‘Culturespeak’ is everywhere: an analysis of culturalist narratives in approaches to sexuality education in Mozambique

ABSTRACT

Building on postcolonial feminist scholars and critical anthropological work, this paper analyses the frequent deployment of the notion of ‘culture’ by decision-makers, educators, international agency staff and young people in the design, delivery and uptake of sexuality and HIV prevention education in Mozambique. The paper presents qualitative data gathered in Maputo, Mozambique to highlight the essentialising nature of culturalist assumptions underpinning in-school sexuality education. I argue that conceptions of ‘traditional’ and ‘modern’ culture are deployed to explain the epidemic, both of which spectacularise and decontextualise phenomena and practices, and perpetuate the western trope of the Third World Woman. The paper concludes by arguing that a singular emphasis on ‘culture’ – in its various guises – diverts attention from structural causes of young Mozambican women and men’s vulnerability to HIV and AIDS, and crucially, rather than problematise gender relationships, reifies and solidifies these. Thus, while sexuality and HIV prevention education cannot be understood or delivered independently of the cultural context in which it is situated, a more nuanced conception of culture is required – that is, one that is attentive to questions of power and specifically, who is in a position to make meanings ‘stick’.

KEYWORDS:

• Culture
• gender
• coloniality
• HIV and AIDS
• sexuality education

Differently positioned actors […] draw on, re-work and stretch in new directions the accumulated meanings of ‘culture’ […]. In a process of claiming power and authority, all are trying to assert different definitions which will have different material outcomes. (Wright 1998, 10)

Introduction

‘Culture’, Grillo (2003) asserts, ‘has run astray’ (157). The abundant use of the notion of ‘culture’ in debates concerning, among other issues, immigration, young people and identity has been problematised by numerous scholars (Behar 1996; Wright 1998; Grillo 2003; Stolcke 2003; Sangari 2008; Anderson-Levitt 2012; Piedalue 2016). In the field of Comparative Education (CE) too, notions of culture, identity and nationhood have long constituted central themes (Klerides 2009). Of particular interest here are efforts within CE to examine the role of culture in the design, delivery and uptake of education, and developments in the field as a result of postcolonial critique. For example, whereas early CE scholarship departed from the premise that education systems and curricula merely reflected pre-existing cultural and national identities, the integral role of schools in processes of ‘narrating’ the nation and inducting young people into ‘doing’ certain cultural identities is now widely acknowledged (Tikly 1999; Klerides 2009).

Crucial too has been the growing recognition of the role of colonial and postcolonial systems of education in disseminating western forms of (cultural) knowledge (Tikly 1999; Alexander and Mohanty 2010). Salient in this regard is scholarly work on the feasibility and desirability of global blueprints for school effectiveness, and the travel and ‘creolisation’ of, and resistance to, global educational policies and pedagogies (Fuller and Clarke 1994; Anderson-Levitt 2003; Crossley and Tikly 2004; Arnove 2009; Chisholm and Leyendecker 2009; Mattheou 2009; Elliot 2014; Tromp 2018).

This paper seeks to contribute to these wide-ranging debates in CE by examining the ways in which the notion of culture itself is deployed in sexuality education policy and practice in the context of Mozambique. In so doing, I respond to the work of critical anthropologists Grillo (2003) and Wright (1998), and, specifically, their argument that the pervasive use of ‘culture’ in scholarly literature, and policy-maker and decision-maker discourse is a cultural phenomenon in and of itself, and should be examined as such.

Background: sexuality education, culture and gender

One of the fields in which the notion of culture has gained particular traction is that of health promotion, and – of specific interest here – sexuality and HIV prevention education.¹ The rationale for attending to ‘culture’ is here held as more or less self-evident. As, for example, Kreuter et al. (2002) state: ‘[i]t is a truism of health education that […] interventions will be more effective when they are culturally appropriate for the populations they serve’ (133). Or, as Airhihenbuwa and DeWitt Webster (2004, 5) argue, ‘[c]ulture has been shown to have both positive and negative influences on health behaviours’, whereby they go on to state that ‘[i]ndeed, culture is often shown to be a factor in the various ways that HIV/AIDS has impacted on the African population’ (emphasis added). From the late 1990s onwards, UN agencies have increasingly paid attention to the cultural dynamics of HIV and AIDS. In a 2001 publication, the United Nations Educational, Scientific and Cultural Organisation (UNESCO), for example, argued for grounding sexuality education on ‘mentalities, traditions, beliefs and value systems […] in so far as they may enhance needed changes, or hamper them, if they are not correctly identified, and will necessarily interfere in the action taken’ (UNESCO 2001, 9).

International development agencies have often used the notion of culture – and particularly ‘traditional’ culture – to explain gender inequality (e.g. UNESCO 2012, 2001; UNFPA 2012). In a similar vein, culture is frequently used to explain the gendered nature of the HIV epidemic – an ever-increasing body of research, policy documents and training manuals detailing the ‘cultural traditions’ and ‘cultural barriers’, and/or ‘cultural problems’ that render women vulnerable to the spread and impact of HIV (Save the Children 2002; Lindgren, Rankin, and Rankin 2005; Safaids 2008; Gender Links 2009; UNAIDS 2014; Avert 2017).

Yet despite the frequent deployment of ‘culture’ in explaining gender inequality and the epidemic, and the broad consensus that HIV prevention education needs to be culturally ‘appropriate’, there is little research that complicates the notion of culture in and of itself. An important exception is Anderson-Levitt (2012), who interrogates how hegemonic power operates in processes of cultural meaning-making and dissemination around the globe. Research by Parkes et al. (2016), provides another example of a more nuanced account of culture, complicating the gendered dichotomy of tradition-modernity.

The lens offered by these authors allows for an analysis of power dynamics underpinning recurrent deployments of the notion of ‘culture’. These approaches are considered in the case study presented here, of various actors involved in the design, delivery and uptake of sexuality education in Maputo, Mozambique. Drawing on data gathered in Maputo between 2010 and 2011, this paper seeks to untangle the hybrid and, on the face of things, contradictory articulations of ‘modern’ and ‘traditional’ ‘culture’, and the seemingly ubiquitous anxiety about culture that appeared to inform the design and delivery of sexuality education in this context. In so doing, the paper asserts that culturalist approaches to sexuality education initiatives fail to address gendered dynamics of intimate and sexual relations between young people in Mozambique, and instead may serve to further entrench existing gender inequalities that drive the epidemic. Crucially, this paper engages with one of the central questions Anderson-Levitt (2012) raises as to how hegemony operates in contestations over cultural meanings, in this case, explanations of gender inequality and the prevention of HIV.

