Backlash in Action? Or Inaction? Stalled Implementation of the Domestic Violence (Prevention and Protection) Act 2010 in Bangladesh

The Bangladesh government adopted the Domestic Violence (Prevention and Protection) Act in 2010. While the formulation and enactment of the Act was an achievement for the government and the coalition that championed it (the Citizens’ Initiative Against Domestic Violence, CIDV), its implementation has been weak. In Bangladesh, dominant social norms have led to stereotyping domestic violence as a trivial, personal matter, which has led to service providers delegitimising and deprioritising the issue. Combined with all the gender biases that operate within the judiciary and other government departments, this has resulted in an unsatisfactory use of the Act for women seeking redress from violence. This paper explores to what extent this lax implementation is due to the weak capacity of implementing agencies versus deliberate/intentional inaction in the form of backlash. It also examines how women’s rights organisations have articulated feminist voices in terms of agenda and framing and used collective agency to counter the pushback.


Introduction
Domestic violence (DV) in Bangladesh is recognised as an offence that the state, society, and families have to prevent and take action against. According to data from the Bangladesh Bureau of Statistics, almost three in every five women (57.7 per cent) have experienced some form of physical, sexual, or emotional violence in their lifetime (BBS and Statistics and Information Division 2015). The Bangladesh government adopted the Domestic Violence (Prevention and Protection) Act (DVPPA) in 2010, which refers to DV 'as any form of physical abuse, psychological abuse, sexual abuse or economic abuse against a woman or a child of a family by any other person of that family with whom the victim is, or has been, in a family relationship' (Government of Bangladesh 2010).
It has been 12 years since the enactment of the Act, yet the struggle for effective implementation continues. DV continues to dominate the private sphere or family space for Bangladeshi women (Sultan et al. 2021;Mahpara 2021). Although progress has been made by introducing legislation, adoption is not enough (Nazneen and Hickey 2019). The lack of progress in implementation of the law demonstrates that political action has fallen short of addressing the problem. The government rhetoric about the importance of addressing violence against women and girls, as reflected in various policy documents and laws, does not translate into the allocation of financial or human resources, or the development of capacity for the prevention or redress of DV.
This paper asks the following questions. In what ways has backlash from oppositional actors against the DVPPA 2010 manifested itself in the weak implementation of the law? What strategies and tactics are women's rights actors deploying to counter this backlash? How successful have women's rights organisations (WROs) been in ensuring the incorporation of feminist voice in terms of setting and framing the DV prevention agenda and using collective agency to counter the pushback?
We argue that the weaknesses in implementation are related to a subtle form of backlash against women's equality among service providers, particularly the judiciary, police, and even the Ministry of Women and Children Affairs (MoWCA) and its staff. The backlash takes the form of deliberate inaction in implementing the provisions of the law; for example, not referring complaints and complainants to the law. The paper also demonstrates the various ways that social norms and ideologies about the family -prevalent among women and men, judiciary, and service agencies such as the police and health services -result in the neglect of protection and redress for women and girls subject to DV.
We conducted a series of interviews with proponents and opponents of the law at national and sub-regional levels, and an analysis of secondary documents. A number of women activists interviewed mentioned that inaction is deliberate and intentional, rather than a lack of capacity or resources. They emphasised that the skills could have been developed and resources allocated if the commitment was present. 1 This was also emphasised by Citizens' Initiative Against Domestic Violence (CIDV) allies: 'Lack of capacity is obviously connected to neglect'. 2 These assertions are corroborated by our observations and interviews with legal practitioners and service providers.
The paper also reviews the strategies taken by CIDV and its members to counter the backlash faced and to strengthen the implementation of the DVPPA 2010. The paper will argue that while CIDV was successful in framing the agenda on the need for and importance of a law on DV, its collective agency and voice have had only a limited impact in setting the agenda in terms of implementation. Similarly, the strategies and collective action that were effective during policy formulation seem to have now lost their novelty and impact at the national level and it is only at the local level that some positive impact can be found in countering the pushback.

Objectives
The question of whether lax implementation of the Act is due to weak capacity or deliberate/intentional inaction is one that needs to be examined. When it comes to understanding how backlash operates in a state such as Bangladesh, which obviously suffers from capacity and resource constraints, it is important to distinguish between a lack of state capacity and deliberate inaction (Flood, Dragiewicz and Pease 2018;Lombardo and Mergaert 2013). Our proposition is that there are deliberate efforts by the actors responsible for implementing the act to delegitimise the issue of DV and deprioritise it by not allocating sufficient financial or human resources (Flood et al. 2018). The Act is also stigmatised by government agencies as an 'NGO law' (Nazneen 2017), as it is known to have been promoted by women's rights and human rights organisations -a considerable disadvantage in a context where non-governmental organisations (NGOs) are often seen negatively. Combined with the gender biases that operate within the judiciary and other government departments, the stigmatisation (Sen, Vallejo and Walsh 2017) of the Act results in its unsatisfactory use by women and their inability to access remedies by using it. Our research tries to understand these biases as stigmas play out in the Act's implementation, taking into account both intention as well as capacity issues.
Both from an academic and activist viewpoint it is important to make the distinction between lack of capacity and deliberate inaction. Academically, identifying the issue as one of deliberate inaction allows us to analyse it within the backlash framework. On the other hand, action to overcome the lax implementation of the DVPPA 2010 will depend on the analysis and identification of the problem. While activists have suspected for some time that it is a matter of deliberate inaction, the present research brings out how delegitimisation, deprioratisation, and trivialisation of DV and its victims constitute the backlash against the Act. However, if the strategies taken by women's rights and human rights organisations to address the lax implementation assume the problem to be one of lack of capacity, they might not be as fruitful. This paper will review the strategies taken by the coalition and its members to counter the backlash in terms of how effectively they have voiced their concerns and demands and their collective action. In section 3 of this paper, after describing the socio-political as well as sociocultural context of the last decade in Bangladesh, we provide some background to the adoption of the DVPPA 2010. In section 4, we briefly discuss the concepts of backlash and resistance, voice and agency, and movement strategies that we use throughout the paper. We describe the methodology in section 5. Section 6 provides information on the progress made on implementation as well as on the duties and responsibilities of the concerned actors. Section 7 presents the findings and analysis. It discusses the legal difficulties of implementing the law, as experienced by legal practitioners and legal aid organisations, and the attitudes of the legal practitioners towards women, DV, and justice. We then review the barriers at the level of the institutional actors responsible for implementation. In section 8, we analyse the strategies taken by the WROs that were the proponents of the law and that are actively seeking its implementation, to overcome the barriers and counter the backlash faced and review the results achieved. Section 9 reviews the findings and analysis in the light of the various concepts used, and the conclusion in section 10 finally reviews the research questions on that basis.

