Conceptualising Subjectivities and Rationalities in Understanding Gendered Violence: Processes of Social and Cultural Change

Abstract

International development policy and programming that is geared towards women’s empowerment through reversing normalization around gendered violence tends to focus either on social norm change (in line with much social science research) or, more commonly, on increasing women’s paid work. In this article, we argue that neither exclusive focus on social norm/mindset change nor on women’s economic engagement can come to grips with what sustains or may end violence. This is because such approaches cannot unpack the intersectional and dynamic interaction of social relationships, power, institutions and environmental dimensions that shape perceptions, attitudes and behaviours. We suggest that Bourdieu’s concept of habitus, combined with a concept of collective agency, pushes us to focus on the dynamics of the different spheres and types of relationships that sustain violence and can help us to untangle them. Drawing on data from Myanmar and Nepal, we argue that understanding the complexity of how different forms of violence feed from each other and link to symbolic, structural and behavioural dimensions is critical alongside a clearer idea of how collective agency can facilitate change.

I Introduction

Arguably, we know more now than ever before about what motivates perpetrators of violence against women and how survivors respond, but these insights are largely focused on the individual. Feminist literature in various disciplines (e.g., including sociology, social work, social policy and geography) explores discrimination through lenses of structural inequality. However, the vast majority of research that aims specifically to impact gendered development policy and practice draws on public health and psychology frameworks. Gendered violence, in particular, gets cast as a consequence of either individual pathology or deeply embedded social norms—a mindset. International development policy and programming that is geared towards women’s empowerment through reversing normalization around gendered violence tends to focus either on social norm change (in line with much social science research) or on increasing women’s paid work. The latter is problematically entwined with an instrumentalization of gender (i.e., the idea that women’s empowerment will contribute to economic development). This is arguably the main perspective and approach among major institutions; it dominates the perspective of the World Bank, it is evident in the United Nations (UN) Women’s ‘Women’s Empowerment Principles’ and it is reflected in many documents produced by organizations like the UK’s Department for International Development (DFID) and the United States Agency for International Development (USAID). In stating this, we are in no way undermining the enormous strides these institutions have taken in mainstreaming gender in terms of gender-inclusive budgeting, women’s employment and mitigation of gender-based violence (GBV).

In this article, we argue that neither exclusive focus on social norm/mindset change nor on women’s economic engagement can come to grips with what sustains or may end violence because such approaches cannot unpack the dynamic interaction of social relationships, power, institutions and environmental dimensions that shape perceptions, attitudes and behaviours. Research has shown that violence exists in a structural institutional sense through the (differentiated) exclusion of women from many domains and positions (e.g., Batliwala, 2012; Bowstead, 2017, 2019; Eisenstein, 2009; Fraser, 2009; Heise, 1998; Larance and Porter, 2004). It also exists physically in various acts of abuse across public and private spaces. We suggest that Bourdieu’s concept of habitus pushes us to focus on the dynamics of the different spheres and types of relationships that sustain violence and untangle them. Habitus can be both an organizing tool and a theoretical lens through which to understand the social realities for different groups of women and to capture how change manifests. This concept reworks functional theories on gendered violence, which, in turn, help to adapt habitus to nuances of GBV. For example, in Heise’s (1998) ecological (functional) model, the social world consists of differing spheres and levels from the individual to household, community and outwards. She argues that violence against women persists as each social layer reinforces norms that, in turn, legitimize violence at the individual level. Understanding the complexity of how different forms of violence feed from each other and link to symbolic, structural and behavioural dimensions is critical if violence is to be replaced with gender equality. Whether drawing on the idea of habitus or functional theories, separately or together, we recognize the need for analytical caution to move beyond a possible circular social logic. We do this by pointing to material and social resources and to contradictions in institutional, social and cultural spheres that allow for and are enabled by collective and individual agency, identifying gaps in the logic as moments when agency and resources can come together to challenge the gender status quo rather than to reinforce it.

The reductive framework of social norms overlooks processual dynamics and, hence, cannot ground a theory or policies for change. Bourdieu’s conceptual tools of habitus, practice and strategy offer us a more nuanced understanding of the web of factors that sustain inequality and violence in women’s lives; their taken-for-grantedness. We argue that utilizing Bourdieu’s work allows us to overcome the lack of understanding in the ecology model of where and how relationships operate to sustain structures of power. In focusing on possible entry points to change, we draw attention to the significance of social and cultural capital and collective agency. While they exist in all contexts, we give examples of peer networks and local advocates, working to build resilience to and prevention of violence. We find evidence of questionings being produced and/or transformed through these networks and discourses. In highlighting this as a form of collective agency, we recognize the ubiquity of possible conflict and contestation—social, cultural and symbolic capital, as much as economic capital, rest on/emerge in difference, inequality and intersectionality—which can vitiate social relations, aspirations and recognition of possibilities even among those who are coeval.

