Many African states have adopted laws that criminalize rape and other forms of gender-based violence (GBV), but the enforcement of such laws is often weak. Many rape cases are never brought to court and victims are frequently encouraged to accept reconciliation instead of prosecution of offenders. Drawing on research from post-conflict Liberia, this article investigates the ability of women's movements to influence the state's implementation of rape law, and seeks to theorize the relationship between women's activism and the enforcement of rape law. It documents the range of strategies adopted by the Liberian women's movement, and argues that these tactics have contributed to an increased referral of rape cases to court. This was made possible by two conditions: a relatively open political environment and political and material support from international organizations, which in turn enabled women's NGOs to gain access to and make an impression on the implementation process. This demonstrates the capacity of civil society organizations in Africa's more open and internationally connected states to influence policy at the implementation stage–even in particularly challenging areas such as women's rights.

Gender-based violence has received increasing attention in recent years, and many African states, often under pressure from local women's movements, have passed laws to criminalize various forms of gender-based violence (GBV), including rape.1 However, the passage of such laws has not always translated into implementation; victims of GBV often face many obstacles within society and in the criminal justice system that render it difficult, and sometimes impossible, for them to benefit from new laws.2 This problem is particularly acute in post-war states that have weak implementing agencies, which often prevent victims from receiving justice.3

In post-war Liberia, where GBV is prevalent, a large proportion of reported cases never advance to the courts.4 Instead, police officers withdraw these cases from the police station, sometimes upon the request of complainants.5 The withdrawal of cases, especially when forced upon complainants, undermines the purpose of GBV legislation and raises serious concerns about the translation of formal policies into substantive changes in the lives of women.6 Yet, despite this problem, scholars of women's movements and women's rights have largely failed to study policy implementation.7

This article addresses this gap in the literatures by interrogating how the women's movement in Liberia has influenced police officers' enforcement of the rape law. The rate of withdrawal of rape cases is significantly lower than other forms of GBV, and is lower now than it was in the pre-war period. To assess whether this is the result of the activism of the women's movement, I conducted one-hundred and fifty interviews in 2010 and 2011 with women's rights advocates, police officers, bureaucrats, staff of international organizations (IOs), and victims of violence in two Liberian counties. Based on this evidence, I document the ways in which, since the end of Liberia's civil war in 2003, women's groups have worked–sometimes independently, but mostly in collaboration with the state and IOs–to amend the rape law, create institutions to address rape, and provide education on rape to police officers and the public. Most significantly, I argue that these measures have contributed to a lower rate of withdrawal of rape cases. The article also demonstrates how the state's openness to the participation of NGOs in the crafting of the national response to rape, combined with the material and political support of IOs, has bolstered NGOs' anti-rape efforts and facilitated their ability to influence police enforcement of the rape law. This goes against the grain of much recent literature that has emphasized the limitations of civil society, especially in areas such as women's rights where limited progress in policy implementation has been made over the past forty years.

The article first discusses the literature on social movements and policy implementation, before turning to African women's movements. After outlining the methodology, the article introduces the problem of GBV in Liberia and analyses the strategies that women's groups have adopted since the end of the civil war, as well as the effects of these strategies on police enforcement of the rape law. I conclude with a discussion of how future studies can further advance knowledge of the implementation of women's rights policies.

Social movements and policy implementation

An extensive body of literature in the fields of public administration, public policy, and other disciplines investigates policy implementation, and although numerous research questions and policy areas are examined the focus is generally on explaining three broad outcomes: why policies are implemented, why policies are not implemented, and how policies are implemented.8 Studies probe how the presence, absence, and characteristics of certain actors, structures, resources, norms, and policy instruments affect one or more of these themes. While earlier studies produced long lists of variables to explain implementation, without specifying the importance and effects of each variable, comparative case studies and large-n quantitative studies conducted over the last two decades have aimed for parsimony in explaining the implementation process, its outputs, and outcomes.9

Scholars have identified four categories of factors that explain the behaviours of policy implementers such as police officers: political control, organizational control, individual-level characteristics, and external pressures.10 Political control is exerted by political officials, while organizational control emanates from actors and conditions within the implementing agency. The individual level refers to the norms and beliefs of individual implementers, while external pressures are produced by actors and conditions that are outside the state and its bureaucracies. IOs and interest groups, such as social movements, are sources of external pressure. While some scholars have argued that individual-level characteristics best explain implementation behaviour, others have found that social movements are important to implementation and that they also shape the individual-level characteristics of implementers.11 The weak implementing capacity within many African states suggests a more prominent role for social movements in policy implementation.