Conceptual premises

Understanding gender and the role of the school

The growing recognition of the centrality of gender to the ways in which the intimate realm ‘relates to the global and everything in between’ has resulted in an expansive corpus of gender scholarship, providing an increasingly nuanced understanding of the gendering of our lived lives, experiences, health and wellbeing (Sangari 2005; Campbell et al. 2009; Parkes et al. 2013; Alexeyeff and Besnier 2014, 1). In what follows, I discuss key feminist scholars who jointly offer an analytical prism to analyse gender and gender scholarship as sites of power.

Poststructuralist thinkers – most notably Judith Butler – have been crucial in shedding light on the ‘performative’ nature of gender (Butler 1999). Rather than an a priori category that is tied to our ‘natural’ sex, Butler conceptualises gender as being brought into being and constituted by the ways in which we conduct ourselves in everyday interactions and settings (Connell 2011; Parkes et al. 2013). Butler’s work has been crucial to much current gender scholarship in that it illuminates that gender is an effect of power rather than an immutable fact. In this conception gender is thus regarded as learned, enacted and policed, whilst also offering the possibility of contestation and transgression.

Raewyn Connell’s relational conception of gender offers another critical set of analytical tools. Connell (2011) contends that gender needs to be understood as a process that involves numerous actors and categories, linking bodies, institutions and sites. Relations between and among men and women are regarded as constituted by, and constitutive of, large-scale patterns that can be found within and across institutions and spaces, including gendered division of labour and notions of appropriate expressions of masculinity versus femininity (Connell 2011). Similarly emphasising the learned nature of gender, Connell asserts that these large-scale gendered patterns or structures concerning (appropriate) ‘femininities’ or ‘masculinities’ are learned in a ‘series of encounters with the constraints and possibilities of the existing gender order’ that are embedded in relationships, law, economic structures, regulations and semiotic codes (2009, 101).

Intersectionality theorists have drawn attention to the critical in ways in which gender interacts with, and can be altered by, other social categories, including race, class, ability and sexuality (Brah and Phoenix 2004; Yuval-Davis 2006). Postcolonial feminist scholarship has been critical in drawing attention to the exclusionary tendencies of universalist assumptions concerning gender oppression and equality within western-centric (second-wave) feminism (Mani 1987; Mohanty 1988, 2003; Spivak 1988; Chowdury 2009). While postcolonial feminism is characterised by a rich array of social and political positions, common concerns relate to colonial underpinnings of understandings and responses to various forms of gendered oppression and discrimination in postcolonial contexts. Of particular importance are postcolonial feminist analyses of contemporary developmental efforts, and specifically their grounding in the western trope of what Mohanty (1988) termed the ‘Third World Woman’, conceived as uniformly uneducated, backward and oppressed (see also Syed and Ali 2011; Roy 2017).

Much research highlighting the ways in which schools function as critical sites for social reproduction, including along gendered lines (Arnot 2006; Valentin 2011; Humphreys 2013; Miedema and Millei 2015). Scholars have detailed the pivotal role served by schools in colonial enterprises, and the colonial vestiges of current education systems and curricula in countries such as Mozambique (Crossley and Tikly 2004; Vavrus 2006; Unterhalter 2009; Meneses 2012). Education, as Connell (2010, 609) notes ‘was not a colonial afterthought’, but instead a crucial element of what Mudimbe (1994) refers to as the ‘colonising apparatus’ that re-fashioned economies, cultures and power, and constructed appropriate colonial (female and male) subject identities. The notion of enculturation acknowledges the pivotal role state supervised schools serve in gendering young people and turning them into citizens, directing our attention to political struggles concerning the knowledge and values that are produced in, and promoted through, schools and classrooms.

Conceptualising ‘culture’

Drawing on Grillo (2003) and Wright’s (1998) stocktaking of the ways in which the notion of ‘culture’ is deployed by policy-makers, the media, development agencies as well as anthropological literature, this paper draws a distinction between three broad conceptions of ‘culture’. The first ‘old’ yet tenacious interpretation of culture – found also in early historical scholarship in the field of CE – is founded on the conception of the world as made of different ‘peoples’ who are defined by a distinct way of life (Grillo 2003; Klerides 2009). In this conception, people are seen to ‘have’ culture, and ‘culture’ is conceived in terms of spatially bounded entities, a particular set of characteristics, with an integrated system of shared meanings, which, furthermore, reproduces itself despite socio-economic or political shifts. A culture and ‘its’ people are, by this definition, static and knowable (Laakso and Olukoshi 1996; Gupta and Ferguson 1997; Taylor 2007). In this view, young people are ‘assimilated’ into a culture through forms of education and socialisation (Tikly 1999).

Conceptual developments in feminist, postcolonial and poststructural anthropology led to the emergence of a second account of culture, in which ‘cultures’ are not perceived as (ever having been) unitary entities, but instead as a site of perpetual struggle, negotiation and contestation. Conceived as an ongoing struggle for hegemony rather than a ‘benign’ system of meaning-making, this understanding of culture gives rise to a different set of questions, the most crucial of which concerns who is in a position to determine ways of thinking and meaning, including within the realms of educational policy-making and pedagogy (Tikly 1999; Crossley and Tikly 2004). Questions regarding what constitutes a ‘tradition’ or the ‘modern’ are then irrelevant, except in analyses of who has power to make meanings ‘stick’, and how these meanings become naturalised: appearing to provide a coherent, consensual account and thereby becoming authoritative (Wright 1998; Grillo 2003; Halualani, Mendoza, and Drzewiecka 2009).

The third interpretation of culture draws on features of both the ‘old’ and ‘new’ accounts presented above. Similar to new accounts, ‘culture’ is seen as a system of meaning that is forever unsettled and ambiguous. At the same time, and it is here the third interpretation resembles the old understanding, the process of constructing culture is perceived as long and difficult, resulting in a system of meaning that is not easily dislodged. Thus, while human beings are understood as active participants in meaning-making processes rather than passive heirs of a particular system of meaning, they are simultaneously thought to be ‘culturally embedded’, neither determined by, nor able to fully transcend culture (Parekh 2000; cited in Grillo 2003, 161).