Socio-political context
At the time of formulation of the Act, Bangladesh had a multiparty democracy. Gender equality was high on the government's agenda, with allies of the women's movement present in key government positions. The government was trying to demonstrate its gender credentials to the international community and nationally, which led to the rapid adoption of the Act in 2010.
Providing an appropriate legal framework is just the beginning, and its effective implementation heavily depends on the priority given by the state to the issue of DV. Unfortunately, since the law's enactment, the issue has lost its priority, at least relatively. Currently, other issues such as child marriage and rape law reform are considered more important than DV (Sultan, Mahpara and Tasnin 2022). Many prominent allies of the DVPPA 2010, including the then minister of MoWCA, who could have worked as champions in ensuring the Act's effective implementation are no longer in place. This has hampered strategic alliances between women leaders in the coalition and in parliament (Mahpara 2020a(Mahpara , 2020b. Moreover, the political landscape of the country has also changed significantly, transitioning into a dominant party rule, where power is configured in a large political party (in our case, the current ruling coalition), with limited opposition from other political parties (Hassan and Raihan 2017). These changes have not only made the state less responsive to civil society and gender justice actors, but also caused these actors to lose their avenues for lobbying alternative political parties to promote their agenda, considerably shrinking their space for negotiation and bargaining (Mahpara 2020a(Mahpara , 2020b.
The present context of politicisation at all levels of the country, within the government and civil society, also spills over into the legal profession. The context is also characterised by various practices of corruption that affect women's access to justice, which became apparent in interviews at the district level with various lawyers and human rights activists.

Sociocultural norms
Bangladesh society is generally characterised by discriminatory and restrictive gender norms, which devalue women and girls. Marriage is the norm, with the man being the provider. Women are responsible for maintaining family honour by safeguarding the family reputation in terms of chastity and behaving like obedient wives and caring mothers (Rozario 2001;White 1992).
Male family members serve as guardians in the family to protect the female members and to discipline them if necessary. Therefore, chastising or even physically beating women is condoned. While this has religious justification, 3 violence against women (VAW) by guardians is traditionally socially acceptable (Gibbs et al. 2020). According to national statistics, women and men widely believe that it is justified for husbands to beat and assault women for reasons such as arguing with their husbands, going out without informing them, burning food, neglecting children, and refusing to have sexual intercourse (BBS and Statistics and Information Division 2015). There is a 'culture of silence' as women feel ashamed to openly acknowledge their experiences of DV; often stemming from their feeling that they are the ones at fault (Naved et al. 2018).
Social and gender norms are important for explaining how individuals, the community, and institutions respond to women who have broken the culture of silence and raised their voices in protest against the violence they have faced at home, who want the violence to be stopped or prevented, or seek redress for the violence they have experienced (Sultan et al. 2021). The influence of norms on the responses of law enforcement, judiciary, and health authorities to DV complaints has been partially documented (Naripokkho 1997;Yasmin 2020).
Research by the BRAC Institute of Governance and Development (BIGD) on access to justice for DV survivors during the Covid-19 pandemic has highlighted the importance of the perceptions and norms of the service providers to women's ability to advance their 'justice journeys ' (Sultan et al. 2021).
The prevalence of gender discriminatory social norms and the never-ending efforts to preserve families at any cost is pervasive. These norms determine the attitudes of the legal practitioners and are also shared by the implementing agencies, allies of the WROs, and even sometimes by the WROs themselves (Naripokkho 1997;Sultan et al. 2021). The norms are used to justify certain forms of action and inaction. Rather than using the Act and its provisions, importance is given to dispute resolutions through mediation, or through village/community mediation (salish) (Sultan et al. 2021). There is still a reluctance for the state to interfere in what is considered to be a private matter between a couple; one of the former finance ministers made the following remark to a WRO when its members went to meet him, which still resonates with many: 'Will you now even enter the spouse's room? (ekhon ki apnara shoamir ghoreo dhukben ni?)' 4

Background of the Act's adoption
The DVPPA 2010 has been praised as an example of outstanding collaboration between the government and the women's movement. The progressive provisions include protection orders for women, the right to reside in the marital 3 Surah An-Nisa allows men to discipline their wives gently (Verse 4:34, Quran n.d. There was opposition from Ministry of Law officials and male Members of Parliament (MPs) with concerns around 'breaking of families', recognition of marital rape, and 'women misusing the law'; however, the adoption process was facilitated by the full support of the Prime Minister, who saw this as a way of gaining international approval as well as pleasing the national women's rights constituency (Nazneen 2017). Her support sped up the parliamentary process and within only three months of placement in parliament, the Act was passed in 2010.

Defining and operationalising concepts
We use the concepts of backlash, resistance, and inaction, as elaborated by Flood et al. (2018) and Lombardo and Mergaert (2013), to analyse the stalling and barriers to the implementation of the DVPPA 2010. We also use the concepts of voice and agency to analyse how WROs promote the implementation of the DV legislation and also counter the backlash faced while doing so. Finally, we will be defining actions, strategies, and repertoires of social movements such as the movement for DVPPA implementation.

Backlash
Let us now look at how the concepts around 'backlash', 'resistance', and 'inaction' help us better analyse the struggles and blockages in the implementation of the DVPPA. Given that manifestation of backlash may be different when policy is formulated, we engage with the backlash literature to identify the ways backlash manifests itself in policy/law implementation processes.
Flood et al. (2018) have defined backlash as 'resistance' against progressive change. They have used the terms 'backlash' and 'resistance' interchangeably to refer to any active form of pushback against progressive and feminist policies, perspectives, and programmes. This resistance is a response to actual or perceived challenges to existing hierarchies of power and, in most cases, it aims to push back progress made on gender equality, civil rights, and social justice (Faludi 1991). This can be done both individually and collectively, and formally and informally (Flood et al. 2018). During progressive change, resistance is manifested by the members of privileged groups (advantaged by the status quo) to restore, maintain, and increase their power position in contexts of inequality and injustice.
Often resistance takes the form of denial of a problem and can be manifested through disavowal -the refusal to accept responsibility for dealing with the change process (ibid.). Backlash does not always necessarily involve direct opposition or hostility to feminist initiatives, but it can take the form of trivialising gender issues through the demonstration of lack of interest, inaction, and nonparticipation. An example of this is how DV is perceived to be a 'trivial issue' in Bangladesh that is best kept private and within the family. This has led to the persistent discomfort and dilemma among state authorities about whether they should intervene in a 'private issue' like DV. Disinterest also refers to inaction by implementing agencies. Lombardo and Mergaert (2013) emphasised how certain relevant stakeholders use different excuses for their inaction. One of these includes not taking responsibility for the change process because of the lack of ownership, and only expecting the ones who mobilised for the change to take responsibility. This was observed when the government implementing agencies chose to label the DVPPA 2010 as an NGO act.
This also includes refusing to implement change by not allocating resources for its implementation, not enforcing new policies, not setting standards or timelines to monitor the change, co-opting the process by delegating the change to those who disagree with it, and actively sabotaging the change process. Non-acting is described by Lombardo and Mergaert (2013) as an implicit form of institutional resistance, distinct from more explicit forms such as policy discourse and actions that are in opposition to the goal of promoting gender equality. 'Resistance' in some accounts includes processes that preserve the status quo such as institutional inertia and lack of support in the form of non-engagement, understaffing, under-budgeting, insufficient gender training, and so on (Davidson and Proudford 2008;Lombardo and Mergaert 2013). Lack of action is also a form of agency. This is applicable in the case of the DVPPA's implementation, as it suggests the agency of service providers in deciding whether to take action and steps to change the status quo. It also includes dismantling change processes that have already begun, such as shutting down of new policies, which is an active form of repression (Agocs 1997).
We have been inspired by various authors who have spoken of strategies of backlash in terms of various types of marginalisation, stigmatisation, and delegitimisation. Sen et al. (2017) identified marginalisation as actions that indirectly target women through tactics such as silencing, stereotyping, and stigmatisation. Hossain et al. (2018) speak of delegitimisation when discussing the closing of civil space and the backlash against NGOs. Van der Borgh and Terwindt (2012) identify actions and policies that are variously used to restrict NGO operational space as backlash. These include stigmatisation and negative labelling, including criminal and social stigmatisation of specific actors such as NGOs.
In the face of such deliberate inaction and disinterest, it remains important to understand how WROs collectively use their voice and agency to push forward their agendas in sustaining movements.

