Mechanisms of adopting and reformulating comprehensive sexuality education policy in Ethiopia

ABSTRACT

Over the past few years, international organisations have advanced Comprehensive Sexuality Education (CSE) as a global policy to promote sexual and reproductive health and rights (SRHR) and address gender-based violence in schools. This paper analyses policy adoption, transfer mechanisms, and reformulation of CSE in Ethiopia, a late adopter of the policy. To do this, we identify education policy transfer mechanisms and apply a gender analysis by focusing on conceptualisations of gender relations in the uptake and reformulation of CSE policies. Drawing on document analysis and stakeholder interviews, the paper reveals that CSE in Ethiopia is largely a donor-driven agenda, advanced through dissemination and networking strategies. CSE is particularly embraced by the Ministry of Health, international organisations and NGOs in Ethiopia, but at the same time, the Ministry of Education and other critics continue to resist adoption, emphasising cultural differences. As a result, CSE in Ethiopia is (re)formulated and reflects narrow conceptualisations of how CSE can address gender-based violence, restricting its focus to health and development outcomes.

KEYWORDS:

• Education policy
• policy transfer mechanisms
• policy reformulation
• comprehensive sexuality education
• gender
• Ethiopia

Introduction

Over the past two decades, Comprehensive Sexuality Education (CSE) has become a well-established global policy. Global guidance documents broadly define CSE as education that equips children and young people with the knowledge, skills, and values that empower them to make informed choices about their health, well-being, sexual and social relationships, and ensures protection of their rights (United Nations Educational, Scientific and Cultural Organisation (UNESCO) 2018a; UNESCO and UN Women 2016).¹ UNESCO’s Global review on CSE highlights that almost 80 per cent of the countries in their assessment have policies or strategies that support CSE (UNESCO 2015). All countries in Eastern and Southern Africa now report to have a policy to promote life skills-based HIV education for young people (which can be CSE or similar), although the full operationalisation of these strategies and policies remains a challenge for many countries (UNESCO 2015).

The rise of CSE should be seen in the light of growing international attention towards preventing HIV/AIDS (UNESCO 2009), and more recently the increase in global declarations and commitments to address gender-based violence (for an overview, see Parkes et al. 2016). Research in Sub-Saharan African countries has revealed that gender-based violence is prevalent in schools, and takes up various forms such as bullying, corporal punishment, and verbal, physical, and sexual abuse (Bhana 2012; Dunne, Humphreys, and Leach 2006; Leach, Dunne, and Salvi 2014; Parkes 2015). Gender-based violence may include sexual, physical, emotional, and symbolic violence, and is embedded in complex webs of power relations and institutional structures that both reproduce, and are affected by, inequalities (Dunne, Humphreys, and Leach 2006; Parkes 2015). Gender relations are understood as sets of social relations between and among men and women that are multidimensional and part of a larger social (gender) structure in society (Connell 2012b). Gender relations are not static and can be contested and re-negotiated at intrapersonal, institutional and societal levels. The increasing focus on gender-based violence in education policies is especially important because education can make substantial contributions to addressing gender-based violence by raising critical consciousness through pedagogy and curricula, and by educating young people about justice and equity (Chege 2007; Connell 2012a; Parkes et al. 2017).

CSE is thus increasingly promoted as an educational programme that can address gender-based violence. However, only a few studies have investigated the contribution CSE can make to this. One such study, focusing on how CSE can prevent violence against women and girls, finds that CSE can (1) promote gender-equitable attitudes among young people; (2) improve young peoples’ gender relations and reduce violence against women and girls; (3) transform attitudes in the wider community; and (4) improve reporting and response mechanisms when violence occurs (Holden, Bell, and Schauerhammer 2015). However, in practice, most CSE programmes are limited in the extent to which they are able to change gender relations to be more equitable (Holden, Bell, and Schauerhammer 2015). In addition, there is limited research on how CSE can prevent other forms of gender-based violence such as violence against men and boys. Nevertheless, the shift in attention towards gender-based violence is evident in the 2018 International Technical Guidelines on Sexuality Education, compared to the first version from 2009 which predominately focused on using CSE as a strategy to prevent HIV/AIDS. UNESCO’s global review on CSE (2015) reiterates that discussing gender relations is a core component of CSE programmes and that, ideally, acceptance of sexual diversity should also be discussed in CSE. The review recognises that there is a ‘lack of appropriate attention to gender’ in the enactment of CSE, but nevertheless, it remains globally recommended due to its potential to include social and gender-related topics.

Despite the growing international promotion of CSE, research on policy developments with regard to CSE, especially on its adoption at national levels, remains scarce. This is surprising given recent global controversies around CSE,² which may affect the processes of policy adoption (see also Theoretical Background). The limited studies that are available in this field address the concerns of CSE adoption in light of opposition. These studies are often supported by bilateral donors or international organisations (e.g. Chau et al. 2016; Huaynoca et al. 2014; Panchaud et al. 2018). Opponents’ concerns about CSE are often framed as conservative and religiously informed (see also Roodsaz 2018). From this perspective, it is especially striking that CSE has been increasingly adopted in highly religious contexts such as Ethiopia. Ethiopia is also a late adopter of CSE: To illustrate, Nigeria has had a clear nation-wide policy on CSE since 2002 (Huaynoca et al. 2014) and Kenya started a sexuality education programme in 1998, which was expanded and supported by national policies three years later (UNESCO 2014). China, Lebanon and Vietnam were also already implementing and scaling up sexuality education programmes in 2008 (UNESCO 2009). In Ethiopia, it was not until 2009 that sexuality education was suggested as a strategy for HIV/AIDS prevention (MoE 2009), and it is only in the latest Ministry of Health adolescent youth health strategy that comprehensive sexuality education is mentioned (MoH 2016). Ethiopia is thus an interesting context to scrutinize what mechanisms and rationales have led to the (late) adoption of CSE as a national policy. Hence, in this paper we address the following questions:

• 1) To what extent and why is CSE adopted as a national policy in Ethiopia?

• 2) Which international and national actors have been influential in CSE policy adoption processes?

• 3) Through which mechanisms of education policy transfer has Ethiopia adopted CSE?