Fultner (2017) argues—drawing on Habermas’ theory of communicative action—that humans are intrinsically intersubjective and cooperative, that individuals accept that working together to maintain ‘society’ is in their best interest. This ‘best interest of society’, we know, obscures the inequalities of advantage and power in society. We are cognizant of Fultner’s (2017) warning that cooperation is not inevitable, not least when it involves the adoption of deviant behaviours that challenge dominant societal structures as set out through the ecology model. She brings together social ontology and critical feminist theory to cite Arendt’s claim that collective political action is not triggered by shared identity, but it is triggered by a shared goal. We emphasize that it is sharing the experience of violence (or in some cases, the perception of one’s vulnerability) that generates the commitment to cooperate towards the shared transformative goal of ending it. To share personal and painful experiences, Fultner (2017) points to the significant trust and openness required, which a collective organization can enable. With sustained sharing, women do not then adopt a shared new identity as the ‘abused’ or ‘survivors’, but they can realize a collective commitment to act to end violence.

Thus, though possibilities to engender change are present, they are not assured, even when mobilized. A range of scholars, Appadurai (2013) among them, argue that ‘development’ policies/programmes have tended to miss the resource represented by social and cultural capital—that it is often undervalued or even missed by stakeholders. Yet, even when taken as the central pillar for a policy, when pinned to a reductive framework, its instrumentalization, rather than transformative possibilities, may be mobilized. We believe that seeking an end to violence against women and girls (hereafter VAWG) through a lens that combines habitus, concepts of social and cultural capital, and collective agency will make a significant contribution to achieving Sustainable Development Goal (SDG) 5.

Using data from Myanmar and Nepal gathered through the DFID-funded research project ‘Women, Work and Violence’, we argue that the links between violence, gender inequalities and social transformation cannot be seen as a series of linear steps or stages of social norm change. We look at how structured environments that enable the sharing of individual experiences of violence, occurring largely within the household and community, can lead to a questioning of habitus and engender a commitment to collective action for change. The focus of this article is specifically on two countries, and our conclusions are, therefore, inevitably context specific. However, it is important to recognize that the theoretical perspective that we advance here could be applied equally well in other contexts. Indeed, we believe that the arguments presented here help to underpin and expand the views projected in other research (e.g., Batliwala, 2012; Bowstead, 2019; Larance and Porter, 2004) that emphasizes the limitations of individualizing and medicalizing our understandings of the perpetration and experience of violence.

The article is structured as follows: we begin by exploring what social norms are and how they are currently being applied in the context of VAWG. In the second section, we present the utility of Bourdieu’s work in moving approaches to end VAWG into a more context-responsive form taking account of feminist critical theory and collective agency. In the third section, we explore our argument through data collected in Myanmar and Nepal across a number of different work contexts. The conclusion then draws out our overarching findings, stressing the need to recognize the complexities of how relationships operate at family, community and organizational levels and through networks. We argue that a perspective on ‘positive deviance’ is critical as a foundation for social change.

II Social norm change and violence against women and girls

Before going further, we need to pose a question: what then is a social norm? According to Paluck and Ball, ‘A social norm is a perception of where a social group is or where the social group ought to be on some dimension of attitude or behaviour’ (2010: 9). Norms are linked to sanctions put in place at various levels that operate to maintain a collective, even if not consensual, normative view or perception. As already stated, a critical focus for campaigns to end VAWG has been the question, ‘how are norms changed?’ According to Marcus and Harper (2014), behavioural change occurs when a new attitudinal norm becomes more widely adopted than an old one. Understanding what is needed to trigger the rise of new norms or how different the new norm must be to be considered changed is far from clear though. For many working in the ‘end VAWG’ space, understanding these triggers has become the most critical issue (see Heise, 2011).

Many VAWG programmes are founded on the Knowledge–Attitudes–Practice (KAP) strategy (earlier seen in population control programmes), which is focused very much on changing individual behaviour. However, research has shown that KAP’s linear process simply does not materialize (see Westoff, 1988: 225). Rates of VAWG remain stubbornly high despite evidence of knowledge and even attitude shifts. Perpetrators of VAWG continue their violence even in contexts where there is some level of public agreement that violence is wrong. Female genital mutilation (FGM) is a good example; a family may commit publicly to end FGM but then go on to cut their daughters in private (Bedri and Bradley, 2017). Mackie’s (1996) analysis of the persistence of FGM and foot-binding shows the importance of moving beyond individual attitudes or ideas to consider the interdependence of decision-making processes. He argues that more focus is needed on the strong role of social sanctions and moral judgements that operate to maintain collective, dominant views and suppress (positive) deviancy at the individual level. Clearly, evidencing sustainable shifts in social norms to include both attitudes and behaviours is complex.

