKER whose nick-name was Mitchell’s Plain, an identity articulated through the name of a poor and coloured township outside of Cape Town, the promise of freely discarding and re-inhabiting identity was already impossible. As a momentary articulation of the emplaced performativity of identity, the groundBREAKER’s perceived identity was embedded in a discursive construction of place because he looked like he came from there. That place followed him into the pedagogical encounter, and was already a constitutive part of struggles over identity refusing discursive and material closure. It is here that any “local” life is both within a place and constructed through notions of place. This includes conceptualizations of a “national” youth subject.   126 Contested national subjects  loveLife was launched in 1999 as a consortium between the South African government, South African research and non-governmental partners, and international funders. Bringing together “local discussions” with “international research” (Robbins, 2010, p. 5), the campaign sought to respond to the failure of abstinence messages in HIV prevention with young people by drawing on a “marketing brand-driven approach” (p. 8).34 The brand was framed as an “essential affirmation of life and of the future” (p. 8) and revolved around notions of a unified South African youth identity. As David Harrison, the CEO of loveLife from 2000 to 2010 told me, the “starting point was to put young people in the middle and to really try and understand what made them tick…what really motivated them and what were the real pressures of their life” (Personal communication, July 15, 2013). That approach aimed to be a celebration of youth identity, and indeed did mark an explicit resistance to determinist accounts of who youth were said to be within dominant HIV prevention discourses.   The “middle” that the youth subject was mobilized within was already a complex and deeply contested space. In the late 1990s, the South African government was facing increasing national and international pressure to respond to HIV and AIDS. When Nelson Mandela took office in 1994, the HIV infection rate was estimated at 7.6% among pregnant women (Malan, 2013). With a presidency focused on nation rebuilding post-apartheid, Mandela was critiqued for not addressing HIV and AIDS during this time, though he later become a vocal activist for HIV                                                 34 This was in response to formative research showing South African youth to be “very brand conscious” (Nwokedi in Robbins, 2010, p. 6).   127 responses within the country.35 In 1998, however, even as the Treatment Action Campaign36 was formed to fight for all South Africans’ right to HIV prevention, treatment and care, Mandela’s successor Thabo Mbeki was aligning himself with AIDS denialists37 and treatment was a highly political topic. While prevention responses were also needed, loveLife’s exclusive focus on prevention distanced the initiative from the treatment activists (Epstein, 2007) and can be seen as the very condition that first allowed loveLife to become a legitimate “national” programme supported by the government. As a national programme, the question then becomes what kinds of “national” youth subject(s) loveLife has drawn on and deployed through its pedagogical imaginaries.   As I explored the “national” loveLife youth subject, I encountered an embattled figure in research literature related to the organization (as introduced in Chapter 2). Broadly, there are two dominant ways of conceptualizing the youth subject within this literature. Firstly, there is a celebratory reading of a new kind of youth identity: the stories of successful groundBREAKERs (Bateman, 2007) or a “meta narrative of hope and action” created through the programme (Etkin, 2013, p. 13). This resonates with official loveLife discourse, such as monitoring and evaluation reports (e.g. loveLife, 2012h). These discourses deploy a supposedly unitary national youth subject who, through participation in the loveLife programmes, is gaining a “sense of purpose in life” (p. 22). Secondly, and in direct opposition to the first construction of loveLife’s youth subject, there is a large body of work that critiques this subject as a neo-liberal and consumerist imaginary (Breidlid, 2009; Epstein, 2003, 2007; O'Toole, 2004; Parker, 2006; Prinsloo, 2007).                                                 35 See, for example, the profile at, accessed January 22, 2015. 36 See, accessed January 22, 2015.  37 AIDS denialism is a view that asserts that there is not a causal link between HIV and AIDS.   128 Primarily focused on loveLife’s public communication campaigns (such as billboards), these critiques argue that loveLife ignores the diversity of youth, as well as their material and political realities. For example, Parker (2006) is stinging in his critique of a “mono-cultural” understanding of youth who are presumably “unified by sexual desire and materialist consumption” (p. 1). Across these critiques, the concern is that a form of liberal individualism (Jeeves & Jolly, 2009) drives the programme and operates as a “form of modern governmentality” (Prinsloo, 2007, p. 33). As a present day articulation of the “imperative of health” (Foucault, 2003, p. 341), sexuality education is a site through which subjects and their relations are disciplined within the interests of global capital and national politics.   These debates are highly divisive. However, instead of trying to decide on the “truth” of these constructions, or becoming immobilized by their binary terms, the constructions themselves can be engaged as symptomatic of the complexity of the youth subject within a “local” place (here, the Nation) that is socially produced through interconnections (Massey, 2005). Focusing on what has been called the “crisis” of South Africa in the transition to democracy following apartheid, Hart (2014) argues that a South African national identity is being constructed through simultaneous moves of de-nationalisation and re-nationalisation. For Hart, de-nationalisation refers to social, cultural and political processes that have re-connected a post-apartheid South Africa to the global economy. These are the moves that are often read as neo-liberalism, such as economic policies. The second move, re-nationalisation, refers to how the transition from apartheid to a South African state required a discursive centering on a unified national identity in order for the very idea of a “nation” to be coherent. One example is Nelson Mandela’s discourse of the “rainbow nation,” which was a necessary imaginary for navigating a landscape where identity was at the heart of the traumas over who was or was not a part of that  129 nation. Hart uses the intertwining of these processes to explore the complexity within conceptualizations of the “nation” and resist explanatory frameworks that take neo-liberalism as a monolithic force shaping current political, economic and social events in South Africa.   Taken as constitutive parts of co-existing discourses constructing “who” a national subject is taken to be, this conceptualization offers a way to think outside of an either/or debate (the celebration of youth potential or the neo-liberal governance of that subject). As articulated through processes of re-nationalisation, all South African youth can be united as “purpose-driven.” Indeed, and as a reaction to a South Africa that was shaped through racialized discourse, a non-racialized discourse of possibility is a condition for the celebration of racialized identity (Soudien, 2009). In other words, the celebratory loveLife subject works to extend the Rainbow Nation discourse, whereby the subject may embody a racial identity, but that identity is ultimately positioned as irrelevant to their potential as a citizen of the nation. At the same time, as a youth identity articulated through processes of de-nationalization, the “future” that loveLife’s subject strives for has been untethered from particular social, economic or political circumstances that may keep youth from accessing this subject position. In this discursive construction of identity hard work and determination are enough to overcome any obstacle. Taken together, the processes of de-nationalisation and re-nationalisation are co-constitutive of the “national” subject, even as they reach across multiple registers to do so.   Taking the youth subject as formed through multiple and simultaneous discursive constructions of the “nation” can be helpful for thinking about some of the tensions within the loveLife programmes. The integrated racialized youth body is celebrated on loveLife  130 publications, even as racial identity is silenced in educational documents.38 For example, in the 191 page love4Life manual (loveLife, 2012f) the word, “race” is only used 3 times: once in relation to the rights of the human race that is “full of diversity” (p. 23), and twice in relation to glossary definitions of segregation (p. 190) and stereotypes (p. 191). Another tension can be seen in the relation to place. While youth are targeted based on their membership in “marginalized communities,” their location within those places can be overcome through learning “the importance of hard work” (loveLife, 2012e, p. 75). Rather than resolve these tensions, reading them as inevitable traces of contradictory constructions of a national youth identity is an illustration of how “sexuality is perhaps the most revealing marker of the complexity and vulnerabilities in the drive to produce a newly democratic unified nation” (Posel, 2002, p. 3). As part of a “national pedagogy” on “youth, sexuality education and HIV and AIDS” (Lesko, 2007, p. 520), loveLife draws on multiple conceptions of the nation, and by extension the national subject. In this way, the “national pedagogy” is also a contested and contradictory space, and the linear transmission of who youth should be (through that pedagogy) is already impossible. Engaging this complexity necessitates moving beyond binary debates over the “truth” of that subject: youth identity cannot be summarized in a billboard, any more than that billboard could represent some “true” identity of youth. Instead, the nation as a pedagogical space is formed through, and already populated with, subjects who are struggling to articulate their identities within contradictory processes that work across spatial and temporal boundaries. Beginning from this struggle, the question then becomes how the pedagogical approach conceptualizes and mobilizes its relation with these already complex and entangled identities.                                                  38 Race has been addressed explicitly in other documents, such as the “UNCUT” magazine that featured a debate on “is being white a privilege” (loveLife, 2012b).   131 Facilitating local lives  When I spoke with loveLife educators during individual interviews, I began by asking them: what does “teaching” mean to you (see Appendix A)? I used the word “teaching” as an invitation to begin to explore the relationships between knowledge, learners, and their own “teaching” identities. Across their responses, the educators immediately resisted my focus on “teaching.” Perhaps this was a small moment of resistance to me as an outsider to loveLife, and a disciplining of my own failure to take up the loveLife imaginary of the teaching relationship, specifically the language of “facilitation.” At the same time, what I also encountered through the responses was “teaching” as a power-laden notion entangled in the histories and imagined futures of what it means to be known in education. Lukanji, for example, told me that teaching is “telling people how things are done” (Interview, August 12, 2013). This was echoed by Varni, introduced in Chapter 3, who spoke of teaching as a “boring routine way of doing things and instructing young people on what to do” (Interview, June 10, 2013). Drifter, another Regional Programme Leader, defined teaching as “standing” and “throwing a whole bunch of words and numbers and knowledge” at learners (Interview, May 28, 2013). Within these responses, the notion of “teaching” can be seen to carry the entangled personal and public histories of education in South Africa.   While education may always be a political project, the education system under apartheid was a central mechanism for the overt control of racialized populations (Soudien, 2009). Curriculum reform after 1994 focused on removing “racist and sexist elements” from the national curriculum, as well as introducing “pedagogical principles such as learner-centeredness” (Chisholm, 2003, p. 1). However, the experience of schooling in South Africa has remained one of classed and racialized divisions and loveLife educators, who were all between 21 and 30 years  132 old at the time of the interviews, have experienced education in the throws of these transitions. Recent statistical research has highlighted the ongoing realities of two school systems in South Africa, with 75% of schools being classified as “dysfunctional” (UNISA, 2014). This is a “high-cost, low-performance education system” with problems that include “teacher shortages, “insufficient resources,” “inadequate infrastructure” and “lack of classroom discipline” (Centre for Education Policy Development, 2009, p. 12). Against this background, it is perhaps unsurprising that loveLife educators resisted the very word “teaching.” Even the two educators who defined teaching positively did so by separating their approach from that of “teachers” and disciplinary practices of teaching.  Drawing on this conceptual division between “facilitation” and “teaching,” one way of understanding loveLife’s pedagogical approach is to position it as resisting power relations that operate through “teaching” in South Africa, as well as negative constructions of the “deviant” youth subject that needs intervention. Across the interviews, educators described facilitation as an “interactive” approach: “involving” young people (Varni, Interview, June 10, 2013);  “creating an awareness” of the issues they face in their communities (Drifter, Interview, May 28, 2013); working with youth to “make their life better” (Nyla, Interview, May 17, 2013). As a form of critical pedagogy, these notions of “facilitation” position teaching as a transformative space where power relations can be acknowledged and challenged (Freire, 2007; Giroux & McLaren, 1986). While loveLife does not explicitly engage with the theoretical underpinnings of this view, “facilitation” draws on concepts,  such as dialogue and learner-directed involvement, that have been used in South Africa to challenge unequal power relations through education. This is reflected in research that has engaged with what it means for youth to participate in sexuality education programmes (e.g. Campbell & MacPhail, 2002) as well as research interested  133 in how teachers can be supported to engage with issues related to HIV prevention (e.g. Wood, 2009). Broadly, across this framing, facilitation is positioned as a pedagogical approach that can identify, address and challenge power relations. This includes the goal of teaching about and promoting gender empowerment (Campbell & MacPhail, 2002).  The discursive separation of “teaching” and “facilitation” is a powerful conceptual imaginary, holding forth the possibility of social transformation through the pedagogical approach. However, the separation of this approach from relations of power and knowledge has already been undone when loveLife’s pedagogical approach is seen to rely upon the very “deviant” youth subjects that it claims to speak back to. Non-governmental organizations, including loveLife, have arguably become the “favoured institutional form through which every social problem is to be addressed” (Choudry & Kapoor, 2013, p. 41). Youth participation, with the goal of “fixing” a problem, requires youth first to be understood as “deviant” (Gibbs, Campbell, Maimane, & Nair, 2010). This was enacted in the pedagogical encounter discussed in Chapter 3, for example, where the issues of the boys were a part of the conditions within which a loveLife educator and I met those boys as learners. It was also part of the texture of the “Gender” session, discussed in Chapter 3, that Lu led with Grade 9 learners. 2013 was the first year that loveLife had worked in that particular high school. It did so as part of a partnership that aimed to coordinate NGO involvement in the school and improve relations with the Department of Education that had recently restricted NGO access to schools. Lu and I attended an initial meeting with the NGO programme coordinator, a white South African woman, and encountered a school supposedly over-run with deviant subjects:  youth have teenage pregnancy rates that are out of control youth have low literacy rates youth have poor English skills  youth escape wildly out of the school gates   134 youth sleep with older men to get extra money youth don’t have parental role models youth come to school just to get a meal youth are undisciplined (Fieldnotes, March 2013)  The coherence of even having this meeting hinged on the implicit separation between the “experts” (programme designer, researcher, and to some extent educator) and the “participants” (youth). Here, the very starting premise of a programme like loveLife, using sexuality education as an HIV prevention strategy, is already in tension with a reading of “facilitation” grounded in critical pedagogy. Following the critical pedagogy theorist Paulo Freire (2007), educating “for” others, with a pre-determined purpose in mind, is never liberation. This plays out in a pedagogical approach premised on knowing whose “local lives” need fixing with sexuality education.  One response to this tension would be to promote a more “faithful” application of facilitation. However, my concern is that this response props-up the illusion that relations of power and knowledge can ultimately be “fixed” through sexuality education, and places an impossible responsibility on both the educator (to faithfully apply a pedagogical approach) and the learner (to take-up the relations offered). I have argued that any pedagogical approach, whether it is called teaching or facilitation, is entangled in the power relations of knowing and being known through education. The language of facilitation, and its methods such as dialogue, hail the youth subject into a supposedly power-neutral space (Ellsworth, 1992). The impossibility of this space was already present in the words of loveLife educators who spoke to me about the challenges of working with youth. Educators marked the very need for education in terms of youth as troublesome subjects: not being “open” to a topic (Nyla, Interview, May 17, 2013); being uncomfortable with a topic because they are “involved” (Gustav, Interview, August 6,  135 2013); being “not that willing to pay attention” (Nkunzi, Interview, June 29, 2013). Furthermore, in sessions that I attended, educators and youth also enacted, and potentially resisted, the very power relations that facilitation was said to overcome in the space of the classroom. In one session with Gustav, for example, youth enacted a farce of discipline when, frustrated that learners were not paying attention to the lesson, Gustav repeatedly sent the learners out to run laps around the inner courtyard of the primary school (Fieldnotes, June 2013). Over the course of the hour, learners sprinted in and out of the classroom, collapsing back into their desks as the rest of the class giggled. This could be an example of groundBREAKERs becoming “pseudo-teachers” and “doing a bad job of being – or trying to be a teacher,” as Grace Matlhape, CEO of loveLife, told me when speaking of some of the sessions that she had observed around the country (Personal communication, August 7, 2013). The “solution” would then be more training, such as training on the techniques of facilitation. The apparent simplicity of this solution is part of its appeal, and loveLife is currently making their training for groundBREAKERs more structured. What gets silenced in this solution is Gustav and the learners may also have been using the farce of discipline to speak back to the very disciplinary practices of schooling that have tried to tell them “who I am.”  It is precisely the tensions with the accounts of educators, and within their pedagogical encounters in particular classrooms, that prompts me to engage calls for more training, and more structured pedagogical approaches, as limiting moves. They are limiting in that they fail to interrogate understandings of self and Other that are already operating within any pedagogical approach. For instance, after the meeting with the NGO programme coordinator described above,  Lu and I were driving back to the clinic where he worked when he was not at the schools.  We travel on the dirt road leading from the high school to the highway (and the more affluent suburbs in the area) and talk about how we’ll work together for the rest of the  136 year. Lu will be working with grade 9 learners on Thursday and I say I can join him each week. He tells me he is glad, adding: with blacks, when a white is there they know it is going places. (Fieldnotes, April 2013)   In the invocation of going places, the privileges of being white were explicitly positioned within the “solutions” that NGOs offer and the promotion of western (research) knowledge to the problems of international development. His statement etched whiteness in histories of white privilege and black oppression, travelling across the boundaries of the nation, as well as the critique that the very notion of “potential” in South Africa remains a “white preserve” to “manage the integration of black people into the hegemonic order” (Soudien, 2009, p. 40). At the time, I did not respond directly to his comment, my mind racing through the places that he evoked. My failure to respond was itself an enactment of the privileges of whiteness, a silencing of direct engagement with what it means to be implicated in these power relations. The stories that white North Americans have told about black South Africans underpin the legitimacy of education’s interventions in South African youths’ lives (Lees, 2008). While I seek to resist these stories through research, they are also an inevitable part of my conceptions of self. Here, silence became a troubling reification of the comment, discussed earlier, that it isn’t racist, that is how it is.  