• 4) How is CSE reformulated as policy at the national level in Ethiopia?

To answer these questions, we draw on qualitative interviews with key CSE policy stakeholders in Ethiopia, as well as a policy document analysis. At the theoretical level, we make use of education policy transfer mechanisms as identified by Dale (1999) and Steiner-Khamsi (2006, 2014), and apply a gender analysis by focusing on conceptualisations of gender relations in the adoption and reformulation of CSE policies (Connell 2012b; Unterhalter and North 2017), as discussed in the next section.

Theoretical background

Adoption mechanisms and policy networks

Despite the availability of global guidelines on CSE, exactly how these directions have been adopted in national policies remains largely unexplored. Dale (1999) describes five mechanisms of external effects on national education policies: harmonisation, dissemination, standardisation, installing interdependence, and imposition. Harmonisation refers to processes where national policy-making capacities are combined at a regional level through collective agreement (for instance in European Union policy-making). Harmonisation mechanisms are initiated collectively by multi-national members and are focused at regional level policies. Dissemination mechanisms are initiated by supranational/international organisations who employ an agenda-setting strategy to convince national bodies to adopt a policy. This can include making publications and technical guidelines available, organising conferences around specific themes, and initiating knowledge platforms/working groups. Standardisation mechanisms are also initiated by supranational/international organisations; however, the aim of standardisation is to develop international norms that countries subscribe to, in order to create common norms and standards across the world (e.g. commitments to Education for All). Installing interdependence refers to processes driven by concerns about issues that extend beyond the scope of a nation state, such as climate change. Its main concern is centred on the particular issue, more than the adoption of a particular policy, and often operates bottom-up. Finally, imposition mechanisms compel recipient countries to adopt particular policies by making them compulsory. Imposition processes are typically initiated by bilateral donors or supranational organisations and are focused on explicitly requesting adoption of specific policy goals, most typically through aid conditionalities.

These different policy adoption mechanisms may operate simultaneously and dynamically, and actors involved in national policy processes may have diverging interests. It is also increasingly difficult to separate the ‘local’ from the ‘national’ and ‘global’, as all social relations and networks cross-cut these boundaries (Ball 2016). Networking spaces where a variety of ‘global’, ‘national’ and ‘local’ policy actors come together can help advance certain policies. Such networking spaces can contribute to multiple actors creating a unified discursive logic to advance a policy and may also open up possibilities for new policy actors to enter education policy spaces (Ball 2016). At the same time, national policies are increasingly contingent on global policy processes and susceptible to dominant discourses in education policy, which are often put forward by powerful international actors (Rizvi and Lingard 2009). Hence, it is important to scrutinize how and by whom the problems are constructed (Rizvi and Lingard 2009). Indeed, sometimes global policies may not be relevant to national or local realities. For example, Fetene and Dimitriadis (2010) argue that some HIV/AIDS policies in Ethiopia have failed to include young peoples’ experiences, and therefore do not adequately identify and respond to the most important issues for youth. It is furthermore important to keep in mind that political and economic incentives might influence policy adoption (Steiner-Khamsi 2006): when donor financing is available for a specific global policy, receiving countries are interested in policy take-up. Hence, understanding the contextual specificities and the relations among policy actors is key to analysing how certain policies are adopted, resisted, and/or re-formulated (Ball 2016; Steiner-Khamsi 2014; Altinyelken 2011).

Reformulation

Policy reformulation and resistance may occur when a global policy is adopted at a national level. Proposed policies may be reformulated to better reflect the most pressing concerns and interests in a particular context (see, e.g. Sun Kim 2017; Altinyelken 2011). To resist policies, critics often (over)emphasise contextual differences, for instance differences in education systems (Steiner-Khamsi 2014), as well as raising questions of appropriateness and feasibility (Ringeling 2005). In the case of CSE, reasons often cited for the resistance or reformulation of the policy are socio-cultural or religious. In Uganda, for instance, sexuality education curricula have been tailored to fit religious affiliations of schools (De Haas 2017), and in Nigeria, some schools and community members resist the implementation of CSE due to socio-cultural and religious values, despite its inclusion in national policies (Huaynoca et al. 2014). A study in Senegal highlights that members of Ministries of Education (MoEs) and Ministries of Health (MoHs) are more constrained by religious influence than Non-Governmental Organisations (NGOs) (Crossouard, Dunne, and Durrani 2017), as NGOs are more likely to reflect global or donor discourses.

One particular reason for resisting CSE, which has been observed across several countries, relates to CSE’s attention to sexual diversity, and discussion of non-marital sexual relationships and sexual pleasure. In the US, conservative groups and opponents of CSE have raised concerns that it could encourage acceptance of sexual and gender diversity or promote promiscuity (Family Watch International 2016; stopcse.org). This is despite research showing the positive health outcomes of CSE including delay in sexual onset (Kirby 2008). Outside the US, studies have similarly described opposition to CSE as being rooted in religious social-cultural norms (Huaynoca et al. 2014; Vanwesenbeeck et al. 2016). Such resistance to CSE can be expressed by parents and communities, as well as by federal and state education authorities, school administrators, and teachers (Huaynoca et al. 2014). From this perspective, a recent study concluded that important ‘levers’ for a favourable policy environment are government willingness to adhere to international commitments in support of CSE and NGOs’ willingness to promote CSE in the face of opposition (Panchaud et al. 2018). In such discussions, CSE is typically framed as a ‘global’, progressive, and secular/scientific policy (also observed by Roodsaz 2018), while resistance to CSE is considered a result of ‘local’ traditional or religious values. Interestingly, however, Weiss and Bosia (2013) argue such resistance against policies associated with sexual diversity is not merely the result of religious beliefs or ‘traditional’ values, as some literature suggests, but can be seen as a form of ‘political homophobia’. Such political homophobia may be employed by states to resist Western agendas and simultaneously seek ways to exercise control (Weiss and Bosia 2013). Hence, the construction of CSE as a ‘progressive’ global policy, resisted by ‘conservative’ and ‘religious’ actors, can be seen as an expression of political agendas. Furthermore, such framing of the modern progressive versus the traditional conservative produces a binary that results in transnational and local processes of ‘Othering’ and might not be helpful in formulating policy priorities that are contextually pertinent (Grewal and Kaplan 2001; Roodsaz 2018).

