While the literature on social movements has shown that women's movements can play a key role in the adoption of women's rights laws (Adomako Ampofo Reference Adomako Ampofo2008; Htun and Weldon Reference Htun and Laurel Weldon2012; Tripp et al. Reference Tripp, Casimiro, Kwesiga and Mungwa2009; Tsikata Reference Tsikata2009), their role in the implementation of such legislation is less well explored. Furthermore, the relatively thin scholarship that does exist is mainly based on Western contexts, with a few exceptions (see Burgess Reference Burgess2012; Jubb Reference Jubb2001; Medie Reference Medie2015, Reference Medie and Cheeseman2018; Walsh Reference Walsh2008). We address this knowledge gap by studying how the women's movement in Ghana has attempted to participate in the implementation of the Domestic Violence Act of 2007 (Act 732, also referred to here as the DV Act) while identifying the factors that explain the movement's lack of impact in this phase of the policy process relative to its success in the adoption of the legislation. We recognize that policy implementation is affected by a multiplicity of factors and that it is primarily the responsibility of government, yet the literature shows that social movements can be important in influencing the implementation of legislation. It is on their role that we focus.
The passage into law in 2007 of the Domestic Violence Bill in Ghana was largely attributed to the vibrant campaign conducted by the National Coalition on Domestic Violence Legislation in Ghana (the Domestic Violence Coalition or DVC), a coalition of civil society organizations that is part of the country's larger women's movement (Adomako Ampofo Reference Adomako Ampofo2008; Crawford and Anyidoho Reference Crawford, Anyidoho, Andreassen and Crawford2013; Fallon Reference Fallon2008; Tsikata Reference Tsikata2009).Footnote 1 Since 2007, however, a picture has emerged of slow implementation of the act by the bodies tasked with setting up the ancillary laws, bodies, and processes required to make the legislation effective. Yet no studies have investigated empirically whether and how the DVC has sustained its influence in the implementation of the law it fought for.
The question of whether social movements can create change has preoccupied researchers, but just as important is the question of how such change can be created (Andrews Reference Andrews2001). In this article, we examine how the DVC as a movement has attempted to influence the implementation of the DV Act, and we consider the factors, both within and outside the movement, that have mediated the DVC's impact on implementation. Specifically, we examine the DVC's role in influencing the establishment and effective operation of the institutions, instruments, and processes that are described in the act as crucial for the delivery of justice and services to survivors of domestic violence, namely, a Legislative Instrument, a National Policy and Plan of Action, the Domestic Violence Management Board, and the Victims of Domestic Violence Support Fund. We also include in our discussion the resourcing of the Domestic Violence and Victim Support Unit (DOVVSU) of Ghana's Police Service. We borrow from Andrews (Reference Andrews2001, 72) in arguing that the “focus on institutional outcomes makes sense because it encompasses the long-term goals of many social movements.” In so doing, we acknowledge the omission of other levels of implementation, including the “street level” (Lipsky Reference Lipsky2010). We argue, however, that given the dearth of research in this area, an initial, even if narrowly focused, examination of overarching structures is important as these influence the working of actors, institutions, and processes at other levels.
We draw on the social movement literature to construct an analytical framework of the key factors that have influenced the impact of movements. This framework consists of three internal factors (strategies, movement infrastructure, and framing) and two external factors (political context and support of allies). We then apply this set of factors to our case study of the DVC in Ghana and assess its relative significance in explaining the DVC's influence. We argue that of the five factors, changes in movement infrastructure have been most decisive in explaining the relative ineffectiveness of the DVC in influencing the implementation phase. This article contributes to the growing literature that has sought to explain the strategies that social movements employ to influence policy implementation and the conditions under which movements can impact the implementation of laws and policies in areas such as women's rights.
We adopted a qualitative approach and collected data through interviews with key informants, supplemented by a review of newspaper articles. Twelve interviews (each lasting between one and three hours) were conducted between June 2009 and November 2010 with past and present coordinators of the DVC, organizational members who had hosted the Domestic Violence (DV) Secretariat, and individuals who led various aspects of the coalition's activities, many of whom were also heads of other women's rights organizations. Between May 2015 and January 2016, we reinterviewed the coordinator and some members of the DVC, as well as the director of the DV Secretariat. In all, seven interviews were conducted with five individuals during this second phase. A coding scheme was developed and applied to the interview transcripts to capture the actions that the DVC has taken in support of the implementation of the DV Act and the challenges encountered in this process. These codes were both inductive and deductive; they were derived from the data as well as informed by a review of literature.
We also conducted a newspaper review aimed at providing more information on the implementation process and triangulated the information elicited through the interviews. We first identified newspaper articles in print media (Daily Graphic and the Chronicle Footnote 2) and online news media (Joy FM and GhanaWebFootnote 3) from 2007 to 2018 that mentioned the act, the DVC, the DV Secretariat, the Domestic Violence Management Board, DOVVSU, or domestic violence. Approximately 300 articles were found in print and online, the contents of which were coded along a number of dimensions (including actors, activities, and challenges). To provide context, we also reviewed minutes of DVC meetings, government reports, background papers, informational pamphlets related to the bill, official reports of parliamentary debates, the text of the Domestic Violence Act of 2007, and the National Policy and Plan of Action.
Following this introduction, the next part of this article examines the literature on the role of social movements in the enforcement of women's rights legislation and constructs an analytical framework of the key factors that influence the impact of such movements. The third part describes the involvement of the DVC in the passage of the Domestic Violence Act of 2007, and the fourth focuses on the DVC's role in the implementation of the act at the national level. Finally, we draw conclusions about the potential of social movements to influence the implementation of domestic violence laws and other women's rights legislation.