We do agree with scholars who state that understanding the link between social norms and violence is critical (e.g., Heise et al., 1999). Our contention is that caution and a complex understanding of social norms are necessary. Most theorizing around social norms, including functional approaches, cannot explain how contradictory norms can be simultaneously upheld. For example, women may be expected not to undertake breadwinning activities in the outside world, as this is an infringement of domesticity expectations. Nonetheless, they are expected to ensure the well-being of their children and families—even if this means they must take paid employment to maintain idealized forms of domestic life. According to the erratic logic of culture, this infringement can then ‘justify’ violence. In general, social norm theory is often embedded in social role theory, which fails to appreciate the ways in which power operates and intersects at multiple levels (Connell, 1995); it does not adequately capture the power exerted by those in authority to maintain a structure from which they benefit. Codes of behaviour and conformity are presented by dominant sections of society as universal morals to ensure compliance. To deviate or challenge will bring sanctioning that few are willing to risk. This lack of attention to power is evident in approaches to social norm or mindset change, which bypass issues of authority and how critical collective agency may disrupt it.

III A different approach to violence against women and girls: With and beyond Bourdieu

Quite in contrast to mainstream social norm approaches, Bourdieu claims that large-scale social inequalities are maintained not at the level of direct discrimination, but through subtle multi-layered power relations inculcated in and imposed upon the bodies and dispositions of individuals (Jenkins, 2002: 75). He uses the terms ‘field’ and ‘habitus’ to describe this process. First, Bourdieu describes the social world as comprising multiple spheres or ‘fields of practice’, such as education, religion or art. Each field has its own set of rules, concepts and valued things/attributes of a person, which make them (un)impressive within a given field (i.e., cultural capital). Fields can overlap (e.g., education with art on an art degree course), but Bourdieu, nevertheless, posits that fields are relatively autonomous of each other. People consistently struggle within the many fields they inhabit to utilize their cultural capital to flourish as best possible and to make sense of the norms that produce negative experiences.

Next, Bourdieu uses the idea of habitus to describe the durable set of characteristics that social beings develop as part of their exposure to particular social structures in the various fields of their experience. The habitus incorporates our instinctive behaviours, beliefs and attitudes, which are reflected in our everyday interactions as well as through our bodies themselves. One example Bourdieu offers is of women maintaining a quick-step walking style suited to high heels and short skirts even when they are wearing trousers and flat shoes (Bourdieu, 2001: 29). Such dispositions for certain types of social action are presented as a ‘sense of the game’, a largely instinctive knowledge of how to function in your own fields of experience, much like a footballer ‘just knows’ how to kick a ball at a goal without considering the mechanics of it. So the dispositions that make up the habitus are unconscious, ingrained responses to social interaction, appearing to the individual as normal and natural rather than arbitrary or chosen.

The habitus of any one individual will reflect her social position as an actor in specific fields. By responding as such, individuals tend to legitimate and reproduce social aspects of domination, obscuring potential awareness of possible alternative power relations. ‘Belief’, Bourdieu says, ‘is an inherent part of belonging to a field’ (1990: 67). In other words, belief (unquestioning faith) in the field, in its social game, is fundamental to belonging in that field in the first place. From this perspective, social inequalities persist through a logic of practice that appears objective, inherent and natural rather than the arbitrary ‘product of obedience to rules’ (Bourdieu, 1990: 53).

Both the dominated and the dominant, who are in ongoing social relations, are conditioned into a habitus, an ‘embodied history, internalized as a second nature and so forgotten as history – [which] is the active [our italics] presence of the whole past of which it is the product. As such, it is what gives practices their relative autonomy with respect to external determinations of the immediate present’ (Bourdieu, 1990: 56). It is this sense of an embodied and a sedimented history, which is active and renewed in the present, which the ideas of social norm and mindset overlook and which gives life and ongoing, ‘natural’ relevance to social norms, even when they are no longer common sense or part of doxa (taken-for-granted truths). This active presence, however, is not only at the level of ideas and discourse, but it perpetuates through ‘conditionings associated with a particular class of conditions of existence’ (Bourdieu, 1990: 53). This implies that to become integral to the logic of practice, new social norms have to become ‘natural’ elements of the habitus, of dispositions, made in practice and situated within ‘real activity’.