The understanding of ‘culture’ that is loose on the streets – including those in the field of international development – by and large continues to be what Grillo (2003) refers to as an ‘old school’ culturalist interpretation, providing an essentialist account of ‘culture’ as defining the identity of individuals and collectivities (158; see also Piedalue 2016). It is this essentialist understanding of culture that is seen to underpin fears with regard to cultural mixing and hybridity, and the loss of cultural coherence and identity as a result of ‘neoliberal globalisation’, transnationalism and migration (Giddens 1990; Snel and Stok 2008).

Postcolonial feminist analysis is critical to analyses of the ‘use and misuse’ of the notion of culture in the field of international development. An important example is the work of Kumkum Sangari (2005, 2008, 2012), who deploys the term ‘culturalism’ to describe a logic that ‘homogenise[s] and pathologise[s] […] non-Western places and peoples’. and attributes phenomena such as gendered discrimination, violence and oppression to ‘“timeless” traditions and anti-modern sentimentality’ (Piedalue 2016, 3). Similar to scholars such as Mohanty (2003, 1988), Chowdury (2009) and Spivak (1988), Sangari’s analysis draws attention to (tenacious) Western conceptions of ‘the third world woman’ and concerns regarding her presumed plight. This critique resonates with postcolonial approaches in the field of comparative education, which have challenged the racialised, gendered and sexualised underpinnings of school curricula and their implication in the maintenance of colonial and neo-colonial power (Crossley and Tikly 2004).

This paper thus conceptualises gender, but equally race and class, as brought into being, as learned, policed and – at times – contested, including in schools and classrooms. Crucially, gender and other social categories are conceived as effects of power rather than unassailable facts. In a similar fashion, culture is regarded as a site of struggle, a critical question being who is in a position to determine what counts as (traditional and modern) culture. Postcolonial feminist scholarship constitutes the overarching prism for the present analysis of perceptions of, and responses to, different forms of gendered oppression and the perceived role of culture therein by different sets of actors, including international aid agencies and national government officials. In particular, I attend to underlying postcolonial feminist concerns with regard to generating research that challenges simplified western-centric understandings of gender inequalities, and the role these play in the spread and impact of the HIV epidemic.

Methodology

This paper, with its focus on problematising the deployment of ideas about gender and culture, extends the work of Miedema, Maxwell, and Aggleton (2011) that was geared to creating greater conceptual clarity in approaches to school-based HIV prevention education. The authors drew a distinction between initiatives that build on notions of science, rights and/or conservative moral values. The tripartite framework was further refined by identifying sub-categories, highlighting the diversity of programmes grouped in one particular cluster. For example, programmes informed by notions of ‘rights’ include those that build on what was defined as formal accountability interpretation of rights and more informal approaches that draw on language of ‘empowerment’ and ‘inclusion’ (Miedema, Maxwell, and Aggleton 2014). It is important to note that in the development of the tripartite framework, emphasis was placed on what was perceived to be the primary focus of a particular sexuality education initiative. While the framework presents three approaches as distinct, in practice, approaches often overlap and different initiatives may draw on, and be reflective of, more than one approach.

The findings presented here emerge from data gathered over a period of seven months between 2010 and 2011 in Maputo, Mozambique’s capital city. Prior to, and at the time of data collection, UNESCO sought to promote a ‘cultural approach’ to HIV and AIDS (UNESCO 2001; Nielse 2011). The UNESCO Maputo office similarly applied this approach, but here the prefix ‘socio’ was added to more explicitly acknowledge the social dimensions of cultural practices, norms and values that were seen to drive the epidemic (Bukali de Graça 2002). UNESCO Maputo started its work on the socio-cultural approach to HIV and AIDS approximately a year before I began work there as junior programme officer (early 2002). Shortly after my arrival, I was appointed the UNESCO Maputo focal point for HIV and AIDS. The title of ‘focal point’ masks the ambivalent role I had within the organisation, that is, while I was accorded considerable responsibility to promote the agency’s approach in external forums, I had less say in defining the actual agenda.

My work involved close collaboration with the Working Group on HIV and AIDS of the Ministry of Education. The working group was composed of representatives of various ministerial departments and a number of bilateral and multilateral agencies, including UNESCO, the Danish Development Cooperation, the United Nations Children’s Fund (UNICEF), the United Nations Population Fund (UNFPA) and, on occasion, the World Bank. Within the framework of the working group, these agencies sought to advance their respective approaches to address the epidemic, ranging from emphasising ‘life skills’, human rights to addressing the socio-cultural dimensions of HIV. Broadly speaking, these approaches seemed to be conceived – or at least presented – as mutually exclusive.

Returning to Mozambique to do my doctoral research and my increasing engagement with postcolonial feminist scholarship during, but particularly following the completion of my doctorate, made for occasionally uncomfortable confrontations with preconceptions I held regarding, among other issues, ‘development’ and my part in UNESCO’s efforts to promote ‘the socio-cultural approach’. Returning to Mozambique as a doctoral student, untethered from, but intimately knowledgeable about, international development circles also offered space to better understand the complex terrain many policy-makers and educators navigated – complying with foreign donors’ demands while striving to pursue a ‘Mozambican’ agenda for education and development. Whilst this paper critically examines narratives concerning gender, culture and development of those taking part in the study, I am cognizant of my own implication in the questions I raise with regard to the coloniality of development and constructions of gender relations (Heron 2007).

I conducted semi-structured interviews with a gender-balanced sample² of policy-makers from the Ministry of Education and Culture (MoEC), staff working for international agencies, educators and young school-going people (15–17 years, Grade nine) from three public secondary schools. Nine educators were recruited on the basis of their involvement in the delivery of HIV prevention education to Grade nine pupils, and eight policy-makers and nine international agency staff participated. These participants were all responsible for coordinating the educational response to HIV and AIDS within their particular departments, directorates or organisations.³

I analysed key MoEC and governmental documents, including those that, in principle, informed the design and delivery of school-based HIV prevention education at the time of data collection. The analysis of documents gathered and transcripts of interviews and FGDs were conducted following a systematic and iterative approach. At first, data were summarised, emerging themes were highlighted, and patterns and relations (between themes and actors, for example) were identified. A thematic analysis was then undertaken to identify sub-themes and refine the analysis of connections across, and patterns in, the data. For example, participants’ discussion of the aims and means of (sexuality) education yielded further categories such as ‘maintaining national culture’, ‘liberation’ and ‘progress’. Separate overviews of thematic clusters were created for the different sets of actors and documents, allowing for a comparison of perspectives concerning, for example, the aims of education within and between actor groups. It was during this process that the pervasive use of the notion of ‘culture’ became apparent.⁴