THE INFLUENCE OF SOCIAL MOVEMENTS ON IMPLEMENTATION PROCESSES
Although the literature is mostly silent on the implementation of women's rights legislation in Africa, there is a larger body of literature in non-African contexts, notably in the fields of public administration and sociology, that investigates both policy implementation generally and the implementation of domestic violence policies and laws specifically. Of particular relevance to this article is the literature that suggests that social movements can impact policy implementation by influencing the establishment, funding, and performance of institutions (Amenta et al. Reference Amenta, Caren, Chiarello and Su2010; Andrews Reference Andrews2001; Medie Reference Medie2013). From this literature, we identify three internal factors (strategies, movement infrastructure, and framing) and two external factors (political context and support of allies) that provide an analytical framework to assess the relative impact of the DVC.
In a review of the literature on how movements engender change, Andrews (Reference Andrews2001) describes an “action-reaction” model that proposes that the use of disruptive and attention-seeking strategies such as protests will elicit responses from powerful actors with influence over the policy process, but he points out that there is little empirical evidence that these movements have much influence beyond the initial agenda-setting stage. He then presents an “access-influence” model that proposes that movements can influence policy formulation and implementation through “the acquisition of routine access to the policy through institutionalized tactics” (Andrews Reference Andrews2001, 75). This model is also consistent with the implementation phase of the policy cycle; in contrast to the policy formulation phase, in which there might be a single concrete goal, routine access to policy arenas would further the more diffuse goal of institution building. Thus, we can conclude that an action-reaction model may be appropriate for agenda setting, but the access-influence model is more appropriate for thinking about the policy implementation phase.
Studies show that in the implementation phase, social movements employ a variety of strategies, including lobbying decision makers, building the capacity of implementers, providing funds to the government, and disseminating information. As an illustration, women's organizations in Liberia used two main strategies in advocating for the establishment of a specialized sex crimes prosecution unit: they lobbied key actors within the government, including senior personnel of the Ministry of Justice, and disseminated the findings of their study on the prevalence of sexual violence to the ministry, in a bid to convey the urgency of the problem (Medie Reference Medie2013). Women's organizations in Guatemala and Nicaragua, working with international organizations, supported the government with information that influenced the creation of women's police stations (Walsh Reference Walsh2015). The access-influence model suggests that movements will move toward “less disruptive tactics” during the implementation phase, such as lobbying, capacity building, and resourcing, while still making an implicit “threat” that they can work outside state structures if compelled to do so (Andrews Reference Andrews2001, 76). Thus, the literature recognizes disruption as a tactical strategy when it is necessitated by circumstances (Andrews Reference Andrews2001). Indeed, it may be argued that such strategies are more natural to social movements, grounded as they are in “contentious” collective action that aims to contest the power of the state and other better-resourced actors (Tarrow Reference Tarrow2011).
The internal organization or infrastructure of movements can influence both the choice and the efficacy of strategies used. Andrews's (Reference Andrews2001) movement infrastructure model builds on the action-reaction and access-influence models by focusing on the internal changes that must occur in a movement if it is to be successful at influencing policy, specifically in regard to organizational structure and resources. He proposes that the confluence of these elements allows the movement to employ a variety of strategies, giving it more influence on policy implementation.
Scholars have pointed to framing as another factor in policy implementation (Cress and Snow Reference Cress and Snow2000; McCammon et al. Reference McCammon, Newman and Terrell2007). Social movements engage in framing when they actively engage in “producing and maintaining meaning for constituents, antagonists, and bystanders” (Cress and Snow Reference Cress and Snow2000, 1071). Cress and Snow (Reference Cress and Snow2000) differentiate between diagnostic and prognostic framing. The former “problematizes and focuses attention on an issue, helps shape how the issue is perceived, and identifies who or what is culpable,” while “prognostic framing is important because it stipulates specific remedies or goals for the [social movement organization] to work toward and the means or tactics for achieving these objectives” (1071). Furthermore, framing is not just about how movements construct their cause; the framing of a movement is simultaneously carried out by the media and the state, among others.
Strategies, movement infrastructure, and framing are factors that are somewhat within the purview of social movements. There is also a set of external variables that impact the ability of movements to influence implementation. In particular, the political context influences social movement mobilization, strategies used, and impact seen (Amenta et al. Reference Amenta, Caren, Chiarello and Su2010). The political context has been used to describe the institutional structure of the political system as well as the “informal structure of power relations that characterize the system” (McAdam Reference McAdam, McAdam, McCarthy and Zald2012, 26). One aspect of this is the openness of the political environment—that is, openness to the articulation of interests and thus to institutional accessibility—which affects a movement's involvement in the implementation process (Kitschelt Reference Kitschelt1986). Kriesi (Reference Kriesi, Snow, Soule and Kriesi2004, 71) explains, based on a study of Western democracies, that “[s]trong states are characterized by institutional structures that limit their accessibility with respect to their environment and make them capable of getting things done, whereas weak states have institutions that open them up, but also limit their capacity to act.”
While states in Africa might not perfectly fit these models, it is clear that some states are more open than others. For example, in Ethiopia, Burgess (Reference Burgess2012) found that domestic violence legislation was not implemented partly because the government had prevented women's movements from contributing to implementation. In contrast, Medie (Reference Medie2013) found that the Liberian government's willingness to work with women's organizations, combined with political and financial support from international organizations, enabled the country's women's movement to have an impact on the implementation of its rape law. Indeed, the political context also describes repression and other forms of social and physical control that paralyze, slow down, or demobilize dissent (McAdam and Tarrow Reference McAdam, Tarrow, Snow, Soule, Kriesi and McCammon2019). Importantly, the political context also affects the kinds of strategies adopted. In a study of antinuclear movements in Europe and North America, Kitschelt (Reference Kitschelt1986) found that those operating in “open” political systems adopted assimilative strategies such as lobbying, while those in “closed” systems opted for more confrontational strategies such as public demonstrations. In the case of Ghana, Tsikata (Reference Tsikata, Hansen and Ninsin1989, Reference Tsikata2009) argues that the contemporary women's movement has become more political in both the issues it takes on and its approach compared to the period of authoritarian and military regimes in the country's history.
