Dynamics of gender justice, conflict and social cohesion: analysing educational reforms in Pakistan

This paper analyses the role of national level reforms in the school curriculum and initial teacher education in gender justice in con ﬂ ict-a ﬀ ected Pakistan, using a multidisciplinary framework applied to multiple data sets from selected teacher education institutions in Sindh. The school curriculum texts analysed potentially perpe- tuate gender injustice and foster con ﬂ ict. While teacher education reforms o ﬀ er the potential for transformative gender justice, gender remains peripheral in initial teacher education curriculum. Furthermore, institutional practices entrench gendered norms. Lecturers ’ and teachers ’ limited understanding of their role and capacity for transformative gender justice pose challenges to education for gender justice, social cohesion and con ﬂ ict mi- tigation. Informed by our understanding of gender as a social construct, multiple strategies within and beyond education are o ﬀ ered towards transformative gender justice.


Introduction
The 2030 sustainable development agenda frames the sustainable development goal (SDG) 4 for education as the key driver for the realisation of the remaining 16 SDGs, including SDG 5 that focuses on achieving gender equality and empowering all women and girls. Furthermore, education is seen crucial to the development of peaceful societies, while the unequal distribution of education is viewed to foster conflict (UNESCO, 2016a). The interplay between education policies, gender (in)equality and conflict is reflected in the global indicator 4.7 which focuses on education for gender equality and the 'promotion of a culture of peace and non-violence', and requires identifying the extent to which national education policies, curricula, teacher education and student assessment mainstream gender equality (ibid: 397).
Against the above backdrop, international analysis indicates a greater likelihood of armed conflict in contexts of high levels of gender inequality in education (FHI 360 EPDC, 2015). Nevertheless, a review of literature on the relationship between gender, education and conflict highlights a range of theoretical and empirical gaps. For example, studies exploring the relationship between education, conflict and social cohesion, with very few exceptions (e.g. Durrani and Dunne, 2010) seldom focus on gender. Likewise, the literature on conflict and peace tends to focus largely on the gendered impacts of conflict but ignore the gendered drivers of conflict (Wright, 2014). For example, much of the literature talks about young men as most likely to be perpetrators and victims of violence, and women to be victims of gender-based violence. This limits understanding of the ways the social construction of gender identities is connected to conflict. Furthermore, limited empirical evidence exists regarding how educational reforms, including those in the curriculum and teacher education, may contribute to social cohesion and conflict mitigation (Horner et al., 2015). In addition, conceptualisation and measurement of gender in global education and development discourses have been critiqued (Dunne, 2009) for being too focused on girls in ways that 'limits the potential for discussing complex gender issues that affect the possibilities for gender equity' (Monkman and Hoffman, 2013: 63). Finally, the need to subject assumptions regarding the positive relationship between education and gender equality to empirical scrutiny is highlighted (Khurshid, 2016). This paper addresses the substantive gaps in the literature in relation to the dynamics between education, gender and conflict by illustrating the interconnections between gender inequality in education and conflict. Theoretically, it addresses narrow conceptualisation of gender in global education policy discourses by utilising a multidisciplinary and innovative framework, drawing on feminist political philosophy and gender, conflict and education studies to analyse the relationship between education, gender justice and social cohesion in the conflict-affected setting of Pakistan. Although the terms 'gender (in) equality ' and 'gender (in)justice' are often used interchangeably, in line with our theoretical framework discussed in Section 2.3, we prefer to use the term 'gender justice' to indicate both the ending of as well as the provision of redress for gender injustices that results in unequal gender power relations (Goetz, 2007). Specifically, we offer empirical insights into the potential of large-scale reforms in the school curriculum and initial teacher education (ITE) in supporting teachers for mitigating gender injustices in Pakistan. The empirical study reported deploys a multi-method approach, generating data from a range of educational stakeholders in Sindh. The findings offered have wider policy implications beyond Pakistan. The paper draws on a larger study designed to explore the ways social cohesion is integrated into the country's education sector at macro and micro level and the role of teachers in promoting social harmony.
The next section reviews the literature, culminating in offering the 4Rs framework of redistribution, recognition, representation and reconciliation. The subsequent two sections analyse gender justice first in the socio-cultural and political economy of Pakistan and then in education system through the lens of the 4Rs. A discussion of the context of education reforms and research methods follows. The findings are offered in three sections. We first analyse the ways gender and conflict are represented in the curriculum texts. Second, we offer an account of the possibilities of teacher education for transformative gender justice. Finally, we present an analysis of how teachers and lecturers understood gender justice and its relationship to education and conflict. The concluding section problematises the limits of educational interventions for transformative gender justice and offers multi-dimensional implications for the role of education in fostering gender justice and social cohesion.

Literature review and theoretical framework
This section first locates social cohesion and gender within educational research literature and concludes with the presentation of a theoretical framework for analysing gender justice in education in conflict-affected settings.

Education, social cohesion and conflict
The promotion of social cohesion is positioned as a key purpose of education, particularly in younger nations where nation-building is a key priority (Heyneman and Todoric-Bebic, 2000), and countries affected by conflict (Tawil et al., 2004). Interpretations of social cohesion are diverse. While some definitions stress 'common identity and a sense of belonging', others emphasise 'an active civil society' or 'equality and social solidarity' (Green and Janmaat, 2016: 171). Most definitions link the term to a societal property, based on the promotion of positive relationships, trust, solidarity, respect, (structural) inclusion, collectivity, and common purpose and associate it with social justice and equity.
In broad terms, education inequality includes inequality in access, quality and outcomes of education; a lack of respect for difference in educational structures, processes and content; and inequality in participation in educational decision-making. Educational inequality, as measured by skill inequality, is positively correlated with violent crime and political unrest and negatively correlated with political and civil liberties (Green et al., 2006). Furthermore, education inequality interacts with wider social disparities, intensifying the risk of conflict (UNESCO, 2016a). Likewise, greater level of gender inequality in education, as calculated from group differences in mean years of schooling, is reported to be associated with intra-and inter-state conflicts (FHI 360 EPDC, 2015). Additionally, countries characterised by high levels of gender inequality, as measured by fertility rates and participation in the labour force, are more likely to experience internal conflict (Caprioli, 2005). What might explain this link? Rather than seeking explanations of conflict and violence in the 'manifestations of gender inequality', analysis needs to focus on 'ideas about masculinities and femininities which are used to justify these inequalities' (Wright, 2014: 5). In other words, masculinities and femininities produce genderings and are product and productive of violence (Sjoberg, 2013). The link between conflict and gender injustice is better understood through Galtung (1990) conceptualisation of conflict which includes not only (i) armed conflict or direct violence but also (ii) structural violence-social injustices embedded into social structures-and (iii) cultural violence-any part of culture that renders direct violence or structural violence acceptable in society. Furthermore, gender is integral to 'structural and cultural violence for gender forms the basis of structural inequality' (Caprioli, 2005: 164).
The impact of education on social cohesion is multifaceted, with education being both a conflict driver and a means to mitigate it (Bush and Saltarelli, 2000;Davies, 2010;Smith et al., 2011). Linking education to conflict, UNESCO (2016a: 103) contends that if curriculum texts and teacher pedagogies 'inculcate prejudice, intolerance and a distorted view of history, they can become breeding grounds for violence'. By contrast, education strengthens social cohesion when it fosters a sense of inclusion, participation and respect for diverse social groups . Likewise, 'the content and quality of education and knowledge provided are key to reducing violence', by supporting 'women, girls, boys and men to understand, question and challenge gendered norms and behaviour that underpin forms of violence' (UNESCO 2016b: 52). Teachers' and educators' role as agents of socialisation is vital in validating masculinities that are non-violent, caring and favour gender justice (Connell, 2002;Wright, 2014).