Over the course of this research project, I met up with educators with(in) place. In each place, at each time, our relations of power, knowledge and identities were thrown together in our encounters with youth. No amount of training could ever provide a pre-packaged solution for how every “local life” was “utterly dispersed, unlocalised, in its sources and in its repercussions” (Massey, 2005, p. 184). This is not only to say that social transformation cannot be scripted. What is at issue is also how a more structured pedagogical approach risks becoming a violent closure of the very possibility of engaging differently with one another within the pedagogical  137 encounter. This is, I argue, what necessitates articulating sexuality education outside of the imperative of knowing who youth are. This gestures towards an articulation of sexuality education where the complexity of every local life may become a site for beginning to refuse who we are said to be (Foucault, 1982) and envision transformative relations of self and Other in excess of the goal of “teaching empowerment.”  Encountering the self within the pedagogical vision of empowerment  During training events, loveLife educators learn a “vision call” and everyone rhythmically chants the words together:  An HIV-free generation of young people in South Africa, who are in charge of their health, lives and futures.   This “vision call” hails loveLife’s youth subject: a national subject who takes charge of his or her own life. The collective identity that loveLife strives towards, an HIV-free generation, requires self-control and self-determination. In other words, youth must be in charge in order to be attain the coveted status of being HIV-free. The promise of autonomy through being in charge, combined with the goal of being HIV-free as a liberal democratic value of good health (Lupton, 1995), makes the vision hard to critique. Indeed, who would not want an HIV-free generation? At the same time, I argue that it is precisely the discursive power of this vision, and its potential silencing of the material effects in lived lives, that makes it a particularly problematic educational aim.   Combined with the discursive centering of a knowable identity within the loveLife programme, as discussed throughout this chapter, the “vision call” carries with it a number of educational assumptions. Although individual autonomy (being in charge) is the goal, the implicit assumption is that youth (as educational subjects) lack autonomy (Gacoin, 2014). As such, the role of education is reified as giving power to subjects who lack that power. Consider,  138 for example, the subjects that are deployed in the vignettes discussed in this chapter: the girls who are told they will be taxi queens or the boys who are very young and need life skills. Likewise, throughout my conversations with loveLife educators, there was a dominant imaginary of youth as “trapped” or “boxed” into being something that was not who they really were. In this way, and secondly, the very “need” for education relies on, and gains coherence from the vulnerability of the educational subject (Gacoin, 2014). The implicit role of education is thus to give the “correct” knowledge in order for the subject to act. Taken together, these assumptions reify the role of education as helping youth to tap into a broadly liberal conception of “power-within,” knowing “Who am I,” in order to exercise “power-over” (Laverack, 2004). Until a youth subject has this self-knowledge, he or she will remain vulnerable to the social environment. In turn, once the youth subject knows “Who am I,” education has done its job and it is that youth subject who carries the weight of responsibility for an HIV-free generation on his or her shoulders.  In this chapter, I have asked what happens when the very question of “Who am I” is seen as a constant site of struggle, not a stable target, within the pedagogical encounter. Articulations of “Who am I” are power-laden constructions of social identities that take form in and through sexuality education as an always political project. These articulations may be momentarily stabilized and held separate from one another, but these are always fragmentary configurations that do not travel unhindered through the pedagogical encounter. In contrast to a political rationality that couples self-knowledge with both action and moral responsibility, the pedagogical encounters engaged in this chapter point to how understandings of self and Other are always being encountered with(in) place: made-sense of, negotiated, re-articulated and potentially shifted.   139 Perhaps this gestures towards (without promising the solution of) a different way of thinking about the identities deployed within empowerment discourses. Any articulation of empowerment discourse is an always unstable articulation of who youth are and these conceptions of identity are just as “crucial to its pedagogy as the content of the information itself” (Ellsworth, 1997, p. 8). The point here is not to replace one way of conceptualizing youth identity with another. Any pedagogical encounter is structured through multiple articulations of “who” youth are said to be, and understandings of identity will always be constitutive of any particular pedagogical approach. At the same time, the inevitable failure of “overcoming” identity invites explicit engagement with what understandings of social difference are playing out in local lives and their relations. This, I argue, presents a “problem” for sexuality education outside of its perceived “solutions”: identity as a site of struggle and a problem for educators, learners and researcher to think about within pedagogical encounters.  My engagement with identity in these encounters draws on conceptions of power and agency that some would argue has destroyed the ability of the subject to act (e.g. Baylies & Bujra, 1995). However, throughout this chapter, I have argued that an educational subject premised on stable knowledge of identity is always already impossible to begin with. This includes, but cannot be reduced to, what comes to be known as gendered identities. It also points to how youth are not the only “I” within a pedagogical encounter. Educators, for example, encounter youth within and through their own identities and relations of power and knowledge. As a researcher entering this encounter, my own identities, as a North American cisgender researcher and co-educator within the sessions, were inextricable from relations of “self” and “Other” that reached across temporal and spatial boundaries. When moral responsibility is untethered from a neo-liberal political rationality, the vulnerability of all educational subjects  140 may be seen as an ethical condition (Butler, 2001). Here, the un-scripted and messy struggle over who “I” am reifies relations of power at the same time as it holds the possibility to re-articulate those relations (Butler, 1992). It was Makaziwe who marked the inherently political nature of this struggle when she explained her own understanding of empowerment as “trying to find ways of living” (Interview, July 15, 2013). This is, perhaps, the greatest challenge for sexuality education: a question of how its encounters might open up, rather than foreclose, ways of living that are more just for more people.    141 Encounters: A pause on gender as a topic to teach in South Africa   Sexuality education is the “cornerstone on which most HIV and AIDS prevention programmes rest” in South Africa (Francis, 2010, p. 314). Since 1994, sexuality education has been provided to learners through the Life Orientation curricular area. Broadly, life orientation “addresses skills, knowledge and values for the personal, social, intellectual, emotional and physical growth of learners” (Department of Basic Education, 2011, p. 8). Life orientation is provided to learners in Grade 7 – 12 (approximately ages 12 to 18) and provides instruction related to sexual and reproductive health within Grade specific topics. In addition to being sensitive to gender as an “issue of diversity” (p. 5), learners are expected to cover topics such as gender constructs (Grade 7), gender equity (Grade 8) and gender roles (Grades 10 and 11). While loveLife is not an official part of the South African school system, loveLife educators who work in schools are commonly paired with a Life Orientation teacher, or teachers, and loveLife content is broadly seen as complementary to the Life Orientation content.  The coherence of gender as a teachable topic for sexuality education in South African educational discourse draws on processes of curricular change post-1994 (Chisholm, 2003). Following apartheid, the first phase of curricular change was focused on “representational issues” and eliminating the “racist and sexist elements” of the apartheid curriculum (p. 1). The second phase introduced what has become known as the highly controversial “outcomes-based education,” a model of educational reform that resulted in a national curriculum statement called Curriculum 2005. Curriculum 2005 was then revised in 2000 following controversy over the implementation of outcomes-based education in South Africa. While the language of outcomes-based education is no longer prominent, it is debatable as to what extent the Revised National Curriculum Statement (re-reviewed in 2009) is a departure from this model. According to the  142 Western Cape Education Department website, the revised curriculum remains focused on “comprehensive outcomes and assessment standards, which indicate the knowledge and skills required for each grade and learning area, and how progress should be assessed.”39   Outcomes-based education (OBE) is broadly focused on measurable student outcomes rather than particular methods or heavily scripted content and draws coherence from multiple intellectual histories. For example, Soudien and Baxen (1997) argue that OBE “has its roots in two educational reforms: mastery learning and competency-based education” (p. 451). These reforms are based, respectively, on the theories of American cognitive psychologist Benjamin Bloom who believed that with the right “input” all learners could “master” outcomes, and on a rhetoric of the “competencies” needed by youth to enter a rapidly changing job market in the United States in the 1960s. Jansen (1998), in contrast, notes other influences on OBE, including the work of American behavioural psychologist BF Skinner and the model of curriculum objectives promoted by Ralph Tyler. At the same time, for Jansen, the most direct influence on OBE came from training debates related to “competency” in Australia and New Zealand. While the multiplicity of these potential influences may be confusing, they are all instances of how South African national curriculum development extends beyond the boundaries of the “nation.” Focusing on one can illustrate how gender is framed as a teachable topic.  In what has become known as the Tyler Rationale (Tyler, 2004 [1949]), curriculum development is a scientific process that is guided by four fundamental and universal questions. These questions are: “1) What educational purposes should the school seek to attain?; 2) What educational experiences can be provided that are likely to attain these purposes? 3) How can                                                 39 See, accessed January 21, 2015.  143 these educational experiences be effectively organized?; [and] 4) How can we determine whether these purposes are being attained?” (p. 51). Reading these questions through the guiding logic of sexuality education as an HIV prevention strategy, they might be answered as follows: 1) The school aims to reduce HIV infection among young people; 2) youth need to be engaged in conversation to understand and challenge gender inequities; 3) notions of gender can be divided into topics such as “gender constructs,” “gender equity” and “gender roles” and 4) the purpose can be measured in terms of attitudes to gender equity, and ultimately reduced HIV infection rates. The coherence of “tasks” and “modules” and “topics” to teach in sexuality education resonates with a view of curriculum as fixed content that can be developed through a precise and objective scientific process (Kliebard, 2004). Within this logic, a definition of gender can be agreed upon before the pedagogical encounter, and the purpose of sexuality education is to guide learners to the right outcome based on that definition. Extending the ideological assumptions implicit in this conceptualization, gender as a teachable topic can be further explored by coming back to the multiple potential influences on OBE. Taken together, these influences are an extension of how “the process of curriculum making has almost always been managed within the framework of a form of internationalization” in South Africa (Soudien, 2009, p. 24, emphasis original). In his historical reading of four key periods of curriculum-making, beginning in the 17th century, Soudien (2009) argues that “social difference” has been, and continues to be, the “central question that drives curriculum development” within the country (p. 20). While this is perhaps most explicit in colonial and apartheid-era educational discourses, Soudien argues that it has continued in a post-apartheid curriculum that has “catalyzed and even amplified the major vectors of discrimination from the past” (p. 43). It does this through addressing a “middle-class citizen” (p. 40) who achieves  144 equality through the “meditational technologies of whiteness” (p. 45). The import of this argument for the topic of “gender” is that failing to problematize relations of social difference already operating within and through that topic risks mobilizing gender as an enactment of that difference, such as particular gendered identities and their relations enacting whiteness. For example, as gender is framed as a “national” topic, the “problem” of gender in HIV and AIDS discourses settles in and through the classed and racialized body. As reported in the 2012 South African National HIV Prevalence, Incidence and Behaviour Survey, the overall HIV prevalence was 15% for black Africans, and 0.3% for whites (Human Sciences Research Council, 2014). Reading curricular articulations of gender through social difference draws attention to how particular subjects (namely black and coloured youth from what loveLife calls marginalized communities) are targeted as “needing” sexuality education as an HIV prevention strategy (as discussed in relation to loveLife’s youth subject in Chapter 4). As long as it is framed as a neutral topic, gender risks mobilizing the very co-constitutive relations of self and Other, and understandings of their differences, that it claims to address.  Furthermore, the topic of gender is entangled in understandings of social difference that extend beyond the imagined boundaries of the “nation.” Global HIV prevention discourses, and the forms they take through international technical working groups, guidelines and experts, can be read as part of the “form of internationalization” that South Africa is currently within. For example, the United States Agency for International Development (USAID) contracted an American “sexual health expert,” Douglas Kirby, to advise the South African Department of Basic Education as they developed a national policy calling for sexuality education to be a  145 mandatory and assessed subject in all South African schools.40 Kirby’s expertise included having developed “17 characteristics” for effective sexuality education curricula development applicable for any national setting (Kirby, Laris, & Rolleri, 2007). The point here is not only that a global expert has had a role in saying what is South African educational policy. It is also necessary to interrogate how this expertise is an enactment of relations of power and knowledge that move across what gets taken to be a “local” curricula or “global” expertise. As the Department of Basic Education begins to revise teacher training curricula to “include HIV and sexuality education” (Department of Basic Education, 2013, p. 29), already multiple and power-laden understandings of sexual subjects and sexual meanings will be deployed. These understandings shape the limits of “gender issues and sensitivities” and are woven into what becomes seen as “both the content and teaching approach” (Senderowitz & Kirby, 2006, p. 36). This is an enactment of social difference playing at the limits of what settles into both content and pedagogical approach.   Working with the notion of “gender” as implicated in diffuse spatial curricular configurations further problematizes the premise that notions of gender can ever be simply transmitted through the pedagogical encounter. Gender (as a problem for youth in global HIV prevention discourses) is an impossible universal topic that settles into the appearance of stable forms of knowledge such as “gender equality.” By engaging these forms within the pedagogical encounter, I am interested in both the limits of these forms, as well as what might happen with(in) sexuality education by thinking at those limits.                                                  40 Kirby was a speaker at a meeting organized by the Department of Basic Education at the 5th South African AIDS Conference. A press release from the event is available at, accessed January 21, 2015.  146 Chapter five: Risky forms of knowledge It was an exchange that could have unfolded in multiple places, and in multiple times, over the past 30 years. An educator stands before a group of learners.  Educator: What is sex?  Learner: Sex is something that a man and woman do together.  Educator: There is safe sex and risky sex. The consequences of risky sex include teenage pregnancy, sexually transmitted infections (STIs), and infection with the human immunodeficiency virus (HIV). Use a condom.  Learner: What should you do if you are coming home late at night with your girlfriend and you have to have sex?  Educator: Don’t have sex without a condom. (Fieldnotes, April 2013)   The coherence of this exchange (and the perceived pedagogical linearity as a series of questions and responses) revolves around how scientific discourses have made sense of (hetero)sexual behaviour (risk/safety) in the context of the HIV and AIDS epidemics (Patton, 1990; Treichler, 1999). In South Africa, as in other parts of Africa and Asia, it is sex between a man and a woman that drives the problem of the “heterosexual epidemic” (Karim & Karim, 2002). These gendered and heterosexual subjects engage in sexual behaviour that puts them at risk of physical consequences. Condoms are a widely accepted HIV prevention method and there is no excuse to have sex without a condom once the risks are known. Indeed, this is part of what speaking about gender roles and power relations within sexuality education aims to achieve: young women will be able to negotiate for condom use to protect themselves from HIV infection.  And yet, within this universal and universalizing narrative, this exchange has already been constituted through its excesses. For example, in the definition of sex, the “facts” of safe sex and risky sex are inseparable from the discursive histories of HIV prevention that have  147 tethered safety/risk to a power-laden binary of heterosexual/homosexual identity (Weeks, 2010), including how the “risk” of homosexuality in early accounts of the epidemic in the United States drew on, at the same time as it reified, the “safety” of a normative heterosexuality. These binary and power-laden constructions continue in HIV-related educational materials that prop-up the discursive norms of heterosexuality (such as the “safety” of marriage, for example) through the implicit risks of homosexuality (Gacoin, 2010a). Attending to the effects of these constructions raises questions as to what kind of “safety” is being offered to those who may not fit its norms. For instance, there have been multiple cases of “corrective” rape to “cure” black women of their sexual orientation. The lived lives of these women who identify as (or are perceived to identify as) lesbian is in excess of the “safety” offered in the pedagogical encounter introduced above. Furthermore, while the term “corrective” rape draws attention to the intersections of homophobia, gender and violence in South Africa, media discourses41  around “corrective” rape implicitly separate rape acted on lesbian bodies from that acted on heterosexual bodies.42 The violence of any act of rape haunts a learner’s question of what to do when you have to have sex. The answer, an incitement to rational decision-making - don’t have sex without a condom - masks how that having is already entangled in power, in emotions, in pleasure, in control. In this exchange, these are excesses that settle in and through both what is known and what kind of relation is imagined to that knowing.                                                  41 See for example a 2013 article in the New York Times entitled “The brutality of ‘corrective rape’,”, accessed January 25, 2015. 42 This is the argument put forward by the Triangle Project, a South African non-governmental organization, who critiques how a sensationalist media discourse of “corrective rape” implies that this is a “new phenomenon that is quite separate from rape perpetrated against women in general.” (See the public statement, archived at, accessed January 21, 2015).   148 The forms of knowing in this exchange are also already entangled in diffuse and power-laden spatial relations of knowledge. The exchange is drawn from a session that took place in 2013, in a black isiXhosa-speaking township outside of Cape Town. Within it, getting the right “facts” across collapsed into telling youth what to do in this local place, such as using a condom. At the same time, those “facts” had already been formed through relations of power and knowledge that exceed place. For example, perhaps the young person who does decide to use a condom will use one of the Government of South Africa’s Choice condoms distributed by non-governmental organizations.43 These “local” condoms are part of a “technical fix” that has been critiqued as underpinned by a global “condom code” driven by lucrative grants from donors, grants that carry with them political incentives to prove the “effectiveness” of condom use as an HIV prevention strategy (Timberg & Halperin, 2012). Attending to these relations necessitates asking how any “fact” (including using a condom) comes to be coherent as a stable form of knowledge. Asking this question within the pedagogical encounter is a way of exploring what happens when the pedagogical approach is engaged as implicated in, rather than external to, what come to be seen as the “facts” of sexuality education as an HIV prevention strategy (and an always political project). At issue are both the effects of those configurations as well as what might happen in moments of knowing that are already in excess of those configurations.  