Another component of the political context is the presence of allies within the state and the implementing agency. Political mediation models are employed to assert that a movement is more likely to produce policy change when actors within the state “see benefit” in addressing the concerns of the group represented by the movement (Amenta et al. Reference Amenta, Caren, Chiarello and Su2010, 298). However, these are not the only allies that matter for implementation. Allies outside the state and the implementing institution also matter for the impact of social movements on the implementation process (Anyidoho and Crawford Reference Anyidoho and Crawford2014; Medie Reference Medie2013; Medie and Walsh Reference Medie and Walsh2019; Montoya Reference Montoya2009; Walsh Reference Walsh2015). In a comparative study of Nicaragua and Guatemala, Walsh (Reference Walsh2015) found that women's police stations were more likely to be constructed when women's organizations worked with individuals in the state and with international organizations and when they encountered an opening in the political opportunity structure.
In summary, despite the complexity of explaining movement impact, our review of the literature suggests factors that we use to frame our analysis. Nonetheless, as Amenta et al. (Reference Amenta, Caren, Chiarello and Su2010, 296) note, “there are no specific organizational forms, strategies, or political contexts that will always help challengers.” Rather, it is the “combinations of specific forms of mobilization, action, and political factors that determine whether movements have consequences” (Amenta and Caren Reference Amenta, Caren, Snow, Soule and Kriesi2004, 469). Our objective in this article is to understand how the DVC has sought to influence policy implementation and how the combination of the factors delineated here influences its effectiveness. This analysis is undertaken in the next section, following a brief account of the DVC's key role in campaigning for the legislation.
THE DVC AND PASSAGE OF THE DOMESTIC VIOLENCE ACT
Drawing from the literature and from primary data, we give an overview of the DVC's role in the passage of the DV Act as background for the analysis of its role in the implementation stage, and then we provide a narrative of the progress of implementation since 2007.
The DVC had its origins in the campaign against serial killings of women in Accra between 1997 and 2001, when women's groups formed a loose coalition called Sisters’ Keepers to confront the president and government agencies about their inaction.Footnote 4 Subsequently, the campaign was expanded to address the broader agenda of violence against women in Ghanaian society. One key issue was the absence of legislation, despite research indicating that one in three women and girls in Ghana suffered some form of domestic violence (Coker-Appiah and Cusack Reference Coker-Appiah and Cusack1999). Thus, in March 2003, representatives from 45 women's rights and other organizations came together and formed the National Coalition on Domestic Violence Legislation in Ghana. As a member of the new coalition put it, “The women's movements [in Ghana] put their little turf wars aside and were all part of the coalition” (interview with TN, July 13, 2009).Footnote 5
Organizational members took turns hosting the DV Secretariat and took the lead on particular activities (such as media campaigns, public education, training, and so on), as dictated by their interests and capacity. Another characteristic of the coalition was its exclusively Ghanaian composition. Despite campaigns for domestic violence legislation being a common struggle among women's rights advocates worldwide, the DVC had few linkages to international organizations, in part because of the DVC's wariness of being portrayed as advocates of an externally imposed agenda (interview with AC, July 29, 2009). As a former coordinator stated, “No international networks were formed … It was an internal fight and internally focused” (interview with OB, July 2, 2009).
Predating the official formation of the coalition, women's legal organizations had been involved in drafting possible domestic violence bills in 2001 and 2002, in collaboration with the Attorney-General's Department. In spite of this early partnership, the bill that was finally presented for passage faced significant resistance within government.Footnote 6 The most vociferous public opposition, surprisingly, came from the minister for women and children's affairs, Gladys Asmah, who repeatedly and publicly contended that parts of the bill were contrary to Ghanaian cultural values (Tsikata Reference Tsikata2009; interview with NE, July 6, 2009). The minister's personal opposition led to what one coalition member described as “a big fight” between her and the DVC that frequently played out in the media (interview with BC, July 29, 2009).
During the 2004 election campaign, activists placed significant pressure on the government to effect changes at the Ministry of Women and Children's Affairs (MOWAC) (Fallon Reference Fallon2008), and they took the appointment of Alima Mahama as minister to be a victory. A lawyer with previous civil society involvement, Mahama was recognized as an ally by the women's movement in Ghana. The DVC was able to lobby her successfully, and she was, in turn, instrumental in influencing the cabinet to support the bill. After approval by the cabinet, the bill then went to the Parliament, where it faced considerable opposition, with prominent male members of Parliament (MPs) expressing concerns in the House that the bill would cause “social dislocation” within Ghanaian society and would lead to men being “trampled upon by their wives and denied their conjugal rights.”Footnote 7
To counter such opposition, the DVC engaged in a variety of strategies in different spaces aimed at raising support in the wider Ghanaian society (Crawford and Anyidoho Reference Crawford, Anyidoho, Andreassen and Crawford2013). These included media campaigns, public demonstrations and marches, and training of coalition members, sympathetic civil society actors, and media personnel. The coalition also embarked on a nationwide public education campaign, holding public meetings in most parliamentary constituencies, hosted by local organizations, and undertaking a signature campaign, with constituents signing a petition in support of the bill. One DVC member observed afterward, “In my lifetime there has not been any other legislation that has generated so much public interest and incorporated so much public opinion” (interview with NE, July 6, 2009).