Voice and agency
WROs are the primary means of expressing opinions, positions, or interests of various women's constituencies. These organisations also seek to influence policies, laws, institutions, and organisations towards gender justice goals by demanding responsiveness and accountability from the government. They can use their collective voice to legitimise claims, build constituencies, and develop strategies to mobilise at local, national, and international levels (Nazneen 2022).
In doing so, they encounter various forms of backlash, which they try to counter by coming up with various strategies using their voice and agency. In contesting power relations, voice can be used to bargain, negotiate, and defend one's interests. The observed countertactics to backlash include framing of the demands, evidence generation, alliance-building, and developing alternative narratives to challenge stereotypes and gender norms, as well as seeking accountability (Lewin 2021;Batliwala 2012;Tarrow 2011; Nazneen and Mahmud 2012; Nazneen and Hickey 2019), which will be discussed in this paper.
The term 'voice' is used as a metaphor for powerful speech and is most often associated with acts or arguments that influence public decisions. Voice can be individual or collective, public or private. For women's public voice to be effective, it needs to be heard by others and acted on (Goetz and Musembi 2008: 3). While voice has a value in itself of allowing women to be able to articulate and be heard (intrinsic), it also has an instrumental value of enabling women to attain various outcomes that are important for substantive equality and enjoyment of rights (Goetz 2003). In this paper, voice would be that of WROs protesting and demanding the elimination of DV.
As described by Goetz and Musembi (2008), voice has two dimensions: contents, or what is said; and performativity, or how it is said (Goetz 2003). Both contents and performativity are influenced by formal rules on voicing, gender norms, and power relations (Oosterom et al. 2022). For example, the content or what may be said is influenced by gendered rules and norms around what is acceptable for women and men to articulate, e.g. if it is considered acceptable to discuss family matters such as DV publicly. The performative aspect or how women voice their opinions and concerns is mediated by gender and social norms on what are the expected gendered patterns of behaviour; for example, women are not expected to assertively demand their rights.
There is a question of whose interests do various voices represent?
If voice is about powerful directed speech that advances the social, economic and political position of the speaker, that articulates the researched preferences of the speaker; if voice should be about building a platform of constructing an interest group capable of demanding responsiveness and accountability, then we will know we are advancing towards it… (Goetz and Musembi 2008: 12) It is recognised that women's collective action has pioneered innovative ways of using voice and has been stronger on voice (mobilising, forming alliances, and raising the profile of various issues) than on building institutions that can endure and engage at the level of formal politics (Goetz and Musembi 2008: 24-25, 31). This paper seeks to assess how effective CIDV's strategies of voicing have been in terms of demanding responsiveness and accountability. In face of cultural values, women's empowerment is dependent on collective voice and solidarity in the public arena, as well as individual assertiveness in the private sphere. Women's organisations and social movements play an important role here to create conditions for change (Kabeer 1999). How far CIDV and its members have been able to do so will be explored further in this paper.
The ability to voice is a form of individual and collective agency. In development literature, agency is the ability to make strategic life choices and take actions to attain desired outcomes (Agarwal 2010;Kabeer 1999). Quoting Kabeer: Agency is the ability to define one's goals and act upon them. Agency is about more than observable action; it also encompasses the meaning, motivation and purpose which individuals bring to their activity, their sense of agency, or 'the power within'. (Ibid.: 438) Kabeer cautions that while agency tends to be operationalised as 'decisionmaking' in the social science literature, it can include bargaining and negotiation, deception and manipulation, subversion and resistance as well as reflection and analysis (ibid.: 438). This paper will be applying these concepts to the analysis of the behaviour of the service providers and the WROs struggling to establish their rights.
While looking at the struggle to protest and end DV, we will examine both individual and collective voice and agency, which will involve understanding: (a) consciousness (perception and understanding of DV); (b) voice (articulating one's experiences and views about DV and best options that are needed to secure protection/redressal/support); and (c) action or lack of action (whether taking steps to change the status quo, the functioning of services, and options available for women).
In section 7, we will analyse the implementation of the DVPPA 2010 from the perspectives of various implementing agencies such as law enforcement, Department of Women's Affairs (DWA), and judiciary and legal practitioners to understand how social norms and backlash and resistance in the form of inaction has played out in stalling the implementation of the Act. We will use the concepts of voice and agency to understand the actions and the reactions of both the WROs and the opposition actors.

Strategies to counter backlash
In section 8, we will analyse strategies taken by gender justice actors to counter backlash. We will use the term 'strategies' as outlined in social movement theory to understand better what sort of actions and responses social movements have taken to both advance their cause and respond to the backlash. The struggle for the implementation of the DVPPA 2010 can be classified as a social movement according to the definition of being 'collective challenges, based on common purpose, and social solidarities in sustained interaction with elites, opponents and authorities' (Tarrow 2011: 9).
Demanding a separate law for DV was a gender-transformative policy change as it challenged women's subordinate status, the traditional understanding of DV as private, and male power to perpetrate violence (Htun and Weldon 2018; Sultan and Mahpara 2020). The struggle for implementation of the Act continues, or even aggravates, the challenge to traditional norms as it entails translating the law into action. In order to translate the law into action, one of the most significant strategies for women's rights activists working on the issue of DV would therefore require coalition-building to overcome opposition from implementation actors wanting to preserve male power and privilege (Nazneen and Hickey 2019). Similarly, Batliwala (2012) states that it is necessary that women's rights activists form collectives to build a mass base in order to amplify their voice, vision, and struggle. Nazneen and Mahmud (2012) also refer to comparative studies which showed that anti-DV reforms required women's rights leaders and activists to adopt strategies in the form of building alliances, using personal networks, framing their issues in a non-contentious way, targeting certain parts of the state bureaucracy, cultivating allies even among male politicians, and establishing their expertise and experience on gender equity agendas. Batliwala (2012) emphasises that movements should focus on public opinionbuilding as well as advocacy strategies, and have clear internal or external targets for engaging in the change process, such as through their own membership. She further argues that one of the key elements of movementbuilding requires women's rights activists to raise their consciousness around women's social, economic, and political structure of oppression, and therefore build awareness for larger societal change. This would entail their framing of the problems they are protesting and the change they want to see in a way that motivates and inspires the membership. Nazneen and Hickey (2019) use Benford and Snow's (2000) framework, stating that women leaders often use motivational framings when packaging demands to generate greater attention and commitment, and linking movements with international movements and conventions to gain widespread support. It is also crucial that movements generate knowledge and evidence, by democratising the processes of learning and knowledge generation within and by their movements. These movements should create the space, respect, and concrete mechanisms for their members to participate in theorising, analysing, monitoring, and evaluating their experiences (Batliwala 2012).
Social movement theorists such as Tarrow (2011) have classified 'repertoires of contention' as ranging from violence and confrontational actions, to disruptive, as well as contained actions. Various movement repertoires include such actions in varying degrees which vary historically according to the political context, depending on the 'political opportunity structure' (Tilly 1978), with contained actions predominating as 'offering the advantage of building on routines that people understand and that elites will accept or even facilitate'. Batliwala (2012) has discussed that to ensure the sustainability of a movement action should change and adapt to new realities, and there should be dynamic action strategies which change over time depending on situations, as static ones reproduce the same analysis and strategies over long periods of time without spaces for critical reflection and re-strategising. Gamson (1989) highlights the importance of the media and how social movements, both contentious or not, use it. 'Media spotlight' makes any confrontation between challengers and the opposition a 'potentially critical event' (ibid.: 459). It can help counter opposition, create sympathy among public opinion, and can contribute to the framing of demands.
Finally, feminist movements should not only be concerned with changes at the formal institutional level, but also prioritise the informal level (Batliwala 2012). The use of informal networks allows women's rights leaders to gain access to policy spaces (Nazneen 2019). These informal practices also influence what kind of institutional and policy changes are possible and the extent to which the policy reforms are successfully implemented (Waylen 2017).
The repertoire of strategies for mobilising support and building constituencies has allowed WROs' greater legitimacy and strength as advocates of women's issues (Nazneen and Hickey 2019).