In this chapter, I take up these questions by thinking with(in) the pedagogical encounters of sexuality education in relation to “knowledge in the making” rather than knowledge as something “already made” (Ellsworth, 2005). A conceptual focus on “knowledge in the making”                                                 43 For example, the Society for Family Health (SFH) is a distributor of the Choice condoms through the government’s Public Sector Distribution program (see, accessed January 25, 2015). SFH is the South African affiliate of the US-based “global health organization” Population Services International (see, accessed January 25, 2015).    149 foregrounds how all knowledge, no matter how stable its appearance as a thing made, is being continually made and re-made as it encounters the presumed subjects and objects of sexuality education within relations of power. Ellsworth (2005) describes this in terms of a relational experience of learning that “never takes place in the absence of bodies, emotions, place, time, sound, image, self-experience, history” (p. 55). In other words, it is always already impossible for the “facts” to travel unmediated through the pedagogical (or research) encounter. It is also impossible to conceptualize those “facts” as fixed or bounded objects that may just be understood differently by different subjects. The very boundaries of the “facts,” including their binary construction of forms of knowledge and identities, are continually being made and re-made within these encounters. Attending to knowledge in the making provides a way to engage with the “facts” that initially appeared to overwhelm the pedagogical encounters in this study: youth needed to learn the “facts” of risk, and there was a pressing “risk of getting the facts wrong” as groundBREAKERs used the knowledge they learned through the loveLife training to work in communities (Grace Matlhape, Personal Communication, August 7, 2013). Those facts were written in manuals, presented on PowerPoint slides, and written on classroom chalkboards. Engaging these facts as momentary effects of knowledge in the making is a way of resisting binaries such as fact/fiction or risk/safety (even as my own habits of thought inevitably pull me back into them). The argument here is not only that the “facts” of sexuality education are formed through relations of power and knowledge within HIV and AIDS discourses (Treichler, 1999), but also that those formations are ongoing sites of struggle within the pedagogical encounters of sexuality education.   150 In this chapter, I draw on two analytic concepts to engage sexuality education in terms of “knowledge in the making.” Firstly, Ellsworth’s (2005) concept of “configured knowledge” provides a way to think about knowledge that has the appearance of being already made while simultaneously resisting the apparent “truth” of those configurations. The appearance of stability is precisely part of the power of configured knowledge, such as the objectives of a lesson plan, the “facts” to explain, or the behavioural outcomes to measure. However, these are ultimately unstable configurations – relations of power and knowledge that have only momentarily coalesced in a particular form. This conceptualization of knowledge draws attention to “strategies of power” that have given rise to particular configurations of knowledge (Foucault, 1998). Simultaneously, and extending the argument made in Chapter 3 that the pedagogical approach is itself a technology of power, this focus interrogates how particular pedagogical practices may work at the limits of those configurations. In this chapter, I develop and use the term configuring pedagogical practices to explore the pedagogical encounter as a site within which notions of the “facts” of sexuality education (from “risk”/”safety” to “gender equity”) are continually being constructed and potentially resisted by educators, learners and myself as researcher.   The second concept, “excess”, draws explicit attention to the silences and ruptures at the limits of configured knowledge. Following Orner, Miller and Ellsworth (1996), excess refers to meanings that exceed the norms of a particular discourse. In other words, excess is not knowledge that was left out of the curriculum, for example, because of time constraints, relevance, or lack of foresight. Instead it is “meaning out of control, meaning that exceeds the norms of ideological control or the requirements of any specific text” (Friske in Orner, Miller, & Ellsworth, 1996, p. 72). The concept of excess is a way of attending to, no matter how fleetingly,  151 the moments when meanings may escape their intentions. Excess also provides a way to think about these moments outside of the failure of meaning as a problem of “implementation,” such as learners do not understand, learners are not paying attention, or the educator is not following the “right” approach. In reading for excess, the focus is on how multiple relations of power are knotted together in what becomes known and knowable. Working with excessive moments is thus a gesture towards the “histories and interests bound up in how and why we do what we do” (p. 73). My focus is what might happen with(in) sexuality education when those histories and interests are engaged as a part of the risk of knowing and being known, rather than something to overcome through knowing about risk.  To problematize particular forms of knowledge as the solution to “risk,” and attend to what happens when these forms encounter and are encountered by learners and educators, this chapter engages configuring pedagogical practices in terms of three of its effects drawn through multiple relations of power and knowledge. These are: the configuration of particular forms of knowledge as “fact,” the configuration of “risky” learners and their perceived pedagogical relation to knowledge, and the configuration of pedagogical spaces as “safe” for encountering understandings of social difference. The discussion of these practices revolves around pedagogical encounters drawn from loveLife training events: the pre-training of Regional Programme Leaders (in which I participated as a trainee) and the initial training camp for new groundBREAKERs (in which I participated as a trainer). Each practice is then doubled through moments of “excess” drawn from multiple pedagogical encounters. These excessive moments “speak” to the preceding discussion of configuring pedagogical practices, at the same time as I ask how these moments may interrupt the linearity of my own argument and work with and through the limits of knowledge as something already made.   152 The last section of this chapter then turns to how the inevitable “inadequacy for life” of any form of configured knowledge, including those deployed through gender empowerment discourse, might be “useful and valuable as a potential provocation to action” (Ellsworth, 2005, p. 164). Here, I ask what might happen to sexuality education within this provocation, and how the impossibilities of knowing might invite possibilities for knowing and being known otherwise.  Configuring the “facts”   Speaking about the “facts” in relation to HIV infection is so commonsense that it may not seem risky. Since the beginning of the epidemics, responses to HIV and AIDS have been driven by scientific understandings of sexuality (Patton, 1990). HIV infection is measured in quantitative data on prevalence, incidence and risk factors (Cáceres, 2000). Behaviour is about “norms,” “phases,” “stages,” targets” and “forces” (Lupton, 1995). These statistics provide a linear story of challenges and progress, making order and rationality over an epidemic that is profoundly unknowable (Treichler, 1999). Taken to HIV prevention with youth, the “facts” are coupled with “risk” in a self-evident circle of truth: begin with the facts of risk of HIV infection, give youth more and more facts depending on their risk factors, and then measure whether youth have understood those facts by whether they have avoided risk. This logic underpinned the training that loveLife provided for Regional Programme Leaders (RPLs) who would, in turn, train new groundBREAKERs for the 2013 programmatic year.  The topics for the morning are treatment, opportunistic infections and medical male circumcision. The trainer tells us: Assume they know nothing. Give the information. Then do an interactive game. (Fieldnotes, January 2013)  The overall aims of the RPL training are to ensure that RPLs have an “understanding of latest developments in the field of HIV and AIDS,” learn “new training methodologies” and discuss “best practice methods” for working with groundBREAKERs (loveLife, 2012g, p. 15). The  153 topics on this day were a part of the facts related to HIV and AIDS: scientific knowledge related to treatment and opportunistic infections and the promotion of medical male circumcision as an HIV prevention method. The trainer’s instructions – to give the information and then do an interactive game can be understood as a methodological approach of involving learners. Coupled with the reminder to assume they know nothing, these instructions can be taken as a practical approach for working with groundBREAKERs who will have different knowledge levels related to HIV. For example, while all groundBREAKERs must have completed high school in order to join loveLife, South Africa’s education system can be seen as in a state of “crisis” with wide discrepancies in the quality of education along the lines of wealth, school location, language and province (Spaull, 2013). The need for this training is thus guided by the concern that groundBREAKERs may not have received instruction based on the “latest developments in the field of HIV and AIDS” that loveLife wants them to have.   In many ways this brief moment may appear unremarkable: it is just a matter of making sure that loveLife educators “have the facts right” (Grace Matlhape, Personal communication, August 7, 2013). However, the very notion of having the “facts right” relies upon and reifies a particular conception of what knowledge is, as well as the presumed relation between subjects and knowledge. Knowledge as something to get “right” deploys the pedagogical encounter as a site for information transmission from educator to learner. In doing so, it masks how the pedagogical approach is itself a constitutive part of what comes to count as knowledge within that encounter.   In the pedagogical encounters of the RPL training, there were already multiple configuring pedagogical practices at work, each shaping what came to be seen as “fact.” Each topic was presented on a series of PowerPoint slides that had been prepared by a scientific expert  154 for loveLife. Framed as presenting the “facts,” the slides themselves enacted a textual expertise that fixed the apparent “truth” of those facts. In other words, presenting these slides was a configuring pedagogical practice that solidified the appearance of particular forms of knowledge as neutral and objective truths (Foucault, 1998). Once the “truth” of the facts was established, a particular relation to those “facts” was then set up through the pedagogical approach. The trainer explained this information to us (as trainees) and we in turn (as educators) would be expected to explain this information to groundBREAKERs. Following Ellsworth (2005), explanation can be understood as another configuring pedagogical practice that works “in the service of knowledge as a thing already made” (p. 160). The supposed neutrality of explanation (of what knowledge is) is a part of its power, and a constitutive part of what gives knowledge a stable appearance. Furthermore, explanation also operates in the “service of the grids of positionality” (p. 160), or who subjects are taken to be and what is their presumed relation to knowledge. In this instance, the relationship between learners and knowledge was defined through deficiency: they know nothing. The coherence of this relation to knowledge draws on and reifies relations of power between educational subjects (teacher/learner) as well as relations of power that take form in understandings of age (youth/adult). It also masks tensions inherent in the relation to knowledge imagined in and through the position of loveLife’s Regional Programme Leader. Regional Programme Leaders are conceptualized within the loveLife programmes as “trainers.” However, and operating within an understanding of pedagogy as information transmission, the RPLs must themselves be carefully trained. In other words, it is not only the learner who knows nothing, but also the educator. These tensions in the RPL role are further extended, at the same time as they are masked, when the role of the educator is centered on leading an interactive game. The focus on an interactive game draws on an ideology of participation, broadly based in critical pedagogy  155 (Freire, 2007), which explicitly seeks to disrupt a pedagogy based in information transmission, and which resonates with loveLife’s discursive centering of “facilitation” as discussed in Chapter 4. However, critiques of development programmes such as loveLife’s have foregrounded how techniques of participation, here the interactive game, give the illusion of participation without addressing underlying power relations of what is known and what kind of relationship to knowledge is being offered (Leal, 2007). In this example, the interactive game encourages participation after youth and educators both have the right “facts.” It does not seek to engage youth or educators in the making of those “facts.” If anything, it continually reduces the facts by making them “manageable,” as seen in the next vignette.  The trainer puts up a PowerPoint slide with a long list of opportunistic infections. She adds: you don’t need to give all this information to the groundBREAKERs. (Fieldnotes, January 2013)  The statement that groundBREAKERs do not need all this information is explicitly a practical concern. The power-point slide was a long list of the medical names of infections. The slide was visually overwhelming, and at some point a comment was made that it was risky to have groundBREAKERs walking around thinking that they could give medical advice. In many ways I agree. At the same time, my agreement (drawing on my own discursive expertise around what “should” be known) is implicated in how simplification operates as a configuring pedagogical practice. Simplification configures knowledge based on who learners are, according to age, and what they are supposed to need based on a developmental understanding of age. Those relations of power and knowledge stretch across moments that are supposedly outside of the pedagogical encounter (such as the PowerPoint slide prepared for the training) as well as within a particular pedagogical encounter (such as what information the educator is expected to ultimately give).   156 Across these readings of configuring pedagogical practices (presenting, explaining and simplifying the “facts”, as well as involving youth in the transmission of the “facts” with a game), there is a tension in arguing that topics such as treatment or opportunistic infections are instances of knowledge in the making. AIDS denialists have appropriated the instability of the “facts” of HIV to argue that “treatment” is not “real,” a claim that has devastating effects on people who need treatment to address the biological effects of both the HI-virus and opportunistic infections. Perhaps, however, this appropriation also points to the political necessity of putting forth alternative conceptual frameworks for thinking about knowledge in the making and its relations of power. AIDS denialist views have been formed in relation to how the promotion of bio-medical “facts” have power-laden histories in South Africa, as seen for example in racialized medical abuses perpetuated by the apartheid regime (Nattrass, 2012). These views are an (albeit extreme) way of understanding, and resisting, conceptions of self and Other that are inseparable from what comes to be known. In this already power-laden discursive context, and under the rhetoric of scientifically based training, loveLife offers youth the “facts” of HIV. My concern here is that these “facts” are offered at the expense of a consideration of how they come to be known as such, and what relations of knowing and being known already operate at and through their boundaries. Following Chilisa (2005), western epistemologies have set the conditions for what counts as knowledge related to HIV transmission and prevention. The notion of “modes of transmission,” for example, defines HIV infection in terms of western understandings of relations between self, Other and knowledge. The violence of this knowledge is not that “context” is ignored, but that a particular understanding of knowledge, and people’s relation to knowledge, is deployed as a universal “truth.” In other words, speaking about “mother to child transmission” or even “sexual transmission” without exploring the power relations within the notions  157 themselves may silence and dismiss how people name and understand their own relations to one another and to disease.  As the “facts” were deployed within the pedagogical encounters of the loveLife training, the configuring pedagogical practices worked to foreclose these potential spaces for making sense of the world within multiple relations of power and knowledge. For example, one activity focused on “agreeing” or “disagreeing” with statements related to HIV.  The educator reads the statement: if you have unprotected sex with someone who is HIV positive you will get HIV. The answer, she tells us, is agree. (Fieldnotes, January 2013)  The purpose of this activity was to involve learners in the “facts” of HIV and correct misconceptions. Within this logic, making a causal link between unprotected sex and HIV infection is unsurprising. However, as we did this activity, the imperative for rational and binary thinking drove what engagement was allowed with the “facts.” The activity was structured as a binary choice between agree/disagree that, depending on the statement, was equivalent to the correct or incorrect answer (a framing that resonates with the agree/disagree “Gender” activity discussed in Chapter 3). Within this structure of knowledge, the “correct” answer was restricted to agree even through the estimated statistical probability of contracting HIV from a single unprotected sexual act is relatively low.44 This could be dismissed as a generalization needed to get across the HIV prevention message, a contradictory condition of getting the “facts” across. However, as a configuring pedagogical practice, certainty operates in the service of scientific                                                 44 For example, the estimated probably of contracting HIV from “receptive penile-vaginal intercourse” is 8 per 10,000 exposures  (see, accessed January 15, 2015). The highest “risk” sexual act is “receptive anal intercourse” with a risk of 138 per 10,000 exposures. While these “risks” are estimates, and will vary greatly depending on how the sexual “act” occurs (e.g. was it consensual or not), they still point to the vast deal of “uncertainty” within the “facts” of HIV.    158 discourses that are themselves filled with uncertainty (Treichler, 1999). In other words, the pedagogical approach is a constitutive part of how facts are “socially and linguistically managed” (p. 16), and this management acts on and through the identities and bodies it targets. In this particular encounter, the imperative for stable knowledge silenced the multiple reasons that youth might have unprotected sex, such as coercion or violence within gendered power relations (Wood, Lambert, & Jewkes, 2007; Wood, Maforah, & Jewkes, 1998). In silencing those reasons, it also silenced “who” is most at risk in unprotected sex: the receptive (male or female) partner. This itself is a relation of power and knowledge, particularly given how discussions of anal sex may be silenced through the discursive and power-laden collapse of anal sex with a “perverse” sexuality (Hardy, 2011). Furthermore, and in line with a neo-liberal political rationality discussed in Chapters 1 and 4, there is a moralizing effect for not taking up the “facts:” you will get HIV (and implicitly have only yourself to blame). This is an incredibly “risky” claim that has the potential effect of extending stigma of people living with HIV, stigma which has been found to be a significant barrier to testing for HIV in South Africa (Kalichman & Simbayi, 2003).   At the same time, while configuring pedagogical practices may work in the service of particular configurations of knowledge, those practices also speak to how the transmission of knowledge is always already impossible. In other words, these practices are necessary because knowledge encounters, and is encountered through, contested relations of power and knowledge, as seen in the following vignette.  We are talking about treatment in small groups. People are angry: there are shortages of antiretroviral (ARV) drugs in rural areas; people are stealing ARVs to get high. The trainer tells us that this is a hard session. Planning is key. We will have 90 minutes to discuss this topic with groundBREAKERs: What are ARVs? How do they fight HIV? What are the types of ARVs? How can you access ARVs? How are they used? Who is eligible for treatment? What are the side effects? (Fieldnotes, January 2013)   159 In this encounter, the discussion in small groups momentarily worked at the limits of its “facts.” Anger points to treatment in South Africa as a profoundly political issue. Since the late 1990s, access to treatment has operated at the fault lines of imperial legacies and globalized imperialism. The former South African president, Thabo Mbeki, infamously endorsed AIDS denialists and the government did not provide ARVs through the public health system until 2003. In an analysis of this period, Marais (2012) argues that, while much analysis tends to either defile or defend Mbeki, “denialism” provided a “grammar” for “national defiance” in response to challenges in South Africa, such as rapid changes in economic policy, as well as past colonial constructions of African sexuality (see also Nattrass, 2012). In the years since these debates, ARVs are being rolled out through a public health system that is chronically underfunded. Media reports have repeatedly highlighted the “crisis” in the health care system, including shortages of ARVs in rural areas (e.g. Bhekisisa Team, 2014; Taylor, 2012). While reports of ARVs being stolen to make street drugs may be exaggerated (Motema, 2013), these stories are also part of how people make sense of a society marked by multiple inequalities, both within the health care system as well as outside of it. When the loveLife trainer referred to this as a hard session, she was perhaps referencing these relations of power and knowledge. She might also have been referring to the emotional challenges of discussing “treatment” when an educator knows someone who is dying, or when that educator might be living with HIV. The acknowledgement of the session as hard speaks to difficulties within any instance of knowing, or claiming to know. The question then becomes what kind of relation is invited to that difficulty through the pedagogical approach.   