Additionally, the DVC engaged with the Parliament in strategic ways, including depositing a copy of the petition with signatures from each MP's constituency into their respective pigeonholes. They targeted the Women's Caucus in Parliament and the Committee for Gender and Children's Affairs and encouraged them to lobby MPs from within. The coalition achieved its immediate goal when Parliament finally passed the bill on February 21, 2007, and enacted it into law in May 2007.
Thus, the coalition and its constituent bodies, as representatives of civil society, played a crucial role in the adoption of domestic violence legislation in Ghana. Their achievement was due in large part to the variety of strategies—both assimilative and confrontational (Kitschelt Reference Kitschelt1986)—that were employed as appropriate for different constituencies and contexts.
After more than a decade, the implementation of the DV Act has been characterized by an apparent lack of government commitment, notably on the part of the main implementing agency, the Ministry of Gender, Children and Social Protection (MoGCSP), which was created by executive instrument to succeed MOWAC in 2013. At a minimum, implementation requires the establishment of those instruments and bodies stipulated in the act, notably, a Legislative Instrument drafted in collaboration with the Ministry of Justice, a Domestic Violence Management Board, and a Victims of Domestic Violence Support Fund. The process of establishing these has been desperately slow.
The Domestic Violence Management Board is responsible for key implementation tasks, including formulating a national plan of action, monitoring and reporting on the implementation of the plan, collecting data on domestic violence, advising the minister on policy matters, and preparing guidelines for the Victims of Domestic Violence Fund (Article 37). The board is to be given administrative and technical support by the DV Secretariat located within MoGCSP (Article 40). Within the first three years of passage of the DV Act (i.e., between 2007 and 2010), MOWAC was headed by three different ministers. As the minister is also the chair of the board, this has meant periods of transition when there was no board in place. When one was constituted, it had difficulty holding regular meetings (interview with AC, June 24, 2015; interview with IH, June 25, 2015). Additionally, the capacity of the DV Secretariat (in terms of staff strength and technical expertise) to support the work of the board has been called into question (Manuh and Dwamena-Aboagye Reference Manuh, Dwamena-Aboagye and Al-Sharmani2013).
The Legislative Instrument (or LI, as it is commonly known in Ghana) supplements the DV Act through specific regulations on a range of pertinent processes (e.g., training for the police and court officials and the provision of shelter, social welfare, health services, and financial assistance for victims) and develops the administrative structures needed for these processes.Footnote 8 The LI is crucial for the implementation of the act, and yet the executive arm of government failed for nine years to bring the LI before Parliament, despite pressure from the DVC, the public, and even parliamentarians (see Amenuveve Reference Amenuveve2012; Daily Graphic 2012, 2014). It was only passed in July 2016.
The act also established a Victims of Domestic Violence Support Fund (Article 29), to be administered by the Domestic Violence Management Board with financing from government and private organizations. It is meant to support the rescue, upkeep, and rehabilitation of victims of domestic violence. The fund was finally launched in 2011 (Quaicoe-Duho Reference Quaicoe-Duho2011), but as of February 2018, no money had been put into it despite a High Court order requiring the government to do so (Ghana News Agency 2018; Joy FM Reference FM2017).
IMPLEMENTING THE 2007 DOMESTIC VIOLENCE ACT
Why has the implementation of the DV Act been so sluggish, given that more than a decade has passed since its enactment? And, although implementation is the responsibility of government, why has civil society seemingly been unsuccessful at spurring the process? We address these questions using the analytical framework that we constructed from the literature. First, we look at factors internal to or within the control of the movement, and then we consider external factors.
Strategies Employed by the DVC to Support Implementation
Here we seek to understand the strategies employed by the DVC and, implicitly, to assess the extent to which it has been able to adapt its strategies to the institutions and processes of implementation. Our thematic analysis of primary data suggests that, broadly, the strategies used by the DVC are participation, cooperation (through funding, capacity building, and information dissemination), and lobbying. Participation represents working within the state, while cooperation and lobbying locate the DVC outside the state, but working to support implementation by the state.
We define participation as having a presence and working within the institutions and processes initiated by and under the control of the state agencies. The DVC is represented by its coordinator on the Domestic Violence Management Board, which means that, formally, it has a seat on the working body that is to guide the implementation of the act. However, the DVC works within other structures of implementation as well. One member of the DVC who previously served on the board was co-opted by a technical team set up by MoGCSP to draft the LI. Her description of the task indicates the level of contribution of the DVC:
A lot of time, personnel, resources [were] put into the drafting of the LI, huge consultations held—all the components brought together, the medics, social services, health, all that came together. We even drafted the forms that will be used for referral. We did a really detailed piece of work. (interview with IH, June 25, 2015)
The DVC was also instrumental in the drafting and adoption of the National Domestic Violence Policy and the Plan of Action to Implement Domestic Violence Act (2009–2019) (the National Policy and Plan of Action, for short), which guide the work of the Domestic Violence Management Board and other implementers of the law (Government of Ghana n.d, Quaicoe-Duho Reference Quaicoe-Duho2012). A leading member of the coalition explained why the activists took on this further task:
After the Bill was passed, of course, there was all this euphoria. And then we realized very quickly that we needed to have a policy that would enable implementation. And that's because the bill or the act is a legal document, it takes care of the law. It doesn't take care of the in-between. It doesn't take care of the fact that organizations and agencies have to work together … in terms of integration and coordination, monitoring, and all. (interview with BE, June 24, 2015)
In other words, after the passage of the act, the DVC shifted its attention to the implementation of the act, and thus key members of the DVC (including the coordinator and others interviewed for this study) worked through the National Advocacy Partnership (NAP) Project Working Group to draft a plan of implementation after the act. The membership of NAP was made up of “the usual suspects” who had been involved in the passage of the act, as a coalition member put it. Formally, however, the group was distinct from the DVC, as she explained:
The coalition was considered an advocacy group. You know, “be out there on the street, keep pressuring the Government that this and that should happen.” If you look at the NAP documents, it was called a working group or a working partnership … It was just the focus of the work was different … This was trying to look at a coordinated approach toward addressing violence against women and children in Ghana, from an institutional point of view, how institutions should work together … Whereas the DV Coalition is the tsooboyi [call for action]. (interview with BE, June 24, 2015)
This quote suggests an awareness within the DVC of the need for different types of strategies. Here, this was signaled by the creation of a separate organization from the DVC that focused on drafting an implementation plan, fronted by MOWAC and eventually adopted in 2008 by the government of Ghana. We categorize the involvement of the DVC (or its members) in this process as participation because, while the NAP working group was instrumental in initiating the plan, its adoption and use was under MoGCSP. There are, however, elements of cooperation—as described in the subsequent paragraph—because of the initial independent action taken by the DVC to begin developing the plan of action.