Against this backdrop, it is important to note that CSE has been defined differently by several actors (Hague, Miedema, and Le Mat 2018), which may explain some of the variations in how the aims of CSE are understood, and how they are adopted, resisted, and/or reformulated by different parties. Gaining conceptual clarity is essential in understanding the adoption and reformulation of global policies at the national level, particularly in the case of gender-related policies (Connell 2012b; Unterhalter and North 2017). Unterhalter and North (2017) argue that policy research has often overlooked the possible variation in what is meant conceptually with specific policy ideas, and the influence of key people therein. Their research in Kenya and South Africa highlights, for instance, that in gender and education policy enactments, policymakers held conflicting interpretations of what gender equality in education means and should achieve. These conflicting interpretations resulted in narrow conceptions of gender in education policies, which limited opportunities for organisations to put forward arguments for gender equality that went beyond indicators of gender parity (Unterhalter and North 2017; see also Silova and Abdushukurova 2009). Studies specifically on gender-based violence have also established that policies should move away from narrow conceptions of gender-based violence as sexual violence, and instead link gender-based violence to issues of corporal punishment, bullying, and re-admission for young mothers (Chege 2007; Leach, Dunne, and Salvi 2014). To address this, Connell (2012b) suggests that gender policies should place the relations between and among women and men at their core. Such a relational framework moves away from a categorical gender approach, where policy is written for specific genders (mostly ‘men’ and ‘women’), and instead focuses on gendered dynamics in institutions and practices. Connell (2012b) also highlights that global gender theory should be mindful of how ‘gender’ has been historically constructed from a Eurocentric point of view, with assumptions that similar notions of gender apply to contexts outside of Europe. This legacy remains today, where many policies reflect Eurocentric or Western ideals. Furthermore, Parkes (2016) concludes, based on an analysis of policy enactment on school-related gender-based violence, that policy enactment is not only a matter of overcoming practical challenges; ‘It is about reflecting on our deeply held beliefs and practices, through which we knowingly and unknowingly collude with gender-based violence’ (p. 103). Thus, policy adoption and reformulation require creating dialogues between actors and institutions at various levels to gain conceptual clarity, while challenging gender and power relations (Unterhalter and North 2017; Connell 2012b).

Contextual background

Over the past years, Ethiopia has made progress in reaching gender parity in primary and secondary schools. The Gender Parity Index (GPI) is 0,91 for primary and 0,96 for secondary schools according to the latest statistics from 2015 – meaning that only slightly more boys than girls are enrolled in schools (UNESCO 2018b). At the same time, gender equality in access to tertiary education and health services remains an important issue on the agenda for the Ethiopian government. Gender-based violence is also mentioned within this agenda, and CSE is one of the (non-formal education) programmes that is implemented in Ethiopia to improve young peoples’ knowledge on their sexual and reproductive health (SRH) and to promote gender-equitable relations (MoE 2016; MoH 2016). It is important to understand policy adoption processes in Ethiopia, including those surrounding CSE and gender-based violence, in the context of high poverty and inequality levels, strong religious values, and donor dependency. Ethiopia had a GDP of USD 47.6 billion in 2013 (World Bank Data 2018). The net overseas development assistance received by Ethiopia made up 8.2% of the government net income in the same year. The largest donors are the World Bank, USAID, DFID, the African Development Bank, the Global Fund, and the European Union (DAG Ethiopia 2015).

Ethiopia is a signatory of what is commonly known as the ‘ESA Commitment 2013ʹ, a commitment signed by Eastern and Southern African Ministries of Education and Health affirming their political will to ensure access to CSE. Ethiopia has also ratified international human rights treaties, including the Convention on the Rights of the Child (CRC), the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW), and has adopted the African Charter on the Rights and Welfare of the Child (ACRWC). However, despite these measures, gender-based violence prevalence remains high: 26.3% of women older than 15 and 12.9% of adolescent girls (aged 15–19) have experienced intimate partner violence. Indeed, Ethiopia is among the 20 countries with the highest levels of gender-based violence, according to a study of 168 countries (Know Violence in Childhood report 2017). Other studies indicate 68% of young women in secondary schools in East Ethiopia have experienced sexual violence (Bekele 2012), 41% of girls marry under 18 (UNFPA 2012), 74% of women have undergone female genital mutilation/cutting (FGM/C) (UNICEF 2013), and women are faced with higher rates of domestic violence, and fewer work opportunities compared to men (UN-HABITAT 2008).

Importantly, Ethiopia’s 2009 Charities Proclamation has restricted organisations that are supported by foreign funds from working on human rights and advocacy. This has affected SRH as well as education initiatives in Ethiopia, where rights-agendas remain implicit (because of this, the final ‘R’ for ‘rights’ in SRHR is omitted in the Ethiopian SRH agenda). This was still the case during fieldwork for this study (November 2016), but recently there has been a discussion of updating the proclamation to allow NGOs to carry out rights-based work.

At the time of data collection, the country had been affected by several protests and riots instigated by economic and political inequalities, leading the government to announce a ‘state of emergency’. Many of the interview participants were concerned that gender-based violence prevalence would increase under these circumstances. In areas affected by protests, schools were temporarily closed and CSE programmes were put on hold during the state of emergency.

Methods

To analyse how CSE is adopted in Ethiopia, we draw on content analysis of 23 policy documents and interviews with 16 stakeholders held in November 2016. Our content analysis was geared towards identifying manifest content of the policy texts and interview transcripts, followed by an exploration of underlying and recurring themes (following Graneheim, Lindgren, and Lundman 2017). That is, in the initial stages of data analysis, explicit descriptions of policy adoption and transfer mechanisms, i.e. manifest content, were identified in policy texts and interview transcripts. Following these concrete descriptions, underlying and recurring themes were uncovered (e.g. debates around cultural differences, see findings section) that allowed to interpret and explain the policy adoption and reformulation process and its implications. At the time of data collection, the Government of Ethiopia had recently updated their four-year policy frameworks, and therefore the document analysis focused on the most recent policy frameworks available. This included policy documents from Ethiopia’s national government and ministries, international organisation guidelines and bilateral policy documents. The inclusion criteria for documents were that they concerned CSE or gender-based violence in education, and were not dated before 2004. For national policies in Ethiopia, at least the most recent (covering 2016–2020) health-sector and education-sector policies were included for analysis, as well as the overarching ‘Growth and Transformation Plan II’ (GTP II) to contextualise findings, even if they did not explicitly mention CSE or gender-based violence in education. Almost all documents were publicly available; some were shared in draft form and in confidence by stakeholders.