Developing the ‘feel for the game’ that is the habitus is like learning a mother tongue, which is instinctive and requires no thought about the rules of grammar. On the other hand, trying to change within a fully formed habitus, to learn new social rules, is as difficult as learning the complex grammar of a foreign tongue even if one very much wants to. McNay argues that part of Bourdieu’s greatest contribution is that he makes clear how complex and difficult it is to achieve change: ‘it provides a corrective to certain theories of reflexive transformation which overestimate the extent to which individuals…are able to reshape identity’ (1999: 113). The habitus concept emphasizes the embeddedness of norms, values and behaviours in our thinking and our bodies. As such, a linear attempt to ‘raise awareness’ in a series of community workshops, for example, would fail to modify the habitus. The critical question for feminist research has been whether there is room within Bourdieu’s theory to position feminist agency and activism, or whether it suggests an infinite loop of current practice and thought.

Gendered violence may not be related to specific norms, but to a complex of norms and dispositions that are tacit and implicit. In the contexts we have studied, we could identify certain practices with context-specific norms articulated in an ‘official’ ideology. For example, men should be the breadwinners, such that women can only work in lower paid roles that require fewer skills, thereby not challenging the breadwinning status of husbands. Women who travel in public without husbands are shameful and likely to be unfaithful and are, therefore, ‘asking for’ harassment. Women should expect to be beaten if they fail to perform their household chores correctly. We can continue to draw from our data the norms that clearly need to change if transformation is to occur. However, we argue that these statements of social norm and action as outcome cannot be taken at face value—as cause and effect. Moreover, if they were cause and effect, social norm theory would not help us understand why these norms are so embedded and so very hard to change. The wider context is significant. Behaviours might be expressed at the individual level, but they are reinforced and enforced by wider spheres and structures and by relationships that bind different spheres together.

It is important to note here Archer’s disagreement with Bourdieu’s statement that ‘The principles [of dispositions as durable, bodily hexis] embodied in this way are placed beyond the grasp of consciousness and, hence, cannot be touched by voluntary, deliberate transformation, cannot even be made explicit’ (1977: 75, cited in Archer, 2007: 44), which she relates to his apparent denial of discursive thinking in the logic of practice. Rather than a complete absence of such questioning, we believe that our data show that the norms underpinning VAWG can change when women have resilience and a collective agency strong enough to respond to the backlash that efforts to change bring. However, as emphasized later in the text, collective agency should not be confused with lasting shared identity. Cooperation and pooling of agency may be a momentary support for an immediate goal. As such, it may not be enough for sustained disruption and transformation of social norms, which require a more lasting, deeper and critical solidarity.

Before we present the data in support of this view, we note that our position finds strong support in Chambers (2005), who sought to identify a feminist mode of resistance within Bourdieu’s theory. In comparing and bringing together Bourdieu’s work with that of the radical feminist Catharine McKinnon, Chambers has identified two modes of potential change-making in gender relations: consciousness raising and critical reflexivity. Importantly, these processes come with caveats from within the theory that the standard approaches miss.

Chambers notes that Bourdieu seems somewhat indecisive about the concept of consciousness raising. At times, he derides its potential, saying, ‘the symbolic revolution called for by the feminist movement cannot be reduced to a simple conversion of consciousness and wills’ (Bourdieu, 2001: 41, cited in Chambers, 2005). However, he also advises methods for change-making, which seem suspiciously redolent of consciousness raising. For instance, he suggests that ‘[women should] invent and impose forms of collective organization and action and effective weapons, especially symbolic ones, capable of shaking the political and legal institutions which play a part in perpetuating their subordination’ (Bourdieu, 2001: ix, cited in Chambers, 2005).

Bourdieu’s theory implies that if ‘we attempt to identify our habitus, to bring it to consciousness, we can start to resist the social structures to which it corresponds’ (Chambers, 2005: 336). This goes much deeper, however, than raising awareness in communities of so-called harmful practices. As Chambers (2005) insists, the symbolic change enabled by habitus-level consciousness raising does not negate the need for institutional change to remove masculine domination. For ‘the relation of complicity that the victims of symbolic domination grant to the dominant can only be broken through a radical transformation of the social conditions of production of the dispositions that lead the dominated to take on the point of view of the dominant on the dominant and on themselves’ (Bourdieu, 2001: 42). Consciousness raising is a required first step towards wider change; if people cannot see the need for change, they will not demand it. Finally, this conscious raising should be in group form, with women sharing ideas about how to address problems and create radical change (e.g., Henry and Derlet, 1993). The strength of relationships between peer networks is, therefore, critical to their success as a vehicle of resilience and support.