In this particular instance, and in line with the centering of rationality discussed throughout this section, the session as hard was met as a problem of planning. As such, the facts would need to be carefully managed: a return to scientific definitions of ARVs, access as a matter  160 of when and where within scientific guidelines and national policies, side effects as physical symptoms that can be medically managed. Of course there was a game: we acted out a battle between the captain (T4 cells), his soldiers (white blood cells) and the intruder (HIV). The army won, but I remain troubled by the causalities of configured knowledge within these struggles.  Excessive facts of violence  A couple is talking – HIV testing, sexual desires, using a condom. There are accusations. Men storm off the scene. A boy is being bullied and is killed. His mourning father breaks down in an exaggerated religion display. He then kills the bully with a gun. A girl has been out partying – alcohol, drugs – her mother yells and forcefully pushes her out of the house. A guy forces himself on his girlfriend – the father gets home and beats him up. A girl tells her parents she is pregnant – the class watches as a physical fight breaks out. (Fieldnotes, February – August 2013).    These brief moments, drawn from multiple role-plays that occurred during loveLife sessions, slip through multiple readings of violence and its effects. In one, violence is endemic in South Africa. In 2012 to 2013, there were an average of 45 murders per day, four and a half times the global average (Africa Check, 2014). The South African Police Service reported that during 2011 to  2012 there were 9,193 cases of sexual offense in the Western Cape (Rape Crisis, 2014). Taking into account the number of cases that go unreported, some estimates put these figures as high as 84,000 in one year in the province. Positioning the role-plays within these statistics is part of the texture of how South Africa is living a “culture of violence” forged through the relations of apartheid and their ongoing formations (Rape Crisis, 2014). Under apartheid, institutionalized racial discrimination played out through apartheid laws and interactions with police, as well as labour migration from rural to urban areas (Barnett & Whiteside, 2002; Jewkes, 2009). Within this context, many men “adopted more exaggerated constructions of masculinity” that included  161 “risk taking” and the “use of violence” (Jewkes, 2009, p. 30).45 Cut off from political and economic achievement, black men found “masculine affirmation in homosocial (sometimes criminal) settings and in their relations with black women” (Jewkes & Morrell, 2010, p. 4). Here the role plays, performed in and through the bodies of black learners, can be read for the traces of a “racialized gender arrangement” (p. 4) created through the relations of apartheid. Violence breaks out between black men – the boy being bullied - and is used by black men on black women – a woman being raped. Violence is a solution to the problem - a response to disobedience when a child is pregnant or stays out late partying - a closure of communication. This reading lends coherence to why loveLife has developed a session on “rape and abuse” (loveLife, 2012f, p. 162) at the same time as the relations of power and knowledge within it destabilizes the possibility that this knowledge can be bounded into objects such as “protection against sexual abuse,” saying “no to drugs” and “managing conflict” with “parents and step-parents” (loveLife, 2012f, p. 162).   Consider, for example, the “problem” of the family as it operates in and through these moments of violence. A dominant reading of this “problem” within HIV and AIDS discourses is that relations of apartheid, and the transition to “modernity,” have caused “traditional” family structures to break down in South Africa (Barnett & Whiteside, 2002). Families were torn apart as men migrated from rural to urban areas to work in mines. This breakdown is extended today as HIV positive adults die and leave children behind (Walker et al., 2004). In the role-plays, these families are ones where parents respond with violence rather than talking to their children. Fathers take justice into their own hands, or are absent as mothers deal with unruly daughters. This reading of the “problem” of family supports an educational aim, as loveLife puts forward,                                                 45 See also Delius & Glaser (2002).  162 of improving “communication between young people and their parents” (loveLife, 2012h, p. 12). At the same time, this aim has already been exceeded by how these moments of family unrest are entangled in multiple (and non-linear) trajectories (Hunter, 2010). Colonialism, Christianity and capitalism came together in notions of the nuclear family that revolved around men’s work and men-led housing. Christianity, for example, discursively ties sex to procreation, and has reified the “proper” expression of sexuality as within the nuclear family. This religious discourse itself operates through, and in the service of, a tradition/modernity binary whereby Christianity is tethered to “modern” ideas of intimacy in opposition to “traditional” constructions of sexuality, particularly among black South Africans (Hunter, 2010; Jewkes and Morrell, 2010; Walker, Reid, Cornell, 2004). While these are often framed as competing discourses, discursive constructions of sexuality within Christianity is an enactment of how mutually constitutive, and power-laden, understanding of “modernity” and “tradition” are lived out in sexual subjectivities and their relations in the present political economy of unrest and unemployment (Hunter, 2010). In one role play, for example, a grieving father turned to religion, perhaps finding comfort in religious discourse as offering “order” in the face of violence and enacting a religiosity entangled with the individual and social effects of HIV in South Africa (Dilger et al., 2010). Yet, as he took up a gun and killed the boy who had killed his son, he also enacted the violence of loss: the emotional loss of his son as well as a social violence that is never an “individual act” (Jewkes & Morrell, 2010). This is knowledge in the making that works at the limits of self and Other.  These moments were all invited through a particular pedagogical approach. According to the loveLife facilitator’s guide, role plays are a facilitation “methodology” that can “help people stay interested in the session or activity, have fun, and help them learn easily” (loveLife, 2012d, p. 31). The educator, for example, might ask learners to create their own role-play based on  163 topics such as teenage pregnancy, peer-pressure or drug and alcohol use. However, the moments of excessive violence evoked above illustrate the impossibility of ever neatly defining any of these topics. The question, then, is not whether the pedagogical approach can “contain” the topic, but how the pedagogical approach meets the impossibility of containing knowledge. For instance, one of these moments of excessive violence was formed through a session with Grade 6 learners at a primary school that I attended with Gustav. The role-plays had been assigned as “home-work” and Gustav focused her feedback on taking your homework seriously or judging the role-plays as practiced or serious (Fieldnotes, April 2013). As she spoke, she drew on my body, and my discursive power as a white North American researcher participating in the session, to reprimand the learners for not following instructions: Andrée and I are disappointed. Here, being disappointed was a violent disciplining act as well as a violent closure of knowledge that had already exceeded the perceived boundaries of the topic. It was also in tension with how Gustav chose “violence” as a topic for another session, saying that it was an important issue to address with learners.  Engaging excessive moments in terms of how they might be a pedagogical resource for sexuality education is not about re-defining the topic (such as addressing gendered power relations in a discussion of teenage pregnancy). Re-defining a topic is a move that (impossibly) re-centers particular forms at the same time as it explains away excessive moments as a problem of learners being “off-topic:” learners didn’t understand the topic (showing family fights instead of telling the class how to prevent teenage pregnancy) or the aim of the activity (yelling at each other instead of demonstrating the communication skills they had been taught to use). My interest here is how these multiple expressions of violence, entangled in relations of self/Other/knowledge, always already exceed both the discursive boundaries of the topic as well  164 as the goal these topics carry with them, such as don’t do drugs or use a condom. Indeed, my contention is that these “goals” are inadequate if sexuality education is to be a potential site of social transformation. There is already violence operating within these “goals”: the violence of linear historical narratives that act upon gendered subjects and ignore how the “family,” including norms of parental communication, is produced through power relations (Hunter, 2010; Wilbraham, 2008); the violent assumption of particular gendered norms to discipline a hetero-normative family (Adams, 1997); a violent reduction of sexuality to the “problem” of teenage pregnancy or disease (Undie & Benaya, 2006). It is these violences, I suggest, that necessitate asking what might happen if sexuality education invited educators and learners to engage with the excess that is always already part of a “topic,” whether it is drug use, teenage pregnancy, or using a condom to prevent HIV. This is an invitation that prompts a re-imagining of the perceived pedagogical relation to knowledge.  Configuring risky bodies and their pedagogical relation to knowledge Amid a “resurgence of biomedical approaches to sexuality” (Vance, 1999, p. 47), the 1990s saw a rich range of public-health focused theoretical work challenging the “facts” of sexual subjects and their behaviours. One strand of critique, underpinned by social constructionist views of sexuality drew attention to the “inter-subjective nature of sexual meanings” (Gagnon & Parker, 1995, p. 11). A second theoretical lens, queer theory, was engaged as a direct response to stigmatizing accounts of homosexuality in early accounts of the HIV and AIDS epidemic in the United States (Davey & Hart, 2002; Weeks, 2010), and drew attention to the multiplicity and fluidity of sexual behaviours and identities and the implication this has for “risk” discourses that are coupled with particular identities (e.g. Jones, 1999). Thirdly, multiple strands of feminist theories were used to challenge the construction of gendered subjects within  165 HIV prevention discourses, particularly in relation to a gendered sexual “normality” (Patton, 1993b; Squire, 1993). Although these theoretical challenges have been driven by different underlying assumptions and concerns, they can be grouped together in how they have drawn attention to how the “facts” of HIV and AIDS discourse rely upon and reify particular subjects (such as women, gay men or youth) and their behaviours.   No matter how “complex” this subject is said to be, failing to interrogate the educational claims made in the name of HIV prevention has, I argue, ultimately blunted the effects of these critiques. As introduced in Chapter 1, there is a conceptual division between educational theories and public-health based critiques, a division that allows education to be held up as a “solution” to the dilemmas of knowing and being known within HIV prevention discourses. This plays out in how, almost 20 years after these critiques, the “risk” of youth, and their attendant need for already configured knowledge, drives sexuality education as an HIV prevention response. Problematizing these risky constructions necessitates explicitly engaging with how subjects, who are already enacting and potentially resisting multiple understandings of “Who am I” as discussed in Chapter 4, are also being produced through their perceived relation to knowledge within the pedagogical encounter. In other words, my interest here is to explore how the pedagogical approach configures subjects and their relation to knowledge. This can be explored by moving back into the middle of the loveLife training for Regional Programme Leaders.  The second day of the training for Regional Programme Leaders starts with a session called “Understanding loveLife.” The message is clear: school leaving is the time of greatest risk. Condom-use drops among school-leaving girls. They are experimenting with new things. (Fieldnotes, January 2013)  The first part of the explanation in this vignette, that school leaving is the time of greatest risk as seen in decreasing condoms use among girls, mobilizes loveLife’s conceptualization of gendered risk in relation to HIV prevention. According to loveLife’s 2007 Annual Monitoring Report, the  166 “sticking point” for decreasing HIV infection is “high rates of new infection among young women” which are the “effect of limited social and economic opportunity on school leavers” (2007 Annual Monitoring Report in Robbins, 2010, p. 245). For example, a young woman may have a transactional sexual relationship with an older man and, due to unequal gender and age related power relations, be unable to negotiate for condom use. Strategically, this “sticking point” underpins loveLife’s programmatic focus on “gender equality” (Robbins, 2010). At the same time, this conceptualization of gendered risk has mobilized an imagined pedagogical relationship between already gendered subjects and forms of knowledge. As a configuring pedagogical practice, explanation centers notions of risk around condom-use and, by extension, to sex without a condom. This centering draws coherence from, at the same time as it reifies, bio-medical discourses that have made risk “objective” by being able to pass judgment on which behaviours are considered “risky” (Lupton, 1995). Once a behaviour is considered risky, particular subject positions can be constructed as “at risk” and targeted with knowledge about that “risk” (Lupton, 1995). In this example, the subject position of school-leaving girls is made risky through implicit understandings of both “age” and “gender.” These girls are experimenting with new things, a discursive site of anxiety for the gendered youth sexual subject in which “experimentation” is collapsed with a transgressive gendered expression of sexuality (Carter, 2001). The “truth” of this subject also mobilizes the discursive construction of the transition from “childhood” to “adulthood” as a time of intensified risk in South Africa (Bhana, 2007). The very premise of the pedagogical approach (to target risky subjects with knowledge) not only relies upon these moves, but also potentially reifies what is already said to be known about risk, and risky subjects, within them. In turn, the issue is that the pedagogical may center an imaginary of linear information transmission at the expense of ways of making sense of the world that may not fit this imaginary.  167 This imaginary is not just about HIV prevention, but extends the “risk” of not taking up the right relation to knowledge into life itself, as seen as the session continued.  We are asked to brainstorm the risk factors for HIV infection among youth. The educator makes a list from the discussion: poverty, society, yourself, environment, life-transition. We divide the risks into individual, social and structural factors. (Fieldnotes, January 2013)  At first this list may seem excessively broad: all of life is a “risk.” At the same time, this is precisely the point. Following a reading of risk discourse as a regulatory form of bio-power, a term used by Foucault (1998) to refer to forms of power concerned with the lives of both individual and social bodies, every aspect of life can be targeted and managed for the health of the individual (Gastaldo, 1997; Lupton, 1995; Rose, 2007). Here, “sexual behavior has been colonized by the discourse of risk” and extended into all aspects of our lives (Lupton, 1995, p. 87). In this pedagogical encounter, the list itself operated as a configuring pedagogical practice that simultaneously reified risk as all encompassing at the same time as it divided those risks into discrete targets for knowledge: individual (e.g. yourself), social (e.g. society) and structural (e.g. poverty) factors.   An educational focus on risk factors is often understood as a shift away from individualistic theories of behaviour change that guided early HIV and AIDS prevention responses (UNAIDS, 2001). loveLife’s own conceptualization of risk has enacted this shift (Robbins, 2010): moving from a theoretical grounding in the AIDS Risk-Reduction Model (Catania, Degeles, & Coates, 1990) to a model of “risk tolerance.” Broadly, the AIDS Risk Reduction Model posits that there are three stages of individual behaviour that are key for reducing risk of HIV/AIDS infection: 1) behaviour labeling through accurate knowledge, 2) commitment to change and 3) taking action. While early evaluative research on loveLife claimed that there was a causal link between participating in loveLife programmes and decreased HIV  168 infection rates, as per this model (Pettifor, Rees, et al., 2005), it also highlighted “socio-economic conditions” in young people’s lives, including poverty, education levels, unemployment and gender power imbalances, and recommended that these factors be addressed in loveLife’s educational programmes (Robbins, 2010). This recommendation underpinned the organization’s paradigmatic shift to “risk-tolerance” in 2005/2006 (Robbins, 2010), where risk-tolerance broadly refers to how willing people are to take a risk because “having weighted up the choices, they decide that the risk is worth taking” (Harrison, 2010, p. 11). For example, a young woman may have knowledge about HIV prevention, but find herself in structural conditions of poverty and rationally decided to exchange (unprotected) sex for material goods. The aim of education is then to build a “perception of real and immediate possibility” (Robbins, 2010, p. 281) that will allow an individual to act based on a “strong sense of purpose, identity and belonging” (p. 284).   A discursive shift to risk tolerance does make the relation between subjects and forms of configured knowledge more complex. However, the “shift” fails to challenge the imagined relation between educator, learner and knowledge. In other words, the coherence of “risk-tolerance” remains reliant on a conceptualization of accurate knowledge that can be transmitted through the pedagogical encounter. What is being challenged here is how an individual perceives already configured knowledge, not the relationship between subjects and forms of knowledge within the pedagogical encounter itself. The effect of this argument is to draw attention to how the “shift” to “risk-tolerance” is not a shift away from relations of power that make particular configurations of knowledge, and their perceived subjects, coherent. For example, the “risk-factor” of life-transition is formed through a conceptualization of youth based on age (as discussed more fully in Chapter 4): experimentation, peer-pressure and out-of-control hormones  169 (Lesko, 2001). In South Africa, these constructions of youth within sexuality education discourses draw on western developmental theories at the same time as they are formed through “an intertwining of apartheid ideology, and historical and cultural practice” (Macleod, 2003, p. 421). This includes the “risk” that has been given to racialized, classed and gendered bodies in historical outbreaks of disease in South Africa (Carton, 2006). Here, the question is not how youth are tolerating risk, but how to disrupt a pedagogical approach that collapses the pedagogical relation to knowledge to power-laden forms of what is already said to be known and knowable.   Taking up this question in relation to the “risk-factor” of gender provides one way to explore some of its potential ramifications.  Gender is added to the list of risk factors. The educator facilitating the session explains: boys have more sex drive, boys are experimental, boys are influenced by the movies. The lead trainer intervenes to speak more about how risk is influenced by gender. Girls are at high risk – they are raped, they are in coercive relationships, they are out there to get money. Guys are influenced by stereotypes – boys don’t do certain things, boys don’t play with dolls. A participant adds, there is power that is given to boys, and then they abuse that power. Everyone seems to agree. The conversation ends there. (Fieldnotes, January 2013)   In this exchange, the pedagogical approach configures a particular relation between gendered subjects and messages related to gender equality. In this configuration, gender is coherent as a “risk-factor” because there are inequalities between boys and girls, with girls at high risk in their interpersonal relationships (rape, coercion) as well as from broader structural factors (poverty and the need to get money). This “risk-factor” reifies already power-laden understandings of who boys and girls are said to be (as argued in Chapter 3) and simultaneously frames power-relations within gendered identities as stereotypes. The discursive framing of these constructions as stereotypes makes them understandable as forms of knowledge (a mind-set) that can be targeted and changed through the pedagogical encounter. Here, “gender” is deployed not only as a known  170 subject position but also as a stable form of knowledge that can ultimately be transferred from educator to learner.   