Contextual backdrop

The borders of Mozambique were determined by the 1891 Anglo-Portuguese agreement, which established Portuguese colonial rule (Newitt 1995). Independence from Portugal was gained in 1975, after almost 10 years of armed struggle led by Frelimo (Mozambican Liberation Front, Frente de Libertação de Moçambique) (Newitt 1995). Following independence, the (one-party) People’s Republic of Mozambique was proclaimed, Frelimo declaring itself as Marxist-Leninist (Casimiro 2004; Sabaratnam 2011). However, in 1977, war broke out between Frelimo and the armed resistance movement Renamo (Resistência Nacional de Moçambique), causing widespread rural devastation across the country, initially particularly along the Mozambican-Zimbabwean border but by the mid-1980s fighting intensified across the country (Bertelsen 2005; Domingues and Barrre 2013; Wiegink 2015).⁵ By the end of the conflict, Mozambique reportedly had the lowest GDP per capita in the world (Braathen and Palmero 2001). The conflict largely put on hold Frelimo’s modernising programme that sought to create a socialist society (Sheldon 2002). The war formally ended with the signing of a peace agreement in 1992, ushering in a new period during which Mozambique transformed into a multi-party democracy and market economy. Since the first democratic elections in 1994, Frelimo has retained firm control over the Mozambican state (Macamo and Neubert 2003; Macamo 2017).

Despite a GDP growth of more than 6% a year at the time of data collection, Mozambique remains one of the world’s poorest countries (Cunguara and Hanlon 2012). In 2011, the country ranked 184 out of 187 countries on the UN Human Development Index list (UNDP 2011).⁶ Although figures vary, official statistics indicate that 54.7% of the Mozambican population (estimated at 20 million at the time of data collection) lived below the poverty line of $1 per day (Cunguara and Hanlon 2010; Alfani et al. 2012). During this time period, almost 50% of Mozambique’s national budget was financed by external sources (Informal Governance Group and Alliance 2015 2010). The capital city Maputo, where this study was carried out, is relatively well developed, particularly when compared with the rest of the country. Maputo city houses all national governmental offices, has good communications and electrical network, tarred roads and a large share of the country’s relatively small number of secondary schools and tertiary institutions. National and international agencies and businesses are generally based in or close to the capital.

While great strides have been made in improving access and quality of education since Mozambique’s independence, critical challenges remain. During Portuguese rule, native Mozambicans and, in particular, young Mozambican women, had extremely restricted access to formal education. At the time of independence, approximately 93% of women and 86% of men were illiterate (Kruks and Wisner 1989; Sheldon 2002). When data were gathered for the current study, gross enrolment rates for secondary school were 32% for young men and 24% for young women, whilst net enrolment rates for young men was 16% and for young women 14% (UIS 2011). In 2010, pupil–teacher ratio at lower secondary education was said to be approximately 55:1 (UNESCO 2012). However, the class size in the three secondary schools included in this study was considerably higher, with an average of 72 learners per class.

The absence of good quality education for native Mozambicans during colonial rule contributed to the post-independence vision of education as critical to social reform and, specifically, to efforts to tackle the ‘internal’ oppression of Mozambicans (Newitt 1995). Frelimo placed particular emphasis on women’s emancipation – their oppression during colonial rule believed to have been compounded by ‘irrational’ traditions such as polygamy, and whose liberation was deemed fundamental to the creation of the new socialist state (Sheldon 2002; Arnfred 2011). As authors such as Arnfred (2011) and Sheldon (2002) argue, Frelimo’s approach to what it perceived as women’s ‘social problems’ and ‘tradition’ more broadly is likely to have been shaped by the educational experience of Frelimo leadership in Protestant mission schools. As the data presented in this paper will show, important continuities exist in terms of the ‘social problems’ that are seen to afflict women in Mozambique.

HIV and AIDS have greatly impacted on the country and its fragile infrastructure. At the time of data collection, a 1.4 million people in Mozambique were reported to be HIV positive and prevalence rates among adults between 15-49 years amounted to 11.5% (UNAIDS 2010). At present, the prevalence rate among the population is reported to be 12.5% (UNAIDS 2017). Most new HIV infections in adults in Mozambique occur through heterosexual transmission (Nalá et al. 2015). The reported prevalence rate masks considerable variation, both between regions (respectively, 9% and 21% in northern and southern Mozambique) and between young women and men (respectively, 11.1% and 3.7%, INSIDA 2009). Gender inequality forms a critical obstacle to tackling the spread and impact of the epidemic, a growing body of research highlighting the centrality of economic inequality in rendering (young) Mozambican women vulnerable to male dominance in (sexual) decision-making, unprotected sexual intercourse and HIV (Vieira Villela and Barber-Madden 2009; Bandali 2011; Underwood et al. 2011; Van Cranenburgh et al. 2017).

In 2000, a National AIDS Council (NAC) was established to coordinate all institutions and organisations working in Mozambique in the field of HIV and AIDS prevention, care and treatment. Every five years a National Strategic Plan (PEN) is developed detailing the national multi-sectoral response. PEN III 2010–2014 – the strategic plan in place at the time of data collection – identifies women as a particularly vulnerable group, and gender inequality as a central driving factor of the epidemic (Council of Ministers 2009). The education sector is broadly recognised as a critical building block in the multi-sectoral response and part of the government’s response has been to infuse sexuality and HIV prevention education across the school curriculum, with more focused integration in specific subjects, such as Biology at the secondary level (MINED 2007).

Culture as cause of HIV and AIDS

[We] have very complex issues, the question of gender, that girls are not able to say ‘no!’ because we have cultural issues that are very strong. (Lina, multilateral agency, 11/11/10)

As the extract above begins to illustrate, participants taking part in the study made repeated reference to ‘culture’, ‘cultural practices’ and ‘socio-cultural factors’ to explain the causes of, and reflect on possible solutions for, the spread and impact of HIV and AIDS in Mozambique. The quote also begins to highlight the emphasis placed on women and their responsibility to ‘say no’ to prevent the spread of HIV. With regard to ‘culture’, while varyingly deployed, the analysis of data revealed that the ways in which participants drew on this notion can be assigned to two intersecting clusters, that is, one the one hand, as being part of the cause or ‘cure’ of the spread and impact of the epidemic, and on the other hand, as relating to the domain of the ‘traditional’ or ‘modern’. The following sections seek to tease out these overlapping clusters, highlighting on the one hand, areas of agreement and disagreement between participants and, on the other hand, the gendered nature of the accounts.