Methodology
We conducted desk reviews of secondary sources to develop a policy map on the DVPPA 2010, reviewing the key current areas of contention, the gains or tangible progress, how these gains are under threat, key actors for and against the gains, and what strategies are being used to counter the threats to the gains. A continuous media tracking exercise was also conducted to track important events/incidents around DV.
To understand the current implementation status, we engaged with CIDV and, more intensively, in a research partnership with the BMP, which is one of the active members of CIDV. BMP is one of the pioneer feminist organisations leading the women's movement in Bangladesh, and the movement against DV is a core area of its work. In consultation with BMP, we selected their Munshiganj branch, where BMP members and their allies have been actively mobilising in the movement against DV, to study the efforts for the implementation of the DVPPA 2010 and the challenges and opposition faced. Additional primary data was collected through key informant interviews (KIIs) with women's rights leaders, stakeholders, and allies who are involved with the movement for the implementation of the DVPPA, both at the national level and in Munshiganj.
In-depth interviews (IDIs) were conducted with seven CIDV members, three BMP members in the central office, and five BMP members in Munshiganj. Fifteen IDIs were done with BMP/CIDV stakeholders in the Act's implementation (allies at national and local levels). Seven interviews were done with actors BMP/CIDV view as opposition at the implementation level. A total of 32 interviews (IDIs and KIIs combined) were carried out and transcribed. In addition, a focus group discussion (FGD) was done with five judicial magistrates and judges with experience in the DVPPA 2010.

Present context of implementation
We will begin with a quick review of the steps taken for implementing the DVPPA 2010. We find that even 12 years since adoption, the effective implementation of the Act is hampered by a number of barriers. In terms of service delivery, there is a lack of infrastructure; for example, inadequate shelter homes, and a shortage of trained and experienced service providers such as medical professionals to counsel violence survivors (Huda 2016). It also seems that there is a lack of conceptual and technical understanding of the DVPPA 2010 amongst legal practitioners, mostly judges and lawyers, who still seem to be unaware of (or feel uncomfortable) using the Act (Sultan et al . 2021). There has also been bureaucratic resistance, such as delays in the formation of rules and forms, inadequately trained enforcement officers in DWA, and MoWCA's limited resources and poor coordination with other ministries (Huda 2016;Yasmin 2020). Although an important part of planning for the DVPPA 2010 included budgeting exercises for resources, after the Act was approved the necessary budget was not allocated.
Social norms also play a crucial role in determining how legal remedies are being used by survivors to seek redress from violence. Survivors often prefer seeking justice through community shalishes, which are influenced by social norms and do not maintain legal parameters. Even if survivors resort to formal channels, there is a tendency amongst service providers like the police to go beyond their mandate and advise DV survivors to try mediation (Sultan et al. 2021;Yasmin 2020). All these factors will be explored in greater detail below.   Table 6.2 provides a brief overview of the roles and responsibilities of implementing agencies as stated under the DVPPA 2010 and the 2013 rules (Government of Bangladesh 2013). We will be referring to these provisions while reviewing their actions or lack thereof in the sections that follow. The Rules (2013) further specify the role of the Enforcement Officer in the 'follow up (of) the progress of the DV cases with the assistance of service providing organisation or by his/her own initiative' (Section 4.d).
Service provider (e.g. legal aid NGOs) (a) record the incident in the prescribed form if the victim so desires; (b) refer the victim for medical examination; (c) if required, refer the victim to a shelter home, and; (d) perform such other duties and responsibilities as may be prescribed by rules.

Medical service provider
On request by a victim or on her behalf, a police officer, an enforcement officer, a service provider or any other person, the person in charge of the medical facility in the hospital, clinic or medical centre shall provide medical aid to the victim.

Shelter home
On request by a victim or on her behalf, a police officer, an enforcement officer or any other person, the authority responsible for a shelter home shall provide shelter to the victim therein.

Backlash to the Act's implementation
We will now present the research findings on the backlash on the implementation of the Act. The barriers to implementation are divided into three categories. First, we discuss the loopholes and issues with the law that have made implementation difficult. Second, we discuss the backlash and resistance to implementation among legal practitioners due to norms translating into various forms of attitudes and actions which are the manifestations of backlash. Third, we discuss the deliberate forms of inaction and the result of backlash against the DVPPA 2010 at the service provider level, of which we focus on the MoWCA, DWA, and the police.

Legal challenges
Several issues with the substance of the Act continue to make implementation difficult. These have been identified by practitioners and other researchers (Huda 2016;Yasmin 2020), and we discuss them in the section below.
The DVPPA 2010 is quasi-civil in nature, i.e., neither civil nor criminal. Instead of giving criminal sanctions, the Act was designed to provide civil remedies. The Act did not want to criminalise DV as survivors often want to continue their marriages (which a criminal case would jeopardise). But it has been seen that the law is neither able to give a civil remedy nor address the gaps in criminal remedy. However, the court that is charged with dealing with DV complaints is the one that deals with criminal offences. 'there is also no proper training, awareness building, or information sharing among the judges to really understand how this law's implementation should be '. 6 There are also procedural shortcomings, like delays in the hearing of cases, because of the large number of cases under the judicial magistrate court. This is problematic and causes further delays in the cases, hampering the protection of victims when they should be given immediate attention. Some interviews with CIDV members brought out the lack of clarity in the Act in describing how the protection orders were to be carried out. Although Section 25 of the Act specifies that 'the process server or the police officer or the enforcement officer shall serve the copy of the order within three working days and submit the service return to the court along with a certificate that such order was duly served', one of the interviewees, who has a legal background, alleged that the Act does not specify which representative should be sent to inform the family after the court issues the protection order. There is also a lack of confidence that a court order for protection or 'right to reside in the shared home' 8 will ensure safety.
In addition to delays, no follow up is done to ensure that the orders are being followed properly by the perpetrators, although this responsibility should be assigned by the court to the enforcement officer, service providers, or the police (Government of Bangladesh 2010). This makes survivors feel unsafe, and it remains unreported whether they face further violence after receiving the protection orders. Talking about one of her clients, a BMP member quoted her as saying: ' In rare cases, when survivors receive protection orders, they feel uncomfortable living in their marital homes. Although this is not mandatory, the respondent below believed it to be, and felt it was uncomfortable for the victims: Victims are primarily scared to stay in the in-law's house while they file a case, which is mandatory for this particular act of the protection order. They are supposed to stay in that house and call when something happens; then we would send a representative to the scene. 10 Some CIDV members reflected that there are no performance indicators that would make the implementing agencies answerable, and therefore there is no accountability. There is a self-criticism from CIDV that this is something they overlooked while developing the Act.
Another difficulty arises when survivors try to seek redress using the DVPPA for psychological abuse as proving this form of abuse is difficult. The definition of economic violence has also proved to be inadequate.