At the end of this particular exchange, a participant stated that there is power that is given to boys, and then they abuse that power. While this moment was seemingly coherent within the flow of the discussion, it can also be read as marking tensions within the relation to knowledge being configured through the pedagogical approach. For one, it is impossible to attend to the gendered abuse of power without also attending to how understandings of other social identities have shaped that abuse in South Africa in multiple relations across time and space. For example, the increase in sexual violence perpetrated by black men on black women in the 1980s (as a form of abusive control) is inseparable from the political, economic, and social disenfranchisement perpetuated on black bodies by the apartheid government (Hunter, 2010). The abuse of power also circulates in and through the figure of the African “big man,” a term popularly used to refer to political leaders (such as Robert Mugabe) who hold onto power through coercive means.46 Engaging these tensions within the pedagogical encounter necessitates exceeding the forms of knowledge being mobilized through the pedagogical approach.  Furthermore, the statement about power being abused disrupted the implicitly positive value given to power (as knowledge) in empowerment discourse in that it foregrounded a relation to knowledge that is always unknown. Within a conceptualization of knowledge in the making, this “unknown” is irreducible to knowledge not being taken up “correctly.” What is at stake is the impossibility of ever knowing how a particular form of knowledge will encounter, and be encountered by, educators and learners. Nyla (a Regional Programme Leader) invoked                                                 46 See for example, a recent media article focused on how Mandela’s presidency worked against this image:, accessed January 22, 2015.   171 the instability of this relation to knowledge when she defined empowerment as “being informed” and then acting on that knowledge, before adding: “it’s actually quite dangerous, also” (Interview, May 17, 2013). She explained this in terms of how, as an educator, “you might think that you’re empowering somebody for one reason” and “then they think that they have all of the power, themselves, and then abuse that power.” Nyla’s comments work within a certain imaginary of the pedagogical relation (where the educator empowers youth with knowledge) at the same time as they mobilize the relation to knowledge as an ongoing struggle. This struggle is entangled with, but irreducible to, the attachments an educator has to particular forms of knowledge (their reason) as well as understandings of self and Other that are formed through a learner’s encounter with that knowledge (what they think, and how they may act).  As we spoke in the interview, Nyla told me that her work as an educator had been “empowering” in her own life, but she did not articulate this empowerment in terms of neatly defined social identities or a linear relation to knowledge. Indeed, she resisted my question related to her own class and race status (as did other educators) because in answering the questions you “box yourself into being just that.” Within the pedagogical encounters discussed in this section, the boxes of identity (to fill up with knowledge) were already troublesome sites. However, the exigencies of explanation, and the need to keep the training moving and go onto the next activity, worked in the service of the boundaries of configured knowledge and the imagined relation of subjects to that knowledge. Indeed, throughout the training, knowledge as something to transmit constituted the terms of our perceived relation to knowledge as educators. As we reached the end of the training we all focused on one question: Will it work with groundBREAKERs? Perhaps this was the most powerful configuring pedagogical practice at play.   172 The question of will it work is only coherent when knowledge about risk is taken as an already configured knowledge. It is only coherent when the role of the educator is to transmit that knowledge, and the role of the learner is to take up that knowledge. Here, the most pressing “risk” becomes educational subjects who are out of control: learners who do not acquire the right information, educators who do not transmit that information to youth. Yet, this “risk” is only possible if educators and learners are already always exceeding forms of knowledge that have already been configured for them. This is a space where the question of “will it work” becomes its own undoing, and may in turn provoke ways of working with knowledge that seek to challenge the boundaries of what is already said to be known about educators, learners and the worlds they live in.  Excessive bodies   I heard about her in hushed tones and conversations behind closed doors in the loveLife offices: the pregnant groundBREAKER. The Regional Programme Leaders who worked with her fought for her, tried to support her, but the loveLife management had decided she was a problem. Indeed she was a problem that re-occurred through the year and the specifics of her situation (whether she was married, where she lived and worked, how long she had worked with the programme) did not seem to matter. What mattered was her body, a physical manifestation of unprotected sex. The message was clear: she could not be a role-model to young people, especially in schools, if she was pregnant. Once discovered, she was asked to resign – handing her loveLife manuals and branded clothing to the new groundBREAKER who would work in the school.   173 This was a troublesome encounter, besides the obvious issue that it was legally problematic in terms of the termination of an employment contract.47 Indeed, the existence of this practice outside of “official” policy is part of how particular kinds of knowledge act to discipline the youth body. In the context of a programme that seeks to reduce rates of teenage pregnancy, pregnancy is a physical manifestation of the consequences of sex that is visible only on the female body. Within normative discourses of teenage pregnancy, this points to a sexual knowledge that the female youth subject is not yet supposed to have, or that they do have but have not used responsibly (Macleod, 2003). At the same time, the focus on pregnancy as consequence erases power relations (such as physical or emotional violence) that may be part of the texture of that consequence. While loveLife’s response to the pregnant body may seem in tension with stressing that “sex is fun” and everyone having the right to decide when you “want to be a parent” (loveLife, 2013b, p. 44), the response is coherent with an underlying developmental view of adolescence positing that knowledge will lead to prevention (Macleod, 2003). Indeed, disciplining the pregnant body, removing it not only from view but also from the very loveLife identity, can be seen as a necessary condition for maintaining the presumed linear relation between educator, knowledge and a specific kind of gendered empowerment.  This vision of empowerment, exceeded by the pregnant body, is mobilized in and through a problematic relation with age. The disciplined groundBREAKERs had met multiple milestones to be understood as an “adult”: they have graduated from high school; many of them have worked or attended university before joining loveLife; they are all 18 years of age or older; they may be married. However, the pedagogical vision of gender empowerment can be seen to                                                 47 The groundBREAKER position is considered a “learnership” but groundBREAKERs do sign a year-long employment contract with loveLife and receive a small monthly stipend. To my knowledge no groundBREAKER took up legal action after the termination of their contract.   174 construct the educator body in a way that is in tension with the discursive and material privileges given to the “adult.” As a role-model for other youth, the educator’s body becomes a part of the “efficacious method” that works to limit the perceived pedagogical relation to knowledge about risk (Macleod, 2003, p. 434). Grounded in rationality, this efficacious (and linear) model takes the educator’s body as a channel: the imperative to transmit the “right” information necessitating a disciplined body. Furthermore, this relation, and the discursive positioning of the educator as not-yet-adult, is scripted within multiple and hierarchical power-relations within loveLife as an organization. groundBREAKERs were the children arriving at camp; the unruly bodies who required surveillance (and a curfew) because they might drink alcohol and have sex during the camp; and the risky educators who would need careful support and guidance as they worked in communities. Instead of an “expert” (or someone who knows), the educator within these discourses was positioned as an unruly educational subject. This was further reified by imperial formations of knowing and being known in the name of HIV prevention that require “proper” educational subjects: ones who have been empowered with particular forms of knowledge. Here, the educators’ identity became a pivotal part of the vision of “leaders for a new society” (Robbins, 2010, p. 244), carrying not the hopes of a nation, but also the dreams of the international donors and technical partners who support loveLife. The pregnant groundBREAKER is therefore a transgression of what “should” be known, her body the ultimate marker of unprotected sex and the failure of knowledge leading to prevention.  Across this reading, the pregnant body exceeds the imaginary of a linear relation to knowledge. The literal expulsion of these bodies from the loveLife acts to protect both what is knowable as well as the assumed relation of subjects to that knowledge. At the same time, the excess of the pregnant body marks her silencing as a constant presence. The excessive body  175 continues to work at the limits of what “risk” is and what happens when a subject does take up the right relation of knowledge to that “risk.” The expulsion of the pregnant body has already exceeded the safety that was supposed to be offered (to educators and to learners) through the pedagogical encounter. Configuring space and a pedagogical relation to social difference loveLife’s first prevention campaign, in 2000, was called “Talk About It.”48 As introduced in Chapter 2, the campaign consisted of a series of controversial billboards that worked within the aim of getting youth to “talk frankly about sex and sexuality” (David Harrison, Personal communication, July 15, 2013). Outside of debates as to the effects of this campaign (as discussed in Chapter 2), “talk about it” is a discursive centering on the educational potential of youth dialogue within a frame of critical pedagogy. The groundBREAKER programme, introduced the following year in 2001, took this potential from public billboards into a particular imaginary of intimate pedagogical spaces. These are spaces where the groundBREAKER, as educator, can say: “Put it on the table, you are protected, you are safe, in this space we are all valuable people and we all respect each other” (Scott Burnett, Personal communication, October 3, 2013). Read alongside the arguments made in Chapters 3 and 4, the configuration of any space as “safe” is already deeply problematic. If power-laden understandings of gender are being deployed through the pedagogical approach, as argued in Chapter 3, then the very space of the pedagogical encounter is already potentially “unsafe” for educators and learners who do not “fit” particular discursive norms of gender. Likewise, and as argued in Chapter 4, if articulations of                                                 48 See, accessed January 23, 2015, for one example of media coverage of this campaign.   176 who youth are seen to be are read as momentary effects of struggles over identity (rather than identity’s “truth”), then the imaginary of a space as safe risks foreclosing engagement with those struggles. This critique of the very possibility of the safety of space resonates with feminist critiques of critical pedagogy that have drawn attention to multiple relations of power and knowledge operating in all pedagogical spaces (e.g. Ellsworth, 1992). What is also at stake in this conceptualization, I argue, is how a particular relation to notions of social difference is deployed through the pedagogical configuration of space. For instance, in 2013, loveLife training events for both Regional Programme Leaders and groundBREAKERs included a series of activities on “creating safe spaces” (loveLife, 2012g, p. 35). The stated aim of these activities, as per the loveLife Master Training Manual, was to enable educators to “host a ‘Safe Space’ in which young people feel supported to share personal reflections and stories” (p. 35). Crucially, this was a non-discriminatory space in which youth feel a “deeper and profound connection to each other and our individual experiences” (p. 35). Coupled with the imperative to know personal identity, the focus on “connection” between known identities works within an empathetic orientation to difference. This orientation to difference, grounded in liberal conceptions of democracy, foregrounds understanding “known” Others who are different from the “known” self (Boler, 1999) and was evident in how loveLife educators articulated how social difference could be engaged within pedagogical spaces. Nkunzi, for example, suggested that negative gendered stereotypes could be addressed by leading a discussion and asking: if people are “saying these bad things about you, how do you feel?” (Interview, June 29, 2013). Nyla spoke about how she would encourage learners to “not discriminate [against] the person because of what you see, you don’t know the person itself” (Interview, May 17, 2013). Lukanji proposed a debate in which, by arguing both sides of an issue,  177 “one would actually see things the way the other one sees things” (Interview, August 12, 2013). Each of these proposals explicitly aims to address social inequalities that youth are facing. However, following Boler (1999), the risk of this approach is that it fails to challenge the very relations of social difference that are making being “known” as self or Other coherent within this space. Exploring this orientation to difference in terms of how loveLife’s pedagogical approach configures a space as safe is not only to argue that this orientation operates within a particular space. It also opens up consideration of what may be foreclosed when the configuration of space works in the service of an empathetic orientation to difference.  At the training, one of the groundBREAKERs has complained that he is uncomfortable with his rooming situation. Everyone seems to know who is making him uncomfortable: it is he/she. This rooming situation, and the person positioned as a he/she within it, is discussed at the evening debrief for educators. One of the trainers interrupts the conversation to ask if he/she is appropriate language to use, and someone explains that it was just a joke. The trainer begins to speak about the use of language and the sense of unease in the room grows. The educators respond: We were not being discriminatory. (Fieldnotes, January 2013)  The language of he/she used within this vignette works within a power-laden articulation of cisgender identity, in which a body’s perceived biological sex is normatively aligned with gendered identity (e.g. being assigned male at birth and subsequently identifying with masculinity). Gendered pronouns (he or she) are part of the intelligibility of normatively cisgender subjects and the choice of dominant or alternative pronouns (such as zie or hir) is a site of political struggle for transgender activists.49 Calling someone he/she is thus a violent mark of gendered identity as Other (to the normative cisgender) and a reification of the normative self.                                                  49 See a recent article in Time magazine for one discussion of language use in relation to trans activism:, accessed January 23, 2015.   178 In this particular encounter, the educators sought to explain away the language of he/she as a joke, and in doing so re-center the valued educator identity of not being discriminatory. This is one way of working with the tension between this encounter and the claim, made by Regional Programme Leaders during the training, that groundBREAKERs felt comfortable expressing their gendered and sexualized identities within the training space. At the same time, the claim by educators that they were not being discriminatory was supported through a particular pedagogical configuration of space. As a rooming situation, the “joke” of he/she was positioned as external to the space of the pedagogical encounter. Indeed, one educator stressed that they would never use this language with groundBREAKERs, and there was a palpable discomfort in the room when a discriminatory act was made explicit. My concern here is not only that this language may have entered the pedagogical encounter, but also that the very subject position offered to educators through the configuring pedagogical practices of space – the “good” educator who must create a safe space by not being discriminatory – foreclosed consideration of how educators are also implicated in constructions of self and Other within the pedagogical encounter. In other words, educators are also entangled in power-laden understandings of self and Other. When the role of the educator is uncoupled from transmitting the “right” message through space (such as “gender equality” or supporting “sexual diversity” – both of which are loveLife aims) the question becomes what kinds of relations may be risked within sexuality education’s spaces (Gilbert, 2007).  We are not being discriminatory is a powerful political claim. However, what I am suggesting is that it is inadequate for risking a relation between self and Other that works at the limits of what is already known about subjects and their relations of social difference. Beginning to take this risk, I argue, necessitates moving sexuality education into a space that is never safe: a  179 space that seeks to hold (no matter how momentarily) the vulnerability of all educational subjects and engage that vulnerability as a resource for re-imagining the responsibilities we have to one another.  Excessive vulnerabilities within space At the primary school there are two parking lots – one in front for visitors and a gated one for staff. The visitor lot is empty, while in the staff lot cars are packed in. As the educator drives into the lot, she asks if she should park in the visitor spot. No, says the guard, it is not safe. (Fieldnotes, March 2013)  I find notions of risk and safety in unexpected places in my field-notes. Risk became an incessant, if often silent, part of negotiating the spaces of the research: where to park, when to go, who to go with - The white female driving through the townships.  An educator asks me if I feel safe driving around. I say yes, and then start to worry that maybe I shouldn’t. (Fieldnotes, April 2013)  Risk solidified into the “ethnographic detail” of place: barbed wire, a broken window, empty streets.  The school grounds are tidy and the caretaker is busy gardening. He pauses to let a car drive in the gates. The fence is lined with barbed wire. (Fieldnotes, April 2013)   Safety was a promise behind the fence, the gaze of a security guard, or a securely locked car.  The entrance of the loveLife Y-Center is dark. There is a desk for the security guard, but he is not yet at work. (Fieldnotes, November 2012)  These moments are seemingly separate from the pedagogical encounter and engage “risk” and “safety” outside of its discursive boundaries. This is not the “risk” of an individual sexual act, or the “safety” of knowledge within a scripted lesson and space. These moments happen on the way between places, and they can be read as forming a textured context for a particular place. At the same time, as discussed in Chapter 2, these were places that I encountered through both thought and un-thought acts of research. The “risk” in the places that I visited (black and coloured  180 townships) is doubled by the “safety” of the place that I called home (in a white affluent suburb). My conceptions of risk were drawn through my own understandings of self and Other at the same time as they encountered the ways that loveLife staff read me: you should be safe going there, we should not take our cell phones into those communities. These relations of risk/safety are formative of the “localized contexts” of this research and my own conceptions of self with(in) its limits.   Excessive moments of “risk” and “safety” also trouble what “fits” in the spaces of the pedagogical encounter. In one moment, two educators and I are driving to a black township outside of Cape Town to pick up a police report from a groundBREAKER.  It is the end of the month, and a groundBREAKER was robbed as he withdrew cash from an ATM. The educators talk about the robbery: the man had been well-dressed, wearing a suit. They are concerned about what the groundBREAKER will do this month: he always sends money to his mother. The car slows to a crawl – the road is littered with debris. Municipal workers have been striking – they want more pay and more sick leave. We weave through burnt tires and pieces of wood. (Fieldnotes, April 2013)  Perhaps what is most striking about this brief moment is how unremarkable it could be. Street robberies are commonplace in South Africa, and increasing (Africa Check, 2014). There are “service delivery protests,” marked by escalating levels of violence, throughout the country (Hart, 2014). These dynamics are supposedly either external to the work of teaching and learning, or part of a “risk environment” that youth can overcome with the right mix of motivation and optimism about their own futures to tolerate risk. Yet, perhaps that is precisely what is remarkable in this moment. The power of “optimism,” celebrated by loveLife, is troubled by the lived texture of a burning tire and the economic inequalities it protests. The imperative for an educator to deliver a script of equality is haunted by his own experiences of dispossession from having his entire month’s salary stolen at an ATM. We drive through these streets as a way to get  181 to the pedagogical encounter, but the lived effects of “safety” and “risk” on those streets inevitably seep into the space of that encounter and implicate us all within them.  