Education and gender justice
Gender justice has been given a renewed emphasis in SDG 4 but concerns remain regarding its measurement. For decades, the dominant method of measuring gender justice has remained the gender parity index in education participation and attainment, which frames gender as a noun (Unterhalter, 2012). This limits the identification of gender as a social process, with potential to both reproduce and transform social injustices (Dunne, 2009). By contrast, Butler (1990) frames gender as a verb, performed discursively within the constraints of particular social contexts and produced within unequal power relations. Kessler et al. (1985) argue that schools as institutions are characterised by a gender regime constituted through everyday practices that construct a range of femininities and masculinities ranked in terms of prestige and power. Indeed, Durrani (2008) highlight, through multi-site ethnographic casestudies of schools in Pakistan that the gender regime in schools maintained, reproduced and reinforced the gender hierarchies that characterise Pakistan. Gender identities are constrained therefore both by access to resources and the formal and informal regulation of social institutions including the school (Dunne, 2009).
Despite the widely held assumptions between education and women's empowerment, more often than not, education, both in the Global North (e.g. Arnot, 2006;Kennelly and Llewellyn, 2011) and the Global South (e.g. Dunne, 2007;Durrani, 2008), sanctions and perpetuates injustices, through the privileging of the traditionally masculine structures, practices, forms of identifications and gendered teacher pedagogies. With respect to masculine structures, there is 'a higher concentration of men in senior management positions in school boards and education ministries in rich and poor countries alike' (UNESCO, 2016b: 43). Even when women are headteachers, governing boards often constitutes predominantly men, as in Karachi, Pakistan, resulting in their restricted decision-making power (Kirk, 2004). Second, school practices sustain the gendered, hierarchical organisational patterns through minimal teacher interaction across gender boundaries, as was the case in Botswana and Ghana (Dunne, 2007). Third, schools actively promote gendered identifications both through the official and the hidden curriculum. For example, while curriculum texts in Pakistan naturalised home as a woman's legitimate, ideological and physical space, female teachers and adolescent girls remained inside the school premises, guarded by a male gate-keeper (Durrani, 2008). By contrast, male teachers moved out of school at will and boys too were allowed to buy snacks from outside during lunch-break. Fourth, while pedagogic practices are central to the promotion of gender justice, teachers may not use appropriate and gender-sensitive language or provide equal opportunities to boys and girls to speak in class (UNESCO, 2016a). In Pakistan, teachers reinforced gendered textbook messages despite being able to identify curricular content as gendered (Durrani, 2008). In urban Sindh, Kirk (2004) shows that female teachers perpetuated existing gender asymmetries between girls and boys. Finally, schools are also institutions where physical (e.g. corporal punishment), symbolic (e.g. gender stereotypes) and sexual violence (e.g. sex for grades) is played out and normalised (UNESCO, 2016b). Likewise, teachers are both victims and perpetrators of (gender) violence in settings both affected and not by war zones (Horner et al., 2015).
When education explicitly aims to promote gender justice through all of its components, it may open up possibilities for transformative gender justice (Durrani et al., 2017a). Of particular relevance to this paper is the ways curriculum and teacher education reforms have supported teachers in mitigating gender injustices. Promoting gender justice requires teachers taking it seriously within their practice and critically interrogating normative gendered discourses within the curriculum and the wider community (Horner et al., 2015). Destabilising the gendered curriculum messages is important, as UNESCO (2016c) notes that the representation of gendered identities and gender norms in textbook remains a strong barrier to gender justice in education.
Although limited, examples of curriculum interventions contributing to shifting attitudes and norms towards gender exist. In India, Bajaj (2012) reported that curricular innovations focusing on gender role and principles of human rights have helped in bringing about positive changes in attitudes towards gender amongst students and teachers. She attributes the success of human rights education (HRE) in the promotion of gender justice to a range of factors. First, HRE is integrated by important national bodies pertinent to the curriculum and teachers. Second, the National Curriculum Framework is informed by principles and values of global citizenship, co-existence and transformative action. Third, effective textbook development and teacher training were crucial in enabling teachers playing an active role in modelling the 'outcomes of HRE through their own changed behaviour' (Bajaj, 2012: 13). Fourth, the purposeful or unintentional linking of HRE to lessons of religious morality was cited as a facilitator. Similar changes in attitudes towards gender are also noted by Murphy-Graham (2008) in Honduras. She links gender empowerment outcomes to a simultaneous focus on knowledge, self-confidence and gender awareness in educational programmes. Mainstreaming gender across all discipline areas and pedagogies focused on dialogue and debate and project-based learning were highlighted as key facilitators.
Professional development opportunities that aim to enhance teachers' agency for gender justice remain vital. In Pakistan, the capacity of teacher educators in teaching issues related to social harmony is limited (Halai and Durrani, 2017). Teacher educators saw contextual issues of social injustice largely at the margins of the curriculum or outside of their remit and therefore to be brushed aside. Even when they considered such issues as significant, they appeared to lack systematic pedagogic strategies to deal with contentious social issues. With a particular focus on teacher agency for gender justice, shifting teacher's own gendered identities remains a serious challenge in Pakistan (Halai, 2011). The impact of training on gender sensitivity remains negligible, unless teachers are able to embody and enact gender justice in everyday interactions, in and outside classroom. However, Halai (2011) also warned that professional development initiatives aiming to enhance teachers' awareness of gender is much more challenging in Pakistan and such initiatives are not likely to be very successful unless teachers' capabilities to question social and cultural hierarchies are enhanced. While training for gender justice is often provided as part of continuing professional development (CPD) or in-service training, often supported or facilitated by international agencies or local non-governmental organisations (Durrani et al., 2017a), the upscaling and institutionalisation of gender sensitive training in pre-service education institutions is a considerable challenge (Education for Change, 2011). A positive example of an intervention that appears to have supported teachers in recognising stereotypical gender beliefs and practices within curriculum texts and society is the Strengthening Teacher Education in Pakistan (STEP) programme implemented in selected districts/schools in Sindh, Balochistan and Gilgit Baltistan. STEP incorporates gender justice across its activities including curriculum and learning materials for professional development, induction of course participants and project staff, documentation and monitoring and evaluation processes and community engagement is strategically utilised to promote gender justice (Ashraf and Waqar, 2012).
In concluding this section, we draw attention to socio-cultural factors including religious beliefs, which may hinder a transformative gender justice agenda. For example, while governments in Nigeria and Tanzania have suggested the incorporation of sex education in the curriculum, teachers have resisted teaching safe sexual practices and religious groups have hindered the integration of sex education in the school curriculum (Education for Change, 2011).

Framework for analysing gender justice in education
A framework of 4Rs-redistribution, recognition, representation and reconciliation-for analysing gender justice in and through education in conflict-affected contexts is presented, drawing on Fraser (1995Fraser ( , 2005 and Novelli et al. (2015).
Nancy Fraser (2005: 73) theorises social justice as 'parity of participation', achievable only in the absence of three analytically distinct yet interconnected forms of injustice. The first kind of injustice is socio-economic, and redressing it requires redistribution of material resources, goods or services. Fraser's second kind of injustice is cultural or symbolic which results from cultural domination, misrecognition and disrespect and remedying it requires the recognition of marginalised social groups and affirmation of difference. The third injustice is misrepresentation, which results from political marginalisation and the exclusion of specific individuals and groups from decision-making processes and institutions and from membership of the political community at local, national or global levels (Fraser, 2005). Remedies based on representation seek a transformative politics of framing to encourage parity of participation. Novelli et al. (2015: 10) add a fourth R-reconciliation-to the above framework by arguing that for 'conflict-affected and post-conflict contexts, there is a need for processes of reconciliation, so that historic and present tensions, grievances and injustices are dealt with to build a more sustainable peaceful society'. Reconciliation is defined as a process of building positive relationships for tackling issues of mistrust, prejudice and intolerance and bringing about substantial cultural, attitudinal, social, economic and political change .
The above framework offers useful insights for analysing gender justice in education in conflict-affected contexts. The division of labour between paid/productive and unpaid/reproductive and domestic labour, with women primarily assigned responsibility for the latter produces gender-specific distributive injustice (Fraser, 1995). Redistributive gender justice in education therefore requires equity and non-discrimination in education access, resources, and outcomes for females and males. This would require affirmative action for the redistribution of resources to the most marginalised girls/women such as fee waiver, free textbooks, uniform, transportation and school meals and stipends.
Gender being a status differentiation constructs norms that privilege attributes and practices seen as masculine and denigrate those viewed as feminine (Fraser, 1998). The institutionalisation of these norms subject women to recognitive injustice whose remedy in education requires respect for and affirmation of inclusive gender identities in education content, structures and processes.
Recognitive injustice contributes to women's 'exclusion or marginalisation in public spheres and deliberative bodies' (Fraser, 1998: 2). Likewise, women's gendered incorporation into various communities of belonging, such as ethnic, religious and national (Durrani et al., 2017b), further accentuates their marginalisation in the political sphere, contributing to representational injustice. Representative gender justice in education would require that both female and male education leaders, curriculum and textbook personnel and teachers have equity in education decision-making. Furthermore, policy decisions relating to gender equality would need to be made in consultation with women and men in the community where those polices are enacted and not solely made by governments and international organisations (DeJaeghere et al., 2013).
With respect to reconciliation, given that 'masculinities are the forms in which many dynamics of violence take shape', a change in the gender norms that underpin conflict would require a change in the identifications of 'ideal' manhood and womanhood along with the structures that both uphold gender norms and are sustained by them (Connell, 2002, 34). Sustainable peace, that is the absence of structural violence and the conditions to eliminate the causes of violence (Smith et al., 2011: 13), requires the disruption of and shift in the 'social shaping of genders' (Cockburn, 2013: 445). Reconciliation with respect to gender relations in education involves replacing hegemonic masculinities linked to violence and domination with masculinities associated with nonviolence, mutual respect and equality.
Forms of injustice and aspects of social justice do not neatly categorise, as the preceding discussion might suggest. The compartmentalisation of social justice along the 4Rs is for analytical clarity only. In practice, axes of injustice intersect. For example, 'maldistribution and misrecognition conspire to subvert the principle of equal political voice for every citizen' and those subjected to 'misrepresentation are vulnerable to injustices of status and class' (Fraser, 2005: 79). By contrast, reconciliation may gloss over difference in pursuit of building harmony. Thus, in practice all 4Rs are interwoven, and can be both reinforcing or in tension. Importantly, gender intersects with other markers of identities, for example, class, race, religion, ethnicity, location and age, resulting in differential experiences of gender injustice. Thus, promoting gender justice requires tackling multiple, overlapping inequities.