Attitudes of legal practitioners justifying inaction
The inadequacies with the law, as described above, are not serious enough to justify the lack of its application. We will now review the attitudes of legal practitioners towards the law and the norms they reveal, which will help us to understand the resultant disavowal, disinterest, inaction, trivialisation, and delegitimisation -manifestations of the backlash against the issue of recognising DV as a human rights violation (Flood et al. 2020;Lombardo and Mergaert 2013). This section will provide practical examples of the manifestation of backlash that was discussed above in section 4.
While almost all the 15 legal practitioners interviewed (who were not direct proponents of the law) provided lip service to the DVPPA 2010, hailing it as 'wonderful', they immediately added that it was unusable or that they had never used it. As explained by CIDV coalition members and other proponents of the Act, they had to spend a lot of time and effort before and during the formulation, and after the enactment of the Act to provide an orientation on the law, its objectives, and its functioning. In this section, we will discuss a few dominant gender norms that explain the attitudes and actions regarding DV cases. We will then discuss some of the forms of backlash that are manifested as a result of these norms.
In general, both male and female lawyers in Munshiganj shared the same norms about the sanctity of marriage and the need to maintain family honour. They revealed their stereotypical ideas -women are responsible for the violence they face, women are women's worst enemies, and men are the victims of women's greed and adultery. As articulated by many, the priority is to save marriages. And this for the sake/benefit of the women and children. Mediation by law enforcement authorities is presented as important to 'protect families'. Also, as articulated by a women's rights activist, 'every victim who is beaten by her husband also wants to live with him. It is our culture '. 11 Hesitancy in coming forward with complaints of DV is due to women and their family members feeling that going public and filing legal cases ruins husbands'/families' honour. There is a stigma attached to litigation, where internal family matters are disclosed in public. Moreover, men see legal complaints by wives as insubordination and a questioning of their authority and react by sending divorce notices. Even filing applications for protection orders results in divorce notices from the husbands. Therefore, though the Act wanted to prevent violence while facilitating the possible preservation of marriages, this has not worked. DVPPA 2010 does not involve filing a legal case but involves a woman making an application seeking a protection order, residence order, or custody order. However, it is misconstrued by the family and even by some of the lawyers interviewed as a legal case. Given the norms and attitudes discussed above, it is not surprising to discover various forms of backlash and resistance among legal practitioners, which will be explored below.
The trivialisation of DV is a common form of backlash, and the dominant attitude is that DV is a personal issue, disavowal as a rights violation and an incident that should not bother the judiciary, which is dealing with more substantial matters and serious crimes. These matters should be sorted out in the family through counselling of the couples and families so that they would learn to solve their disputes on their own and accept their socially determined roles within the family and society.

Alternative options should have been suggested as these are family matters. The conflict is not with outside people. It is within the family. A man or woman in the family is being abused by another member of the family. So the matter should stay in the family. If the matter should stay within the family, then I do not think it is rational to bring the matter to the court of a first-class magistrate. 14
In Munshiganj, a tendency observed during interviews with people with direct involvement in cases was victim blaming and declaring that the women were at fault. In a few cases, they were portrayed as adulterous and or someone scheming to get her husband's money through false cases. Women were accused of falsely filing cases for their own benefit (receiving dower money) as litigations lead to divorce. In some cases, other women in the family were blamed for creating conflict.

When we [mediators] try to solve matters, another woman is mostly responsible for not being able to resolve [the situation].
We see other women encouraging survivors to go for a legal case. When we try to reach an agreement/make a compromise [Aposh], the women who surround her, like her mother, mother-in-law, sister, or sister-in-law, obstruct things. 15 There were also the general stereotypes of women committing or abetting violence, i.e. the mother-in-law or the daughter-in-law being at fault for not wanting to live with the in-laws (especially if the husband is a migrant).
Women married to migrant workers were accused of engaging in extramarital affairs in an effort to explain why they filed DV cases: 'Women are weak and commit adultery. It is evident that lawyers would have a financial interest in filing and dragging the case on for longer, so that they get more fees. 20 They have an interest in preserving the status quo of how DV cases are treated, either through mediation or through litigation, which would involve lawyers and court fees. They would have less financial interest in promoting a law that seeks to solve matters through a civil procedure.
Yet another factor which reduces the importance given to the DVPPA 2010 by lawyers and judges is that it is perceived as an 'NGO law'. It is a way of delegitimising the Act as 'NGO' activities are seen as self-serving and anti-state by government authorities, another common backlash strategy (van der Borgh and Terwindt 2012). This was echoed by several members of CIDV and also in discussions with judicial magistrates: 'But they do have this perception that NGOs are the ones who did this law. The court's perception is that the DVPPA is an NGO act', 21 and also that 'this is NGO led'. 22 In fact, one of the judges went so far as to say that it was drafted by NGOs who know nothing about the practicalities of the law, which is why it had such procedural gaps and lapses. 23

Role of MoWCA/DWA: deprioritisation of DV
In this subsection and the next, we will explore how backlash manifested itself at the level of two sets of service providers: the MoWCA/DWA and the police. Their proposed roles and functions were reviewed in section 6 and here we present the research findings with regard to their attitudes and roles.
Although the law was proposed by the DWA and screened by the Ministry of Law, Justice and Parliamentary Affairs, there is disavowal or lack of institutional ownership of the law and its implementation is characterised by inaction, disinterest, and deprioritisation (Flood et al. 2020).
The only role prescribed for the DWA in the 2013 rules of procedure is to provide training to enforcement officers, police officers, service-providing organisations, and related persons. While there is no mention of any kind of monitoring, the rules provide a format in which the DWA offices would collect information and provide it to their headquarters and this would be compiled. 24 Also, the district and upazila DWA officials are to serve as enforcement officers. Most of the CIDV members felt that there is a tendency towards neglect amongst enforcement officers and service providers, which mainly comes from their lack of capacity and resources. Some CIDV members felt that the neglect results from a lack of capacity: have not done so in spite of repeated follow-ups from CIDV. The serviceproviding organisations are supposed to assist the enforcement officers in carrying out their duties and support the women and children. The judicial magistrates interviewed mentioned the delay in selecting/or the lack of designated service organisations as a reason for the lack of implementation of the DVPPA. 26 Another judge in the same discussion felt that one of the reasons for the failure of the Act was the 'failure of the service providing organisations' and his perception of the typical NGO was a micro-credit organisation which had no interest in dealing with DV faced by women.
The DWA officers interviewed did not speak clearly or confidently of their roles as enforcement officers and neither did the other actors in the judiciary and police have a clear understanding of the role the DWA was to play. All the participants of the FGD held with the magistrates and judges were uncertain if enforcement officers had even been assigned.
On the contrary, the DWA officers spoke of their role in advising women and their families of the various options available to them to address violence and abuse, including the DVPPA 2010, the Nari o Shishu Nirjatan Daman Ain, 27 the Dowry Act, and provisions of the Family Law in order to receive dower money or compensation. Their role in carrying out mediation, sending out notices, and in a few cases carrying out enquiries for the court was mentioned as their main role in preventing or redressal of DV or even VAW. In general, there was a reluctance among DWA officials to avail of legal measures to remedy DV. Their perception was that the use of the DVPPA would lead to divorce, and they wanted to avoid divorce as they felt women have no alternative to marriage. 28 A common complaint of the DWA district and upazila staff was the lack of resources -financial and human -to be able to devote enough time to their functions with regards to VAW. During the Covid-19 pandemic, they received calls from emergency hotlines regarding cases of VAW and child marriage which they would try and respond to with the help of their local administration colleagues and even family members. 29 However, the DWA and MoWCA are no longer resource-starved agencies and the lack of human and financial resources for dealing with DV is the result of a lack of priority given by these agencies to the implementation of the DVPPA 2010. 26 Group discussion with judicial magistrates.