Stakeholder interviews included questions on the development of these policies and strategies. The stakeholders who participated in this study were all Ethiopian nationals and were affiliated with Ethiopian and International NGOs, Universities, the Ministry of Health, or Bilateral or International Organisations (for a detailed overview, see Table 1). All participants saw value in CSE and advocated for the adoption of CSE as a national policy, although a small number expressed some reservations regarding national ownership and local priorities. Stakeholders were approached via the professional network of the first author, who collaborated with a local NGO for this study. Inclusion criteria for stakeholders were that they are professionals who work on CSE, gender-based violence, education and/or SRH in Ethiopia. To make sure no important stakeholders were excluded, other participants were approached through snowballing, whereby some interview participants linked the researcher to other relevant stakeholders that met the inclusion criteria. In selecting stakeholders, the researcher aimed to have a balanced representation of organisational affiliations and gender; this balance was monitored and ensured as much as possible during data collection through purposive sampling. Anonymity and full confidentiality were ensured prior to interviews, and for this reason, organisational affiliation of the participants is not revealed in this article. Interviews with stakeholders were conducted in English and audio recorded with permission. Ten interviews were held individually, and three sessions included two participants. Most interviews were held in participants’ own offices, and some in nearby cafés based on participants’ preference. Stakeholder interviews were geared towards identifying how CSE and gender-based violence is understood conceptually, how CSE as a global policy is adopted and re-formulated, and understanding possible important moments of acceptance of and resistance to CSE in the Ethiopian context.

Table 1. Overview of interview participants, their gender and organisation affiliation.

Affiliation	Interview participants
National CSO/NGO	7 (all male)
International NGO	2 (both female)
University staff	2 (1 male, 1 female)
International Organisation	3 (2 male, 1 female)
Bilateral organisation	1 (female)
Ministry of Health	1 (female)
Total	16 (10 male; 6 female)

The procedure for data analysis followed a constant comparison analysis approach (Leech and Onwuegbuzie 2007; Miles and Huberman 1994). First, an initial coding scheme was developed based on our theoretical framework and research questions. Codes included: policy transfer mechanisms; key actors in the policy adoption process; conceptualising CSE; conceptualising gender-based violence; and re-formulation of CSE policy. These codes and sub-codes were developed based on the theoretical framework; for instance, the code ‘policy transfer mechanisms’ was sub-divided in the five policy transfer mechanisms identified by Dale (1999). The use of these five different mechanisms enabled an analysis of how respondents explained CSE was adopted and reformulated as a national policy, and what its implications were. During analysis, this coding scheme was further developed. For instance, the code ‘networking mechanisms’ was developed as an additional policy transfer mechanism sub-code based on the responses of the participants. In addition, when it became apparent that very little information could be found on conceptualisations of gender-based violence in national policies, additional codes were developed that would gather information on how ‘gender’ is understood within key policy documents. Coding and analysis identified the most important mechanisms of policy adoption, and recurring themes relevant to understanding policy adoption and reformulation in this context. Both interviews and key policy documents were analysed systematically by the first author using these codes in Atlas.ti, and the analysis of selected codes was performed by the second author and then discussed to ensure internal coherence and consistency.

Findings

CSE in Ethiopia: actors and adoption

CSE was pioneered in Western Europe, particularly by Sweden, Norway, and the Netherlands (UNESCO 2018a). Because of its positive health outcomes, CSE was taken up as a global policy by international organisations such as UNESCO, and the United Nations Populations Fund (UNFPA), which are now guiding actors on CSE. This is reflected at the national level in Ethiopia; interview participants considered UNFPA, UNESCO and International Planned Parenthood Federation the most influential actors in policy processes, supported by European bilateral partners, particularly Sweden and the Netherlands.

Many national and international NGOs working on CSE are supported by these international and bilateral organisations. There are numerous national NGOs in Ethiopia,³ including in the education and SRH sectors. Most of these NGOs are largely, if not fully, dependent on external finance from international and bilateral actors. Many NGO-staff emphasised it was important for NGOs to collaborate in order to influence national policies. However, collaboration is difficult in practice as there is increasing competition among Ethiopian NGOs to gain funds due to donors’ financing systems. According to some participants, this raises concerns over transparency; with NGOs becoming less transparent about the successes and challenges of their projects out of fear of losing funds. This, in turn, limits shared learning and dissemination efforts. Interview participants from NGOs agreed that donor partners should be more transparent about the basis on which funding is allocated to projects, in order to improve coordination. Furthermore, due to the dependency on foreign funds, donor agendas are reflected more strongly than local priorities in the work of NGOs (Dupuy, Ron, and Prakash 2015; Stone 2004). This makes their involvement in policy adoption processes more contentious due to fears of imposing or advocating for imported agendas (Crossouard, Dunne, and Durrani 2017; Steiner-Khamsi 2014).

At the national government level, the most important actors in adopting CSE are the MoE and the MoH. A list of the current most relevant policy documents per Ministry is provided in Table 2.

Table 2. Overview of policies and ministries relevant to adopting CSE as a policy to address gender-based violence in education.