Social movements' influence on implementation is first determined by the political context; the openness of the political environment conditions movements' involvement in the implementation process. A democracy, therefore, offers more opportunities for movements to influence policy implementation than an authoritarian regime.12 Furthermore, the political opportunity structure affects the strategies adopted by movements. In a study of anti-nuclear movements in Europe and America, Herbert Kitschelt found that movements 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.13 Movements also find that the conditions and strategies that make it possible for them to influence agenda setting and policy adoption do not necessarily support their involvement in policy implementation.14

Once movements gain access to the political system, they typically try to influence implementation from the policy design stage.15 By directly participating in policy design or lobbying lawmakers, movements attempt to imbue policies with their ideas of how a problem should be addressed, which agency should be responsible for addressing this problem, and which instruments should be employed. In doing so, they shape the directives that implementers eventually receive and contribute to setting the organizational rules and procedures that guide implementation. They also attempt to shape the organizational structure and capacity of implementing agencies, thus moulding the systems and processes through which policies are delivered.16 Opportunities to participate in the development or restructuring of implementing agencies, therefore, present opportunities to influence the implementation process.

Social movements also seek to affect policy implementation from the bottom up, exerting influence on both policy implementers and beneficiaries. They employ strategies such as lawsuits and negotiations to compel implementers to change how they implement policies,17 and they educate policy beneficiaries on their rights and help them in formulating the demands that they make of implementers.18

Scholars, focusing mostly on the US and Europe, have found that the effects of these mobilization strategies vary across policy issue areas, because influence on implementation is not determined solely by the strategies adopted but is contingent upon other factors.19 In a study of how the creation of agency-level hate crime policies affects police officers' official recording of such crimes in California, Ryken Grattet and Valerie Jenness found that the effect of policies depends on characteristics of the police force (such as police integration in the community) and the presence of human relations commissions (organizations that serve as community watchdogs).20 Thus, ‘Communities with a human relations commission and neighbourhood-based police stations experience the largest payoff for creating a policy.’21 In a study of the civil rights movement's influence on the implementation of poverty policies in Mississippi, Kenneth Andrews found that communities with strong movement infrastructures (leadership, organizational capacity, and resources) generated greater funding for programme implementation and succeeded in shaping programme content.22 In South Africa, the Soul City Institute for Health and Development Communication in partnership with the National Network on Violence against Women used strategies such as lobbying and media advocacy successfully to pressure the government to implement the 1998 Domestic Violence Act.23 After this intervention, a ‘growing sense of urgency was discernible from within government ranks to move the process forward’.24 However, advocates were excluded from many aspects of how the act was implemented, including the development of the national instructions and the training of police officers, and had less impact in this regard.

These findings show that movements' influence is conditioned by their organizational characteristics, the mobilization strategies adopted, the policy issue area, the structure of implementing agencies, and the political environment, factors which vary temporally within and across states. However, ‘there are no specific organizational forms, strategies, or political contexts that will always help challengers’.25 The political, social, and economic conditions in Liberia, and many other African countries, differ significantly from those of the developed countries addressed in the literature, providing an opportunity to examine these issues in a new context.

African women's movements and policy implementation

I adopt Myra Max Ferree's definition of a women's movement as one that ‘organizes women as women to make social change’.26 Such movements are characterized by women's mobilization around a shared gender identity and a focus on the concerns of women in specific issue areas. In Africa, women's groups have a long history; Filomena Steady, for example, describes how 10,000 Sierra Leonean women under British rule in 1951 staged a demonstration against a proposed increase in market dues and the steep rise in food prices caused by the monopolization of wholesale food distribution by Lebanese traders.27 The women subsequently won the rights to purchase directly from the government and coalesced to form the Sierra Leone Women's Movement (SLWM). Despite some weaknesses and constraints, women's movements have continued to protest political and socio-economic injustices in the post-colonial period and in transitions to democracy.28 However, with the end of colonial rule, many women's movements were co-opted into ruling parties and several one-party states established women's wings. Similar forms of co-option and incorporation occurred in the immediate aftermath of transitions to democracy,29 and this absorption of women's movements by ruling parties weakened these groups' ability to hold the state accountable as they began to function as extensions of governments.30