From the above-mentioned discussion, it is clear that while the body of work around social norm theory has helped us to identify processes of normalization around VAWG, it tends to be reductive. It is largely blind to the complexities of dynamic interactions between norms; between norms and social relations; and between norms, social relations, power and context, which bind and are bound by social practice, to the manner in which consciousness, habitus and social logic around violence and gender relations are intertwined. While violence exists globally, the specific social ecology in which VAWG flourishes will differ from place to place. Religious and cultural practices will play a stronger role in promoting violence in some parts of the world than in others. Using the habitus lens helps us to consider how and where change might be possible and to ensure (from a programming perspective) that sufficient resilience exists to weather the likely backlash.

In applying this concept to our research on VAWG, we need to be wary of projecting passive representations of agency, especially of women’s agency. If habitus is viewed as practical knowledge, then in Archer’s (2007) terms, it is procedural, implicit and tacit. Our data suggest that explicit statements of alternative norms may not question the power of the tacit. All our participants express levels of agency by trying to improve their lives through income generation and in expressing the view that certain contexts and behaviours are harmful. Despite this agency, there nevertheless remains for most women a level of acceptance of violence as an integral part of their social world.

Applying Bourdieu’s theory, we start to see certain gendered behaviours and perceptions as symbolic violence because of how they ultimately render women inferior and vulnerable. The hidden power relations are hard to see, and the violence may not manifest itself in clear acts that can be singled out—misrecognition is integral to unequal gender relations. Below, relating the experiences captured in our data to habitus makes it possible to understand the intersections of the symbolic and physical realities of how violence plays out, regardless of how and when agency is exercised.

IV Exploring the data in Myanmar and Nepal

We demonstrate in this section that data gathered during the DFID-funded ‘Women, Work and Violence’ project support the views of change presented earlier. Here, we look at a few key findings that highlight the contradictory ways in which attitudes and norms operate, drawing attention to the need for anti-VAWG programming to look towards a more holistic approach that builds on and supports collective agency.

In Myanmar, we conducted research first in the capital city, Yangon. A total of 50 semi-structured interviews were conducted with women in low-paid, unskilled job roles, most of whom are employed in the city’s garment factories. Around 90% of this workforce is female, but the work is exploitative and consistently contravenes basic concepts of health and safety (echoing Pearson and Elson, 1981). We also interviewed 10 women benefiting from the work of the women-focused non-governmental organization (NGO) Akhaya and 75 women participating in the United Nations Development Programme (UNDP’s) Rural Microcredit Programme, which involves all-women support groups. In Nepal, we conducted 32 interviews on the outskirts of Kathmandu, with women working as unskilled labourers in the poorly paid construction industry. We conducted eight interviews with men from the same community, also working in construction (mostly as masons). The interviews were semi-structured to allow for open-ended conversations. Time was spent before the interviews to build rapport with the participants and to ensure that a safe place was found to conduct them in. More sensitive questions around violence came later in the interviews once comfort levels were well established. The research was conducted in line with a strict ethical protocol, which consisted of follow-up check-ins by professional women’s organizations to ensure no lasting trauma was triggered. The interviews were thematically coded by hand in order to draw in nuances and to compare across intersectional features and country contexts.

Both Myanmar and Nepal have undergone significant conflict and natural disasters (the 2015 earthquake in Nepal and a series of cyclones in Myanmar, the worst on record being Nargis in 2008). The female participants in this study have had to navigate multiple instances of hardship and vulnerabilities to violence. In addition to widespread violence in the domestic sphere, many women were systematically raped and abused during Myanmar’s military rule and Nepal’s civil war/Maoist resurgence (Human Rights Watch, 2014, 2019). In more recent times, the economies of both countries have had to adjust to imperatives of growth. It was critical in our study to capture and consider whether earning an income brought with it any empowering impact. In Myanmar, women are entering the paid workforce through the garment sector for the first time and in significant numbers. This offers them secure income (albeit in poor working conditions) that, in turn, brings a degree of status and freedom. Many of the garment workers have migrated from rural to urban areas and live with other women rather than their families. In Nepal, the construction sector has grown with the post-earthquake reconstruction in which many women have found unskilled employment. Opportunities for women to earn a living have, therefore, expanded in recent years.

In development responses to gender inequalities, microcredit programmes, coupled with self-help groups, continue to be popular in both countries. Microfinance is used to support women in setting up small home-based businesses. Support groups are designed to help develop their capacity as business owners and in problem-solving. Understanding the benefit of such schemes in building resilience to violence was another central research interest. Much has been written about the potential limitations of such schemes, yet they continue to represent a ‘mainstream’ approach to gender empowerment (Goetz and Sen Gupta, 1996).

V Does paid employment impact on social norms around VAWG?