Role of police: delegitimisation and disqualification of DV complaints
In the case of the DVPPA 2010, the role of the police is limited to providing information and advice and referring the victims to different services. This is something that the police officials and others interviewed have highlighted. Under this Act, there is no filing of a case but rather an application is given to the judicial magistrate for an interim protection order or a protection order. The difference between an application (which is an administrative matter) and filing a case (which is a legal matter) is not realised by many of the actors involved at the implementation level, with some of the WROs and lawyer respondents mentioning the filing of cases. None of the police officials interviewed seemed to be interested in referring victims to the DVPPA 2010 procedures as they felt they could not take any punitive action, which they thought would be wanted by the victims. It is unclear whether it is the police who think that the victims want the perpetrators to be punished (by beating them up, keeping them in jail, and having them face the court) or whether it is really the victims' desire to have the perpetrators punished.

They told us many women wanted to punish their husbands and inlaws immediately. But there is no immediate punishment and arrest under this law. That's why women are not interested in this law. 30
Also, in spite of having special desks opened up in most police stations for women, children, people with disabilities and elderly people, the police feel that DV is not a priority and not a legitimate claim to make to the police. It was also recounted by BMP activists in Munshiganj that one of the interventions they sometimes had to do was to ensure that the police would register applications or cases that women wanted to file. The BMP member explained it as the police not wanting to increase the number of cases filed in the police station (even though DVPPA 2010 applications would not appear as cases filed!). 30 Interview, CIDV member, 1 April 2022. 31 Interview, CIDV member, 9 April 2022. Advocates and allies of the BMP in Munshiganj also mentioned that in some cases the police would be reluctant to take cases filed by women if the perpetrator had political influence or could bribe them. The police would then keep discouraging the woman and ask her to resolve the disputes within the family: So, the girl wanted to file a case. But the boy had some connection with the police or maybe he gave some money. Therefore, no matter how many times the girl wanted to file the case, the police did not file it.

WRO strategies to counter backlash
This section will discuss CIDV partners' key strategies on voice (articulation and framing) and agency (collective action) to ensure the Act's implementation and to counter the backlash of intentional inaction by the state, including the police, judiciary, and MoWCA. We will address the second two research questions: What strategies and tactics are women's rights actors deploying to counter this pushback? How successful have WROs been in ensuring the incorporation of feminist voice in terms of setting and framing the DV prevention agenda and using collective agency to counter the pushback?

Strategies used
While many strategies and tactics were used, not all were equally effective in countering resistance and pushback. A great deal of time and energy has been spent at the national level with little visible impact. But at the sub-national or individual case level, the strength, social networks, and reputations of each of the WROs made a positive difference. Also, the use of contemporary events and crisis points such as the pandemic seems to have allowed greater scope to make changes.

Capacity-building
Capacity-building was done by CIDV members in collaboration with the MoWCA, which included preparing training modules and conducting training sessions with enforcement officers (DWA) and service providers (police). From 2012 onwards, the aim of these training sessions was to orient the DWA officers and police (sub-inspectors and officers in charge) on their responsibilities as stated under the DVPPA. CIDV members also trained the lawyers they worked with at different levels.
CIDV also prepared lists of proposed service agencies for enlistment by the MoWCA, but these lists were never finalised or approved. Mapping and listing of resources were also done as part of DWA capacity-building. Various collaborative and institutional initiatives were undertaken by CIDV partners. Naripokkho prepared lists for their programme areas; and BRAC launched the Shongjog app to compile a list of upazila-level services for victims/survivors of violence to facilitate their access to information and services. The services mainly consist of mental or medical health services, shelter home information, local police information, etc.

Advocacy
To develop a shared consciousness of the issue's importance (Batliwala 2012), the coalition conducted sub-national meetings with judges and lawyers to discuss the clauses and provisions under DVPPA 2010. The idea of this advocacy was to increase awareness of DVPPA 2010 among the wider population, especially the legal stakeholders (Huda 2016). According to CIDV members, raising awareness was a significant factor in ensuring the effective implementation of the Act. Our research seems to suggest that while there is a greater awareness of the existence of the DVPPA 2010, this has not translated into a thorough understanding of the Act and its contents.
To make use of the 'political opportunity' (Tarrow 2011;Tilly 1978) provided by the pandemic, advocacy put greater stress on the effect of Covid-19 and lockdowns on the implementation of the Act. Naripokkho continued their observations at the field level on the kinds of barriers faced by violence survivors in the community in five districts. BLAST conducted meetings at the district level with their panel lawyers to understand the legal barriers associated with the DVPPA 2010 and shared their experiences of DV cases in meetings with stakeholders during the pandemic. 34 Our interviews with CIDV members and BMP representatives revealed the importance of organisational acceptance and strength in carrying out advocacy. As a well-known coalition, CIDV was accepted nationally by the government and others working on VAW.

Demanding accountability
CIDV and several of its member organisations have adopted an institutional and contained strategy (Tarrow 2011) of demanding accountability for gaps in, and lack of implementation of, the Act. This has been done by feeding back to the authorities the relevant findings -the problems encountered and the procedural gaps. Organisations such as MJF and BLAST collected these results, and Naripokkho systematically monitored the implementation of the Act and fed back information to the government. The mechanisms for accountability have included holding seminars with MoWCA and the judiciary to present the results and recommendations for improvement. In addition, BLAST, the present secretariat of CIDV, has filed a Right to Information (RTI) application to know how many cases of DVPPA 2010 have been submitted to the magistrate courts for interim or permanent protection orders, or are on police records.