When reflecting on ‘culture’ as a critical explanatory factor of the spread and impact of HIV and AIDS, ‘tradition’ formed the foremost focus in participant accounts. Across the different actor groups, the most frequently mentioned examples of cultural practices seen to contribute to the spread and impact of HIV and AIDS were initiation rites, early marriage, traditional healing and widow cleansing.⁷ Expressing her concerns regarding the spread of HIV, Sra. Vânia (MoEC, 03/12/10), for example, spoke of the ‘many inhibiting socio-cultural barriers’ that still existed, emphasising that these barriers obstructed ‘the empowerment of women’ in particular. In a similar vein, Sr. Carlos (MoEC, 02/12/10), building on the notion of culture as ‘barrier’, articulated his concerns regarding ‘traditional cultural beliefs and structures’ in terms of gender inequality. ‘In our society’, Sr. Carlos explained, ‘many of our cultural practices end up benefiting the boy more than the girl’ (MoEC, 02/12/10).

To reduce the spread and impact of HIV and AIDS educators and policy-makers posited, it was crucial to address what, for example, Sr. António (KaPfumo school, 22/10/10) described as women’s ‘inferiority syndrome … [whereby] the woman always submits herself to the man’. The conception of women underlying these participants’ narratives shows clear parallels with international development discourse in which non-Western women are, more often than not, depicted as victims of men and culture (Vavrus 2006; Chowdury 2009; Groes-Green 2011a, 2011b; Shirazi 2016).

Illustrative in this regard were comments of a senior staff member of a multilateral agency:

The whole gender issue, the role of sexuality, of sex, […] and having children being a blessing: the more, the better. The issue when you are married to a man, you cannot deny, you cannot demand. […] The gender issue is a socio-cultural construct, and [while] people might understand your message […] they are actually pretty powerless to implement that message because it goes against all the constructions about […] men and women. (Giulia, 16/12/2010)

Salient in the excerpt above is the way in which gender is perceived as a ‘socio-cultural construct’, suggesting recognition of gender and culture in relational and dynamic terms, but such constructs as running so deep as to render women and men powerless to act upon or contest them. To a degree, this conception resonates with the third-way definition of ‘culture’, that is, as both perpetually in flux yet extremely difficult to dislodge. However, as will be highlighted below, running through Giulia’s account was an essentialist conception of ‘culture’ as bounded and static, and once sufficiently analysed, as potentially modifiable through educational interventions.

Young people also spoke of problems posed by ‘traditional cultural practices’. The following excerpt – taken from a focus group discussion with young people – highlights the perceived public health risks posed by one such practice, that is, that of polygamy:

In relation to polygamy, our fathers there in the province, they stay with three to five women. And that is natural to them, and in that way they are valuing tradition … [but] in this case I think that if the husband has various wives he must take care because he can be infected and because the wives are all his, hey, [they have] to have sex with [him]. […] I am not saying it is wrong, we must not change tradition, but if he has three wives, he needs to take care and know how to protect himself. (Enoque, 15, Maxaquene school, 25/10/10)

Whilst Enoque concludes that there is a need for change – the polygamous man having to protect his wives from HIV – he also suggests that it is important to safeguard traditional practices. A practice such as polygamy is conceptualised as ‘natural’, it is part of who ‘the fathers’ are (and, by implication, ‘the mothers’?), defining them as a people. This excerpt is illustrative of a conceptualisation of culture in essentialist terms, whereby people are seen to ‘have’ culture, making them and their culture knowable (Grillo 2003; see also Gupta and Ferguson 1997; Miedema and Millei 2015). Enoque’s reference to ‘the provinces’ is reflective of the tendency to yoke ‘tradition’ to the rural, with both conceived as backward and static (see also Miedema and Oduro 2016; Parkes et al. 2016).

In addition to ‘traditional’ culture, participants – particularly policy-makers and educators – regarded the ‘modern’ to be central to the spread of the epidemic. The multiculturalism of Maputo and other large Mozambican cities – as a result of rural to urban migration, the presence of a large expatriate community and foreign media – was deemed explanatory of the unravelling of the social fabric of Mozambique (for a more in-depth discussion of this topic see Miedema and Millei 2015; Miedema 2016). Policy-makers and educators made frequent references to current-day Maputo as a cultural melting pot and what the MoEC strategy document describes as the ‘alienating influence of external cultures’ (PEEC II 2006, 111). As Sra. Regina (MoEC, 02/12/10) observed: ‘[values] were lost as a result of the mixture of cultures’. Young people voiced similar concerns. During a single-sex focus group discussion (KaPfumo school, 25/10/10), Angélica (14 years), for example, mentioned:

[From a] very early [age], adolescents are involving themselves with men, I think it is because of the modernisation of our country, when people see something, they want to imitate … which is why we don’t succeed to reduce the number of HIV/AIDS.

In addition to invoking the modern as a cause of a loss of moral moorings, and similar to the concerns raised by Lina at the beginning of this section, Angélica’s remarks are suggestive of the widely held and insidious conception that women are central to the spread and prevention of HIV (Faria 2008). However, young women and men also challenged ‘traditional’ constellations of gender relations and expectations. During the same group discussion referred to above, Aissa (14 years), for example, indicated that:

A long time ago, right? It was really difficult for you to find a girl talking with a boy. Like my mother says: the girl is never friend of the man and the man is never a friend of the girl, but now this is normal, I prefer to hang out with men.

The excerpts above illustrate the important ways in which young people appropriated the language and views used by educators and policy-makers, that is, that problematising young people’s behaviour. However, as Aissa’s remarks illustrate, while young people may tap into discourse regarding problematic modern norms and practices, they also appreciated the greater freedom provided by the shifting of values.

The loss of a cohesive cultural fabric was seen to affect young women in particular, given that they were more likely, in the words of an educator, ‘to live a, let’s say, undignified life … let[ting] themselves be taken’ when ‘their parents were not in a position’ to cater to their desire ‘to dress well, have a beautiful dress’ (Sr. Augusto, KaPfumo school, 14/02/11). Important in Sr. Augusto’s remarks regarding ‘undignified lives’ is the allusion to parents’ socio-economic position, which could be interpreted as an allusion to an awareness of intersecting categories of inequality, in this case gender and class. While occasional reference was made by policy-makers, educators and international agency staff regarding the lack of resources and services that was ‘characteristic of poor countries’ (Sra. Viviane, MoEC, 24/01/11) or ‘urban poverty’ (Albertina, bilateral agency, 13/12/10), there was a striking lack of engagement with what Aggleton, Bell and Mane (2006) refer to as the ‘systematic structuring of vulnerability’ and structural violence, including poverty, hunger, displacement and discrimination.