The growth of civil society over the last two decades has expanded the political space outside of state control, and challenged governments' ability to control women's movements.31 Consequently, some movements have managed to effect important changes despite the barriers that they continue to face. One important aspect of this activism has been women's efforts to promote the adoption of laws and institutions to overturn existing structures and practices that discriminate against women. For example, a collection of gender and women's rights advocates successfully lobbied for the passage of Ghana's 2007 Domestic Violence Bill.32 However, it is clear that policy adoption is not sufficient for policy implementation.33 Many governments have been quick to pass women's rights legislation, but lack the political will and/or the resources to effectively deliver and implement. This is evident in the implementation of GBV policies; despite laws that criminalize rape and other forms of GBV, police officers encourage and sometimes compel victims to accept reconciliation instead of pursuing the prosecution of the offenders.34 Describing police response to GBV in Ethiopia, Gemma Lucy Burgess states that:

Police often do not follow correct procedure when a rape is reported. One activist said that police think it is better if the woman marries her rapist than pursues a criminal case. Rape cases are often dismissed when the victim is not a virgin.35

This gap in law enforcement demonstrates that getting GBV legislation on the statute book is only half the struggle; where progressive policies have been introduced the question becomes how state agents can translate and deliver GBV policies in ways that will protect and empower women, and how women's movements can help them to achieve that goal. The movement literature demonstrates that the influence of social movements does not end at the point of policy adoption, but extends to the implementation phase and is contingent on the mobilization strategies, the political environment, the characteristics and structures of implementing agencies and of social movements, and the issue area. It is, therefore, necessary to examine the influence of women's movements on the implementation of GBV and other women's rights policies.

In post-conflict Liberia, the rate of withdrawal of rape cases has declined and in order to assess whether the activism of the Liberian women's movement contributed to this decline, I employed the method of structured, focused comparison in two locations: Monrovia and Gbarnga.36 In this method the questions, reflecting the research objectives, are standardized and focused across cases. Monrovia and Gbarnga were selected to allow for variations in the levels of a key factor, the presence of women's NGOs. There is a greater NGO presence in Monrovia than in Gbarnga and most organizations are headquartered in the capital, but the women's movement has nevertheless been able to organize and use its advocacy strategies to influence implementation in both locations.

Data were collected through semi-structured interviews, participant observation, and reviews of official documents and grey literature. I observed police officers as they interacted with complainants and collected documents produced by government agencies, media houses, IOs, and NGOs. Interviewees were identified through purposive and snowball sampling techniques, and included staff of the ministries of Gender and Development (MoGD), Justice (MoJ), and Labor (MoL); staff of IOs; and GBV survivors. They also included 28 women's rights advocates from 12 women's NGOs and 50 police officers. Information on the role of the women's movement was gathered directly from the advocates as well as from the other interviewees, enabling the production of a holistic picture of the movement's structures and activities and its impact on police responses and on the behaviour of survivors of violence. I also observed the staff of women's NGOs at work and attended meetings of the national GBV taskforce. The deliberations of the taskforce shed light on the movement's involvement in the anti-rape efforts in Liberia, its weaknesses and constraints, and its influence on how the state and its agents have responded to rape. I found this influence to be both direct and indirect, defining direct influence as the outcome of NGOs' interaction with police officers and indirect influence as a product of the movement's efforts at changing law enforcement procedures and institutions as well as target group behaviour.

Gender-based violence in Liberia

All warring factions in Liberia's civil war targeted violence at women.37 Although the level of violence has fallen since the war's end, physical and sexual violence remain major problems and the victims in most reported cases of rape are minors. Some women's rights advocates argue that the incidence of GBV in post-war Liberia is higher than before the war, while others contend that it is not the incidence, but the reporting of GBV that has increased.38 The absence of pre-war data makes it difficult to establish the validity of either claim, but reports show that rape was the most reported crime in 2007 and currently, rape and domestic violence are two of the crimes most reported to the Women and Children Protection Section (WACPS) of the Liberia National Police (LNP).39 Nonetheless, the majority of crimes are not reported to the police. Instead, victims turn to family palavers and customary courts for adjudication.