As we drive around the corner, the community that we are visiting that day stretches out in front of us, bounded by mountains and ocean. The educator, a black isiXhosa-speaking man, points out the different areas: poor coloured, rich white, poor black. He points out the houses where the rich white people live: they cost at least 1 million rand. I find myself thinking that is where I would live. My own house in another suburb costs twice as much as that. I don’t say anything. We continue driving into the black area. (Fieldnotes, November 2012)  In this moment, notions of race and class gave form to the landscape before us and I found my identities and forms of knowledge already entangled in this mapping of place. While we had not discussed my own class status, it was implicit from my position as a white foreigner who had the privilege to do research in South Africa. In response to his mapping, I did not say anything. I did not wish to speak the distance between our social identities, even as those identities played out in the landscape before us. I wished to keep the space of my research “safe,” even as who I was and what I claimed to know were already entangled within, and exceeded, the places that we travelled. On that day, the educator and I drove onto the black area – driving into the context where youth were at risk of HIV, and would be empowered with the “facts” in the “safe” space of sexuality education as an HIV prevention strategy.  The excessive configurations of knowing empowerment   About half way through this research project, I led a workshop with loveLife staff on the notion of “empowerment.” The workshop was a part of my research design, and I had proposed it in response to the suggestion from loveLife staff that my research could help them to “extract  182 the DNA” of the loveLife model (Meeting notes, November 13, 2011).50 At the beginning of the workshop I spoke about the different theoretical models that underpin dominant conceptions of empowerment within health education discourses (see Chapter 1). I highlighted how the love4Life script draws on the language of empowerment by discursively constructing power as an object and something that youth can “have” through “knowledge, attitudes and skills” and “loveLife messages” (loveLife, 2012f, p. 4). I then asked the educators to think about what power means to them and they worked in groups to create sculptures to represent ways of understanding power. While this activity was scripted within the loveLife imaginary of an interactive facilitator, as will be discussed further below, the words used by groups to discuss their sculptures also enacted power as “exercised from innumerable points, in the interplay of nonegalitarian and mobile relations” (Foucault, 1998, p. 94):  energy, influence, leading, taking charge, positive, negative, freedom of choice, freedom of spirit, age, position, respect, inside, to act, money, gender, relational, not fixed (Fieldnotes, April 2013)  Read together, these words enact some of the “intensity and texture” that makes relations of power “habitable and animate” (Stewart, 2007, p. 4). Here, forms of power (such as racism, sexism, or homophobia) operate as a “live texture in the composition of the subject” (Stewart, 2007, p. 107). They play out who is able to enact influence or taking charge. They are simultaneously positive and negative. They take gender as but one mark of this intensity.   The sculptures themselves were also a visual form of some of these relations. One sculpture, for example, was entitled the “struggle structure,” a naming located within the                                                 50 This phrase was used by senior loveLife staff in our meeting to refer to the theoretical frameworks guiding programme design as well as key concepts (such as “gender empowerment”) being used in the programme (Meeting notes, November 13, 2011).     183 discursive use of “struggle” to resist the racialized power relations of apartheid.51 At the same time, that struggle was not between discrete subjects (individuals), but was instead an inter-relation of life, fire and water (described as symbolizing that which is fluid and contextual). This works against western epistemologies that have tethered power to individual bodies and forms of knowledge at the expense of relational understandings of personhood and power, including within research (Mkhwanzi, 2005). Another sculpture, entitled the power tree, constructed a tree out of two interwoven threads representing the Regional Programme Leader and a groundBREAKER. Presenting the sculpture, the group described how, over the course of the year, both subjects will grow and change through their relationship within the loveLife programme. Here, power was entangled in the intimacy of the pedagogical relation: the mutually constitutive relations of change, the breaking apart (and coming together) of understandings of identities and forms of knowledge that is never complete.   The sculptures were also invited through a particular encounter inscribed within the relations of power in both research and pedagogical practice. As a research encounter, I had created the conditions for the workshop at the same time as my own theoretical understandings of “power” were in tension with the very idea of having a workshop on “empowerment” (in terms of something to know). As a pedagogical encounter, my enactment of “facilitator” was drawn through my past experiences of developing and working with sexuality education curricula as well as my training with loveLife and my desire to be a “good” educator across these multiple registers (seen for example in the “interactive” sculpture activity). These tensions are                                                 51 For example, Mandela (1995) uses the language of “struggle” throughout his autobiography to speak about the fight for freedom from social, political and economic oppression. See also a press statement released in 1961 by the African National Congress in which Mandela states that “the struggle is my life,”, accessed January 26, 2015.   184 irresolvable, an effect of the inevitable discursive centering of what can/should/must be known through both pedagogical and research practice, even if that centering is itself an impossible task. Indeed, I became ensnared in the problematic relations of empowerment discourse as the workshop continued. Following the sculpture activity, I asked the groups to, in light of their multiple understandings of power, write down on small pieces of paper what they thought youth need in order to be empowered (as loveLife’s stated aim). Overwhelmed by the volume of pieces of paper that soon covered the room’s wall, and wanting to make this “useful” for loveLife as per my conversations with them when designing the workshop, I started trying to get people to help me group the pieces of paper into categories of what youth “have,” should “improve” or “need.” The resistance to me and to the activity was overwhelming. On the one hand, senior staff members and partners became increasingly vocal in their complaints that things were too abstract, pointing to the inevitably reductive moves that occur when something becomes known as a programmatic target. The Regional Programme Leaders, on the other hand, stopped talking. This silencing was coherent within hierarchical power relations between head office staff (who design programmes) and regional staff (who implement programmes). Yet, it also marked a space of resistance to these relations. For instance, later in the day, one Regional Programme Leader angrily asked head office staff: how can you design programmes if you don’t know what is going on? This anger was also evident in a discussion I had with another educator after the workshop. The workshop had been useful, she told me, because head office staff needed to hear about realities on the ground (Fieldnotes, April 2013).  The distance between the “concept” and the “reality” can certainly appear overwhelming in a national HIV prevention programme. Orange, a senior staff member at loveLife, explained the “concept” to be that groundBREAKERs would work with small groups of learners in order to  185 let “different people share their stories” (Interview, November 18, 2013). The “reality,” for Orange, was a problem of implementation, such as having to work within a classroom environment or dealing with large class sizes. Without dismissing the day-to-day negotiation of these structural challenges, the tensions between the concept and the reality is also irreducible to them. The binary of concept/reality relies on the assumption that it is ultimately possible (even if difficult) to get the “right” concept that will empower youth within the complexities of reality. Within this binary logic, the anger expressed by educators over how head office staff don’t know what is going on could be resolved through knowledge: more complete knowledge of the “reality”, a new structure for training, an alternative model for implementation. This is precisely the direction currently being taken by loveLife. However, and alongside the pedagogical encounters engaged throughout this study, I also read the educators’ anger as an effect of the struggle over knowledge already happening within sexuality education. This is a struggle where “reality” is a deeply contested space. At stake in this struggle is not only what knowledge can/should/must be implemented, but how sexuality education can attend to the limits of any form of configured knowledge: excessive moments of the violent relations of knowing and being known, the bodies that are expelled from sexuality education’s “safe” space for their transgressions, the not so “safe” places within which we are all thrown together. This is a struggle where the concept will never be able to “fix” the reality.  Throughout this chapter, I have argued that loveLife’s pedagogical approach, located within sexuality education as an HIV prevention strategy, relies upon and redeploys particular forms of knowledge that come to be known as the “facts” to be taught. This is always a political project, inseparable from contested histories of who knows and what is known that have played out in HIV prevention discourses. Furthermore, these forms of knowledge are not external to the  186 pedagogical encounter, but rather animated through it. Configuring pedagogical practices (such as explanation, simplification and generalization) work in the service of knowledge as something already made as well as the presumed relations between educational subjects and knowledge. Configuring the “safety” of pedagogical space also shapes a perceived educational relation to social difference, a configuration that may foreclose the very possibility of risking different kinds of relations between self and Other.  In making these argument, I am suggesting that the “facts,” as forms of configured knowledge, are already a part of how we make sense of the world, and a necessary condition of our desire for understanding in the face of uncertainty. Sexuality education will never be able to get rid of the “facts,” or find some ultimately stable form of those “facts.” At the same time, I am also suggesting that it is a political necessity to work towards articulations of sexuality education that may resist relations of power and knowledge settling within those facts. As long as knowledge is conceptualized as something already made, the “failure” of knowledge settles on and through the bodies that bear its effects: the educator who is not “interactive” enough, the learner who does not “understand.” Shifting the pedagogical imaginary from knowledge as something already made to knowledge in the making is a crucial challenge to these relations of blame. When knowledge is engaged as an ongoing struggle within the pedagogical encounter, “failure” can be seen as an inevitable effect of the impossibility of developing a national campaign that aims to empower youth with the “facts” while simultaneously claiming that it will not tell young people what to do. Reducing this impossibility to the “failure” of educators and youth to transmit or acquire knowledge is, I argue, a deeply irresponsible act.     187 Chapter six: Re-encountering gender for a feminist politics with(in) sexuality education   This study began, as did my research interests, with gender as an “organizing principle and a rallying call” within international development discourses (Cornwall, Harrison, & Whitehead, 2004, p. 1). Within these discourses, gender has become a broad stand-in for a particular version of (cisgender and heterosexual) “women” (Smyth, 2007). Gender is quantifiable into a series of “fragmented topics,” such as gender roles and expectations, that can target and purportedly solve the “problem” of unequal gender power relations (Lewis, 2005). Despite feminist critiques of constructions of women within these discourses, especially in terms of essentialized identities and Othered (non-western) forms of knowledge (Mohanty, 2002; Parpart & Marchand, 1995; Undie & Benaya, 2006), an “oversimplified theory of gender” drives the coherence of gender and the resulting policy and programmatic support of initiatives to address gender, including within sexuality education as an HIV prevention strategy (Humphreys et al., 2008, p. 14). The encounters in this dissertation are thus also inextricably entangled in implications and contradictions. While I critique the discursive centering of gender, and its effects, that centering is foundational to gender as a research topic. I critique particular forms of gender as configured knowledge, even as those forms have given me the legitimacy (a white, North American female-identified “gender expert”) to conduct this research. Indeed, I started this research project trying to “find” gender. In my research proposal I gave analytic priority to the notion of gender, even as I gestured towards other social identities. I held meetings with loveLife staff to conceptualize the study, and was excited by how my initial research questions resonated with loveLife’s focus on the empowerment of young women as a central component of their 2012-2014 strategic plan (Grace Matlhape, Personal communication, November 2, 2011). I  188 introduced myself and my proposed project to the staff educators by speaking about my interest in gender and how it had come from my experiences teaching and developing sexuality education curricula. I’ve repeated this framing as I’ve written up this study. However, it has been precisely when I’ve lost gender (no matter how fleeting or fragmentary those moments are) that I’ve begun to think differently about it.  This is itself a contradictory move where “lost” and “found” inevitably re-settle into an analytic imaginary of the discovery of research. Losing gender can become its own kind of certainty, one that is reified within the discursive norms of a “conclusion” for a doctoral research study and tightened around what I can/should/want to know. At the same time, I argue that the struggles over knowing and being known explored within this study point toward the necessity of suggesting potential ways that sexuality education as an HIV strategy might be re-articulated. This, I argue, is different from suggesting ways for how sexuality education can/should/must be “done.” The “doing” of sexuality education is an effect of how power-laden understandings of self, Other and knowledge take form within its conceptual boundaries. For example, the interactive activity on “gender” discussed in Chapter 3, with its aims to challenge gender roles and expectations, was an inevitable repetition of power-laden understandings of gender identities and their constitutive relations. I do not think that this can be “fixed.” Indeed, my concern is that a myopic focus on “doing” sexuality education differently (in terms of revised curricula or more educator training) is itself a power-laden act that serves to mask the very relations that make that “doing” coherent.  I have sought to problematize the relations that settle into commonsense understandings of the “doing” of sexuality education by explicitly turning to its constitutive imaginaries: identity and knowledge. Guided by the opening propositions of this study, I have traced struggles over  189 identity and knowledge through particular pedagogical encounters with(in) sexuality education as an HIV prevention strategy. These struggles took place in a particular place, even as they exceeded linear spatial and temporal registers. I have also argued that these struggles are not a problem to “solve,” through either an empowered identity or a particular form of configured knowledge. Solving this struggle is impossible at the same time as the solution impossibly holds educators and youth responsible for its power-laden configurations. What is at stake is how sexuality education could be articulated in such a way that it might, no matter how momentarily, provide a site within which educators and learners mobilize these struggles over identities and forms of knowledge in ways that are more just for more people.  Concluding encounters with sexuality education  A political rationality of vulnerability   I am writing a draft of this dissertation when I come across a news article titled 6 ways to fix South Africa’s HIV response. Citing recent research that one-quarter of all new infections are in young women between the ages of 15 to 19 years, Dr. Fareed Abdullah, CEO of the South African National AIDS Council, warns: we’re seeing a more dramatic feminisation of the epidemic.52   The first opening proposition of this study was that sexuality education is always a political project. This proposition foregrounds how the urgency of responding to HIV infection rates among young women continues to drive sexuality education as an HIV prevention strategy at the same time the political nature of this strategy is often masked. Working with sexuality education as an always political project has drawn attention to relations of power and knowledge being deployed within it. For example, as argued in Chapter 3, the “problem” of gender supports a discursive reification of hetero-normative gendered identities (boys and girls) whose coherence                                                 52 This article is archived at, accessed November 25, 2014.   190 is formed through (rather than a means to neutrally challenge) unequal relations of power. Furthermore, as explored in Chapter 4, any identity as target (including gender) draws on multiple discursive constructions of who youth are said to be, and these constructions are the always unstable effects of relations of power stretching across time and place. Particularly troubling in the lived landscape of South Africa, this includes a national subject whose present coherence is simultaneously formed through the historical legacies of apartheid’s violent control of local subjects through racialized identities as well as the ongoing reification of those identities within the “potential” of imagined globalized futures. These subjects then encounter forms of knowledge, as seen in Chapter 5, that are particular configurations of what can/should/must be known about the world and its relations.  Challenging these effects necessitates turning to the political rationality giving form to any particular articulation of sexuality education. While political relations in South Africa cannot be reduced to neo-liberal ideologies (as discussed in Chapter 4), the effects glimpsed within the pedagogical encounters of this study point to a neo-liberal political rationality guiding sexuality education as an HIV prevention strategy. This is a rationality that, by collapsing moral responsibility with individual and rational action (Brown, 2003), allows youth to be blamed when they do not act to protect themselves from HIV. This is not only the abdication of responsibility by those who deploy sexuality education as an HIV prevention strategy. The argument that I have built through these chapters is that the action being promised to youth is already impossible: youth are expected to overcome the very understandings of social identities that make self, Other and their relations intelligible, and adopt forms of configured knowledge that can never “fit” the complexity of lived lives.   191   Holding sexuality education responsible to those lives necessitates proposing alternative political rationalities that may guide its encounters. One way forward is to engage the struggles over identities and knowledge already occurring within sexuality education. A neo-liberal political rationality is centered around a self who is separate from others. The pedagogical encounters engaged through this study, in contrast, point to the limits of self-knowledge. When understandings of self are understood as intimately entangled with understandings of Other, this co-constitutive relation becomes the condition for responsibility to one another (Butler, 2001). This shifts the very grounds of sexuality education from a site to know self and Other, as seen in the empathetic orientation to difference discussed in Chapter 5, to a site to where the inevitable limits of self-understanding hold the possibility for challenging relations where certain lives are said to count more than others. This is a political rationality that seeks to work at the limits of what makes lives livable, rather than telling youth what makes a livable life.  