Young women voiced very different views, problematising the morality of older, supposedly responsible men, the inequitable power dynamics of their relationships with young women. Ruth (17 years, 29/10/10), for example, stated:

There are those fathers who … go in search of Catorzinhas (fourteen year olds) and they are senhores with big bellies, senhores with big children … [They] end up teaching [girls] things [they] did not know how to do […] and ruining [their] body.

While other actor groups homed in on young women’s responsibility (to ‘say no’) and young women’s perceived moral weakness to resist the trappings of modern life, the excerpt above is illustrative of the ways in which young women complicated the notion of responsibility, alluding to sexual violence by older men against young women. In addition, young women justified ‘interesseiras’ behaviour in light of the long-term benefit they could potentially gain, such as payment of their school fees. Despite being light-hearted in tone, the following quote is similarly suggestive of the structural factors driving young women’s involvement with more wealthy men: ‘there are girls who say [that] having a poor father is destiny, but having a poor father in law is stupidity’ (Stélia, 14 years, mixed focus group, 25/10/10). Girls with ‘poor fathers’ had to be strategic, in other words, and take opportunities where they arose.

In closing, across the different sets of participants involved in the study, ‘culture’ was largely conceived as relating to ‘traditional’, local and largely static practices and beliefs that posed a barrier to progress, particularly in terms of achieving gender equality (see also Parkes et al. 2016). In the case of policy-makers and educators, and to a degree young people, this conception of ‘traditional’ culture existed alongside a perception of culture in terms of ‘modern’ lifestyle and norms. Here issue was taken with, among other things, the culturally heterogeneous and changeable – the ‘multicultural’ – character of the city, which was seen to imply a lack of a unified culture in the sense of shared moral values. The absence of such a stable moral ‘culture’ was considered to negatively affect young women in particular and, by extension, fuel the HIV epidemic. As discussed in Miedema and Millei (2015) and Miedema (2016), school-based education was posited as central to instilling moral values, with a focus on young women to restore a sense of communality and, specifically, (re)create ‘culture’ in the singular sense (see also Vavrus 2006). As the next section clarifies, establishing linkages between schools and ‘culture makers’ was conceived as an important means to reshape traditional cultural practices, maintaining those deemed to be positive and modernising those considered to be problematic.

Culture as ‘entry point’ for tackling HIV and AIDS

[When] you’re talking about behaviour change … the most important part is taking into account people’s social and cultural dimensions. So that behaviour change can really be effective. (Giulia, multilateral agency, 16/12/10)

The preceding section attended to participants’ conceptualisation of ‘culture’ in largely essentialist terms, whereby ‘traditional’ practices and ‘modern’ values (or ‘culture-less’ young urbanites) were Othered and pathologised. The quote given above alludes to the widespread expectation underpinning participants’ accounts with regard to the potential of turning ‘problematic’ traditional cultural practices into opportunities, to be specific, using these as ‘entry points’ (Albertina, 13/12/10) for delivering ‘modern’ ‘scientific’ HIV-related education and thereby bring about change.

Participant narratives suggest that using ‘culture’ to address the epidemic required two key steps. The first step in this process was identifying the ‘negative dimensions’ of cultural practices with ‘the makers of culture’ (Sr. Carlos, MoEC, 02/12/10), defined as elderly community leaders, religious leaders and traditional healers, as well as parents and caregivers. Participants often referred to the need to conduct research with these actors in order to identify these ‘negative dimensions’ and to design culturally appropriate approaches to HIV-related education. As Sr. Sergio (MoEC, 04/11/10) pointed out, for instance: ‘at the national level we need to redesign messages: it is important to develop messages in line with the target group’. Asking him how we could come to know about these different target groups, Sr. Sergio replied: ‘Well, it’s those pieces of research. Research can show which strategy we need to adopt’.

In a similar fashion, Sra. Viviane (MoEC, 24/01/11) observed that research was critical in that it provided ‘advice as to the best paths to follow’. She went on to argue that parents and caregivers needed to ‘have an active voice’ in research so that they could ‘say that: “in our area, the way in which this programme is structured, perhaps it can be changed a little like this.” So that [the programme] is effective in that region’. Research was thus deemed important to efforts to develop educational contents that were aligned with the specific context in which these were delivered.

In the same interview referred to above, Sr. Sergio spoke at length about the need to design appropriate prevention initiatives whereby attention needed to be paid to rural-urban divides and differences between genders:

[M]en are more laid back […] in the sense of … they don’t have that shyness, of being closed … while girls aren’t like that. Especially in rural areas. [Girls] are shy, reserved, you know … education, culture, cultural habits. For this reason, the approach to this issue, the way in which we address it here in the city, what we teach in the city cannot be the same in the countryside. In the countryside there are peculiarities, traditions. [For example] when you say something [about sexual relations], you can shock people.

It was particularly easy to shock young women, Sr. Sergio explained, but also younger age groups. In a similar vein, Sr. Alberto (Maxaquene school, 04/01/11) talked about the barriers schools faced in educating young people about sexual relations due to caregivers’ sensibilities:

This matter is a little controversial … . There exists a split in terms of opinions, people disagree, there is a ‘fight’ between ideas. From a scientific perspective within the context of the school, we can tackle [these issues], but we will have a problem [with] parents and caregivers. There are some who will agree it is good to talk about sexuality [with young people] but there are also those who say ‘no, no, no, no!’ […] because of different ideas – ethical, religious. So our pupils may receive very different forms of orientation, socialization […] In this multi-cultural society, it is a little difficult. But in the school I do not see any impediment [to such discussions].

Participant narratives thus suggest that HIV prevention needed to be designed in culturally appropriate or sensitive manners, taking into account intersections of, among other issues, gender, age and degree of urbanisation/rurality. In important ways, the education system is hereby framed as scientific modern arbiter in a ‘multicultural’ society in which some cultures/cultural groups were less scientifically inclined.