Informal means of adjudication are preferred for several reasons. Some view the court system as divisive as it focuses on establishing guilt and punishment instead of reconciling disputing parties.40 People also view the criminal justice system as corrupt.41 The unofficial fees levied on complainants by the police and the courts dissuade people from reporting GBV. The length of trials and the time that it takes away from work is also a burden on survivors. The fear of retaliation, stigmatization, and being accused of lying (in the case of rape) also discourages reporting.42 Some survivors and/or their families also accept payment from rapists in exchange for keeping cases out of the police stations and courts. Liberia's crippling poverty has made this a common practice.43 These factors also contribute to the withdrawal of cases after they have been reported to the police.

Public attitudes toward GBV, though improving, are largely negative. A high percentage of men (30) and women (59) believe that domestic violence is justifiable.44 In the case of rape, people tend to blame the victim. In a 2008 survey of 1,000 households in all 15 counties, 69 percent of respondents agreed that ‘women contribute to rape by flirting with men’ and 61.9 percent agreed that ‘women contribute to rape by being alone with a man in a room’.45 Victims are also stigmatized and viewed as ‘damaged goods’ and ‘leftovers’.46 Nonetheless, 96.3 percent of respondents agreed that ‘rape is a serious crime’ and 90.5 percent agreed that ‘men who commit rape should be given harsh sentences’.47 Liberians, however, do not rank all types of rape equally. Rape against children is viewed as serious whereas the rape of women, especially where they are perceived to have caused it, is seen as less serious. Some people think that the law should allow for ‘restorative remedies’ for these ‘less egregious’ forms of rape.48 These negative attitudes toward rape and other forms of GBV have made it harder to promote women's rights, but have contributed to the mobilization of the women's movement.

Women organizing against rape

The women's movement in post-war Liberia, though facing internal challenges, has organized to prevent rape and to seek justice for survivors.49 Historically, Liberian women have mobilized to challenge policies and institutions that negatively affected them.50 There were instances where women mobilized against forms of GBV prior to the 14-year civil war, but these were typically not sustained and inclusive national campaigns. Instead, their activism was usually in reaction to a series of public and particularly violent attacks against women and involved a few communities protesting for a short period.51 The outbreak of the war in 1989 led women's groups to mount national campaigns against GBV and other injustices. Starting in early 1990, female politicians and women's organizations began to appeal to the warring factions to come to a peace agreement, and, as the conflict progressed, the various groups coalesced into a women's movement, pressuring both national and international actors to end the war.52

During the war, some groups within the movement were active for short periods of time and on a small scale, while others functioned at the national level for several years.53 Since the end of the war, some groups have continued within the movement, although its main focus has shifted from conflict resolution to the promotion of women's rights. The post-war movement consists of small, medium, and large NGOs led and mainly staffed by women, and national networks such as the Women's NGO Secretariat, which maintain links between the various groups. While organizations such as the Women in Peacebuilding Network (WIPNET) have branches across the country, most are only present in Monrovia or a few of the larger towns. Similarly, organizations such as the Association of Female Lawyers of Liberia (AFELL) and Touching Humanity in Need of Kindness (THINK) are more active and influential than others, drawing on their greater expertise and capacity to provide services.54 Given this diversity, rape has emerged as one of the principal issues around which the movement has organized in a more coherent fashion.

The political opportunity structure has favoured women's organizing and provided openings for NGOs to have a voice in policy design and implementation.55 Many lawmakers in the transitional government, cognizant of the prevalence and brutality of rape during the civil war, supported the adoption of the rape law.56 This reflects a commitment and willingness to combat rape that did not exist prior to the war.57 President Ellen Johnson Sirleaf encouraged the efforts of women's NGOs and voiced her commitment to combating GBV.58 Although this has not translated into steady and adequate funding for anti-rape initiatives or the commitment of all bureaucrats and implementers, it has created a political and bureaucratic environment that is mostly open to and sometimes welcoming of NGO involvement. On the other hand, some interviewees complained that the MoGD has attempted to dominate the GBV conversation and the activities of all organizations, showing favouritism towards some NGOs and marginalizing others who express dissenting views.59 This has created resentment amongst some advocates and hindered the extent to which they have shared ideas and information with other organizations.