Cooperation represents a slightly different approach from participating in activities and systems set up by the state; it involves the DVC taking independent initiative in activities that support the state in implementing the legislation. In our analysis, we found three main avenues of cooperation: resourcing, capacity building, and information dissemination. Although all of these tasks are part of the mandate of the Domestic Violence Management Board under the oversight of MoGCSP, they have been taken up by the DVC either because of the lack of capacity or slow pace of the mandated state structures.
Inadequate resourcing challenges the implementation of the DVC and other policies. As a coalition member observed,
This is Ghana. I mean, laws can be passed, policies can be [passed,] and then putting the money into it becomes a problem. Government still expects donors to fund these things, [but the] coordinating institutions like MOWAC, or now [the] Gender Ministry, do not have the staff capacity or the resources. (interview with BE, June 24, 2015)
Therefore, in support of the implementation of the DV Act, the DVC has, on occasion, provided funds to the Domestic Violence Management Board to hold meetings and to pay for sitting allowances of Technical Committee members (interview with IH, June 25, 2015).
Capacity building involves building up the state's ability to enact the domestic violence legislation. Among other capacity-building activities, the DVC organized a “consultative seminar” for DOVVSU regional officers and distributed copies of the DV Act to officers (Quaicoe-Duho Reference Quaicoe-Duho2008). The coalition also planned a national consultation dialogue for the judiciary and DOVVSU officers, among others (interview with IH, June 25, 2015; interview with WO, May 26, 2015). Again, these are activities that the Domestic Violence Management Board is mandated to carry out but that were initiated by the DVC.
Finally, information dissemination is another one of the responsibilities of the Domestic Violence Management Board that the DVC, in a spirit of collaboration, has gotten involved in. The DVC works to inform and educate the public as well as units and individuals within the public service about their role in the prevention or prosecution of domestic violence cases. For instance, the DVC organized education programs for media practitioners (Agyekum-Gyasi and Kyei-Boateng Reference Agyekum-Gyasi and Kyei-Boateng2011) and the general public (Glover Reference Glover2010) in different parts of the country in 2009 and 2010.
In contrast to participation and cooperation, lobbying as a strategy presents civil society organizations as independent of, even oppositional to, state implementing agencies. The DVC has lobbied the Ghana Police Service, the Minister of Interior, and the Minister of Gender, Children and Social Protection regarding the quantity and quality of its human resources at the DOVVSU (interview with AC, June 24, 2015) and petitioned the Minister of Gender to make the case for professionalizing the DOVVSU (interview with IH, June 25, 2015). Indeed, as a matter of course, the coalition would meet with any newly appointed Minister of Gender to brief her on the coalition's concerns, including the implementation of the act (interview with IH, June 25, 2015). Additionally, the DVC has issued press releases, op-ed pieces, and serialized newspaper articles on the content and potential impact of the DV Act, all attempting to use civic education to generate public pressure to motivate the government to act.
In sum, the DVC as a social movement organization has attempted to adapt itself to the processes of implementation. While the period before the passage of the act was marked much more by lobbying, since 2007, the DVC has worked more formally with the state through participation and collaboration. Its continued use of lobbying as a strategy, though in a more limited way, highlights the DVC's position as an outsider in implementation processes and structures. Also notable is the absence of the more “disruptive” or “contentious” strategies demonstrated in the policy formulation stage (see Crawford and Anyidoho Reference Crawford, Anyidoho, Andreassen and Crawford2013). The DVC has not engaged in the protests, marches, or provocative public performances that it previously undertook to such great effect (Crawford and Anyidoho Reference Crawford, Anyidoho, Andreassen and Crawford2013).
There are several explanations for this. One is that the DVC may have been deradicalized by its association with the government. There are some who maintain that the ideological positions and histories of civil society organizations in the global South make them natural enemies of the state and that social movements are most effective when they maintain an oppositional stance (e.g., Busch Reference Busch1992; Tarrow Reference Tarrow2011). Thus, there is a certain caution with which some women's movements approach the state, partly because of the unequal power relationship between these two parties (especially in the context of policy formulation and implementation, over which the state has greater control) and partly because of the fear that movements might lose their feminist radical edge (Basu Reference Basu2010). That said, an argument can be made that engagement in policy processes does not automatically lead to co-optation (Weldon Reference Weldon2011). There is nothing in the interviews to suggest that the DVC has been deradicalized or co-opted by the state; members still see their organization as apart from the state, and they still regard it as their duty to push the state to act. Alternative explanations for the slow pace of implementation may lie in the other elements of our conceptual framework, specifically movement infrastructure (specifically questions of resources and organizational structure) and framing. We now turn to these two factors and discuss their interaction with strategy.