Gender-justice in the social, cultural and political economy of Pakistan
Pakistan is a federal parliamentary republic, divided into four provinces-Punjab, Sindh, Khyber Pakhtunkhwa (KP) and Balochistan-a capital territory, Islamabad, two autonomous territories-Gilgit-Baltistan and Azad Jammu and Kashmir-and Federally Administered Tribal Areas (FATA). The population derives its commonality from Islamic faith, followed by 96% of Pakistan's estimated 193 million citizens (NEMIS-AEPAM, 2016).
Since its traumatic partition from British India along religious lines in 1947, amidst mass migration and communal violence, Pakistan has experienced conflict several times. The two nuclear-armed countries, Pakistan and India, have clashed violently on four occasions-1948, 1965, 1971 and 1999 and have an enduring rivalry, particularly over the status of Kashmir. These historical and on-going relations with India make religion/Islam and the military the key features of Pakistani identity. The military has directly ruled the country on four occasions and it remains very powerful during civilian rule. In 2017, Pakistan ranked 17th on the Fragile States Index, prepared by the Fund for Peace and has remained between 9th and 17th place since 2006. This instability relates to Pakistan's role as a frontline state in the 'War on Terror' and the ensuing terrorism and militancy. According to the South Asian Terrorism Portal, between 2003 and 24th September 2017 Pakistan has suffered a total of 62483 fatalities including 21912 civilian losses. 1 The 4Rs framework, presented in Section 2.3, offers a productive lens for situating gender-justice in Pakistan's socio-cultural and political economy. With respect to redistribution, huge disparities in access to resources based on gender exist which also intersect with wealth, ethnicity, region/province, and location (urban vs rural). Distributive gender injustices marginalise women, particularly poor women in rural areas, with respect to human development, access to services and economic opportunities (Durrani et al., 2017a). With only 25% of women, compared to 85% of men, being part of the labour force, Pakistan holds the second last position on the Economic Participation and Opportunity subindex of the Gender Gap Index (World Economic Forum, 2016). This estimation, of course, does not take into account the valuable yet 'unrecognised labour' that women render within households and other spaces. As such, the Gender Gap Index is a poor measure of women's labour.
Recognitive injustice relates to the strong overlap between Pakistani identity and Islam which has led to the management of religious thought and practice as a key mechanism in the gendered production of the ideal Pakistani citizen. Dominant gender norms associate valorised masculinities with combat and the military and the protection of the territorial and ideological boundaries of the nation and idealised femininities with carrier of 'honour' (Durrani et al., 2017b). Gendered norms are therefore violently enforced on women.
With respect to representative gender injustice, affirmative action has improved the political representation of women in the national, provincial and local governments. However, they continue to have a lower voter turnout and reduced likelihood of winning open seats. Furthermore, the dominance of the military reinforces dominant masculinities with the 'protector' and put women 'in a subordinate position of dependence and obedience' (Young, 2005: 17).
Reconciliatory gender injustice relates to 'a covert gendered social contract' between the Pakistani state and men, which sub-contracts the monopoly over violence to men allowing them to 'discipline, punish, and even kill' women for the preservation of the Pakistani moral code (Brohi, 2013: 214 and 216). The violent regulation of women is manifested in mob lynching, acid attacks, abduction, murder, (gang) rape and harassment at the workplace, despite the existence of protective legislation (Durrani et al., 2017a).

Education, conflict and gender justice
A review of Pakistan's educational context indicates the ways gender injustice is linked to conflict and education as discussed below. With 838 attacks on education between 2009 and 2012, Pakistan has experienced the highest number of targeted attacks against educational institutions (GCPEA, 2014). Schools have been closed, burned or blown up and students and teachers have been abducted and killed.
The violence against education illustrates its gendered dimension. In 2009, the Taliban militants banned girls' schooling in Swat Valley in KP, resulting in the exclusion of around 120,000 girls and 8000 female teachers from schools (GCPEA, 2014). By contrast, male students have been 'recruited, lured or abducted' to perpetrate violence (ibid: 19). Additionally, in the attack on Army Public School in Peshawar in 2014, in which 132 students lost their lives, the Taliban killed youthful boys in middle and upper secondary classes en masse, considering them as legitimate targets but female students both children and adolescents were largely spared.
Moving specifically to a gender justice lens in education, Pakistan has taken a range of measures to promote redistributive gender justice, including waiver of tuition fees, free distribution of textbooks, and stipend for adolescent girls. Since 2010, through the insertion of Article 25-A in the Constitution, free and compulsory access to education for children aged 5-16 years is mandatory. Between 1998-2015, gender disparity in primary education has reduced across Pakistan (Umar and Asghar, 2017). While Punjab has ensured parity, Sindh is close to achieving parity, KP is lagging behind and Balochistan and FATA are considerably behind; urban Pakistan has already achieved parity but considerable progress is needed in rural areas (ibid). Although national comparisons of the impact of stipends on redistributive gender justice are not available, in Punjab evaluation of the programme after four years of implementation indicated that female enrolment rates had increased from 11% to 32% and girls in stipend districts were more likely to progress through and complete middle school (Independent Evaluation Group, 2011). Evaluation of the programme in KP found rather small impacts, with the programme increasing female education only by a modest 0.03% (Ahmad and Zeeshan, 2014).
A gender-desegregated analysis of a range of education indicators reveals disparities at the expense of girls (see Table 1). However, national averages mask the intersection of gender with other markers of marginalisation-poverty, rurality and regional disparities-which, when taken into account, identify poor girls in rural locations, particularly in Balochistan, FATA and KP as most marginalised (Durrani et al., 2017a).
With the exception of Islamabad, gender parity in enrolment in primary and secondary education remains elusive, with greater gender disparities observed in regions affected by conflict-FATA, KP and Balochistan (NEMIS-AEPAM, 2016). Furthermore, boys outperform girls in all subjects but these gaps are much wider among children from the poorest households where poor girls are learning the least (ITA, 2015), as they are required to contribute to household chores and care of younger siblings, which restricts the time they spend on learning and results in irregular attendance at school.
A pro-male bias exists in parental decision to enrol and educational spending (Aslam and Kingdon, 2008). Because in Pakistan sons are responsible for the welfare of parents in old-age, the rates of return for daughters' education are low. Additionally, the family's 'respectability' is at stake in sending daughters to schools, particularly after puberty, in rural communities because gender norms require that females observe modest behaviour (Purewal and Hashmi, 2014). Contradicting widely held beliefs about the incompatibility between Islamic values and women's education, Bradley and Saigol (2012: 686) report that in Pakistan 'both religious and secular respondents support women's education, but not always for the same reasons'.
Access to education is also contingent upon the availability of a school within commutable distance. The government schools, particularly at secondary level, are typically segregated into those for boys with male teachers and for girls with female teachers. Boys' schools outnumber girls' schools nationally, provincially and across urban and rural locations (UNESCO, 2010). Arguments for this segregation relate to the perceived cultural traditions of Pakistani society. It is believed that the prevailing social norms would not allow girls and boys to study together and that parents would not send their daughters to school if they were taught by a male teacher. Though both arguments seem plausible, they are empirically untested (Gazdar, 1999). While for learners this may reinforce gender norms regarding spatial boundaries between males and females, an additional impact is that teachers and managers think that once boys and girls have access to schools there is no gender injustice in the process of curriculum delivery (Halai, 2011).
With respect to recognitive gender justice, UNESCO (2016c) reports pervasive gender bias in Pakistani textbooks. Textbooks depict a strong gender bias favouring males, reinforce familial hierarchies that subordinate women to men and portray women in restricted and homogenised ways (Durrani, 2008). For example, women/girls are exclusively depicted wearing the national dress, with their heads covered and in domestic spaces often engaged in domestic work or as mothers. Ullah and Skelton (2013) report gender biases in 24 textbooks (Urdu, English and Social Studies), implying that over a decade long efforts by the Ministry of Education and international organisations to eliminate gender bias in education have had negligible impact.
A particular challenge concerning representative gender justice is gender parity in the workforce, as the availability of female teachers correlate positively with girls' enrolment (Durrani et al., 2017a). However, at all levels of compulsory education, the proportion of female teachers in state schools is low (Durrani et al., 2017a). The regions with wide gender gaps in enrolment-FATA, Balochistan and KP-have the lowest proportions of female teachers (NEMIS-AEPAM, 2016).
Challenges of reconciliation arise from the ways the curriculum texts draw boundaries between the Pakistani Muslim-self and the antagonistic non-Muslim 'other' for forging national unity, associating Pakistani identity with the military and the readiness to fight and linking dominant masculinities with the 'protector' (Durrani and Dunne, 2010). Venerated national icons are all men engaged in discursive or violent battles with the antagonistic external 'other'--India-to protect the 'motherland' or religious leaders tasked with protecting the ideological boundaries of the nation. Teachers have a key role in subverting such valorised masculinities. However, because of gender-segregated schooling in Pakistan, teachers tend to believe gender issues are tackled once learners access school (Halai, 2011). Importantly, teachers in Pakistan have a very strong sense of religious identity, which defines their professional identity (Halai and Durrani, 2017). In a cultural milieu underpinned by strong conservative gender practices, it is more likely that teachers would be agents of socialisation of the pre-existing practices reflected within the school curriculum. Thus, they may become part of the systemic reproduction and perpetuation of gender injustice.