Name of the policy	Mentioning of CSE	Responsible Ministry
Growth Transformation Plan II (GTP II) (2016–2020)	No	Overarching policy framework of the Government of Ethiopia
Health Sector Transformation Plan V (HSTP V) (2016–2020)	No	Ministry of Health (MoH)
National Adolescent and Youth Health Strategy (2016–2020)	CSE to promote SRH, incl. to develop attitudes against gender-based violence
Education Sector Development Plan V (ESDP V) (2016–2020)	Sexuality and life skill education to prevent HIV/AIDS and increase knowledge about SRH	Ministry of Education (MoE)
The Education Sector Policy and Strategy on HIV/AIDS (2009)	Sexuality education to prevent HIV/AIDS

A representative from the MoH emphasised that Ethiopia was among the first in the region to develop a strategy specific to young peoples’ SRH (2006–2015), even though CSE was not mentioned in that strategy. The strategy is now updated to a national adolescent and youth health strategy (2016–2020), which CSE is included in. In this latest MoH-strategy, CSE is seen as a means to promote SRH, including to increase understanding of the causes of gender-based violence and develop attitudes to alleviate it. In MoE-policies, however, the inclusion of CSE remains limited to ‘sexuality and life skills education’ alone, omitting the ‘comprehensive’ element; much to the frustration of the participants who strongly advocated for CSE. According to these participants, the MoE believes that the existing curriculum already includes all of the essential information about SRH, making CSE unnecessary. Within existing MoE policies, sexuality education is seen as a means to prevent HIV/AIDS and increase knowledge about SRH but is not linked to addressing or preventing gender-based violence. Gender-based violence is predominantly mentioned as a barrier to educational access, positive health outcomes, and development processes, and not as a violation of human rights (also noted by Parkes et al. 2016).

The participants all emphasised that MoE and other federal level institutions are the most important actors in adopting CSE as a national policy, especially to integrate it into the formal education system. However, this is also where advocates of CSE encountered the most challenges in the adoption process. Participants noted the large differences between the two Ministries in terms of their commitment to CSE:

Sometimes it feels like we are working in two different countries. MoH is so progressive and open about so many issues. Even in the current adolescent and youth [health] strategy that we are developing, CSE is clearly indicated that it should be promoted among school- and out-of-school youth. But MoE is very closed about it. (International Organisation, Female, 10)

As this quote illustrates, national policy adoption is a dynamic process where multiple actors can have diverging interests (as observed by others as well: Rizvi and Lingard 2009; Ball 2016). In addition, the quote articulates the opposing views of ‘progressive’ and ‘open’ proponents (MoH) of the policy, vis-à-vis the ‘closed’ opponents (MoE). National actors who support CSE framed the MoE as the conservative ‘Other’ outside the ‘progressive’ network, who should be convinced of the benefits of the policy (see also Roodsaz 2018) – a binary that can also be observed in global discussions about CSE. The difference between the approach of the MoE and the MoH is striking, and similar findings have been reported in studies in Ghana and Kenya where the MoE was more reluctant than the MoH to adopt CSE (Panchaud et al. 2018). One explanation of this difference might be that the discursive logic (following Ball 2016) that promotes CSE based on health incentives is embraced by the MoH but might appeal less to the MoE, whose agenda is predominantly concerned with educational outcomes. Consequently, CSE continues to remain mostly dependent on NGO-led extra-curricular activities, despite the uptake of CSE in MoH-policy and support of sexuality and life skills education in MoE-policies.

Finally, it is significant to note that, overall, interview participants saw little problem with policy texts in themselves, but stressed that the enactment of policies is problematic due to limited capacity and personal or political will (in line with Parkes 2016). Participants considered that problems in enactment were first of all due to the hierarchical nature of how institutions are organised:

In an African context, including Ethiopia, a person is like an institution. This means, if someone, leave alone the minister, a strong director says ‘no’, everybody says ‘no’. In other contexts, if people disagree, they might argue. This is why I say here, a person is an institution. Because if the minister says no – there is no point in arguing. The ministry also says no. Or it could be any other institution. And I think they should be separated. (…) My view as a person and my institution view may be different. (International Organisation, Male, 6)

The participant quoted above argued that despite individual opinions and concerns, institutions should be able to accept the positive outcomes of CSE as a national policy. However, some participants were more critical about this, and highlighted that a sense of commitment at the personal level influences the policy adoption and implementation, especially in policies related to gender justice:

There are also attitudinal challenges. The policy is there, for instance my boss has a policy [about gender equality in the work place], but if he thinks I cannot perform well because I’m just a women he can easily kick me out. Despite of the policy. So those things are the challenges. It’s all about the character, individual level behavioural change is the challenge. (University, Female, 2)

Several of the female interview participants who worked on advancing CSE policies discussed personally experiencing discrimination, including by experts who are responsible for addressing gender discrimination and violence in schools – both male and female. These female participants attributed their lack of influence in the policy process and the discrimination they faced partly to the controversial nature of CSE, but also to their status as (young, unmarried) women. Thus, adopting CSE as a national policy does not merely mean ensuring that it is mentioned in policy text, but also requires the continuous challenging of hierarchical social and gender regimes at all levels of the policy adoption process (Connell 2012b; Parkes 2016).

Mechanisms of CSE policy transfer

Efforts for the adoption of CSE at the national level in Ethiopia have been particularly made through dissemination mechanisms; including networking, organising trainings, and participation in national steering groups on CSE. Participants highlighted that they set agendas and use research findings to make a case for CSE, specifically emphasising the health benefits to young people. The research findings they use include Ethiopian research, but due to the limited capacity of Ethiopian universities, NGOs often draw on research in the region funded by international donors. However, even though many participants believed the dissemination of research evidence on CSE could lead to the adoption of the policy, they found that some evidence remained ignored by the MoE, due to the challenge of limited research budgets, and the disconnect between academia and policy-making.