Social transformation and a pedagogical relation in the making   The new loveLife brand is being launched in the Western Cape: Powering the Future. Adults sit, listening to speeches about how a new community center is for youth and how youth will bring energy to the center and the community. The youth present, groundBREAKERs, stand at the side of the room wearing t-shirts Celebrating the Power of Youth. (Fieldnotes, July 2013)  The second opening proposition of this study, that sexuality education as a potential site for social transformation necessitates engaging with(in) the complexity of its pedagogical encounters, was engaged to resist a particular conceptualization of sexuality education as offering a “solution” to the dilemmas of social identities and their relations. In particular, I have engaged an understanding of pedagogy as a “performative act” (Ellsworth, 1997, p. 154) to resist the imaginary of a linear and rational relation between educator, learner and knowledge. By reading loveLife’s pedagogical approach through its encounters with educators and learners, I  192 have argued that the pedagogical approach is a constitutive part of what come to be known and knowable as social identities at the same time as it serves to mask entanglements between those identities. I have also argued that loveLife’s “interactive” pedagogical approach is implicated in how particular forms of knowledge, woven through relations of power, come to have the appearance of a neutral and stable form. These arguments problematize loveLife’s pedagogical approach, with its claim to be Celebrating the Power of Youth, as already drawing on power-laden constructions of who youth are and working towards particular versions of who they should be. Given the power-laden effects of racialized and classed identities in South Africa, drawn through the imperial legacies of apartheid, any vision of gender empowerment scripted within this approach may have already foreclosed the future it promises for many youth subjects.   At the same time, this reading of pedagogy resists reducing the pedagogical encounter to its approach. Reading pedagogical encounters through the concept of the emplaced weight of performativity has drawn attention to struggles over identity happening as the pedagogical approach encounters, and is encountered by, educators and learners with(in) place. For example, as explored in Chapter 4, understandings of place were part of the texture of who learners were said to be, at the same time as learners took up place to articulate identities that exceeded those constructions. I have also pointed to moments of excess (in Chapter 5) that work at the limits of any configuring pedagogical practices at play. These struggles and excessive moments cannot be re-scripted into the “right” approach. There is no “solution” to the dilemmas of knowing and being known within sexuality education as an HIV prevention strategy. There is no “magic bullet,” either in terms of what youth know or how educators should teach.   What I am proposing is that there may be different possibilities within sexuality education if it loosens its hold on a future known in advance. I recognize, in this proposal, the  193 tension in then arguing that the sexuality education is a potential site for social transformation (which inevitably carries with it political imaginaries of how the world should be). However, and building on a re-articulation of the political rationality of sexuality education discussed above, what this proposal foregrounds is a pedagogical relation to knowing and being known that is always in the making. In other words, it is precisely the impossibility of scripting who youth are and what they should know that may mobilize sexuality education’s scripts as ongoing provocations to consider what effects particular configurations of knowledge and identity have on people’s lives. Perhaps, within the lived relations of the pedagogical encounter, these configurations will begin to shift in ways that open to the future to more ways of living.    Productive impossibilities of empowerment discourse  I ask Nkunzi what empowerment means to him. It depends, he says: when you talk about empowerment, you’re empowering them to be what in what way? He reminds me: listen to me when I’m talking about speaking with them. (Interview, June 29, 2013)  The third opening proposition that I have worked with in this study is that empowerment discourse in sexuality education as an HIV prevention strategy risks shutting down, rather than opening up, the pedagogical encounter as a potential site for social transformation. My concern, and the argument that I have built through this dissertation, is that empowerment discourse is an impossible “solution” to the dilemmas of knowing and being known within sexuality education. As explored in Chapters 3 and 4, a pedagogical vision of empowerment is centered on identity as a target: a self to know and potentially change. When this vision marks gender as an identity to empower, it does so at the expense of a consideration of the co-constitutive and yet irreducible relation of gender to other social identities. Furthermore, and as explored in Chapter 5, empowerment discourse mobilizes particular forms of knowledge such as gender equity as the solution to relations of power and knowledge. The educational aim thus becomes a problematic  194 reification of the imaginary of a linear pedagogical relation at the same time as it holds its presumed educational subjects (known educators and learners) personally responsible for reaching that aim.  The question that Nkunzi asked me during the interview - when you talk about empowerment, you’re empowering them to be what in what way? – marked multiple relations of power and knowledge already at stake within empowerment discourse. This includes who is doing the empowering, who is being targeted, and what vision of knowing and being known is mobilized through it. Attending to this question, I argue, is not about its finding its answer. Instead, I am suggesting that the question itself could be a powerful pedagogical resource. In the space of the pedagogical encounter, the question could invite educators and youth to explicitly engage with who empowerment discourse, embedded in western epistemologies, takes them to be. The power in the question, I argue, is that it works at the boundaries of sexuality education as an HIV prevention strategy: it invites consideration of “global” formations of knowledge that shape the boundaries of livable lives at the same time as it attends to the “local” lives that encounter those forms of knowledge. Part of the texture of this question, in this instance, was Nkunzi’s reminder to me to listen to him when he talked about speaking with youth. He said this several times during the interview, and I take it both as part of his articulation of his pedagogical approach as well as a resistance to how the interview itself became a way to speak about youth. I return to this in terms of the limitations of this study below. What I wish to point to here is how speaking with youth is not a “solution” to either the “problem” of empowerment or the dilemmas of knowing and being known within sexuality education as an HIV prevention strategy. The pedagogical encounters of this study resist a simplistic slide from speaking with youth to identifying the “right” pedagogical  195 approach, such as facilitating a dialogue around the topic of gender. Instead, the challenge is to articulate sexuality education in such a way that can hold the vulnerability that speaking with entails. Speaking with is a moment when conceptions of self and Other are put at risk, including the selves of the educator. It is a moment when action may be invited by refusing to put what is already “known” about gender empowerment back together.  The contributions of this research project   There is a claim that continues to trouble me as I think about what it might mean to re-articulate sexuality education as an HIV prevention strategy. I was speaking with Grace Matlhape and our conversation in many ways worked within the complexities of trying to teach about gender within a national HIV prevention programme. For example, even as she centered notions of gender around measurable programmatic targets, such as increasing access to female condoms, she spoke about her understanding of gender empowerment in her own life as when “I no longer feel the fear” and said that “we are very, very far from being able to measure that with our target group ” (Personal communication, August 7, 2013). She also recognized forms of power that play out in educational spaces, such as the shift from the racialized divisions of apartheid to divisions “around class-lines” in a post-apartheid South Africa. However, speaking about loveLife’s work, she told me: “we are not in education.”  This claim to be “not in education” marks, at the same time as it reifies, the conceptual division between the fields of education and public health, a division that first led me to pursue my doctoral studies in education after years of working in public health. The claim is a constitutive part of what allows sexuality education to be deployed as the “solution” for the dilemmas of knowing and being known within HIV prevention. The risk of this deployment, I have argued, is that the educational claims folded into this “solution” carry with them particular  196 power-laden imaginaries of the pedagogical relation. These imaginaries are scripted into the curricula that loveLife prints, woven into the training it provides, and are a constitutive part of the approach that educators are expected to take up to implement the programme with youth. Claiming to empower youth (as loveLife does) but not taking empowerment as an educational site is, I argue, a violent closure of what might be imagined in and through sexuality education.   In this study, I have engaged sexuality education as an HIV prevention strategy through educational theories. As discussed in Chapter 5, there is a rich range of theoretical work that has challenged how sexual subjects and the relation to knowledge has been conceptualized within the bio-medical imperatives of HIV and AIDS discourses. However, I have argued that failing to engage these subjects and forms of knowledge within the pedagogical encounter risks reifying their normative constructions within HIV prevention discourses. These normative constructions are then deployed when research conceptualizes education as the “solution” for the dilemmas of knowing and being known. For example, research related to the effects of loveLife’s programmes has identified gender as a “social factor” influencing whether or not youth apply HIV prevention messages, and solidified this “factor” into programmatic recommendations related to addressing “gender inequalities” (e.g. Pettifor, Rees, et al., 2005) or specifically targeting information to gendered audiences (e.g. Seutlwadi, Peltzer, & Mchunu, 2012). Even in critiques of loveLife’s “mono-cultural” approach (Parker, 2006), there remains an underlying assumption that there are particular educational subjects (produced through their cultures) that can be appropriately targeted with HIV prevention messages.  To explicitly engage with the claims made (either explicitly or implicitly) in the name of education, I have asked what it might mean to think about pedagogy in relation to identity as a site of struggle (as explored in Chapter 4) and knowledge in the making (as explored in Chapter  197 5). I have developed the concept of the emplaced weight of performativity as a way to engage place as a constitutive site through which understandings of self and Other take form at the same time as they exceed the perceived boundaries of any particular place. This is not a solution to the dilemmas of identity, but rather a lens through which to engage with the lived weight of the effects of power relations without collapsing identity into those effects. Those effects include, at the same time as they are irreducible to, the relations of power and knowledge that my body marked with(in) the pedagogical encounters of this research project. Furthermore, I have mobilized knowledge as entangled within those effects by exploring what I have called configuring pedagogical practices. Here, I have further disrupted the imaginary of pedagogy as information transmission by working with how pedagogy is a constitutive part of what becomes known and knowable. Beyond de-stabilizing the imaginary of information transmission that underpins the coherence of sexuality education as an HIV prevention strategy, these concepts contribute to an analytic space within which sexuality education as an HIV prevention strategy might be re-articulated (as discussed in the previous section).  Working with the pedagogical encounter also marks a contribution to troubling the coherence of gender as a universal concept to deploy within international development discourses. I have argued that the pedagogical encounter disrupts any premise of a universal notion of gender, and is an enactment of how gender “can only be charged and enacted in the sticky materiality of practical encounters” (Tsing, 2005, p. 1). At the same time, these notions of gender are also irreducible to any particular encounter. This is a crucial resistance to the global/local binary (and its constitutive relations of power and knowledge) that settles in and through gender as something to address through sexuality education as an HIV prevention strategy. There is no “global” problem any more than there is a “local” subject to target through  198 education. Instead, and as discussed in the previous section, the pedagogical encounter is a site within which the relations that make these claims coherent may potentially be problematized. This is the site where educators and learners may encounter, and potentially resist, imperial imperatives of sexuality education that continue to seek to tell them who they are and what they can/should/must know.  Research limitations and the limitations of research   Meanings made within the pedagogical encounters of this study are further problematized by a number of limitations within the acts of research. As pointed to above, one of these is the lack of youth perspectives on the encounters themselves. As discussed in the methodology section, this is partly a limit of design (I did not propose to interview learners) and a problematic effect of what is conceptualized as the research design and what happens during the research (I anticipated interviewing younger loveLife peer educators but was not able to do so). However, I also see it as problematic to dismiss this tension as a logistical problem.  Part of my own journey in this study has been working with how my own habits of thought are inextricable from the western epistemological frameworks that I also aim to critique. In conceptualising this study, I implicitly drew on an imaginary of information transmission when I chose to center the interviews around educators rather than learners. While I can justify this in terms of access and working within the organizational structure of loveLife, this decision reified multiple relations of power and knowledge that work to speak for youth. This was enacted, for example, when the interviews became a space for speaking about what empowerment meant to or for youth (even as educators also resisted this focus, such as when Nkunzi said to listen to him about speaking with youth). It is also part of the problematic texture of the vignettes woven around impersonal learners. Who these learners saw themselves as is part of the “loose ends and  199 missing links” (Massey, 2005, p. 9) entangled in this research’s pedagogical encounters. These loose ends and missing links are part of the always-problematic practices of representation in research. In the previous chapters I have worked with a theoretical conception of the pedagogical encounter that resists identity as a target or knowledge as a transmissible object. Putting youth perspectives into this encounter would not have solved the problem of who youth are or what forms of knowledge they are said to need (even as it does not excuse their absence). Indeed, it was only by problematizing identity as a stable target and knowledge as a transmissible object that I was able to begin to realize how my own research design had been implicated in the discursive constructions of those identities and forms of knowledge.  Relations of implication point to a second limitation of this research: who was I to be engaged in this encounter? As explored throughout this dissertation, my identities (white, North American, cisgender) are implicated in the relations of power and knowledge that have taken “gender” as something to be taught through sexuality education as an HIV prevention strategy. Indeed, my academic training in gender theories is a constitutive part of how the notion of gender has been deployed within international development discourses: the empowered self in relation to the Other non-western Woman (Parpart & Marchand, 1995); the gender “expert” who speaks the truth of gender power (Parpart, 1995); the classed and racialized identities that settle into the historical desires of my heterosexual privilege (Stoler, 1995). These powerful relations of privilege settled in and through the English language that I encountered this study through. These identities, and the attachments to knowledge they are entangled in, are always a limiting act. This limit is not only what I could understand (through language, for example) but also the limits in terms of the forms of knowledge that my own ways of knowing and being known had already foreclosed.   200 I have engaged the multiple textures of this limitation by starting from the view that the problematic relations of who I am could never be resolved in the pedagogical encounter by getting rid of my body or being able to speak in “their” language. Following Massey (2005), the “very acknowledgement of our constitutive interrelatedness implies a spatiality; and that in turn implies that the nature of that spatiality should be a crucial avenue of enquiry and political engagement” (p. 189). Part of that spatiality in the context of sexuality education as an HIV prevention strategy has been how white North American experts have told South African educators and youth what they should know and how they should be known (Lees, 2008). As such, part of my task in this study has been to explore the traces of “constitutive interrelatedness” within relations of power and knowledge that my body momentarily made visible: the white lady whose gaze disciplined black youth; the moments of discomfort when my desire for knowledge (such as knowable contexts or finding gender) collided with its own limiting configurations; the moments when my silences were implicated in the discursive practices of places and the identities constituted within them. Refusing to “solve” this limitation is itself a political statement that there is not a “global” truth of sexuality education at the same time as it is impossible to hold forward a “local” place (with “local” subjects) as an educational solution to what it means to know and be known with(in) sexuality education as an HIV prevention strategy. This kind of work is already happening everyday – the question is how we might all think within it to potentially articulate its relations of knowing and being known so that they are more just for more people.  Thinking with(in) sexuality education as an HIV prevention strategy invites a different relation to research (as discussed in terms of the limitations of my methodological approach in Chapter 2). This research is not generalizable, something that is considered a limitation in the  201 imperatives for evidence-based research (Lather, 2007). The articulations of sexuality education that I have gestured towards cannot be scripted into a curriculum (for learners) or mapped into a more efficient training module (for educators). However, I take this as an invitation of this research. By interrogating what education already promises, I have sought to work within a different promise of education (Britzman, 2009): sexuality education as a site where educators, learners and researchers may speak with one another, risking relations to one another that are more just across lived spatial and temporal moments.  The question of future research directions   The pedagogical encounters of this study have also prompted my curiosity about where those encounters could go. As discussed in the previous section, my research is underpinned by a commitment to interdisciplinary work as a condition for being responsible to the educational claims made for and about youth in a world with HIV. While I have focused on particular pedagogical encounters in the Western Cape region of South Africa, the relations of these encounters already gesture to multiple spatialized relations. These relations could be followed into globalized pedagogical spaces (such as international youth forums) as well as into other localized “places.” For example, in Canada sexuality education is positioned as a “right” of all youth (Public Health Agency of Canada, 2008). At the same time, youth who are discursively constructed as most at risk of HIV infection, street-based sex workers,53 will encounter, and be encountered by, sexuality education within multiple relations of power and knowledge. My                                                 53 There is a range of research related to the vulnerability of sex workers, particularly in terms of structural determinates of HIV risk, available through the Gender and Sexual Health Initiative of the BC Centre for Excellence in HIV/AIDS:, accessed January 27, 2015.  202 ongoing interest is how “risk” may come together with struggles over identities and forms of knowledge with(in) “risky” places.   Methodologically, “place” could be further engaged as a problematic site within ethnographic “methods,” and a potential site for momentarily resisting the inevitable analytic closure that these enact. As discussed in Chapter 4, place took me by surprise as I worked with and through the data of this study. Starting from place as a key site through which the entanglements of identity are articulated, I am interested in how spatialized methodologies (such as mapping) might provide another way of working with contested understandings of identities across space and time. This includes trying to disrupt how imaginaries of “risk” settle in and through research designs and, in turn, working to invite more loose ends and missing links that may further problematize relations of knowing and being known with(in) sexuality education.  A return to the invitation   At the heart of this study is a belief that thinking differently about gender is a necessary political act in light of the power relations that circulate in and through notions of gender as sexuality education is promoted as an HIV prevention strategy around the world. In many ways this is not “new.” Responses to HIV have a long history of being a site for a feminist politics committed to social transformation. However, what I have argued throughout this dissertation is that this politics (and its always present implication in western knowledge systems) must begin from the unsettling position of trying to lose what gender “is.” Perhaps this is an impossible task, but that impossibility is precisely what I argue makes it so necessary. It is an invitation to think about “gender” within the threads of the stories of selves of this research project, and in doing so invite configurations of knowing and being known that may exceed available conceptual boundaries. These are the limits with(in) which we are all trying to find ways of living.   