The study
This study seeks to explore: to what extent national level educational reforms in the curriculum and ITE have supported teachers in mitigating multiple gender injustices in Pakistan? This overarching aim is addressed through three interrelated questions: i. to what extent the school curriculum in Pakistan contributes towards gender (in)justice?; ii in what ways ITE in Pakistan prepares teachers for transformative gender justice?; iii and what are teachers' and teacher educators' perspectives on the interrelationships between conflict, education and gender justice in Pakistan? An overview of the reform context is offered, before discussing the study.

Educational reforms
Pakistan revised its National Curriculum, funded by USAID, in 2006 to promote education quality by replacing a content-driven curriculum with a competency-based curriculum. An additional objective was 'to make the whole education purposeful and to create a just civil society that respects diversity of views, beliefs and faiths' (MoE, 2009: 32). The curriculum was revised after national consultations. At the provincial level committees were established to oversee the translation of the curriculum into textbooks. Parallel reforms in textbook production were introduced which shrunk the role of the provincial textbook N. Durrani, A. Halai International Journal of Educational Development 61 (2018) 27-39 boards managed by the government and allowed market competition. Teacher guides were produced based on the revised curriculum and large scale in-service training was offered to support teachers in the delivery of the curriculum. However, the funding support of USAID in the context of the 'War on Terror' politicised the curriculum revision and alienated a range of stakeholders including curriculum and textbook personnel and teachers (Durrani et al., 2017a). A stringent review process ensures that all textbooks comply with the national curriculum.
With the devolution of education from the federation to provinces in 2010, curriculum development is delegated to provinces. However, all provinces have endorsed the 2006 curriculum, with some minor adjustments (Durrani et al., 2017a). The devolution of education has strengthened representative justice for provinces, all of whom have developed their education sector plans. The Sindh Education Sector Plan (SESP) stands out for including social cohesion and gender justice as crosscutting themes (Durrani et al., 2017a). The SESP acknowledges that the curriculum and overall teaching and learning environment lacks gender sensitivity and sets out specific targets for textbook revision and the integration of gender sensitivity in teacher education programmes (GoS, 2014). Alongside school curriculum, significant investment in ITE has been made with the aim to improve the quality of education through enhancing teacher education (MoE, 2009). Much of this reform was driven by funding support and technical advice from international donors and development agencies, notably the USAID (Durrani et al., 2017a). The curriculum of education was revised and schemes of study were developed for the newly introduced two-year Associate Degree of Education and four-year B.Ed. (Hons.) programmes. The revised curriculum of education positions teachers as reflective practitioners who engage in critical thinking and analysis to develop their practice and 'facilitate the process of multiculturalism and pluralism … to bring about social transformation' (HEC, 2010: 15). Although, gender is not explicitly mentioned, pluralism may implicitly include equality between women and men. Secondly, the revised ITE curriculum aimed at a closer alignment with the reality of the schools and classrooms through an enhanced and extended focus on the teaching practice component.
The premise above could potentially support the role of teachers as agents of change by enabling them to be open-minded and pluralistic in their practice through approaches such as 'action research', 'teacher learning' and 'critical thinking'. An inclusive pedagogy and a disposition of constructive critique could be developed through such an orientation of teaching as reflective practice. Over time, it could lead to an education for reducing and mitigating (gender) injustices. However, there was an assumption that teacher educators in the prevailing system would have the capacity to deliver a curriculum that was a paradigm shift from the traditional practice of 'lectures' provided by the teacher as an 'expert'. Indeed, Durrani et al. (2017a) found that for the promise of the revised curriculum to be realised there was need for a sustained and rigorous capacity development of teacher educators.

Research sites, participants and methods
The study was carried out in Sindh province, which includes some of the most poverty-and conflict-affected areas. Sindh is Pakistan's most diverse province in terms of religion and ethnicity, although the Mohajirs predominate in urban areas and Sindhis in rural areas (GoS, 2014). Karachi, the capital city of Sindh, reflects key conflict-drivers-ethnic, political and sectarian violence-and both Karachi and rural Sindh exhibit structural violence. While the study was conducted in Sindh, the issues raised are widely applicable in Pakistan. The National Curriculum 2006 is applied across schools in the country and the curriculum of ITE is a higher education matter and overseen by the federal government. The construction of gender in the curriculum has therefore implications for the entire education system in the country.
Four teacher education institutions (TEIs) were purposively selected to represent diversity in class, ethnicity and faith. Two TEIs were selected from Karachi. One was a private sector premier institution and another a government TEI which covered some of the poorest areas of the city alongside an affluent middle class neighbourhood. Two other government TEIs in smaller towns were selected because their intake of student teachers was from all districts in Sindh including far-flung rural areas, some with larger non-Muslim population. For comprehensive exploration, multiple data collection methods were used to seek perspectives of stakeholders with responsibilities for curriculum implementation and enactment including curriculum and textbook personnel, teachers, student teachers and teacher educators (henceforth lecturers) (see Table 2). Depth of insights was sought by interviews and focus group discussions with teachers, student teachers, lecturers and curriculum and textbook personnel. Breadth of perspectives was sought through a survey completed by teachers and student teachers. The survey included both structured and open questions and was administered in face-to-face group settings inside educational institutions. A purposeful non-probability sampling was used, with our interest not in generalising the findings to other TEIs but to gather data from a much larger pool of participants to provide critical insights from a range of stakeholders such as male/female, urban/rural, affluent/ lower socio-economic class, religious majority/minority. The questionnaire covered a range of topics relevant to the larger study. In this paper, we only draw on responses to questions that are pertinent to gender justice. These included structured questions about the provision of opportunities offered for gender sensitivity and experience of different forms of violence. The open questions explored the nature of gender sensitive education, forms of violence and perceptions of the links between education, conflict and gender justice. In addition, a limited number of classroom observations of lecturers were carried out in TEIs. For the validation of findings, five workshops were held with a range of stakeholders, including one specifically with teacher educators and those engaged in the governance of teachers. Data generated in these workshops was recorded and has also been utilised in this paper.
In the illustrative quotes offered in the next section, we identify each participant by a unique identification number in which the first letter specifies the method of data collection, followed by participants' category and gender. For example, I for interview, FGD for focus group discussions and Q for questionnaire.
Besides the primary data, we analyse the National Curriculum and textbook in Pakistan Studies (IX & X), a subject introduced after the loss of East Pakistan in 1971 and dedicated to promote national integration, solidarity and citizenship. The subject is compulsory and counts towards the final marks for all young people between the ages of 14 and 17 in their IX and X grades, in both public and private schools.
We bring together qualitative and quantitative data from all sources to study the impact of education reforms on gender justice as theorised in Section 2.3. The qualitative data from all sources is analysed in NVivo, guided by a dialectic interaction between our theoretical framework and the themes emerging from the data. The quantitative data is analysed in SPSS for frequency distribution. Given the negligible number of affirmative answers to the structured questions, it was not relevant to analyse the data further for bivariate analyses.