Use of contemporary events
Another strategy has been using contemporary events to frame (Tarrow 2011;Tilly 1978; Nazneen and Hickey 2019) the Act's implementation as an urgent matter. The pandemic acted as a critical juncture to build awareness around DV. It was feared that as more areas went under lockdown with stay-at-home orders, financial and mental distress would contribute to the rise of violence at home. Several studies were conducted to understand the pandemic's impact on the issue (by BIGD, MJF and ActionAid Bangladesh (AAB)). Many of them were conducted and commissioned by CIDV and its members to surface the problems in implementing DVPPA 2010. Bangladesh National Woman Lawyers' Association (BNWLA), in collaboration with Plan International, conducted a study to understand the status of the DVPPA's implementation five years after its adoption (Huda 2016). A similar study was conducted by AAB to understand the implementation status after ten years (Yasmin 2020). The evidence generated through these studies has helped raise demands with state actors to facilitate victims' access to courts/police/hospitals during the pandemic. Warnings from United Nations (UN) bodies and international media increased the awareness of the MoWCA, National Human Rights Commission (NHRC), and NGOs active on VAW, and the efforts to increase awareness led to greater sensitivity of the issue.
CIDV members used the 'shadow pandemic' packaging (Nazneen and Hickey 2019;Batliwala 2012) to garner greater attention to DV. Different organisations pressed demands in the form of consultations, webinars, and evidence to the state in an effort to ensure that DV was recognised and treated as a state emergency during the pandemic. Members of CIDV also mobilised for innovative measures such as holding virtual courts, filing online GDs, and conducting telephone mediations. In addition to the government's national helpline for DV, other legal aid and development organisations under CIDV also activated helplines for prompt action (Mahpara 2020a(Mahpara , 2020b. Virtual courts started proceedings with bail issues and 'urgent' applications. Even before the pandemic, there were attempts by both parliament and the supreme court to digitise the judiciary in the context of VAW, but these were never put into action. Though the introduction of such a system during the pandemic was seen as a historic achievement in the judiciary, it did not promise much in terms of DV survivors' access to justice as the pleas made by the coalitions were never considered by the state. Another important strategy was that of building alliances. CIDV has developed strategic alliances across many key stakeholders associated with the movement against DV. This includes partnerships with academics, donors, local NGOs, lawyers, magistrates, as well as government institutions like MoWCA, the Multisectoral Project on Violence against Women, the NHRC, and the police (Victim Support Centres, VSCs).

Networks and alliances
In addition to CIDV research programmes, academics, institutes such as BIGD, International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b), MJF, CARE, ActionAid, and the Law Department at DU have shown relatively high interest in VAW. CIDV has worked closely with most of them to try to build up a larger consensus (Nazneen and Mahmud 2012;Nazneen and Sultan 2010) on what needs to be done to implement the Act. However, the organisations part of this broader alliance have had limited influence on policy implementation. Development partners and donor agencies such as UN Women, Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), and Royal Netherlands Embassy have commissioned programmes and research, and engaged with activities related to DV for a significant period in Bangladesh.
Most members of CIDV such as MJF, BLAST, and Naripokkho have been working in partnership with local NGOs at the field level for addressing DV. This is mostly through project-based interventions or research partnerships, and include activities around awareness building and referral systems in the local communities.
CIDV members have also been working in partnership with the police. After the formulation of rules in 2013, they have conducted multiple training sessions on the provisions and clauses of DVPPA 2010 with the police. Members are still working in partnership with the VSCs. It is at the local level that partners have been able to put pressure on implementing agencies to be responsive to DVPPA cases.
Several CIDV member organisations, including BLAST, Ain o Salish Kendra (ASK), BMP, and Naripokkho, have worked with individual magistrates and panel lawyers to facilitate the use of the DVPPA. 35 The organisations all found that they had to spend a lot of time and effort to explain the procedures to be used, and in some cases, they were ultimately successful.
If we look at BMP Munshiganj as an example of local alliance-building (Nazneen and Mahmud 2012;Nazneen and  individuals who can speak out against injustices and wrongdoings against women and girls.
Munshiganj Mahila Parishad has a district committee, 35 village committees, and six or seven ward committees with 1,050 members. Their activities include awareness raising, arranging regular training for committee members and courtyard meetings for awareness raising, taking stands against social injustice and standing by helpless women, arranging dowry-less marriages, and promoting women's health. They also mediate and mitigate instances of dowry and DV.
At the district level, BMP maintains good working relations with the local administration, local government representatives, government departments, the judiciary, the press, and the police. They are well known and well connected with the community and district-level stakeholders. Often, active BMP members are also members of various other cultural and social associations. Within the District BMP Committee, there are different political identities. But they stated that it does not divide them as they are united to work for women's rights.
Some of the allies are locally elected representatives. They may also be experienced in mediation and have knowledge of the law. However, they find themselves in a delicate situation when a dispute is brought to them as often both sides are their voters/constituents. The strategy taken by a public representative in such a case might be to provide support and advice behind the scenes. Another important strategy is to protest or take action jointly with others. This was mentioned both by the BMP members and their allies. At times the district BMP mobilises to bring attention to a case and reduce biases of vested interests by arranging rallies, press conferences, and human chains and submitting memorandums. This is a common repertoire of actions for such movements (Tarrow 2011;Tilly 1978;Batliwala 2012). This strategy of BMPbuilding and using their alliances (Nazneen and Mahmud 2012; Nazneen and Sultan 2010) -has been very effective in many cases, allowing them to respond to women's demands for access to justice or mediation. 36 Ally, local, 9 May 2022. Women's and human rights organisations who promoted the formulation of DVPPA 2010 saw the Act as a means of changing social attitudes, norms, and behaviour -a way of building a society where DV is seen as an offence, where women can claim protection from the state, and where demanding the right to a violence-free marriage would not result in the breakdown of a marriage. It involves or presupposes an effort to bring about social change in a desired direction. This can be done through education on the law and its spirit (of intention), as well as through its implementation.
Therefore, CIDV and BMP see the struggle to counter the backlash faced by the DVPPA 2010 as a long-term one. Members are tired and disappointed in the struggle but still believe that the formulation of the law is a step towards changing dominant perceptions about women's roles and status, violence, and how it should be dealt with.

Incorporating feminist voice and using collective agency
We have established that there is backlash to the DVPPA 2010 (Flood et al. 2020;Lombardo and Mergeart 2013). The attitudes and inaction from legal practitioners and service agencies are manifestations of backlash in the form of denial, disavowal, and inaction, with stigmatisation of women, trivialisation of DV, and delegitimisation of the law itself.
We have also reviewed the strategies undertaken by WROs to counter this backlash. The question that then arises is how effective was the coalition in countering the backlash? More specifically, as stated in our fourth research question, how successful have WROs been in ensuring the incorporation of 37 National BMP member, 26 April 2022. feminist voice in terms of setting and framing the DV prevention agenda and using collective agency to counter the pushback?
We will briefly discuss the successes and failures of the various strategies. Though capacity-building was a necessary condition, this strategy was not sufficient to overcome the lack of will to implement at various levels. It is likely to increase knowledge among some government agencies and legal professionals, but the knowledge does not seem to translate into practice for want of demand and/or lack of confidence in the process, or lack of commitment to ensure justice for DV victims. Contributing to knowledge does not necessarily translate into agenda setting. In terms of advocacy for strengthening implementation, personal-level follow-up and organisational esteem have helped achieve some gains for individual cases. Various agencies such as the police and the judiciary seem to be more open and responsive to dealing with cases of violence, as observed in Munshiganj and the Dhaka VSC, as a result of organisational alliance-building. But it is doubtful whether this has translated into formal measures to use DVPPA 2010.
CIDV has developed the capacity and an approach to demanding accountability of the state (feminist voice), both by using research evidence (Batliwala 2012) and organising regular events where government agencies are invited to explain their actions or inaction. However, how effective is this voice in the sense of Goetz and Musembi (2008)? Except for the results related to the pandemic, it is difficult to say if the research results have served the purpose of strengthening the implementation of DVPPA 2010. CIDV members and other stakeholders have been better informed and equipped to hold discussions and negotiations with the government on the subject, but there is a tendency for the latter not to give importance to sources of evidence other than their own (Sultan et al. 2021).
The strategy to increase accountability does not seem to have yielded many results, as MoWCA does not seem to feel accountable for the lack of implementation of the DVPPA 2010. 38 This may be due to their lack of interest in responding to these demands.
The strategies of working with lawyers and the various systems to overcome implementation barriers and to build alliances at various levels (Nazneen and Mahmud 2012;Nazneen and Sultan 2010) have helped the coalition members to deal with individual cases and achieve results. However, one-to-one work with magistrates and legal professionals is not a sustainable strategy. There is a limit to how many magistrates and lawyers the CIDV staff can mentor.
For BMP, the achievements of developing collective agency in the form of alliance-building at the district level are reflected in the reputation the organisation enjoys, its access to government departments and decision makers, SPC and Shamajik Anachar committee members are considered as allies, who support BMP, in terms of both access and assistance, in dealing with individual cases of VAW. But the same allies can become an opposition in their attitudes and handling of cases of DV, sharing the prevalent discriminatory gender norms of victim blaming, desire to preserve marriages at all cost, and considering DV as trivial family matters which should not be brought to public forums or given importance by the formal justice system. Allies' perceptions and values may not match those of BMP and the alliance may have more 'political value' than substantive value.
CIDV has been less successful in deploying/exhibiting collective agency (Kabeer 1999;Goetz and Musembi 2008) in bringing about a shift in norms and attitudes regarding DV. One of the most significant gains made since the Act's adoption is people's awareness that psychological and financial abuse are punishable offences under the law. 40 Another gain is that DV is now being talked about openly in public. This has resulted in women being increasingly willing to speak out about this form of violence. This has also enabled lawyers, human rights organisations, and WROs to refer to DV as a violation of rights.
Our interviews have shown that DV continues to be viewed as a 'trivial matter', and family disputes are seen as a private matter to be resolved through counselling or mediation, with the objective of restoring marital relationships. The violence faced by women is hardly recognised, acknowledged, or validated as an offence. None of the interviews dwelt on this, preferring to speak of quarrels and slaps. While this could be understood if these views were expressed by persons openly against women's rights, it is more worrying when they are expressed by those who claim to be in favour of women's rights.
Of greater concern is when women's rights activists themselves are sometimes caught up in traditional and discriminatory gender norms. It cannot be denied that the women's rights activists are themselves embedded in society and share