In both contexts, the data expose fundamental problems with any assumption that economic engagement leads to women’s empowerment, including resistance to abuse. All low-paid interviewees noted recent generational shifts in social norm perception about women’s exclusively domestic role, with the contemporary view being that women should—indeed must—provide financially for their families, much like men do. Both men and women confirmed this dominant opinion, although in Myanmar, men also tended to emphasize their own desire to be the primary (albeit not only) breadwinner. The social norm shift here relates to attitude as much as behaviour; while families in poverty have usually been forced to rely partially on women’s incomes, the idea that this is desirable is a notable move away from traditional norms that placed women firmly in domestic roles. Indeed, a large percentage of the women interviewed stated that they would choose to work even if they did not need to, noting that earning an income increased their self-esteem and reduced personal ‘sensitivities’. For example, in Myanmar, one woman stated:

I have to deal with many people when I am working. So I know more about the personalities of people. I also dare to deal with people more. Before, I was just hiding at home.

In Nepal, a woman was very clear:

To me, even if it [the financial situation] is not difficult, working is better because it is satisfactory to my body too. It ‘exercises’ my body…I am ready to do anything; it may be carrying loads or anything else, I am ready to work.

The data from Yangon and Kathmandu demonstrated an increased likelihood of decision-making power among employed women. A significant majority of married women stated that they decide jointly with their husbands how to spend the household income, while noting that non-employed women did not. In Kathmandu, a typical view shared was that:

It is a very obvious thing that if one earns, others will not say anything but if one doesn’t then others try to dominate you.

However, despite these apparently positive signs, the data make it clear that employment has little effect on levels of intimate partner violence (IPV) for poor women. Our interviews revealed no significant disparities (in frequency or form of violence or in willingness to challenge it), even though women earning an income said that they were more likely than their non-employed contemporaries to believe that IPV is unacceptable. Moreover, they specifically correlated the shift in their attitudes with becoming income earners. Poor men with employed wives were also more likely than those with non-employed wives to state that IPV is unacceptable, although they were no less likely to commit it.

Many women recounted extreme and regular IPV. For example, one woman in Kathmandu shared this story, which was by no means isolated among her peers:

‘He started talking to the other girl again. Then he beat me up saying that there was nothing wrong with him just talking to her. So I stayed at my parents’ house for a week. He had said that he did not need me and that I should get out of his house. And if I say anything against his affair then my husband beats me! I have so many bruises on my body.

Women who remained in a marriage with violent husbands talked of regular beatings as a normal occurrence between man and wife. Many women in both countries maintained that husbands beating their wives was just what happens; if a wife does something wrong she should expect to get beaten.

Women in Kathmandu and Yangon also stated that they felt deeply unsafe walking between their workplaces and the community living areas. In Yangon, fear of rape and robbery was common. This was intensified by the long hours garment workers endure, often finishing after dark. This also came through strongly in the Nepal data, with examples of female construction workers organizing groups to walk home together to minimize the risk of rape. A majority of these women also claimed to suffer from ‘depression’ as a result of long hours and poor conditions. There was little evidence of lessening violence in these narratives.

Nevertheless, women in both Myanmar and Nepal report higher levels of self-esteem as a direct result of their employment. The women were clear that the contact they made with the outside world through employment had enabled them to become stronger and more outspoken when necessary. This translates into an increased likelihood of challenging violence in public, including retaliation to violence against oneself and against friends:

Participant: If I have money, I dare to speak up.

Interviewer: If you don’t have money?

P: I don’t dare.

I: How about violence that your friends experience?

P: If I have money, I will challenge violence that my friend experiences. If I have no money, I will not.

Thus, women frequently stated that their experiences of employment have given them the confidence to speak up for a mistreated friend, especially when that mistreatment comes from outside the person’s family (e.g., from superiors at work or in the streets). They also spoke of their own increased capacity to respond to personal experiences of being touched, grabbed or ‘rubbed against’ on public transport:

I hit them with my elbow or sometimes I stamp on their foot.

My friends suggested that I bring a pin (to stab gropers on the bus).

What we see here is an increased confidence to respond to behaviours that may be common, but that are already socially acknowledged to be wrong; it is widely accepted by all (even by perpetrators) that groping strange women is not ‘nice’ behaviour. Paid work increases confidence to resist such male behaviour. However, while these forms of resistance are underpinned by a new gendered confidence, they also fit with the faithful and chaste wife stereotype. The traditional view of IPV as normal is arguably beginning to shift, but it is extremely slow. IPV does not contradict traditional views around gendered morality within personal relationships, and as a result, resistance has not yet been translated into shifts in behaviour. Our data highlight the challenges to change that Bourdieu’s theory of habitus posits, and Fultner’s question that ‘if individual action is seen as socially constituted, that is conditioned by the possibilities available in a given society and culture, how is change possible?’ (2017: 520).