Related to dissemination mechanisms, international organisations appeared to promote policy adoption through facilitating networking between different actors operating at various levels, including federal levels, district offices and schools (c.f. Stone 2004). Even though such networking and coordination were challenging at times, most participants considered this approach the most promising:

This is a time when all actors need to come together and act as one. So that we can get the desirable change that we are aspiring to. All those actors need to come together, need to lobby, advocate, on behalf of implementing those policies and laws that are existing. (International NGO, Female, 11)

Networking also took place in the form of task forces and CSE steering groups (as recommended by CSE global guidance documents). For instance, task forces on CSE were set up in Ethiopia (mostly by international organisations), with participants from international organisations, universities, bilateral organisations and NGOs. These task forces typically disseminate evidence and share knowledge among members and use this to advocate for CSE in national ministries. They also seek multi-sectoral collaboration and support for CSE, in order to harmonise efforts within national policy adoption processes, and synchronise the national policies of different ministries. As such, the task forces and networking spaces are used to create a unified discursive logic about the benefits of CSE in the country (following Ball 2016), and allow new education policy actors, such as the MoH, UNFPA, and NGOs, to enter the debate. Some participants highlighted that the inclusion of young people and university staff should be better facilitated and institutionalised in these groups. They believed this would strengthen the ownership and development of policies and strategies based on young peoples’ views and Ethiopian universities’ evidence base. Youth participation was emphasised especially in reference to recent protests, where it were mainly young people who joined and voiced their socio-economic concerns. In the opinion of several stakeholders, this is evidence that policies should be more inclusive and responsive to youth priorities.

Harmonisation is another transfer mechanism through which the adoption of CSE has been advanced; referring to the fact that the MoH and the MoE are signatories of the Eastern and Southern African CSE commitment in 2013. In this commitment, ministries affirm their political will to ensure access to CSE, and to provide CSE in schools using international standards. Several actors viewed this as an important entry point to hold the ministries accountable. However, being a signatory to such a collective agreement highlighting the importance of CSE is not sufficient:

But after that [signing of the commitment], what it means is not clear… We are really asking the MoH to refer to this document. If the MoH really accepts it, there is no way the MoE can reject [CSE] anymore. (International Organisation, Male, 6)

Thanks to the signing of agreement of the CSE commitment in South Africa where 22-23 countries signed that commitment… The government intends to meet it. But unfortunately the MoE didn’t go far on the improvement itself. (International Organisation Male, 8)

Indeed, for most interview participants, their main concern remained focused on whether and to what extent the MoE will embrace CSE as a national policy.

While the strategies used by several actors in Ethiopia to advocate for CSE as a national policy are in line with harmonisation, dissemination and networking mechanisms, the same processes remained contentious. In fact, some participants considered CSE agendas a form of imposition, and were concerned it does not respond to young peoples’ priorities:

[Donors] come with agendas but it may not merge in here. They came from abroad. We really need to have our own agendas, from their local environment. But most of the time the agendas are imported. That’s a challenge because it might not resonate with the young people here. (National NGO, Male, 4)

Another participant also suggested that CSE is based on the concerns of the Western institutions and is not reflective of the Ethiopian national context:

Now, looking into gender, SRHR, where did it come from? Does this come from a minister, an NGO, UN? Which institution has brought it in? And under circumstances where you have such knowledge and experiences coming from somebody else, from somewhere else, then it’s an imposition. You are lobbying government persons and asking for it to be accepted. So then it’s not yours but somebody else’s. (…) When we talk about CSE, it was brought in and then there was reaction, and then you start negotiating. It’s not something that has come around [meaning: it is not something that has been initiated locally]. (University, Male, 13)

Participants argued that the fact that national ownership of CSE was weak in Ethiopia would have negative implications for implementation processes at various levels as well as for the sustainability of the policy. It was highlighted that this form of imposition is a result of the fact that international organisations have the funding and therefore the power to influence the government in specific ways:

Other actors, like the UN ones, are influential because they have money. I mean, that’s the reality. It’s a black and white truth. They have the money and therefore can direct the government in a certain way. (International NGO, Female, 2)

Nevertheless, other participants suggested that even if the government is susceptible to foreign influence because of donor dependency, it remains a very strong government. If ministries are not convinced about the importance of a policy they would, therefore, be likely to reject it, irrespective of where the policy has come from. Within this context, stakeholders from bilateral and international organisations mentioned that they ensure that their strategies align with, and support, government policy. As such, organisations working on CSE in Ethiopia design their projects to be ‘culturally relevant’ and ‘legally sound’, as requested by the government (see below).

Reformulation of CSE

To understand reformulation of CSE policy at the national level in Ethiopia it is important to note that conceptualisations of CSE in the MoH and MoE-policies are predominantly focused on the instrumental use of CSE: CSE ought to deliver information about adolescent SRH, which should lead to positive health outcomes. However, CSE is rarely linked to addressing gender-based violence in Ethiopian policies (whereas, in some other country contexts, gender-based violence seemed to be the main reason for adopting CSE, see Panchaud et al. 2018). In fact, concerns with gender more broadly do not go beyond access to health and education services. Gender is mostly referred to in terms of gender differences; the adolescent youth health strategy (MoH 2016) states that approaches need to be adequate for both gender categories, and the MoE (2016) policy states sexual and life skills education needs to ‘address the needs of both males and females’. Within a policy framework that is largely based upon a categorical understanding of gender (following Connell 2012b), CSE is reformulated to fit gender parity and equal health outcomes goals. Moreover, these policy goals are part of a wider developmental agenda that emphasises participation in, and access to, economic welfare and development as a policy priority for women, children, and youth (GoE 2016). Within this context, CSE is thus reformulated as a programme contributing to socio-economic development through better health outcomes, a logic that is also reflected at the school-level (see, e.g. Le Mat 2016; Schaapveld 2013).

Reformulation of CSE was furthermore influenced by what some interview participants referred to as the lack of legitimacy of national NGOs in advancing certain agendas relevant to CSE (specifically on human rights, sexual diversity, and gender equality).^4,5 Many national NGOs who used foreign funding had reformulated their activities and agendas to reflect a development instead of a rights agenda (see also Dupuy, Ron, and Prakash 2015). In the case of CSE, NGOs and other actors reformulated their CSE policy to reflect a health and development goal, rather than the advancement of rights and promotion of gender equality.⁶ Interview participants from NGOs as well as other affiliations highlighted that they re-named their organisation, project proposals, and goals in order to reflect government policy. Yet, in day-to-day practice NGOs found some space to include gender equality agendas in, for instance, trainings, task forces, and partnering with local and district level government actors.