Discussion
Our analysis in sections 7 and 8 has shown how backlash is played out in the DVPPA 2010 in terms of inaction, disinterest, deprioratisation, and delegitimisation/disqualification (Flood et al. 2020;Lombardo and Mergaert 2013) on the part of the authorities responsible for its implementation. On the other hand, the women and their complaints of DV are stigmatised, stereotypes of women being responsible for VAW are used (victim blaming), and the experience of DV itself is trivialised and not considered worthy of the time and attention of lawyers, magistrates, and law enforcement authorities. The gender norms which normalise and deprioritise women's experiences of violence influence both those seeking justice, as well as those responsible for providing redress.
Marriage is seen as the norm, emphasis is put on the restitution of marriage and therefore women are expected to live with violence and abuse in order to save their marriage. Legal remedies such as the Nari o Shishu Nirjatan Daman Ain or the Dowry Prohibition Act, are generally seen as ways of threatening the husband in order to restitute the marital relationship. DVPPA 2010 with its philosophy of encouraging the couple to continue married life without violence, without bringing in punitive measures, is not understood and its use is not promoted by practitioners. As the Act seeks to transform gender relations through the law it aggravates the challenge to traditional norms (Htun and Weldon 2018).
In response to the inaction and deliberate delays, CIDV and its member organisations have mobilised and undertaken various strategies to counter the backlash and resistance faced. The repertoire includes collective action to raise consciousness through national and local-level engagement (Batliwala 2012;Nazneen and Mahmud 2012), framing of DV as a rights violation (Benford and Snow 2000; Nazneen and Hickey 2019), generating knowledge and evidence to convince stakeholders (Batliwala 2012), seeking media coverage to influence public opinion (Gamson 1989), capacity-building, advocacy, and use of informal networks (Nazneen 2019;Waylen 2017). The repertoire at the local level included more direct interactions with the implementers (e.g. magistrates, lawyers, law enforcement, DWA) in order to influence their actions. In general, the approach was collaborative but in cases where the expected support was not forthcoming, the activities could take a more confrontational approach, through protests and complaints to public representatives.
We found that coalition strategies have not changed much over time, but rather have been static. The forms of strategies and actions have become conventionalised and do not have the same impact of shock value as more disruptive and innovative activities (Tarrow 2011;Batliwala 2012). CIDV has traditionally relied on advocacy, research, evidence generation, workshops, and informal networks. This worked well during adoption (Waylen 2017;Nazneen 2019; Sultan and Mahpara 2020) and the same repertoire has continued during implementation. However, to sustain the movement, CIDV should have restrategised and used more dynamic strategies which would adapt to new realities during the implementation (Batliwala 2012), built alliances with new sets of actors, and found new champions.
The strategies adopted by CIDV and its members to counter the backlash have been only partially successful. Despite the strength and reputation of the coalition and its individual members, it has not been able to effectively build a voice powerful enough to demand accountability and responsiveness of the decision makers and government agencies and attain the outcomes that are important for substantive equality and enjoyment of rights (Goetz 2003). The coalition is able to bring the different stakeholders 'to the table' but has not been able to bring about changes in outcome. Most of the CIDV members interviewed emphasised that the capacity to implement the provisions of the law could have been developed if the state's commitment was present. However, they were not able to facilitate the creation of this commitment.

Conclusion
This paper set out to understand why the implementation of the DVPPA 2010 has not lived up to expectations and what were the barriers and challenges. We sought to understand how the backlash against gender equality was playing out at the level of implementation. To do so we asked the following questions. In what ways has backlash from oppositional actors against the DVPPA 2010 manifested itself in the weak implementation of the law? What strategies and tactics are women's rights actors deploying to counter this backlash? How successful have WROs been in ensuring the incorporation of feminist voice in terms of setting and framing the DV prevention agenda and using collective agency to counter the pushback?
We have found that the socio-political as well as sociocultural contexts have greatly influenced both the adoption and the implementation of the Act. The socio-political context has changed from a multiparty democracy open to addressing women's priorities and demands, to a dominant party rule which is less responsive to civil society demands and more suspicious of offers of collaboration. The sociocultural context and the dominant gender norms have come into play at the implementation level with both justice seekers, promoters/activists, and implementing agencies being confronted with their personal norms and beliefs about women's role and position in the family and society, the sanctity of marriage, as well as their attitudes towards VAW. With DV continuing to be viewed as a personal matter and less important than 'real crimes', it continues to be seen as a matter to be resolved informally, within the couple and without government intervention for redress.
The implementation of the DVPPA 2010 has been subject to deliberate inaction, which is the result of a backlash against the principles and values that the Act stands for. There is a refusal to acknowledge or accept DV as a human rights violation that must be taken as seriously as any other rights violation. This translates into inaction by the implementing agencies, who would rather send women back to their homes and marriages instead of disrupting the status quo and giving the responsibility of safeguarding the marriages to women (Flood et al. 2018;Lombardo and Mergaert 2013;Davidson and Proudford 2008).
The repertoire of actions and strategies by CIDV has included consciousness raising, capacity-building, using the media to influence public opinion, generating evidence, building alliances and networks, using opportunities presented by critical events, and using informal networks. It has sought to hold the government to account by holding regular workshops where the implementation challenges would be highlighted and solutions sought. At the local level, more intensive follow-up and advocacy has been done with those responsible for