Nonetheless, NGOs have worked with allies in government to achieve policy and institutional changes, but have also pressured these entities when their efforts have been resisted. They have also worked closely with IOs (the UN–the organization with the largest presence in the country–as well as other multilateral agencies, bilateral agencies, and international NGOs) who have prioritized the issue of rape, relying on them for financial and technical support to lobby the state and to fund programmes. The movement, often in collaboration with the state and IOs, has focused its anti-rape efforts on three areas: legal and policy changes, institutional and programmatic changes, and education and awareness-raising.

Laws and policies: Women's NGOs initiated action against rape through their participation in the creation of several women's rights policy documents including the 2006 GBV National Action Plan. The movement has also engendered legal change, lobbying for the amendment of the rape law in 2004 in response to frequent reports of rape, including gang rape and statutory rapes. AFELL identified the absence of a comprehensive rape law as an impediment to providing justice to victims and to deterring the crime, and began to work with other NGOs, IOs, and the state towards strengthening the existing law.60 The amended law significantly enhanced the pre-existing statute on rape in the penal code, broadening the categories of acts that are included in the legal definition of rape, categorizing some forms of rapes as first-degree felonies, and mandating harsher punishment for those found guilty.

However, the amendment of the law highlighted the weaknesses of the judicial system. Rape cases sent to court were often left pending on the docket. Liberia lacked sufficient lawyers, judges, and courtrooms to adequately handle the wave of cases. This led to the prioritization of offences such as murder and treason over cases of rape and other forms of GBV.61 Men accused of rape were sometimes held in custody for extended periods of time without trial and eventually released on humanitarian grounds.62 Advocates also complained that rapists were bribing their way out of police custody.63 This state of affairs underscored the need for the creation of new anti-rape institutions and programmes.

Creating new institutions and programmes: The women's movement has also advocated for the creation of new state agencies to address rape, while its members have developed and implemented programmes to prevent rape and support survivors. Within the government, the GBV Secretariat of the MoGD oversees all anti-GBV activity in the country and also chairs the National GBV Taskforce. Established in 2006 by a coalition of women's NGOs, IOs, and the government, the taskforce meets monthly in every county to strategize on plans of action, develop policies, evaluate existing programmes, and report on the successes and challenges that members experience in their day-to-day operations. NGOs in Monrovia and Gbarnga use the taskforce as a forum to discuss problems related to the police force and to bring their concerns to the attention of representatives of the WACPS, who sometimes attend taskforce meetings.64 The taskforce provides a reasonably united front of activists advocating for change. It is relatively easy for a corrupt police officer to ignore the demands of a small NGO but significantly more difficult to disregard the taskforce.

The women's movement has also contributed to changes in the judicial system. In 2009, AFELL, with the support of several IOs, advocated for the passage of an act to create Criminal Court E, a court that is mandated exclusively to prosecute all cases of sexual violence, in Montserrado County. As of January 2011, there was one courtroom dedicated to hearing cases of sexual violence and trials were conducted in camera to protect the victim's confidentiality. Once this had been achieved, women's NGOs publicized the court and its function to the public. However, the prosecution of cases has proceeded at a glacial pace because there is only one courtroom dedicated to rape cases. Furthermore, as of August 2011, funding constraints had restricted the court to Montserrado County, meaning that survivors in the other 13 counties continue to face major barriers to justice.

AFELL, in collaboration with the MoJ and donors, was also a central figure in the creation of the ministry's Sexual and Gender Based Violence (SGBV) Crimes Unit in 2009 to complement Criminal Court E and the WACPS. The unit is mandated to assist in the investigation and prosecution of sexual offences but is ‘concerned about the whole being of the victim’.65 It provides counsellors to work with victims during trial. The unit also supports the WACPS and complements its effort. WACPS officers liaise with and receive legal advice and guidance from staff of the unit when they investigate sexual violence cases. As of August 2011, however, the unit remained a pilot project that was operational only in Montserrado County.

Women's NGOs have not limited their efforts to the state. Organizations such as Women Aid Incorporated and THINK run safe homes, which, although severely under-resourced, shelter women and protect them from their abusers and the public. They sometimes resettle survivors in new communities and, when they are willing, NGOs guide them through the criminal justice system to reduce their fear of cooperating with the police and going to court. Staffs of some women's NGOs visit police stations to pressure the police to prioritize reported cases and to follow up on those that are pending. NGOs also serve as liaisons between complainants and the police, and their efforts facilitate the investigations. They sometimes provide transportation and phone airtime to enable investigating officers to follow up on cases. Although they are heavily underfunded and the services they provide are limited, they fill a crucial gap, as the state largely fails to provide these services.