Nevertheless, reformulation of CSE should be understood in light of resistance at federal as well as local levels. As Steiner-Khamsi (2014) theorises, the resistance of emerging policies also goes hand in hand with emphasising differences. In the case of CSE in Ethiopia, cultural and religious differences are emphasised:

But from federal level there is so much resistance. That’s the difference we see in Ethiopia. Some of the CSE [initiatives] are really successful and working and accepted by the woreda [district] level officials. But then at the federal level… (…) the argument they give is that it’s already there and they don’t want too much Western influence and the LGTBI come into picture obviously. So they don’t want to hear about it. “It’s not our culture, it’s illegal, it’s Western and doesn’t go with our culture” etc. (Bilateral Organization, Female, 3)

[CSE] is sensitive against a very sensitive society where orthodox Christianity and Muslim traditions are so strong. There are misunderstandings. All our [organisation’s] documents state that CSE should be age-appropriate, culturally sound and content-relevant. But there is always a hesitation because of the area of sexual orientation. That area has been twisted, misunderstood and misinterpreted by so many actors as [CSE] has been taken as a culture to promote homosexuality. (International Organisation, Male, 6)

Indeed, issues of sexual diversity, often referred to as an issue of ‘cultural relevance’, seem most pressing for the adoption of CSE as national education policy. Interestingly, participants reported that most resistance is met at the federal level, whereas woreda (district) level government offices are often favourable towards adopting CSE. The above quotes also illustrate how CSE has become conflated with programmes promoting sexual diversity. Indeed, the mentioning of sexual diversity in the global CSE policy might trigger political resistance to the policy, as part of an agenda to counter Western imposition and exercise control (following Weiss and Bosia 2013). Arguably, opponents of the policy may overemphasise the ‘cultural differences’ related to homosexuality in order to reject the policy, at the cost of creating opportunities for a dialogue on what policy priorities should be and how CSE may respond to this. Interestingly, global guidelines legitimise such cultural difference arguments by emphasising the ‘progressive’ and ‘scientific’ norms of the policy, which these documents contrast to resistance from ‘traditional’ actors – depoliticising any other agendas for resistance.

Within this context, some interview participants suggested re-naming CSE to, for instance, ‘life skills’ education; a name that would take away some of its controversies in order to avoid resistance and increase the likelihood of CSE being adopted as national education policy. For interview participants, the controversy of CSE also almost solely concerned issues of sexual diversity. One participant, therefore, considered not mentioning diversity at all, if that would allow the remaining of the CSE curriculum to be integrated into the formal curriculum:

The international community accepts [homosexuality] as a right, but the Ethiopian government has criminalized it. The Europeans expect something different than the Ethiopian government. You better leave these kinds of things, and focus on other parts of CSE. Otherwise, a single word may spoil it. (International Organisation, Male, 8)

Others were more hesitant about removing homosexuality from the curriculum or the ‘comprehensive’ from the name ‘CSE’. They considered it essential to convince all actors about the importance of including information about sexual diversity in the curriculum and believed that re-naming the educational programme would not avoid this controversy. There is therefore continued dialogue to define, conceptualise and contextualise CSE. This is heavily influenced by debates among influential actors about national ownership and the politics of diversity, and engages little with young men and young women’s own priorities. In the meantime, the implementation of CSE programmes remains largely NGO-initiated, limited in scope and often dependent on short-term donor funding.

Discussion

This paper presented how and why CSE is adopted in Ethiopia. Our analysis has revealed that CSE is a largely donor-driven agenda in Ethiopia, and is rarely mentioned in national policies. Importantly, by including a conceptual analysis of adoption and reformulation of CSE policy, the paper identifies that national policies on CSE restrict their aims to promoting health and development outcomes. This neglects the ways CSE could address gender-based violence in schools and bypasses priorities voiced by young men and young women.

International actors such as UNESCO and UNFPA have advanced CSE as a global policy, with the potential to address gender-based violence. Dissemination in global guidance documents, regional harmonisation and national networking strategies has been most influential in the policy adoption of CSE at the national level in Ethiopia, in particular, the MoH’s adolescent and youth health policy. However, dissemination and harmonisation strategies are at times considered as an imposition in a context where CSE is so donor-driven (Dale 1999; Steiner-Khamsi 2014).

It is for similar reasons that CSE is resisted, with an emphasis on cultural, religious, and legal differences, particularly in reference to sexual diversity. This is not unique to Ethiopia (see De Haas 2017; Huaynoca et al. 2014; Crossouard, Dunne, and Durrani 2017). Participants highlighted that arguments of cultural and religious difference were particularly advanced by the MoE at the federal level, revealing a discrepancy with what they considered a more progressive MoH and offices at district and local levels. This inconsistency confirms again that policy adoption processes are diverse and can be contested within nation-states (c.f. Stone 2004). It also reflects a wider tendency of MoEs being more likely to reject CSE than MoHs (c.f. Panchaud et al. 2018). One explanation for this might be that the policy networks that create a ‘unified discursive logic’ (following Ball 2016) on the benefits of CSE advance a logic based on health outcomes that might be more appealing to the MoH than the MoE; the latter being more concerned with educational outcomes. Rejection of policies related to sexual diversity might furthermore emerge from a political agenda to reject Western imposition and exercise domestic control (Weiss and Bosia 2013). However, such ‘political homophobia’ (Weiss and Bosia 2013) might be concealed in an argument of ‘cultural difference’. This ‘cultural difference’ argument is made possible by the global framing of CSE as a ‘progressive’ policy resisted by the ‘traditional Other’, at the cost of focusing on the policy priorities that CSE might be able to address (see also Grewal and Kaplan 2001; Roodsaz 2018).