203 References  Adams, M. L. (1997). The trouble with normal: Postwar youth and the making of heterosexuality. Toronto, Canada: University of Toronto Press. Africa Check. (2014). Factsheet South Africa: Official crime statistics for 2012/2013. Retrieved August 3, 2014, from Aggleton, P., & Crewe, M. (2005). Effects and effectiveness in sex and relationships education. Sex Education, 5(4), 303-306. doi: 10.1080/14681810500278030 Alaimo, S., & Hekman, S. (2008). Introduction: Emerging models of materiality in feminist theory. In S. Alaimo & S. Hekman (Eds.), Material feminisms (pp. 1-19). Bloomington, IN: Indiana Univerisity Press. Alcoff, L. (1988). Cultural feminism versus post-structuralism: The identity crisis in feminist theory. Signs, 13(3), 405-436.  Allen, L. (2006). Trying not to think 'straight': Conducting focus groups with lesbian and gay youth. International Journal of Qualitative Studies in Education, 19(2), 163-176. doi: 10.1080/09518390600576012 Allen, L. (2008). Poles apart? Gender differences in proposals for sexuality education content. Gender & Education, 20(5), 435-450. doi: 10.1080/09540250701829995 Altman, D. (1995). Political sexualities: Meanings and identities in the time of AIDS. In D. Altman, R. G. Parker, & J. H. Gagnon (Eds.), Conceiving sexuality: Approaches to sex research in a post modern world (pp. 97-108). New York: Routledge. Apple, M. W. (2004). Ideology and curriculum (Third ed.). New York, NY: RoutledgeFalmer. Arnfred, S. (2004). Re-thinking sexualities in Africa: Introduction. In S. Arnfred (Ed.), Re-thinking sexualities in Africa (pp. 7-29). Uppsala, Sweden: Nordiska Afrikainstitutet. Barnett, T., & Whiteside, A. (2002). AIDS in the twenty-first century: Disease and globalization. New York: Palgrave MacMillan. Bateman, C. (2007). The loveLife-saving 'ripple effect': groundBREAKERs seed hopeful attitudes. South African Medical Journal, 97(3), 162-166.  Bateson, G., & Mead, M. (1942). Balinese character: A photographic analysis. New York, NY: The New York Academy of Sciences. Baxen, J. (2010). Performative praxis: Teacher identity and teaching in the context of HIV/AIDS. New York: Peter Lang. Baxen, J., & Breidlid, A. (2004). Researching HIV/AIDS and education in sub-Saharan Africa: Examining the gaps and challenges. Journal of Education, 34, 9-29.   204 Baylies, C. (2000). Perspectives on gender and AIDS in Africa. In C. Baylies & J. Bujra (Eds.), AIDS, sexuality and gender in Africa: Collective strategies in Tanzania and Zambia (pp. 1-24). London: Routledge. Baylies, C. (2004). Cultural hazards facing young people in the era of HIV/AIDS: Specificity and change. In N. Poku & A. Whiteside (Eds.), The political economy of AIDS in Africa (pp. 71-83). Surrey, England: Ashgate Publishing Ltd. Baylies, C., & Bujra, J. (1995). Discourses of power and empowerment in the fight against HIV/AIDS in Africa. In P. Aggleton, P. Davies, & G. Hart (Eds.), AIDS: Safety, sexuality and risk (pp. 194-222). London, England: Taylor & Francis. Behar, R. (1995). Introduction: Out of exile. In R. Behar & D. Gordon (Eds.), Women writing culture (pp. 1-32). Berkeley, CA: University of California Press. Bermudes, C. C., & Da Cruz, R. (2004). From policy to practice: The anthropology of condom use. In K. D. Kauffman & D. L. Lindauer (Eds.), AIDS and South Africa: The social expression of a pandemic (pp. 136-159). Basingstoke: Palgrave Macmillan Ltd. Bhana, D. (2007). Childhood sexuality and rights in the context of HIV/AIDS. Culture, Health & Sexuality, 9(3), 309-324. doi: 10.1080/13691050601116686 Bhana, D. (2008). Discourses of childhood innocence in primary school HIV/AIDS education. African Journal of AIDS Research, 7(1), 149-158. doi: 10.2989/AJAR.2008.  Bhana, D., & Anderson, B. (2013). Desire and constraint in the construction of South African teenage women's sexualities. Sexualities, 16(5/6), 548-564. doi: 10.1177/1363460713487366 Bhana, D., & Pattman, R. (2009). Researching South African youth, gender and sexuality within the context of HIV/AIDS. Development, 52(1), 68-74. doi: 10.1057/dev.2008.75 Bhekisisa Team (Producer). (2014, August 3, 2014). ARV-shortages uncovered: Bhekisisa director speaks on VoW FM. Retrieved January 25, 2015 from Bobbitt, F. (2004). Scientific method in curriculum-making. In D. J. Flinders & S. J. Thornton (Eds.), The curriculum studies reader (Second ed., pp. 9-16). New York: Routledge. Boler, M. (1999). Feeling power: Emotions and education. New York: Routledge. Boler, T., & Aggleton, P. (2005). Life skills education for HIV prevention: A critical analysis. London: Save the Children and ActionAid International. Boyce, P., Huang Soo Lee, M., Jenkins, C., Mohamed, S., Overs, C., Paiva, V., . . . Aggleton, P. (2007). Putting sexuality (back) into HIV/AIDS: Issues, theory and practice. Global Public Health, 2(1), 1-34. doi: 10.1080/17441690600899362  205 Bragg, S. (2007). 'Student voice' and governmentality: The production of enterprising subjects. Discourse: Studies in the Cultural Politics of Education, 28(3), 343-358. doi: 10.1080/01596300701458905 Breidlid, A. (2009). HIV/AIDS, cultural constraints and educational intervention strategies. In J. Baxen & A. Breidlid (Eds.), HIV/AIDS in Sub-Saharan Africa (pp. 21-32). New York: United Nations University Press. Britzman, D. P. (1997). The tangles of implication. Qualitative Studies in Education, 10(1), 31-37.  Britzman, D. P. (1998). Lost subjects, contested objects. Albany: State University of New York Press. Britzman, D. P. (2000). 'The question of belief': Writing poststructural ethnography. In E. A. St. Pierre & W. S. Pillow (Eds.), Working the ruins: Feminist poststructural theory and methods in education (pp. 27-40). New York, NY: Routledge. Britzman, D. P. (2009). The very thought of education: Psychoanalysis and the impossible professions. Albany, NY: State University of New York Press. Brown, J., Sorrell, J., & Raffaelli, M. (2005). An exploratory study of constructions of masculinity, sexuality and HIV/AIDS in Namibia, Southern Africa. Culture, Health & Sexuality, 7(6), 585-598. doi: 10.1080/13691050500250198 Brown, W. (2003). Neo-liberalism and the end of liberal democracy. Theory & Event, 7(1).  doi:10.1353/tae.2003.0020 Butler, J. (1990). Gender trouble: Feminism and the subversion of identity. New York: Routledge, Chapman and Hall, Inc. Butler, J. (1992). Contingent foundations: Feminism and the question of 'post modernism'. In J. Butler & J. W. Scott (Eds.), Feminists theorize the political (pp. 3-21). New York: Routledge. Butler, J. (1997a). Excitable speech: A politics of the performative. New York, NY: Routledge. Butler, J. (1997b). The psychic life of power: Theories in subjection. Stanford, CA: Stanford University Press. Butler, J. (2001). Giving an account of oneself. diacritics, 31(4), 22-40.  Butler, J. (2004). Undoing gender. New York, NY: Routledge. Cáceres, C. F. (2000). Afterword - The production of knowledge on sexuality in the AIDS era: Some issues, opportunities and challenges. In R. Parker, R. M. Barbosa, & P. Aggleton (Eds.), Framing the sexual subject: The politics of gender, sexuality and power (pp. 241-259). Berkeley: University of California Press.  206 Campbell, C., & MacPhail, C. (2002). Peer education, gender and the development of critical consciousness: Participatory HIV prevention by South African youth. Social Science & Medicine, 55, 331-345.  Carter, J. B. (2001). Birds, bees, and venereal disease: Toward an intellectual history of sex education. Journal of the History of Sexuality, 10(2), 213-249. doi: 10.1353/sex.2001.0022 Carton, B. (2006). 'We are made quiet by this annihilation': Historicizing concepts of bodily pollution and dangerous sexuality in South Africa. International Journal of African Historical Studies, 39(1), 85-106.  Catania, J. A., Degeles, S. M., & Coates, T. J. (1990). Towards an understanding of risk behaviour: An AIDS risk reduction model (ARRM). Health Education Quarterly, 17(1), 53-72.  Centre for Education Policy Development. (2009). Challenges facing education in South Africa. Johannesburg, South Africa: Centre for Education Policy Development. Chilisa, B. (2005). Educational research within postcolonial Africa: A critique of HIV/AIDS research in Botswana. International Journal of Qualitative Studies in Education, 18(6), 659-684. doi: 10.1080/09518390500298170 Chilisa, B., & Ntseane, G. (2010). Resisting dominant discourses: Implications of indigenous, African feminist theory and methods for gender and education research. Gender and Education, 22(6), 617 - 632. doi: 10.1080/09540253.2010.519578 Chisholm, L. (2003). Gender equity and curriculum 2005. Paper presented at the University of London Institute of Education Seminar: Beyond Access: Curriculum for Gender Equality and Quality Basic Education in Schools, London.  Choudry, A., & Kapoor, D. (2013). NGOization: Complicity, contradictions and prospects. In A. Choudry & D. Kapoor (Eds.), NGOization: Complicity, contradictions and prospects (pp. 1-23). London: Zed Books Ltd. Clair, R. P. (2003). Expressions of Ethnography: Novel approaches to qualitative methods. Albany, N.Y.: State University of New York Press. Clifford, J. (1986). Introduction: Partial truths. In J. Clifford & G. E. Marcus (Eds.), Writing culture: The poetics and politics of ethnography (pp. 1-26). Berkeley, CA: University of California Press. Collins, P. H. (2000). Black feminist thought:  Knowledge, consciousness and the politics of empowerment (10th anniversary ed.). New York, NY: Unwin Hyman. Connell, R. (1987). Gender and power. Stanford, CT: Stanford University Press.  207 Cornwall, A. (2007). Buzzwords and fuzzwords: Desconstructing development discourse. Development in Practice, 17(4), 471-484. doi: 10.1080/09614520701469302 Cornwall, A., Harrison, E., & Whitehead, A. (2004). Introduction: Repositioning feminisms in gender and development. IDS Bulletin, 35(4), 1-10.  Cornwall, A., Harrison, E., & Whitehead, A. (2007). Introduction - Feminisms in development: Contradictions, contestations and challenges. In A. Cornwall, E. Harrison, & A. Whitehead (Eds.), Feminisms in development: Contradictions, contestations & challenges (pp. 1-17). London: Zed Books. Crone, E. T., Gibbs, A., & Willan, S. (2011). From talk to action: Review of women, girls and gender equality in NSPs in Southern and Eastern Africa. Durban: HEARD. Davey, B., & Hart, G. (2002). HIV and AIDS. In B. Davey & C. Seale (Eds.), Experiencing and explaining disease (pp. 86-136). Philidelphia, PA: The Open University. Davies, B., Browne, J., Gannon, S., Honan, E., Laws, C., Mueller-Rockstroh, B., & Petersen, E. B. (2004). The ambivalent practices of reflexivity. Qualitative Inquiry, 10(3), 360 - 389. doi: 10.1177/1077800403257638 Delius, P., & Glaser, C. (2002). Sexual socialisation in South Africa: A historical perspective. African Studies, 61(1), 27-54. doi: 10.1080/00020180220140064 Department of Basic Education. (2011). Curriculum and Assessment Policy Statement: Life Orientation (Further Education and Training, Grades 10-12). Pretoria, South Africa: Department of Basic Education. Department of Basic Education. (2013). Draft integrated strategy on HIV, STIs and TB 2012-2016 (summary version).  Pretoria, South Africa: Department of Basic Education. Retrieved January 25, 2015 from Department of Basic Education. (n.d.). Schedule: National guidelines on school uniforms.  Pretoria, South Africa: Department of Basic Education. Retrieved January 25, 2015 from Department of Education. (2002). Protecting the right to innocence: The importance of sexuality education.  Report of the 'Protecting the right to innocence: Conference on sexualtiy education (19-21 August 2011)'. Pretoria, South Africa: Department of Education. Diko, N. (2012). The gender politics of the school uniform. In R. Moletsane, C. Mitchell, & A. Smith (Eds.), Was it something I wore? Dress, Identity, Materiality (pp. 208-224). Cape Town, South Africa: HSRC Press. Retrieved January 25, 2015 from   208 Dilger, H., Burchardt, M., & van Dijk, R. (2010). Introduction - The redemptive moment: HIV treatments and the production of new religious spaces. African Journal of AIDS Research, 9(4), 373-383. doi: 10.2989/16085906.2010.545643 Dwyer, S. C., & Buckle, J. L. (2009). The space between: On being an insider-outsider in qualitative research. International Journal of Qualitative Methods, 8(1), 54-63.  Eisenhart, M. (2001). Educational ethnography past, present and future: Ideas to think with. Educational Researcher, 30(8), 16-27. doi: 10.3102/0013189X030008016 Ellsworth, E. (1992). Why Doesn't this Feel Empowering? Working Through the Repressive Myths of Critical Pedagogy. In C. Luke & J. Gore (Eds.), Feminisms and Critical Pedagogy (pp. 90-119). New York, NY: Routledge. Ellsworth, E. (1997). Teaching Positions: Difference, Pedagogy and the Power of Address. New York, NY: Teachers College Press. Ellsworth, E. (2005). Places of learning: Media, architecture, pedagogy. New York, NY: Routledge. Emerson, R. M., Fretz, R. I., & Shaw, L. L. (1995). Writing ethnographic fieldnotes. Chicago, IL: The University of Chicago Press. Epstein, H. (2003). AIDS in South Africa: The invisible cure. The New York Review of Books, July 17, 2003.  Epstein, H. (2007). The invisible cure: Africa, the west and the fight against AIDS. London, England: Penguin Books. Erel, U., Haritaworn, J., Rodriguez, E. G., & Klesse, C. (2010). On the depoliticisation of intersectionality talk: Conceptualising multiple oppressions in critical sexuality studies. In Y. Taylor, S. Hines, & M. Casey (Eds.), Theorizing intersectionality and sexualtiy (pp. 56-77). New York, NY: Palgrave Macmillan. Etkin, E. (2013). Using narrative communication in a mass-produced youth magazine as an HIV prevention intervention. Journal of African Media Studies, 5(1), 3-17. doi: 10.1386/jams.5.1.3_1 FHI 360. (2013). It about more than just sex. Durham, NC: Interagency Youth Working Group. Fields, J. (2008). Risky lessons: Sex education and social inequality. New Brunswick, N.J.: Rutgers University Press. Fine, M. (1988). Sexuality, schooling and adolesent females: The missing discourse of desire. Harvard Educational Review, 58(1), 29-53.   209 Fine, M. (2006). Bearing witness: Methods for researching oppression and resistance - A textbook for critical research. Social Justice Research, 19(1), 83-108. doi: 10.1007/s11211-006-0001-0 Foley, D., & Valenzulela, A. (2005). Critical ethnography: The politics of collaboration. In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE Handbook of Qualitative Research (3rd ed., pp. 217-234). Thousand Oaks, CA: SAGE Publications. Foucault, M. (1982). The subject and power. Critical Inquiry, 8(4), 777-795.  Foucault, M. (1995). Discipline and punish: The birth of the prison (A. Sheridan, Trans.). New York, NY: Random House, Inc. Foucault, M. (1998). The will to knowledge; The history of sexuality: 1 (R. Hurley, Trans.). London, England: Penguin Books. Foucault, M. (2003). The politics of health in the eighteenth century. In P. Rabinow & N. Rose (Eds.), The essential Foucault: Selections from essential works of Foucault, 1954-1984 (pp. 338-350). New York, NY: The New Press. Francis, D. (2010). Sexuality education in South Africa: Three essential questions. International Journal of Educational Development, 30, 314-319. doi: 10.1016/j.ijedudev.2009.12.003 Freire, P. (2007). Pedagogy of the oppressed (M. B. Ramos, Trans. 30th Anniversary Edition ed.). New York, NY: Continuum. Gacoin, A. (2010a). Sexuality, gendered identities and exclusion: The deployment of proper (hetero)sexuality within an HIV-prevention text from South Africa. Culture, Health and Sexuality, 12(4), 429 - 444. doi: 10.1080/13691051003602596 Gacoin, A. (2010b). Youth voice and HIV prevention: Discursive opportunities, limitations and productive impossibilities. Discourse: Studies in the Cultural Politics of Education, 31(2), 165-178. doi: 10.1080/01596301003679685 Gacoin, A. (2014). Teaching empowerment? Power and agency in sexuality education discourses in the context of HIV prevention programmes in South Africa. South African Review of Sociology, 45(3), 53-73. doi: 10.1080/21528586.2014.945944 Gagnon, J. H., & Parker, R. G. (1995). Conceiving sexuality. In J. H. Gagnon & R. G. Parker (Eds.), Conceiving sexuality: Approaches to sex research in a post-modern world (pp. 3-18). New York: Routledge. Gastaldo, D. (1997). Is health education good for you? Re-thinking health education through the concept of bio-power. In A. Petersen & R. Bunton (Eds.), Foucault, health and medicine (pp. 113-133). London, England: Routledge.  210 Gay, J., Hardee, K., Croce-Galis, M., Kowalski, S., Gutari, C., Wingfield, C., . . . Berzin, K. (2010). What works for women and girls: Evidence for HIV/AIDS interventions. New York, NY: Open Society Institute. Geertz, C. (2000). The interpretation of culture. New York, NY: Basic Books. Gibbs, A., Campbell, C., Maimane, S., & Nair, Y. (2010). Mismatches between youth aspirations and participatory HIV/AIDS programmes in South Africa. African Journal of AIDS Research, 9(2), 153-163. doi: 10.2989/16085906.2010.517482 Gilbert, J. (2007). Risking a relation: Sex education and adolescent development. Sex Education, 7(1), 47-61. doi: 10.1080/14681810601134736 Gilbert, L., & Selikow, T.-A. (2011). 'The epidemic in this country has the face of a woman': Gender and HIV/AIDS in South Africa. African Journal of AIDS Research, 10(Sup1), 325-334. doi: 10.2989/16085906.2011.637732 Giroux, H. A., & McLaren, P. (1986). Teacher education and the politics of engagement: The case for democratic schooling. Harvard Educational Review, 56(3), 213 - 238.  Glaser, C. (2000). Bo-tsotsi : the youth gangs of Soweto, 1935-1976. Portsmouth, NH: Heinemann. Glaser, C. (2005). Managing the sexuality of urban youth: Johannesburg, 1920s - 1960s. International Journal of African Historical Studies, 38(2), 301-327.  Gore, J. (1992). What we can do for you!  What can 'we' do for 'you'?  Struggling over empowerment in critical and feminist pedagogy. In C. Luke & J. Gore (Eds.), Feminisms and critical pedagogy (pp. 54-73). New York, NY: Routledge. Gore, J. (1998). Disciplining bodies: On the continuity of power relations in pedagogy. In T. P. Popkewitz & M. Brennan (Eds.), Foucault's challenge: Discourse, knowledge, and power in education (pp. 231-251). New York: Teachers College Press. Guba, E. G., & Lincoln, Y. S. (2004). Competing paradigms in qualitative research: Theories and issues. In S. N. Hesse-Biber & P. Leavy (Eds.), Approaches to qualitative research: A reader on theory and practice (pp. 17-38). New York: Oxford University Press. Gupta, G. R., Parkhurst, J. O., Ogden, J. A., Aggleton, P., & Mahal, A. (2008). Structural approaches to HIV prevention. Lancet, 372, 764-775. doi: 10.1016/S0140- 6736(08)60887-9 Hall, S. (1992). Cultural studies and its theoretical legacies. In L. Grossberg, C. Nelson, & P. Treichler (Eds.), Cultural Studies (pp. 277-285). New York: Routledge. Hammersley, M., & Atkinson, P. (2007). Ethnography: Principles in practice. London, England: Routledge.  211 Hammett, D., & Steaeheli, L. (2013). Transition and the education of the new South African citizen. Comparative Education Review, 57(2), 309-331.  Haraway, D. (1991). Simians, cyborgs and women: The reinvention of nature. New York, NY: Routledge. Hardy, S. (2011). Anal sex: Phallic and other meanings. In S. Seidman, N. Fischer, & C. Meeks (Eds.), Introducing the new sexuality studies (Second ed., pp. 106-112). New York, NY: Routledge. Harrison, D. (2010). The use of tropes: Beyond belief, it's experience that counts.  Hart, G. (2014). Rethinking the South African Crisis: Nationalism, populism, hegemony. Athens: The University of Georgia Press. Hatch, J. A. (2002). Doing qualitative research in education settings. Albany, NY: State University of New York Press. Hawkes, S., & Buse, K. (2013). Gender and global health: Evidence, policy, and inconvenient truths. Lancet, 381(May 18, 2013), 1783-1787.  Helleve, A., Flisher, A. J., Onya, H., Mukoma, W., & Klepp, K.-I. (2009). South African teachers' reflections on the impact of culture on their teaching of sexuality and HIV/AIDS. Culture, Health and Sexuality, 11(2), 189 - 204.  Hoeken, H., Swanepoel, P., Saal, E., & Jansen, C. (2009). Using message form to stimulate conversations: The case of tropes. Communication Theory, 19, 49-65.  Holland, J., Ramazanoglu, C., Scott, S., & Thomson, R. (1994). Desire, risk and control: The body as a site of contestation. In L. Doyal, J. Naidoo, & T. Wilton (Eds.), AIDS: Setting a feminist agenda. London: Taylor and Francis. Holland, J., Ramazanoglu, C., Sharpe, S., & Thomson, R. (1998). The male in the head: Young people, heterosexuality and power. London: Tufnell Press. Holstein, J. A., & Gubrium, J. F. (2004). The active interview. In D. Silverman (Ed.), Qualitative research: Theory, method and practice (2nd ed., pp. 140-161). Thousand Oaks, CA: Sage. hooks, b. (1994). Teaching to transgress: Education as the practice of freedom. New York: Routledge. Human Sciences Research Council. (2014). South African National HIV, Behaviour and Health Study 2012. Pretoria, South Africa: Human Sciences Research Council. Humphreys, S., Undie, C.-C., & Dunne, M. (2008). Gender, sexuality and development:  Key issues in education and society in sub-Saharan Africa. In M. Dunne (Ed.), Gender, sexuality and development: Education and society in sub-Saharan Africa (pp. 7-38). Rotterdam: Sense Publishers.  212 Hunter, M. (2010). Love in the time of AIDS. Bloomington, IN: Indiana University Press. Interagency Youth Working Group. (2014). About. Retrieved August 3, 2014, from Jansen, C., & Janssen, I. (2010). Talk about it: The effects of cryptic HIV/AIDS billboards. Communicatio: South African Journal of Communication Theory and Research, 31(1), 130-141. doi: 10.1080/02500160903525072 Jansen, J. D. (1998). Curriculum reform in South Africa: A critical analysis of outcomes-based education. Cambridge Journal of Education, 28(3), 321-331. doi: 10.1080/0305764980280305 Jeeves, A. (2001). Introduction: Histories of reproductive health and the control of sexually transmitted disease in Southern Africa: A century of controversy. Southern African Historical Journal, 45, 1-10. doi: 10.1080/02582470108671399 Jeeves, A., & Jolly, R. (2009). Sexually transmitted infections and