Movement Infrastructure
At the height of its advocacy for the domestic violence law, the DVC had a membership of about 100 organizations and individuals, a steering committee, and a secretariat in Accra, hosted by a membership organization and run by a coordinator who was the only full-time staff member. In the first phase of the research in 2009 and 2010 and in the second phase in 2015, members acknowledged that their energies waned in the aftermath of the passage of the bill. There were a number of reasons for this.
To begin with, it was difficult to maintain the momentum of the advocacy work that the coalition had undertaken to help push the Domestic Violence Bill through Parliament. As discussed previously, the DV Act became a reality only after a sustained campaign, which was particularly intense in the three years leading up to the passage of the act. It is not surprising that the DVC thereafter “lost steam,” as one member put it. She explained that this was a result of fatigue and change of focus among some members:
The leaders from 1999 till, you know, 2007, are getting on and many people are changing focus, you know, many people have moved on; the younger ones that we worked with, many have moved on into corporate organizations. (interview with BE, June 24, 2015)
This combination of fatigue and an aging membership with fewer younger members to take over from the old guard (interview with BE, June 24, 2015; interview with IH, June 25, 2015) may explain the stance of some members that the passage of the act should mark the end of their efforts as a coalition, even while others argued for the need to extend their efforts into making sure that the law was actually implemented. Another member described this debate at a DVC meeting held after the act was passed:
First and foremost, the DVC was set up as a force to ensure the passage of the act. So, in 2007, DVC had succeeded. It could have closed shop. After May 2007 when the act was passed members of DVC now came together and said, “Okay, legally we don't exist, we have died. How can we ensure that we are still useful?” … And at this meeting there were some movers and shakers who thought, “We have finished our work, let us just close shop,” and other movers and shakers who said, “No, we know our history in Ghana. We pass laws and they just end up on the shelf so then adwuma a yɛayɛ no, na yɛabrɛ; agu [Akan: “All our work would have been for nothing”] so by all means let us keep it going. Let's decide on what we need to do, first of all, to stay relevant as a coalition but, most of all, to ensure that what we set out to do—that the reason for our trying to get the act passed—is sustained.” (interview with BC, June 15, 2015)
Yet it does not appear that this tension was resolved fully, as another member implied:
I think the coalition should have had this conversation but I am not aware we've had it in a very structured way. Really asked ourselves the hard questions: “Where are we now? Where do we go from here? How do we make that progress?” And the even more fundamental question, “Are we relevant?” (interview with IH, June 25, 2015)
Not surprisingly, given these dynamics, respondents described the coalition as weaker and less unified after 2007. A fragmentation of efforts was observed by members, whereby similar activities have been undertaken by member organizations of the DVC, although not necessarily under its aegis. For example, in their own right, the Network for Women's Rights in Ghana (Netright), Abantu for Development, the Gender Studies and Human Right Documentation Centre (or Gender Centre), LAWA-Ghana, and Women in Law and Development in Africa (WiLDAF-Ghana) have all regularly played advocacy roles or engaged with various stakeholders in issues related to the DV Act.
The availability of resources is an important component of movement infrastructure and the efficacy of movements in effecting change. In many respects, the DVC was remarkable in that it was able to achieve so much with so few external resources and largely relied on contributions from its member organizations. This was partly a result of circumstance, as the DVC at the time was an unregistered entity and could not officially raise or receive funds, but it was also a conscious choice to retain autonomy (interview with AC, July 29, 2009). Thus, for the duration of its advocacy for the Domestic Violence Bill, office space and administrative support for the coordinator (the sole permanent staff member) was provided on a rotating basis by member organizations. The DVC could have such a skeletal administrative setup because members provided the time, expertise, and physical labor required at no cost. In addition, members funded many of the activities the coalition undertook, since many of these fell within their mandates or areas of operation and they could raise funds through their organizations. For instance, the Gender Centre received money from WomanKind to support a national educational campaign, while grants from the donor-funded Rights and Voice Initiative came through the Ark Foundation and paid for the launch of a DVC-produced documentary on domestic violence and other media activities (interview with AC, July 31, 2009; interview with BC, July 29, 2009). While the subsequent registration of the coalition as a legal entity made it eligible to solicit funds, the funding landscape has changed in the years since the act was passed. Our respondents spoke of the lack of resources coming in directly to the DVC and indirectly through its organizational members, explaining this as a result of a shift in donor interest from domestic violence issues and a general decrease in funds for nongovernmental organization (NGO) work.
This observation is borne out by empirical research. Apusigah, Tsikata, and Mukhopadyay (Reference Apusigah, Tsikata and Mukhopadyay2011) point to a marked shift in the agenda and areas of funding by aid agencies that left donor-dependent organizations floundering. They add, “The new aid modalities and the global financial crisis have further served to change donor agendas in directions inimical to women's rights activism” (xv–xvi). They note particular new requirements for more formalization and bureaucratization that have made it more difficult for organizations to operate within “loose entities, networks or coalitions” and have had the effect of “removing the fluidity of operational dynamics, and taking out personal commitments and passions, sometimes even compromising on [organizations’] feminist politics” (64). It is clear, then, that the relative lack of funding, and the new requirements for what little funding exists, have constrained the coalition's ability to access financial resources that could have invigorated its inputs into policy implementation. Somewhat paradoxically, the dynamism that characterized the operations of the DVC before 2007 was achieved without significant external resources. However, the funding space has become more constrained at the very time that financial resources have become more necessary. Moreover, it is difficult to sustain that level of voluntary effort for long-term policy implementation, which has made the relative unavailability of external funds more debilitating than it was during the adoption process. This is especially challenging because significantly more resources are required to implement the act than were required for it to be passed. In sum, the combination of fatigue, attrition of members, changed focus, and constricted funding means that the DVC does not have the human and financial resources to carry out as many activities as broadly and intensely as it did before.