Results
Findings are presented in three sub-sections, starting with representation of gender and conflict in the curriculum texts. Second, the promise of teacher education for gender justice is analysed. Finally, we offer an analysis of how teachers and lecturers understood gender justice.

Representations of gender and conflict in school curriculum texts
The Pakistan Studies curriculum uses an explicit assimilationist agenda in its objectives for the promotion of social cohesion, predominantly using Islam as a marker of unity and deploying a nationalist discourse for invoking patriotic feelings and promoting affective attachment to the nation, as suggested below (MoE, 2006: 1).
• Underscore the importance of national integration, cohesion and patriotism.
• Promote an ideology of Pakistan, the Muslim struggle for independence and endeavours for establishing a modern welfare state.
The intended outcome of the above strategies is to promote social cohesion. However, both of the above strategies hamper social cohesion rather than foster it (Naseem, 2014). While the overlap of religion with nation excludes a range of citizen groups from the definition of the legitimate/ideal citizen, for example non-Muslim Pakistanis and Muslims whose practices are viewed as non-mainstream, discussed at length by Halai and Durrani (2017), in this paper, we contend that the invocation of the 'other', as illustrated in the ensuing discussion, tend to associate hegemonic masculinity with the protector of the ideological and territorial integrity of the Pakistani/Muslim nation in ways that perpetuate conflict and has serious implications for gender justice, social cohesion and conflict that cut across religious, ethnic and regional boundaries.
The 'struggles' alluded to in the curriculum objective are highlighted with reference to an external 'other'-Hindus/India-within the Pakistan Studies textbook. While often the hostile 'other' is depicted to seek the destruction of Pakistan predominantly militarily, other means of infiltrating Pakistan are also deployed including the cultural, economic and ideological annihilation of the country and its (Muslim) citizens. The textbook seeks to normalise xenophobia and the need for protection from the (non-Muslim) aggressor. A Muslim protagonist venerated at length is Syed Ahmad Shaheed (martyr) (1786-1831) who is praised for mobilising a struggle against the 'evil forces in the subcontinent' (STBJ, 2013: 15). He founded the Tahreek-e-Mujahideen (The Struggle of Freedom Fighters) which sought among other objectives to 'protect Muslims against such acts and ideas as are contrary to Islamic values' and to 'preach Jihad because it was not possible to get freedom from evil forces without armed struggle' (ibid: 15). Another Muslim icon whose representation serves to constitute Muslim masculinity as the 'protector' of the ideological frontiers of the nation is Haji Shariatullah (1781-1840), founder of the Fariazi (pertaining to obligations) Movement in British Bengal which sought to 'eliminate the unIslamic customs and traditions which prevailed among the Bengali Muslims' (ibid: 17). The role of the masculine protector also extends to the political arena in which male Muslim political leaders take on to defend the nation against 'the selfish designs of the Hindus' who wished 'to hold supremacy over the Muslims by establishing Hindu rule in the sub-continent once the British left' (ibid: 21). 2 Likewise, the initial problems mentioned in the textbook that Pakistan faced included the development of national integration in the face of Hindu/Indian leaders' prediction and expectation that 'Pakistan would soon collapse and the sub-continent will again be reunited' (ibid: 35).
While the aggression/masculinity of the 'other' is constructed as evil, the same characteristics are celebrated when Muslim conquest is narrated: Thereupon, Ahmed Shah Abdali, in 1761, inflicted a crushing defeat on the Marathas at the Third Battle of Panipat, from which the Marathas never recovered (ibid: 15).
The construction of hegemonic masculinity as the protector of the ideological and territorial integrity of the Pakistani/Muslim nation perpetuates redistributive gender injustice and the erasure of women from official history, with only two women out of 94 national icons cited only in passing. These being Banul Dashtyari, mentioned in a sentence listing Baloch poets and the 'brave' Sugra Fatima, who removed the Union Jack 'from the Punjab Secretariat and in its place installed the flag of Muslim League' (ibid: 30). Another mechanism used for the obliteration of women is through the use of universal pronoun 'he' and the use of gendered terms such as 'brother/brotherhood' which exclude 'sisters/sisterhood' from the faith community: 'Muslim is a brother to another Muslim. He does not betray him' (ibid: 9).
Furthermore, the textbook perpetuates the cultural and economic subordination of women through emphasised femininities and the construction a gendered national imaginary where men 'lead' and women work 'within the four walls', normalising existing power differentials between men and women: Male member has acquired a unique status in Pakistani culture. He is the head of the family. He is the dominant member. But a woman is also considered an important part of the family who governs and manages all family affairs within the four walls. Household keeping and upbringing of children is entrusted to her (ibid: 134).
By delimiting women to the domestic sphere, tasked predominantly with the reproduction and upbringing of the future generation, the textbook perpetuates structural violence against females. Gender roles and expectations, as discussed above, seriously curtail the life opportunities of women, normalises gender power hierarchies in the consciousness of both women and men, and legitimises female subordination and their secondary status. Importantly, it may present alternative femininities as a risk to the nation and therefore to be suppressed, rendering the performance of alternative femininities liable to direct violence.
Cultural norms are central to the creation and sustenance of structural violence and the latter serves to institutionalise cultural violence. The textbook draws on Pakistani culture to perpetuate cultural violence by rendering structural violence against women acceptable, 'More male children are considered to be an asset for the family and a sign of pride among the kith and kin' (ibid: 121), even though such cultural/gender norms could have been strongly critiqued by drawing on Islamic principles of justice, non-discrimination and equity discussed elsewhere in the textbook at length (ibid: 7-9). By linking the application of the above Islamic principles to everyday gendered practices in Pakistan, rather than discussing them in generic terms, the textbook discourse could have potentially destabilised the gender/cultural norms that are central to gender injustice.
With very few exceptions, the 266 teachers/student teachers responding to the questionnaire, showed little understanding of the gendered nature of the curriculum texts. By contrast, discussions with curriculum personnel in Sindh indicated that the gendered national imaginary was linked to the political nation-building project to portray the military as the main signifier of Pakistan: 'The glorification of military and all these things require a political support' (I12, textbook personnel, male). The deployment of a muscular national identity was considered central to recognitive gender injustice, which fosters reconciliatory and representative injustice and supports conflict. N. Durrani, A. Halai International Journal of Educational Development 61 (2018) 27-39 You hardly find women in Pakistan Studies textbooks because, of course, when we are reading about men who fought all these wars you are not going to find any women there. (I11, textbook writer/curriculum expert, female) In line with the recommendations in the SESP (GoS, 2014), a 13 member committee had been established to review textbooks in the province, only two of whom were women. 3 This had opened spaces for representative justice in educational decision making and for shifting hegemonic masculinities to strengthen social cohesion and promote recognitive and reconciliatory gender justice: "We suggested removing gender bias … and there was too much of glamourising the military and the militarised culture" (I13, textbook personnel, male). While the vocal critique offered by curriculum and textbook personnel including members of the textbook review committee is laudable, curriculum revision in Pakistan has proven to be highly politicised and contentious. Textbook revision that is seen in contention with the 'ideology of Pakistan' is hotly debated in public fora, including television and the print media and instigates mass protests, predominantly at the behest of parties with religious/right wing orientation. While textbook revision in Sindh was in its infancy at the time of fieldwork, expectations of substantial change in gender justice may not realise given the politicisation of curriculum.
The analysed curriculum texts deploy exclusivist discourse to construct obedient and loyal citizens governed through ideological discourses of the nation, ostensibly driven by the goal of national unity and solidarity. The gendered representations of the nation and gendered citizenship identities that students are encouraged to take up potentially perpetuate recognitive injustice against women and push male citizens towards hegemonic masculinities that are counterproductive to sustainable peacebuilding and social cohesion.