Once the negative dimensions of culture had been identified and a clearer understanding had been developed as to cultural sensitivities, the second step in the process of using culture as a point of departure entailed persuading makers of culture to alter risky aspects of particular ‘traditional’ practices. Similar to Sr. Alberto, Sra. Viviane (MoEC, 24/01/11) spoke of difficulties parents experienced in terms of talking about sexuality with their children, explaining this in terms of the limited education they themselves had received: ‘at that time […], there was no profound discussion of your body, today the programmes are much more in-depth and when we talk about these things, it is inevitable that we talk about the relationship [between] men and women’. Yoking the past to a lack of ‘profound’ education about male and female bodies and relationships, like Sr. Alberto, Sra. Viviane juxtaposed this situation with the modern-day school. She then went on to highlight what needed to be done to resolve the matter:

Within our school programmes […] we study the human body, demonstrate how it is constituted and functions, [but] this is not regarded positively by society. They say: ‘but what kind of education are you providing our children?’ That is, in the time of our grandparents, parents, it was a taboo to talk about the vagina, the penis, sexual relations. In our modern world we adopt these approaches. […] So we need to raise awareness, mobilise all of society because education cannot be done by one group alone.

The emphasis within these participant accounts is on what is deemed traditional and the need to find ways to modernise traditional cultural beliefs and practices to reduce young people’s vulnerability. Arguably, the prominence attached to traditional culture, albeit at times implicitly, precludes a more critical analysis of the underlying gender order that defines, among other things, perceptions of appropriate masculinity and femininity which, in crucial ways structures young people’s vulnerability to the epidemic. Additionally, these accounts serve to ‘Other’ those who were seen as not having benefited from modern-day schooling, framing them as backward and in need to ‘awareness raising’.

Not only did policy-makers, educators and international agency staff regard the involvement of parents and caregivers as crucial, traditional healers, religious leaders and initiation rites leaders also needed to be involved. On this view, a cultural practice, such as initiation rites and community members in charge of conducting these rites, could thus be the ‘entry point’ referred to earlier, offering a concrete venue for the transfer of ‘scientific’ information about HIV and AIDS to young people, as well as the ‘makers of culture’. In Sra. Viviane’s words again (MoEC, 24/01/11):

Sra. Viviane: Within those cultural aspects, there are things that can even be positive and so perhaps [we can] benefit from these cultural aspects of a region […] to introduce this component [of Sexuality education]. But here we cannot generalise. When we talk about initiation rites, [these] are not conducted all over the country. […] But the leaders of [a particular] region, they know very well how they conduct those rituals and how one could incorporate this component [of HIV-related education] […]

Esther: So should the school work with community leaders?

Sra. Viviane: I think so. If there isn’t dialogue, […] we will not be able to penetrate, we will not be able to reach our objectives. Because they have those practices … and they are radical, radical. They won’t accept the elimination of these practices, so there has to be a dialogue, a mobilisation, a sensibilisation so that they themselves start […] incorporating our objectives.

This excerpt too reflects the dichotomy between tradition and modernity that Parkes et al. (2016) refer to, with rural young people, and girls in particular, seen as victims of the backward ways of rural communities. The ‘modern’, in the sense of scientific knowledge, is posited as the positive other, offering a means to alter traditional structures, practices and tackling ‘irrational’ beliefs and myths in such a way as to reduce the risk they posed in the transmission of HIV. The final section below illustrates how the ‘modern’ was also interpreted in terms of what were defined as modern-day norms and values.

As was discussed in the preceding section, policy-makers, educators and to a degree, young people, perceived cultural hybridity as signifying a loss of community and shared cultural identity, thereby arguably manifesting the ‘cultural anxiety’ that Grillo (2003) refers to. ‘One voice’ (Sra. Viviane, MoEC, 24/01/11), a shared culture, was posited as critical to addressing the epidemic, reflecting not only an apparent desire for shared cultural roots and culture as bounded but also a conception of HIV, and its spread and impact, in largely social terms. The section above clarifies that these same actors as well as international agency staff, believed that closer involvement of various ‘makers of culture’ (including parents and caregivers) was required in order to create the required united voice in the response to HIV and AIDS, and reducing young people’s vulnerability. To create the necessary ‘unity in diversity’ (MoEC Strategic Plan for Education 2006–2011⁸/PEEC II 2006, 131) to jointly tackle the threat posed by HIV, these same sets of participants called for a stronger grounding of school-based (Sexuality) education in a unified set of norms and values.

The following statement from the PEEC II (2006, 14) illustrates that the normative canon that policy-makers and educators called for drew on a mixture of what was conceived as ‘traditional’ and ‘modern’:

[Mozambique] needs to develop an education that is based on the values of the family, respect for African traditions and a compatibility with universally recognised modern values.

Examples of traditional values given by policy-makers, educators and, to a degree, young people, were ‘respect’, particularly for elders (Sr. Sergio, 04/11/10; Sra. Vânia, 03/12/10), solidarity and ‘knowing how to live together’ (Sr. Reís, 19/10/10; Sra. Vânia, 03/12/10). All participants, with the important exception of young people, voiced their belief in ‘modern’ values of monogamy, fidelity, rationality and crucially, girls’ right to ‘say no’ as essential components of an informed ‘evidence-based’ response to HIV and AIDS. Young people’s narratives were once again less morally inflected than those of most other participants, dwelling instead on day-to-day ‘practical’ questions about relationships. Speaking about the aforementioned issue of ‘interesseiras’, Enoque (15 years, 15/12/10), for example, complained that ‘girls nowadays’ were ‘very easy’ and how difficult it was to know whether a girl was an ‘interesseira’ or not:

She will say “no, I am not [an interesseira]", she will swear on all that is sacred that she is not, but at some point she will start demanding money […], so they lie a lot. So I just want to hear from them, [what they are] really.

Therefore, while young people’s accounts were more layered, in the sense of complicating notions of morality and responsibility as well as being more geared to concrete questions about sexual relations, on the view of policy-makers, educators and international agency staff, Sexuality education was framed as requiring a hybrid of specific ‘traditional’ and ‘modern’ values and norms. Despite differences in views as to notions of culture and processes of change, the data illuminate that those involved in the design and delivery of HIV prevention education largely built on reified understandings of ‘culture’, whereby it became a ‘thing’ to be possessed and manipulated in the response to HIV and AIDS (Vavrus 2006; Taylor 2007, 967).