Framing
The DVC was able to frame the passage of the domestic violence law as an event that was important to women's physical and political well-being. It was successful in both the diagnostic framing of domestic violence as a problem affecting women's physical and mental well-being in Ghanaian society and the prognostic framing of the Domestic Violence Bill as the solution.
After 2007, the framing of both the cause and the movement became less clear and forceful. The process of the act's implementation has lacked the trenchant debate about Ghanaian cultural values that held the attention of the public and kept the issue of domestic violence in the public discourse. The framing of the DV Act as an important piece of legislation has not changed. However, during the implementation phase, what constitutes a solution to domestic violence is not as clear and compelling as the act of passing legislation, given that there are many more institutional actors, processes, and systems involved in actually implementing the law. The DVC's original goal, its raison d’être, was the passage of a law on domestic violence. In the aftermath of achieving this, the movement had an internal conflict about its own purposes and relevance, which has undermined the clear framing of its overall purpose. Framing is often discussed in the context of mobilization of citizens to join social movements. One might argue that this factor is not as important in the implementation phase, which involves interaction mainly with politicians and technocrats who have different motivations than the public and to whom the “emotional and cognitive appeals” through which social movements might ordinarily frame their causes (Weldon Reference Weldon2006, 58) might matter less than political interests and expediency. Nonetheless, the DVC's ability to project unity and strength, and to demonstrate that it has a constituency, would strengthen its influence on the state institutions that have greater control over the implementation process. Thus, while we would argue that strategy and movement infrastructure seem to be more significant factors, framing also matters.
Political Context
In contrast to the internal factors of strategies, movement infrastructure, and framing, the overall political context has remained largely unchanged, despite the changes in government. There still exists an open political environment in which civil society is able to express opposition to the government in the media and to lobby the government. On the other hand, the main challenge faced by the DVC in the legislative phase remains—that is, the tendency of the government and opposition parties to not give adequate weight to domestic violence and women's rights in the policy processes and in budgeting.
There is agreement within the DVC that the state has not sufficiently invested in the act's implementation and that it is still necessary for the DVC to exert continuous pressure on the government. As one interviewee explained,
There have been many social legislative efforts that have happened in this country and if people don't keep pushing, it doesn't happen. Children's Act, Intestate Succession Law, name them. So we thought, “We can't let it go.” (interview with BE, on June 24, 2015)
However, while an open political context might have made it possible for the DVC to effectively advocate for new legislation by using public pressure on politicians, such advocacy may not be sufficient for it to change the behavior of the technocrats who are key to implementation, particularly when implementation is happening in a context of limited resources.
DVC members have expressed frustration with how the DV Secretariat within MoGCSP has been managed, but they have had little control over these bureaucratic systems and processes, which are susceptible to partisan politicking. The coalition's attempts to participate in and shape the work of the Domestic Violence Management Board were described by members as an exercise in frustration. According to our interviews, there have been at least four iterations of the board, as it had to be reconstituted each time a new minster was appointed. The board is chaired by the minister, and it appears that in some cases, meetings are set up and then canceled when the minister is not available (interview with IH, June 25, 2015). While the board managed to produce a National Policy and Plan of Action (Government of Ghana n.d.), a past DVC representative on the board spoke about other ways it could have been more effective—for instance, by documenting and archiving information on domestic violence and by producing publications and reports on these (interview with IH, June 25, 2015). Interviews with other DVC members support this blunt assessment by Manuh and Dwamena-Aboagye:
[T]he greatest challenges for the effective implementation of the law and for the future remain the political will to enhance the capacity of all institutional stakeholders with the mandate to implement the law; the allocation of the necessary budgetary support; a robust monitoring and evaluation mechanism that ensures compliance and sanctioning for recalcitrant institutions; and the continued advocacy of civil society organizations … at various levels, from the community to the national level. (2013, 205; emphasis added)
Thus, while the overall political context has remained largely unchanged and relatively favorable to civil society activity, political will on the part of successive governments has not been evident.
Support from External Allies
We have noted that before 2007, the DVC had few linkages to international organizations inside or outside the country. By contrast, the implementation phase has involved a wider range of organizations, with a greater role in particular for international organizations, both governmental and nongovernmental. United Nations agencies such as the United Nations Population Fund, United Nations Children's Fund (UNICEF), and UN Women are more involved, as are bilateral aid agencies such as the Danish International Development Agency (DANIDA) and the country programs of international NGOs such as ActionAid Ghana and Plan Ghana. For instance, UNICEF, at one point, supported the DV Secretariat with staff, while DANIDA supports the DV Secretariat's programming on gender-based violence (interview with WO, May 26, 2015). The Embassy of the Kingdom of the Netherlands and the U.K. government's Department for International Development have been among others who have provided support for the development of the National Policy and Plan of Action for the DV Act (Quaicoe-Duho Reference Quaicoe-Duho2012).
However, there has been little evidence of these international organizations purposefully partnering with the DVC specifically to influence national-level implementing structures.
In the absence of strong and sustained partnerships with international organizations, the DVC has likely missed opportunities to influence implementation. Studies have shown that collaboration between women's movements and international organizations enables the former to impact the implementation process (Anyidoho and Crawford Reference Anyidoho and Crawford2014; Medie Reference Medie2013; Medie and Walsh Reference Medie and Walsh2019). This was the case in Liberia, where women's organizations were supported by and collaborated with international organizations such as the UN (Medie Reference Medie2013). Montoya (Reference Montoya2009), in her study of the European Union, argues that international organizations can build the capacity of domestic organizations working on violence against women by giving them resources, which is especially important for the sustained engagement required in the implementation phase.