The fact that gender-based violence in education needs to be addressed in Ethiopia is undeniable (Altinyelken and Le Mat 2018). However, uptake of CSE in national policies in Ethiopia in its current form reflects the health-based SRHR agenda CSE originally emerged from, and discussion of gender relations seems to remain an ‘add on’ rather than a core component of the policy. In fact, CSE as a policy in Ethiopia is typically conflated with a programme to promote sexual diversity or narrowed down to merely increase knowledge about reproductive health. Wider application of addressing gender-based violence through CSE remains absent, possibly limiting its emancipatory potential in terms of contextualised engagement with gender and power relations and addressing gender-based violence. Without doubt, a legal and policy environment where work on gender equality, children’s and human rights are limited to developmental goals, leaves little space for educational programmes to address the root causes of gender-based violence. This is especially unfortunate because education can make substantial contributions to addressing gender-based violence through critical consciousness raising in curriculum and pedagogy, and educating youth about justice and equity (Chege 2007; Connell 2012a; Parkes et al. 2017). Hence, ways need to be found to advance an emancipatory educational agenda that can include CSE, in order to better address gender-based violence in education. Interestingly, a key concern in gender-based violence policy research has been failures in translating policies from national to local levels. In Ethiopia the opposite seems to be the case; our findings indicated that youth groups and NGOs are seeking ways for their agendas to be taken up by national policies, despite the restrictive legislative context. Recent discussion about updating the 2009 Charities Proclamation to allow NGOs to work from rights-based agendas might thus be a step in a positive direction.

Gaining conceptual clarity (Connell 2012b; Unterhalter and North 2017) may lead to more coherent and harmonised policy adoption processes. However, definitional clarity provided in global guideline documents (see UNESCO 2018a; UNESCO and UN Women 2016) does not necessarily mean agendas are understood in similar ways, responsive to local (young peoples’) priorities (Fetene and Dimitriadis 2010; Steiner-Khamsi 2014) or adopted with similar rationales. Indeed, to resist the imposition of foreign policies, much debate in Ethiopia concerns the defining, naming, and re-formulation of CSE in order to align with national agendas. Importantly, it is in moments of resistance that the debates around CSE as a national policy have increased, which may possibly open up ways to include discussions about gender, gender relations and gender-based violence. In fact, it is through networking strategies that NGOs, universities, youth groups, and other partner organisations increasingly collaborate in the form of steering groups. This has resulted in strengthened capacity to advance concerns that would formally go beyond the scope of their mandate. In this way, concerns about what ‘comprehensive’ means or ought to mean may provide the necessary space for multiple actors to come together and identify which problems need to be addressed and how the policy responds to that (Steiner-Khamsi 2014). Importantly, such spaces should not only focus on questions of CSE adoption and reformulation alone, but especially concern addressing the priorities of those CSE is designed for: young men and young women. This can only be done by including their participation in debates, and future efforts must pay attention to how young men and young women can be systematically and significantly included in steering groups and platforms at governmental levels.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Marielle L. J. Le Mat

Marielle L. J. Le Mat is a PhD candidate at the University of Amsterdam’s Child Development and Education department. Her research explores how comprehensive sexuality education might address gender based violence in schools in Ethiopia. Her main research interests include the promotion of gender equality and addressing of gender-based violence through formal and non-formal education, global education policies, critical discussions of gender and education in its broadest sense, and its relations to citizenship, social justice, and social exclusion.

Hülya K. Altinyelken

Hülya K. Altinyelken is an Assistant Professor of Comparative Education at the Department of Child Development and Education, the University of Amsterdam. Her research interests cover a wide range of topics including Global Education Reform Movement with a focus on adoption and implementation of child-centred pedagogy in different country contexts, analysing in particular how teacher agency and context modifies global education reforms. Her current interests and projects particularly focus on citizenship education, identity, sense of belonging and social integration of Muslim youth in Europe.

Henny M. W. Bos

Henny M. W. Bos is professor of Sexual and Gender Diversity in Families and Youth at the University of Amsterdam. Henny Bos carries out research into same-sex parenting families and sexual minority youth. Bos is the co-investigator of the U.S. National Longitudinal Lesbian Family Study (NLLFS), the longest-running prospective study on lesbian mothers who had children as the result of donor insemination during the lesbian baby boom of the 1980s. Main areas in her research are  lesbian, gay and transgender parenthood, and  sexual minority and gender nonconforming youth, in the U.S. and the Netherlands.

Monique L. L. Volman

Monique L. L. Volman is a full professor of education. She is the program leader of the Educational Sciences research program of the Research Institute of Child Development and Education at the University of Amsterdam. Main areas in her research are learning environments for meaningful learning, diversity and the use of ICT in education, issues which she approaches from a socio-cultural theoretical perspective. In her work she aims to build bridges between educational theory and practice by applying and further developing methodologies for collaborative design research, in which teachers and researchers collaboratively develop and evaluate theoretically-informed innovations.

Notes

1. UNESCO’s International Technical Guidelines on Sexuality Education defines CSE as: ‘[CSE] aims to equip children and young people with knowledge, skills, attitudes and values that will empower them to: realize their health, well-being and dignity; develop respectful social and sexual relationships; consider how their choices affect their own well-being and that of others; and, understand and ensure the protection of their rights throughout their lives.’ (UNESCO 2018a, 16); Global Guidance on SRGBV defines CSE as: ‘[CSE] aims to equip children and young people with the knowledge, skills and values about relationships, gender, sexuality and violence to make informed and healthy choices about their sexual and social relationships’ (UNESCO and UN Women 2016, 63).

2. CSE has been criticised for promoting open discussions about sensitive issues such as masturbation and homosexuality, and blamed for ‘sexualising’ young people.

3. In 2009, there were 2275 local NGOs and 266 International NGOs registered in the country. After the adoption of the ‘CSO proclamation 2010ʹ, this number dropped with 45% to 1701 local NGOs and 262 International NGOs (Dupuy, Ron, and Prakash 2015).

4. Referring to the Proclamation No. 621/2009, national charities and societies (including NGOs and CSOs) that receive more than 90% of their funding from abroad are limited in working on: the advancement of human and democratic rights (Art.14, j); the promotion of equality of nations, nationalities and peoples and that of gender and religion (Art.14,k); the promotion of the rights of the disabled and children’s rights (Art.14,l) the promotion of conflict resolution or reconciliation (Art.14,m); and the promotion of the efficiency of the justice and law enforcement services (Art.14,n).

5. At the time of writing this paper, there is a discussion of updating the proclamation and allowing NGOs to work from rights-based perspectives. This was not yet the case however at the time of data collection and analysis – hence this paper reports on the situation under the Proclamation No. 621/2009.

6. In light of Proclamation Art.14 (j), (k), and (l).