Conclusion

Wright (1998) asserts that instead of being a useful analytical tool, it is the notion of ‘culture’ that needs to be explained, and that its use and misuse constitute important objects of study (see also Anderson-Levitt 2012). Building on qualitative data, this paper elucidates the recurrent deployment of ‘culture’ by different sets of actors in explanations of, and proposed solutions for, the HIV epidemic in the context of Maputo, Mozambique. Specifically, the paper highlights the interrelated gendered binaries invoked in participant narratives and MoEC documents with regard to culture, that is, of victim–perpetrator, rural–urban and tradition–modernity. These accounts speak to the monolithic image of the Third World Woman as uniformly oppressed by men and by culture, and, more broadly, to essentialising and pathologising discourses of non-Western people and places. Postcolonial feminists have criticised this narrative since Mohanty’s seminal work on this subject in the 1980s (Mohanty 1988, 2003; Sangari 2005; MacDonald 2015; Roy 2017). In the field of CE, scholars drawing on postcolonial theory, such as Crossley and Tikly (2004), have exposed the gendered and racialised underpinnings of existing educational policies and curricula.

As the data revealed, the frame of the Third World Woman also informed conceptions of the ‘interesseira’ – the young woman who seeks material gain from intimate and sexual relationships. In this case, young women were depicted as victims of their own moral weakness and inability to consider long-term consequences of their behaviour. In his study on middle-class young people in Maputo, Groes-Green (2011b) asserts that the category of the ‘interesseira’ is a classed one. According to Groes-Greene, the notion of the interesseira is, furthermore, reflective of colonial and postcolonial distinctions between femininities: the interesseira a lower class woman regarded as uneducated, openly sexual and morally weak, the upper-class woman embodying the emblems of good femininity, that is, education, purity and decency. In other words, regardless of whether explanations of gender inequality and the spread of HIV were yoked to traditional or modern culture, women were portrayed as being determined by backward cultural frames of reference.

The paper thus sheds light on the often contradictory conceptions of ‘culture’ underpinning educational decision-makers, teachers and other actors’ ideas of the educated person in a world with HIV and AIDS. In addition to the three approaches to HIV prevention education discussed in Miedema, Maxwell, and Aggleton (2011), this paper identifies a fourth ‘culturalist’ approach to sexuality and HIV prevention education. As Grillo (2003) argues, this approach constitutes an ‘old school’ conception of culture, underpinned by essentialist interpretations of ‘culture’, and one which, as postcolonial scholars have compellingly argued, continues to inform education and international development policies and practice. The data suggest that solutions to the epidemic were sought in hybrids of the traditional (i.e. moral cohesion), and the modern (e.g. rationality, science, monogamy). This fourth culturalist approach to sexuality education thus requires a further sub-division, drawing a distinction between on the one hand, culture defined in terms of ‘traditional cultural practices’ and on the other, culture conceived in relation to ‘modern-day’ norms and values, whereby both were seen as having positive and negative connotations. Common to these various articulations of culture was their gendered underpinnings and the reification of Mozambique as a developing ‘pre-modern’ country.

Piedalue (2016) contends that culture is ‘far too complex and live a phenomenon to be solely an obstacle to women’s freedom’ (10, emphasis in original). Arguably, ‘culture’ also provides far too limited an explanation for the spread and impact of HIV and AIDS, and its resolution. This is not to say that education can be atomised from the broader cultural context in which it is designed, delivered and taken up. Refusing a simplified culturalist understanding of HIV and AIDS does not mean abandoning critical analysis of, or initiatives to, address the roots of gender and sexual inequalities that, to an important degree, drive the epidemic (Piedalue 2016). To paraphrase Sangari (2005), culturalist interpretations of the role of ‘harmful traditional practices’ and gender inequality in the spread and impact of HIV and AIDS either redeem these practices as ‘sacred tradition[s] or condemn them’ from a ‘neo-orientalist’ interventionist gaze that has the tendency to both ‘spectacularise and decontextualise’ socio-historically constructed phenomena and practices (159). The emphasis on ‘culture’ in sexuality and HIV prevention education – in its various traditional and modern guises – diverts attention from structural causes of young Mozambican women and men’s vulnerability to HIV and AIDS, and, crucially, rather than problematise gender relationships, reifies and solidifies these. Considering culture in sexuality education entails complicating ‘culture’, examining the gendered and classed process of cultural meaning-making in relation to questions of health and sexual relations, and, crucially, who decides what is culturally ‘appropriate’.

Acknowledgements

The author is very grateful to the young people, educators, policy-makers and staff of international agencies who took part in this study. I would also like to thank Dr Winny Koster, Drs Marielle le Mat and Dr Huub Dijstelbloem for the helpful comments they provided on earlier versions of this article. Finally, I would like to thank the three reviewers, particularly Professor Elaine Unterhalter, for their insightful feedback, which pushed me to consider a range of issues far more carefully.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Esther Miedema is research associate and lecturer at the Amsterdam Institute of Social Studies at the University of Amsterdam. Her teaching and research related to education, gender, sexuality and international development. She is the co-principal investigator on the ‘HER CHOICE’ early marriage research programme in 10 countries in Sub-Saharan Africa and South Asia (2016–2020).

ORCID

Esther Miedema http://orcid.org/0000-0001-5654-0987

Additional information

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

1. I use the terms sexuality education, HIV prevention education and sexuality and HIV prevention education interchangeably in this paper. I do so given that, at the time of data collection, in-school learning about sexuality and sexual relations in Mozambique took place within the framework of HIV prevention education.

2. Creating a gender-balanced sample was not possible in relation to the staff working for international agencies (all but two of the participants being female).

3. All interviews lasted between one and two hours and were, for the most part, conducted in Portuguese. Interviews were audio-recorded following participants’ consent and transcribed ad verbatim. Translations are the researcher’s own, whereby support from native speakers was sought in case of doubt or where verification was deemed appropriate.

4. Pseudonyms are used when citing participants and, for reasons of etiquette, the titles ‘Sr.’ (Mr) or ‘Sra.’ (Mrs) are used when citing educators and policy-makers. These titles are not used when referring to participants working for international agencies or for young people. A quote from an interview with a policy-maker is followed by the acronym ‘MoEC’. For reasons of confidentiality, no reference is made to the directorate/department or organisation to which policy-makers or international agency staff were connected. Fictive names are used to refer to schools and all citations are followed by the date of the interview.

5. For more background on the war and its impact in Mozambique, including more detailed discussion of the critical ties between Renamo, the Southern Rhodesian Army and, following Zimbabwe’s independence in 1980, the South African Defense Forces, see, for example, Finnegan (1992), Bertelsen (2005), Domingues and Barrre (2013), Wiegink (2015).

6. In 2015, Mozambique ranked 181 out of 188 countries on the HDI (UNDP 2016).

7. Widow cleansing refers to the sexual act of purifying the wife of a deceased man through semen entering her body. The purification is carried out by a male relative of the deceased (UNDP 2012).

8. Henceforth referred to as ‘PEEC II, 2006’.