Before 2007, the absence of international linkages was an asset to the DVC, as it provided the coalition with legitimacy when countering arguments that the bill was contrary to Ghanaian cultural values (Crawford and Anyidoho Reference Crawford, Anyidoho, Andreassen and Crawford2013). A coalition member was able to say, with pride, about the advocacy for the act, “It was an internal (local) fight and internally focused” (OB, July 2, 2009). After 2007, in the implementation phase, the increased involvement of international organizations could be expected to strengthen the position of local advocacy organizations such as the DVC through the provision of international support for implementation of the act. Yet such expectations have largely remained unfulfilled; there is little evidence that the DVC has made the effort to directly cultivate the support of allied international actors.
CONCLUSION
After the elation experienced by members of the Domestic Violence Coalition in Ghana in 2007 at the culmination of their advocacy for a Domestic Violence Act, the subsequent years have seen slow and limited progress in implementation of the act. This article has examined how the DVC has participated in the implementation process and sought to identify the factors that explain the coalition's apparent lack of impact at this stage. We applied an analytical framework consisting of three internal factors (strategies, movement infrastructure, and framing) and two external factors (political context and support of allies). Our overall findings are that, of the five factors, the change in the movement infrastructure is the most significant in explaining the relative ineffectiveness of the DVC in the implementation stage. However, the findings also highlight the interconnections among the five factors. We elaborate these findings here.
In terms of strategies, the DVC has shown awareness of the need to shift strategy from the legislative to the implementation stage. The more disruptive or contentious strategies that were characteristic of the legislative phase have largely been discarded. The DVC has focused on strategies of participation within state processes and cooperation with state agencies, including provision of funds, capacity building, and information dissemination. Such changes in strategy are consistent with the shift from an action-reaction model to an access-influence model in which actors attempt to gain “routine access” to the policy space to engage in institution building (Andrews Reference Andrews2001).
However, the potential impact of this shift in strategy has been moderated by less salutary changes in the coalition infrastructure, the most salient being a decline in the active participation of member organizations and to a constraint on funding of their activities, which has adversely affected the overall cohesion and capacity of the coalition. Thus, while the DVC did adapt its strategic approach to one appropriate for the implementation stage, the changes in movement infrastructure to a looser and weaker coalition structure have militated against its effective application.
The weakening of the movement's infrastructure has also had an adverse effect on framing, again highlighting the interconnection of factors. While the coalition's diagnostic and prognostic framing of issues of domestic violence was unambiguous in the legislative phase (essentially, domestic violence could be addressed with the passage of a DV Act) and successfully communicated to the wider society, its framing is less clear and forceful in the implementation phase, partly due to unresolved internal conflicts about its own relevance and purpose in the postlegislation stage.
While the overall political context has largely remained unchanged and relatively favorable for civil society activity, the same constraints and challenges endure from the legislative phase—notably, the lack of political will. Additionally, engagement in policy implementation processes is more complicated, given the greater numbers of actors, processes and systems involved in implementing the law, compared with the legislative stage. Since 2007, the DVC's organizational capacity for advocacy has proved insufficient for the sustained pressure necessary to push government to undertake the more challenging and longer-term tasks of policy implementation.
Finally, analysis of the DVC's support from allies also provides useful findings. The DVC's initial success was attained without significant international support. However, the post-2007 environment has been characterized by greater involvement of international actors, both UN agencies and international NGOs. The experience in Ghana suggests that the involvement of international allies does not automatically facilitate the advocacy efforts of local movements; there needs to be a concerted effort on the part of both the movement and international actors to build partnerships based on shared goals that can intensify the pressure on government to act (see Anyidoho and Crawford Reference Anyidoho and Crawford2014). This has not happened in the case of the DVC and the DV Act.
In sum, the study demonstrates the importance of movement infrastructure in explaining the DVC's (lack of) impact on implementation of the DV Act. Additionally, our findings highlight the interconnections among the five factors of our analytical framework, rather than their separateness. The study illustrates how changes in key elements of the movement infrastructure, in this instance a decline in its strength and cohesiveness, affected the DVC's capacity to decisively frame the debate, limited its capacity to effectively implement a change in strategic approach, and undermined its ability to negotiate the political context, which includes the presence of potential allies.
Such findings add to the growing literature that has sought to explain the conditions under which women's movements can impact the implementation of women's rights laws and policies, especially in the context of the global South. Previous research (e.g., Burgess Reference Burgess2012; Medie Reference Medie2013; Medie and Walsh Reference Medie and Walsh2019) has shown that civil society organizations can have a positive impact on the protection of women's rights in Africa. Our findings add weight to the literature that highlights the vital role of civil society organizations in advocating for legal protection against domestic violence and for women's rights more generally. However, the implementation of such laws, including the creation of relevant structures and processes, is the crucial next step if legal protection is to be meaningful. This case study of the DVC in Ghana demonstrates the important role of women's movements in maintaining pressure for implementation and also demonstrates the potential weakness of coalitions. The initial strength of the DVC was in unifying a cross section of women's rights organizations around the goal of enacting a domestic violence legislation. Yet this strength became a liability when, after passage of the act, the DVC experienced fragmentation, with many coalition partners being compelled, mainly for financial reasons, to return to their core activities. In turn, this weakening of the movement infrastructure had an adverse effect on other key factors of impact (strategies, framing and alliance-building), and the overall influence of the DVC declined. The case of Ghana's Domestic Violence Act starkly demonstrates the challenge for civil society groups of continuously re-creating themselves to respond to changing internal and external contexts in order to ensure that women's rights legislation is implemented.