ITE curriculum
The gendered citizenship identities sketched in the preceding section would require a sustained focus on gender within ITE curriculum if education has to play a potent role in transforming gender relations. Although, the ITE curriculum includes no dedicated course/module on gender out of a total 45 modules covered in 4 years, two courses have at least a unit or a topic relevant to gender justice. The 'Foundation of Education' course lists 'Gender and Education' among a range of 'Problems and Issues in Education in Pakistan' (HEC, 2010: 80), while 'Contemporary Issues and Trends in Education' has one whole unit on gender that covers the 'Concept of gender equality', 'Factors affecting the role of women' and 'Steps towards reducing gender disparity' (ibid: 58-60). Thus, gender justice receives a peripheral focus in ITE curriculum. Such a tokenistic attention paid to gender within ITE curriculum is unlikely to have any significant impact on transformative gender justice.
The marginal attention to gender in the ITE curriculum is substantiated by findings from the questionnaire. Out of 266 respondents, only eight, three student teachers and five teachers indicated that there were particular modules/curriculum topics that specifically dealt with gender justice and gender-based violence. Student teachers and lecturers pointed that the Islamiat textbook was a significant discursive space for promoting gender equality 4 : In this subject [Islamiat] there is brotherhood [sic] and about equality, rights of women, what are women's rights in our society? (I8, lecturer/ principal, male) Thus, textbooks including Pakistan Studies could draw on religion/ Islam for subverting the unjust gender order that exists in Pakistan and offer an alternative discourse from which gender injustices can be resisted. Such a possibility has been reported in KP, Pakistan where some young men in higher education vocally decoupled gender injustice and Islam and drew on the Qura'n and other Islamic texts to disrupt hierarchical gender relations within the local/national context (Durrani et al., 2017b). There was little evidence of any tangible support being offered to student teachers in developing their capacity to question gender hierarchies as experienced in everyday life: 'I think none of the colleges are working on this crucial [gender] issue although it is the need of time' (QPINS17, teacher, female).
Another issue with respect to the effective enactment of the revised curriculum, in general and with respect to gender justice in specific, is teacher educators' capacity to deliver a curriculum that was considered as a paradigm shift from the traditional practice of 'lectures'. Against this backdrop, participants in the validation workshops maintained that 'even teacher education faculty in the departments of education in the universities find it difficult to interpret the curriculum and develop a syllabus as there are no textbooks for the revised curriculum of education and there is a need to enhance capacity of such faculty' (Stakeholders' Validation Workshop, November 5, 2015).

Pedagogic practices
Responses across the different data sets indicated that the predominant focus of ITE reform has been on shifting pedagogic approaches from more teacher-dominated practices to open and collaborative practices centred on learners. Of course, such pedagogies are better suited to transformative gender justice as was observed by Murphy-Graham (2008). This change in teacher's role was therefore a step in the positive direction, particularly since all three public sector TEIs had enrolled both male and female student teachers and could therefore make use of co-education setting to disrupt gender binaries and promote interaction across gender boundaries in ways that could promote recgonitive and reconciliatory justice.
The lecturers interviewed acknowledged that the TEIs mainly focused on the development of generic pedagogic skills aimed at developing teacher capacity for learner-centred practice such as teamwork, independent learning and inquiry, cooperative/collaborative learning and problem solving. The three classroom observations conducted indicated that lecturers were using some of the above strategies and their practice was not confined to lecturing or explication. Lecturers saw their practice vital to the promotion of social cohesion, although gender was again incorporated implicitly: So as a teacher, my role would be to create a kind of an egalitarian environment in my classroom, where everybody is treated with dignity and respect regardless of any identity or any affiliation (I1, lecturer, female).
Student-teachers concurred that pedagogic practices such as group tasks, teamwork and cooperative learning used in TEIs were conducive to fostering social cohesion, albeit their focus was not on gender relations: Here we have the best example of team work, that make two or more persons work together, and you cannot work together until and unless you understand each other's thoughts and follow the suggestions of your group members. It means that you can't create cohesion until you understand and accept each other. (FGD2-P4, student teacher, male) Not withstanding the widespread acknowledgement of the above pedagogic practices to the promotion of social cohesion, student teachers' perspectives indicated variability in the effectiveness with which these practices were being modelled and used across the four TEIs N. Durrani, A. Halai International Journal of Educational Development 61 (2018) 27-39 studied, as well as a lack of explicit focus on gender.

Gendered institutional practices
At least the field evidence showed that since 2010 public sector TEIs in Sindh moved to co-education from a gender-segregated organisational structure. This may potentially support redistributive gender justice by offering access to both male and female student teachers to TEIs in a context in which female student teachers are under-represented. For example, of the 0.722 million enrolled student teachers only 34% are female (NEMIS-AEPAM, 2016). Furthermore, this policy may facilitate redistribution of lecturers to female student teachers. Although NEMIS-AEPAM (2016) does not offer a gender-segregated analysis of lecturers across Pakistan, male lecturers (64%) outnumber female lectures (34%) in Government Elementary Colleges of Education (GECE) in Sindh (GoS, 2014). In the three government TEIs studied, female lectures accounted for only 18% of a total 56 lecturers.
Nevertheless, in two of the three public sector TEIs, despite enrolling male and female student teachers, it was deemed culturally appropriate to organise separate classes for male and female students, though teachers remained the same. The gender-segregated provision was arranged to mitigate redistributive injustice for female student teachers: So the main purpose is education, so for that purpose we have separated classes of boys and girls which make them [parents] happy and they enrol their daughters here. (I4, lecturer, male) Pakistan is a highly gender segregated society with clearly genderdefined roles and responsibilities. The quote above is a reflection of this cultural practice. Whether or not it is a hindrance to redistributive justice can be argued both ways. A gender-segregated classroom could facilitate redistributive justice if it means more women and girls participate in education. However, gender segregated provision would be a hindrance to redistributive justice if women and girls were denied access to education in the absence of a college designated for them. In this case, the lecturers' practice could be interpreted as an attempt to accommodate women and girls within the cultural norms so that they are able to participate in education.
Even in the TEI where classes were of mixed-gender, gender boundaries were rigidly maintained by teachers either to 'protect' female students, or to maintain the gendered norms and the moral code. This is not to suggest that transgression of those boundaries by male and female student teachers could be ruled out. Indeed, teacher-students acknowledged that both males and females transgressed those boundaries outside classrooms. 'Protecting females' positions females as needing protection and unable to look after themselves and serves as structural violence since such protective behaviour may become a hindrance for them to advance in their careers and social life. Additionally, it entrenches cultural violence against both males and females as TEIs reinforce the normalistaion of gender boundaries existing in society and legitimised by the school curriculum texts, despite the potential that co-education provision offered in diffusing gender boundaries in ways that could contribute to a more egalitarian gender regime. This cultural violence can potentially incite direct violence against those who are seen as transgressors of accepted gender boundaries.
We usually hear that […] there are boundaries between girls and boys, and there should be no communication and you should not cross those boundaries. (FGD1A6, student teacher, male) More importantly, the cooperative, collaborative and communicative approaches that the revised curriculum of education sought to inculcate clashed with the gender regimes practised in TEIs, with lecturers policing the moral boundaries between the 'good' and the 'bad' men and women through the erection of spatial boundaries. This supports cultural violence and potentially direct violence: Initially, girls and boys behave very well, worked together, but our teachers have made it difficult for us. If they see any male talking to a female then they behave very strangely. (FGD1A1, student teacher, male) These disciplinary regimes along with providing gender-segregated education within a co-education structure was intended to 'protect' females from gender-based violence but inevitably establishes their subordination and perpetuates structural violence against them. On the positive side, both qualitative and quantitative data showed very little acknowledgement of gender-based violence in TEIs. The majority (94%) of questionnaire-respondents denied having experienced genderbased violence. Only two male student teachers and 11 male and two female teachers claimed being subjected to gender-based violence. When asked to explicate on the forms of violence, these largely related to the conflict-affected environment in which schools operate, such as being threatened to shut the school in support of political protest and were not examples of gender-based violence. However, lecturers, student teachers and teachers acknowledged 'stare' is a powerful form of violence to which female student teachers are subjected by male peers, even in the presence of spatial boundaries.
Furthermore, the disciplinary regimes deployed by teachers to ensure compliance to gendered norms has to be understood against the backdrop of a social setting where gender symbolism offers the basis of the construction of the nation (Durrani, 2008) and where education is deeply politicised in the construction of the national imaginary (Durrani et al., 2017b). Therefore, in order to create spaces for gender justice teachers/ lecturers have to tread cautiously so that their practice is not seen as a transgression of national/local culture. For example, a principal shared that in a short course for teachers, the course leader asked her class to wish a male teacher well on his birthday as this was seen as a social gesture that showed respect and care towards peers. However, this practice stirred a huge crisis for the principal/TEI concerned: 'I started getting calls from the President House and from the news channel' (I2, lecturer/principal, female).
Upon inquiry it turned out that some young women in the aforementioned class refused to wish their male colleague well on his birthday and complained about the incident to parents. The course leader's efforts, however well intended, backfired simply because the strategy for promoting social interaction and respect for the 'other' was not relevant to the context. In a highly gender segregated environment a social interaction of birthday celebration among men and women was seen as culturally unacceptable. The episode had unintended impact as the course leader was fired and the male trainee was expelled from the short course because of media and community pressure. This episode indicates when local cultural practices conflict with those underpinning the teaching methods the consequences are friction and not cohesion. Transformative gender justice would require therefore pedagogies that are contextually relevant alongside long-term investment in changing cultural norms about acceptable gender behaviour by institutions beyond schools and teacher education.