Returning to habitus, we do find shifts; not all aspects of VAWG remain in doxa. Communities and households consist of sets of relationships shaped by history, which, we emphasize, are fundamentally gendered. These relationships, in turn, embed behaviours and attitudes that are reproduced through relational structures. At the same time, the ordering of these relationships and the way in which roles and responsibilities are divided are in a constant state of flux and vulnerable to external shifts and shocks (e.g., economic downturns, environmental disasters, food insecurity as well as development programming). The dynamic feedback loop of the perceptions, attitudes and behaviours forming and formed in habitus suggest the impoverishment of the frequently used idea of ‘mindset’, with its resonance of a writing on stone that can only be chiselled out. Rather, the arbitrary is constantly under threat of being uncovered, limiting the terrain of doxa and what can be left unsaid; the organizing structure of habitus has to be reasserted through orthodoxy. Attitudes are not static, nor do they correlate directly with individual behaviours; both are influenced by a wider structuring of values and beliefs that shift and re-pattern over time. The social norm of women’s employment appears to correlate with increased negative attitudes towards VAWG, but it is yet to be pervasive enough to influence actual occurrences or responses to violence in the home.

We see then not only the positive impact of women’s income generation but also its limitations and consequential costs. Mackie et al. (2015: 16) demonstrate that prevalent attitudes have to change in a large number of people before associated behaviour change starts to take place. As such, we may conclude that communicative action and human cooperation by working together (collective agency) can push towards that greater shift in habitus that makes individual behavioural transformation possible. If we return to Heise’s (1998) ecology model in which the individual, household, community and societal spheres shape each other, individual change is possible with and through intentional cooperation in these multiple spheres. We address this issue in the next section.

VI The influence of women’s groups and social mobilisers

In data from our interviews in Myanmar with women participating in the NGO Akhaya and in the UNDP’s Rural Microcredit Programme, which involves all-women support groups, the importance of bringing women together to share difficult experiences of violence and, subsequently, to work together to build more resilient futures was evident:

In the past, I did not know what to say when someone insults me. After I have attended workshops at Akhaya, I know it is called sexual harassment. I came to know that there is law on that. If someone attacked me, I should not be afraid. I know now what to do, when that happens.

The importance of networks also emerged from the Myanmar UNDP’s self-reliance groups, with members informing us of how the groups operate as a source of advice and guidance when problems arise, including violence:

At first, money was lent and paid back. That would be it. But as time goes by, the small SRG group united the members and really created some positive energy. It created the energy ‘If one falls, then the others have to help that one up.

The data strongly suggest that membership of women’s organizations is linked to greater solidarity among women, who offer each other support when violence becomes an issue. The ability to elaborate the contradictions among work, responsibilities and violence in one’s life initiates a process of attitudinal change. Many women report that before their association with these groups, they would have felt too ashamed to leave a violent marriage and, indeed, would negatively judge others who did so. As associational capital builds, membership of women’s organizations impacts on this tendency strongly, and women speak of learning to offer refuge and emotional support to other women who need to leave abusive relationships, of taking heterodox or deviant stances.

The data from Nepal also reveal that peer networks and specialist local organizations are critical to both the denormalization of violence and the process of building resilience. Strikingly, only those women, who attended cooperative meetings supported by donors like the UNDP (about twenty of the thirty-three women we interviewed), stated that they would discuss instances of violence with other women. Attendees said these peer groups helped them work through what to do and made them feel that they were not alone. Notably, none of the women talked in this way about the empowering potential of sharing experiences with other family members. In these experiences and reflections, social capital emerges as a critical dimension. Independent income generation coupled with supportive social spaces produces or sparks the recognition of a shared/similar experience that makes intentional cooperation and, in turn, collective agency possible. A new feedback loop challenging and imagining a different habitus is in process.

Additionally, a significant number of women also highlighted the important role of social mobilizers, local women who have been identified by organizations as possessing the agency and motivation to support other women. Many talked about them as the first point of contact if violence at home became ‘too much’. As the following quotation suggests, mobilizers can influence women’s actions. In this case, it was the intervention of a mobilizer that encouraged the woman to stand up to her husband:

I did not do anything like… leave my husband. I used to think that I should rather die. Once I talked about this issue to the social mobiliser of this place, she came and asked him not to repeat such acts in the future. But the very next day, he beat me again; he beat me in the middle of the road in front of the other people. Then I threatened him that I would file a case against him and get a divorce.

Supporting, training and resourcing designated mobilizers are simple but effective forms of intervention. However, these individuals cannot be left floating; they must be encased within new gendered networks—organizational structures to counter the hegemonic dominance of aggressive masculinity. These new structures, founded in peer bonds and connections between women who share common experiences, are fundamentally relational as is their habitus; a new drive to change the world they inhabit can disturb the latter. Linking individuals to mobilizers and then into a solid and permanent women’s organization holds the potential to create the kind of disruption needed: If it happens nearby we tell the social mobiliser of this place. They belong to the same organization. There will be discussion at their place and then reporting happens.