Teachers' perspectives on gender (in)justice
When asked about the challenges of social cohesion in Pakistan, the majority of participants cited at least two or more conflict drivers, demonstrating a nuanced understanding of the multiple and intersecting sources of conflict. The most frequently cited sources were related to reconciliation between diverse Muslim sects and ethnic groups, and to a lesser extent inter-faith reconciliation. Participants linked issues of reconciliation and recognition more explicitly to violent conflict and social tensions: Once you label yourself with an identity, the same moment you find your enemy.
[…] if you label yourself Shia, then Sunni are your enemy, other N. Durrani, A. Halai International Journal of Educational Development 61 (2018)  Redistributive injustice based on class/income, as well as other social markers, was also highlighted by a large number of participants and seen detrimental to social cohesion. Furthermore, redistributive injustice was seen intricately linked to recognitive and representative injustice and all three were seen as perpetuating structural violence and exclusion: Every community in our society is treated differently in accordance with their religion, caste, sect, language, rural/urban location, dual educational system, availability of educational and work opportunities. (QGCES27, student teacher, female) Despite the gendered conflict drivers in Pakistan, discussed in Section 3, rather surprisingly only few participants mentioned gender as an issue of social cohesion and these were largely, though not exclusively, females. Furthermore, with one exception, issues of gender justice were framed in terms of redistribution, i.e. girls' inequitable access to education because of poverty or cultural/gendered norms: Our country is facing some difficulties in gender gap. Most people living in rural areas do not want to send their girls to school because of izat or ghairat [honour]. male) Girls get married at an early age because of which they cannot get education. (QPADE4, student teacher, female) None of the ten lecturers interviewed explicitly mentioned gender as a cause of conflict or a challenge for social cohesion, casting doubts on the role of (teacher) education in promoting gender justice. Nevertheless, when asked specifically to comment on the ways gender is linked to social cohesion, redistributive injustice in access to education for girls was identified as a challenge. However, a dominant perception among lecturers was that gender parity in access was tantamount to gender justice: I think that gender is not an issue, it is just exaggerated. Allah has made men and women. He has put different potential in both of them [-] When I am sitting here both male and female are equal for me. (I3, lecturer/ principal, female) In other words, despite the perceived 'essential' difference between men and women, they become equal once they access the TEIs. There was limited evidence to suggest lecturers and teachers understood gender as a power asymmetry constructed through social practices within and outside educational institutions. With very few exceptions, there was little indication that teachers understood the recognitive and reconciliatory gender injustices perpetuated by the educational discourse illustrated in Section 6.1. Those who were able to identify the complicity of educational discourse in the normalisation of gendered identities and power hierarchies were largely female student teachers/ teachers. In the absence of a gender transformative curriculum, the identification of this link was crucially significant for teachers to enact their agency in mitigating gender injustice: When I will become a teacher I will take out all the topics from the curriculum that indicate gender discrimination trends. […] I will publish Islamic and Moral literature, which will develop a sense of respect towards women among people. (QGCEL89, student teacher, female) The capacity of lecturers, or lack of it, for a transformative gender justice is a serious challenge as the practices of lecturers could potentially have an impact on teachers, as acknowledged by a student teacher: We will try to learn according to our observation and from books, like what the book is telling us and we would spread it, we would share it with others and we will try to act upon it, so that will have an impact on our society. (FGD1A5, student teacher male) In summary, gender justice was not prioritised in the ITE curriculum. Nevertheless, the learner-active pedagogic practices pushed by the reform offered possibilities for developing student teachers' understanding of the constitution of gender and the transformation of gender relations in schools and community. However, the institutional gender regime, while supportive of redistributive and recognitive gender justice, offered little scope for shifting gendered norms. Lecturers' capacity to understand the construction of gender and its link to conflict appeared constrained, leaving little optimism regarding teacher education playing any substantial role in promoting gender justice.

Discussion and conclusion
This paper explored the extent to which reforms in the curriculum and ITE have enabled teachers in mitigating gender injustice and creating spaces for transforming gender relations within the highly gendered and conflict-affected setting of Pakistan. The analysed reforms, undertaken to enhance quality, mitigate inequities and promote social transformation, were national in scale. Thus, the study context offered opportunities for gaining useful insights into the rather under-researched area of the role education might play in transformative gender justice for sustainable peacebuilding and augmenting social harmony.
Evidence of teachers being supported in ways that could promote gender justice for social cohesion appeared limited. A deconstruction of the curriculum texts revealed that they potentially perpetuate gender injustice within the domains of redistribution, recognition and representation. More importantly, they entrench reconciliatory gender injustice through the construction of hegemonic masculinities and emphasised femininities in ways that foster conflict. The confluence of nationalism, patriotism, and citizenship within the educational discourse tends to normalise militarism, authority and discipline, particularly when instantiated through religion.
The gendered educational discourse, infused with conflict, thus poses teacher education a momentous task to support teachers for gender justice. Despite some success in expanding the pedagogic palette of teachers, teacher education offered little space for transformative gender justice because gender remains peripheral within the ITE curriculum. On the positive side, unlike the sexual harassment and violence against female student teachers such as that perpetrated in colleges of education in sub-Saharan Africa (Leach and Bakari, 2008), the TEIs studied took considerable efforts to safeguard the recognitive justice of female students, simultaneously enhancing their redistributive justice. Nevertheless, 'protecting' females through erecting spatial boundaries between male and female student teachers entrenched structural and cultural violence and missed opportunities for developing trusting relationships between them for fostering reconciliation and shifting gendered norms.
More significantly lecturers could neither identify the links between gender, education and conflict in Pakistan nor showed an understanding of the ways gender is constructed in everyday practices. The collaborative, cooperative and interactive pedagogic practices promoted in the ITE curriculum and TEIs was a step in the right direction. However, for the promise of the revised ITE curriculum to be realised, a sustained and rigorous capacity development of teacher educators would be indispensible.
Sindh has explicitly included gender justice and social cohesion in the SESP 2014-18. At the time of fieldwork, this plan had been in place for only a year. Encouragingly, the actors responsible for curriculum texts in the province offered some critique of the gender-nation couplet and its intersection with conflict. The success of these critical voices in shifting dominant masculinities and emphasised femininities in the educational discourse is yet to be established and would warrant further investigation.
Pakistan is not unique in the deployment of education to construct compliant (gendered) citizens. As a state institution, education actively constructs gender and produces the subordination of women and the school curriculum is crucially 'connected to the power relations of gender in the larger society' (Kessler et al., 1985: 43). For example, the civics curricula in Canada maintain masculinist biases underpinning the concept of citizenship in ways that exclude women from the category of the good active citizen (Kennelly and Llewellyn, 2011). Likewise, school curricula in Turkey are reported to foster 'the notion that every male Turk is born a soldier' (Altinay, 2004, cited in Cockburn, 2013. The capacity of teachers to critique and subvert the national(ist) educational agendas that subordinate women is also not distinctive to Pakistan. In a cross-national European study, student teachers showed limited capacity to question the masculinisation of public life despite a strong awareness of sex discrimination (Arnot, 2006). Similarly, Greek-Cypriot teachers failed to recognise the gendered politics of their national identity and appeared unwilling to question the gendered national norms (Panteli and Zembylas, 2013). Thus, the findings of the current paper have implications beyond Pakistan.
Multi-dimensional insights for policy and practice in Pakistan emerge from the findings, which are also significant for other conflict contexts characterised by striking gender inequities. Recognitive and reconciliatory justice would require the production of gender-transformative curriculum in school and ITE, as this can support learners and teachers to not only understand but also question and challenge gendered norms and behaviour that are linked to violence (UNESCO, 2016b: 52). As discussed at length in Section 2.2, Bajaj's (2012) research in India illuminates the potential of how schools can become sites for promoting egalitarian gender attitudes and behaviour. However, she also indicates that for such a change to happen important conditions need to be met including 'deep engagement with stakeholders … rang[ing] from classroom teachers and headteachers, to state and national level policy-makers' for securing their support and the translation of the curriculum into relevant and quality textbooks and effective training and support for teachers (ibid: 11). Such deep engagement would also be key to any attempts at the promotion of gender justice in and through education in Pakistan where teachers' and curriculum development personnel's professional identity, like elsewhere, is shaped by their cultural/religious identifications. Hence, their 'coalitional agency' would need to be secured in order to mitigate any tensions between their socio-cultural positioning and educational reforms.
Given the legitimacy that Islam carries in Pakistan, an explicit focus on gender equality in Islam would offer a space from which to challenge authoritatively gender injustice, as was indicated by Durrani et al.'s (2017b) research in KP, Pakistan, with young people using Islamic discourse as a form of resistance to disrupt local cultural practices that subordinated women and perpetuated structural, cultural and direct violence against them. However, challenging normative gendered norms and hierarchies existing at the macro-and micro-level in Pakistan would require centring the discourse of gender justice on how gender is constituted through everyday practices. For example, in Rwanda, curriculum texts promote gender justice by drawing on human rights and development discourses, rather than 'critiquing gendered and unequal societal structures and examining existing gendered practices and assumptions' (Russell, 2016: 509). Not surprisingly, therefore, gendered hierarchies and biases persisted in Rwandan students' attitudes and beliefs, despite increasing gender awareness promoted by textbooks.
Furthermore, building teachers' and teacher educators' capacity in the use of gender-transformative and conflict-sensitive pedagogies that engage with the implications of dominant masculinities and narrow femininities on gender and wider social relations would support transformative gender justice. However, the impact of training on gender sensitivity remains negligible, unless teachers are able to embody and enact gender justice in everyday interactions in and outside classroom (Halai, 2011). Supporting teachers to reflect on their 'everyday lives to recognise and understand the social construction of gender and patriarchy and how it is reflected in small everyday actions' has been found helpful in India (Achyut et al., 2016: 3) and Pakistan (Ashraf and Waqar, 2012) in the context of in-service training. These insights could be usefully incorporated in ITE. Furthermore, Durrani et al. (2017a) reported that in Sindh, CPD opportunities for teachers incorporated modules looking at social cohesion including gender justice. During these CPD projects, teachers were supported to implement these concepts at the classroom level. However, for long term impact and sustainability of ideas such as gender justice, they need to be incorporated into syllabi and teaching plans at the grassroots level. In addition, Durrani et al. (2017a) reported that CPD opportunities were unevenly distributed and therefore the inclusion of these modules into ITE curriculum would ensure greater reach and impact.
While the representation of an adequately qualified female workforce in schools and TEIs would strengthen representative justice and have a positive impact on redistributive gender justice, the capacity development of both male and female teachers is important because what they say and do impact what kind of masculinity becomes hegemonic and what type of femininity gets emphasised in school. At the same time, teachers working conditions is an important consideration as 'poorly paid, demoralised and over-worked teachers are unlikely to become change agents' (Novelli and Sayed, 2016: 34).
Nevertheless, it is not necessarily possible to induce the transformation of entrenched gendered norms through classroom and curriculum inputs alone. Gender norms in the community may conflict with the egalitarian gender relations that schools seek to promote, constraining teacher capacity for gender justice. For sustainable transformation such dissonance between school and community would need to be resolved. This would require schools engaging with civil society and community mobilisation groups to shift discriminatory gendered practices. Given the relatively low social status of teachers/headteachers it remains doubtful if they would feel empowered to engage in social transformation, beyond their specific curricular role. Therefore, strengthening the school-community link would need the support from multiple stakeholders across government and the community (El-Bushra and Smith, 2016). Finally, transforming gender relations would require a significant cultural shift in society at large. Education reform has to be supported by other elements in the socio-cultural setting.
The implications of the study findings for SDG 4 relate to a reconceptualisation of the ways gender is understood and gender equality is measured and monitored. While gender parity index in access, participation, completion and outcomes of education offer useful information on redistributive gender justice at macro-level, these figures need further segregation by other important context-specific markers such as ethnicity, race, religion, region, urban/rural location, economic background to identify particular gender groups experiencing overlapping and multiple disadvantages in specific contexts. Progress towards recognitive and reconciliatry gender justice can be monitored by analysing school curriculum and textbooks to identify not only the distribution of textbook characters by gender but also reveal how gender is performed and constructed within the textbooks to encourage particular masculinities and femininities, with a particular focus on their implications for social cohesion and conflict mitigation. Given the slippage that may happen between the official curriculum and its enactment by teachers and the reception of the official and enacted curriculum and negotiation by students, as well as the significance of the hidden curriculum on students' gendered identifications, qualitative methods, particularly ethnographies of everyday school life would generate much useful insights. Additionally, it would allow observing teachers' practice in actual classrooms to illuminate if and how teacher education have impacted on teachers' practice for gender justice, something that the current study could not look at. School ethnographies may also generate useful insights regarding representative gender justice and school structures that have a bearing on gender relations to complement macro-level data on the distribution of key educational positions by gender.
The theoretical and methodological contributions of the paper lie in offering a multidisciplinary framework for analysing gender justice in and through education for sustainable peacebuilding in conflict-affected settings and illustrating how it can be systematically applied to the analysis of political economy of gender justice and conflict, macro educational context, curriculum texts and meso-and micro-level data collected in the TEIs. As a heuristic device, the framework was helpful in capturing injustices in the 4Rs, particularly, in understanding redistributive injustice, which predominantly treats gender as a noun and is therefore more amenable to measurement. In highly gender unequal contexts such as Pakistan, this focus is desirable, as policy actors and stakeholders more readily acknowledge challenges of redistributive gender justice. Nevertheless, if sustainable peacebuilding is to be promoted through gender justice, issues of reconciliatory, recognitive and representative gender justice are crucial and require understanding gender as a verb. In practice, disentangling these domains for analysis was hard due to their intersecting nature. However, conceptualisation of gender as performative was productive in understanding how masculinities and femininities were constructed and underpinned by social practices within TEIs. It is the constant production of gender in everyday practices that gives hope for transformation in gender relations even when change is slow to the extent that it appears invisible.