The combined data suggest that, as Bourdieu argues, the best way to change habitus is for women to ‘invent and impose forms of collective organization’ that create effective, symbolic weapons to shake the institutions that undergird their subordination (Bourdieu, 2001: ix). Research participants who did not belong to women’s collectives struggled to make the mental shift towards challenging the dominant patriarchal tenets of the domestic sphere, even when they moved to do so in public. If habitus is like a tightly woven cloth, a thread pulled out does not unravel it. Patriarchal habitus has not weakened to the extent that these women have stopped normalizing and accepting violence in the home. Yet, supportive, women-dominated environments can help the gradual fraying to continue. The research participants belonging to women’s collectives report a firm awareness of rights and, more importantly, a collective will to see them enforced. The dilution of the effects of social sanctions through the presence of a supportive group allows for a synergy in challenges to habitus.

VII Conclusion

The study presented here gives a small but significant insight into the complexities of how norms around violence play out in the everyday lives of women. We have argued that by applying Bourdieu’s habitus concept alongside understandings of collective agency, we are able to appreciate more closely why women seem to comply and even maintain structures that leave them subject to ongoing abuse. We highlight the need to see violence as a key strand locking and maintaining relational structures in place. Our data show moments when individual women confront the pressures of the status quo and deviate to resist violence. We present differences between women who accept that IPV is a fact of life and others who are defiant and prepared to challenge violence.

Two dimensions emerge as key to these individual differences in preparedness. One is recognition of the range of violence that women suffer. Gender violence as a whole is no longer doxa, as women acknowledge the contradictions between the need for their earnings, their responsibilities, their desire to work and being subjected to public and private violence. Norms of harassment at work and while travelling to and from work, in particular, are questioned and refused. The second dimension involves access to a supportive network or organization that provides safe spaces, where women can share experiences, receive advice and work through the implications of the forms of violence they are exposed to, including IPV and other forms of domestic violence. Associational experience helps in recognizing the range of violence women suffer, in fracturing the morality of extant gender relations and its power—including the acceptability of violence—and engenders the motivation to seek change through cooperation with others.

Others, including Htun and Weldon (2012), have also argued for the critical importance of women’s organizations working to end violence. Shifts in understanding of violence and organization/network membership seem often to go hand in hand. The recognition of violence may also be aided by the discussions and mutual sharing that take place in women-only spaces. The courage to face it and talk about it may come about through mutual support among women (Keck and Sikkink, 1998; Weldon, 2006)—a building of associational capital through making the tacit explicit and enabling the elaboration of cultural dissonances and felt hurts and injustices.

A focus on positive and accumulative deviance through the concepts of habitus, doxa, orthodoxy and heterodoxy has been useful here. ¹ Stories of positive deviance do not derive their power from externally advocated norms, but from the real contradictions and conflicts within practice that social logic is not always able to explain away. In our data, we see examples of women who must navigate between (a) the need to build a sustainable income for family security and the consumption desires of their families and (b) accusations of inappropriate and ‘unwomanly’ behaviour (see also Nussbaum and Glover, 1995). Figures who are resolutely vocal against such accusations are vital, not as external figures though they may be such but because they elaborate and make explicit contradictions in and through organizational work. The local organizations we have spoken of are usually run by an individual whose life story reveals her as a positive deviant. Such women either find support in or have managed to build collective networks that push women to recognize that violence, in all its forms, is wrong, and that they should not tolerate it. The mobilizer and the organization are able to assert and argue that the violence is not individual deviance on the part of victim or perpetrator, but an unacceptable individual act and a social pathology that must be changed. They then can motivate others who are less empowered and more fearful of the consequences of their income activities and success.

However, what is also striking is the continuing persistence of oppressive and exploitative work conditions, poverty and the power of toxic gender relations within the home and public domains, bolstered by macroeconomic processes and political pressures. These undergird the backlash to small shifts and changes and undermine the orientation and ability to resist and transform. Without further institutional change, it is the limited regulated liberties of some individuals that have emerged. New social norms have yet to become ‘natural’ elements of the habitus, of dispositions made in practice and situated within ‘real activity’.

Yet, these figures, organizations and regulated liberties are critical in questioning doxa and the organizing structures of the habitus. These questionings and initiatives, enabled when women are also earning for themselves (though it may be little), part of supportive organizations and in sustained contact with positive deviants, lay the grounds for further struggles for change and transformation to end violence and for gender equality.