Education and awareness-raising: Women's NGOs have also participated in the development of several awareness-raising programmes, including the 2007 Anti-Rape Campaign. NGOs also have myriad community education programmes. Development Education Network–Liberia organizes community workshops and dramas to raise awareness of rape in Gbarnga, while the Bassa Women Development Organization airs radio programmes that educate on a range of women's rights issues, including rape. With support from donors, NGOs have also trained journalists on rape reporting, and newspapers now regularly report on rape crimes although the majority of these stories focus on the rape of minors.66 Their reports not only frame rape as a serious problem but also aim to encourage those affected to seek justice through the state. Police officers are exposed to these messages and they also receive training from NGOs such as AFELL and THINK. During these training sessions, they are taught the importance of and procedures for ensuring prosecution. Moreover, NGOs reach police officers through their participation in the crafting of police training documents. For example, AFELL contributed to the development of the MoJ's 2009 Sexual Assault and Abuse Police Handbook, which has been used to train police officers on how to respond to sexual violence cases.

Overall, it is clear that women's NGOs have launched anti-rape initiatives in a wide range of areas that influence policy implementation.67 Their design of policies and institutions has affected the political and bureaucratic structures. They have trained and monitored implementers and implementing agencies, and they have educated and encouraged victims of violence to seek prosecution. They have, therefore, reached police officers through multiple pathways. Opportunities in the state's political and bureaucratic structures to participate in policy and programme design and support from IOs have facilitated NGOs' ability to engage the wide range of factors categorized in the literature as having an effect on policy implementation.68

The women's movement is, however, beset by challenges. Most NGOs face severe financial constraints and are heavily dependent on donor funding. This has enabled donors to set the agenda of some organizations. Donors have been overwhelmingly focused on sexual violence, leading one staff of an international NGO to comment that donors' attention to sexual violence has created a situation in which women's NGOs are ‘blinded’ by rape and mostly ignore other prevalent forms of GBV such as domestic violence.69 Indeed I noticed that staff of some NGOs used the terms ‘rape’ and ‘GBV’ interchangeably. Other researchers have noted this problem.70 This raises questions about the relationship between IOs and women's NGOs and the latter's ability to address non-sexual forms of GBV.

The GBV taskforce is also beset by dissent. Staff of some women's NGOs accused the MoGD, which chairs the taskforce, of punishing them for expressing opposing views. As mentioned above, they claimed that they have been sidelined from planning discussions and have been excluded from representing the taskforce at national and international anti-GBV meetings. This has led some member NGOs to respond negatively by failing to attend meetings, refusing to contribute to policy discussions, and failing to report on the progress of their respective programmes. This problem stems from the fierce competition for resources and recognition that exists among some activists and institutions. It hinders collaboration and limits the degree to which the movement can address rape. Nonetheless, the steps taken so far have affected how police officers enforce the rape law.

The influence of the women's movement on police officers

By amending the rape law and introducing new rape policies, creating programmes and institutions, and educating the police and the public about rape, the women's movement has contributed to increased police referral of rape cases to court. The influence of women's NGOs on police officers has been facilitated by the political and material support provided by IOs and the state's openness to NGOs' participation in policy development. My suggestion is not that the withdrawal of rape cases is no longer a concern in Liberia, but that the rate of withdrawal of rape cases has fallen–relative both to the past and to other forms of GBV. Between January and December 2009, police referred 40 percent of rape cases to court, withdrew 4 percent, transferred 3 percent to other agencies, and left 53 percent pending. Over that same period, police referred 36.5 percent of domestic violence cases to court (prosecuted as aggravated assault), withdrew 29.6 percent, transferred 1 percent, and left 33.3 percent pending. Furthermore, the proportion of rape cases referred to court is significantly higher than prior to the civil war. None of the 49 rapes reported to the Montserrado County police from July 1967 to August 1968 were referred to court in this period.71 In 1972 there were only two rape cases on record in the courts and four in 1977: these figures do not reflect a low incidence of rape but, rather, low reporting and referral to court.72

The interviews revealed that there are two factors that lead the police to respond differently to rape than other forms of GBV, and explain the lower rate of withdrawal. One is the perception that rape is a crime whose adjudication is above the jurisdiction of the police. The other is the existence of a non-withdrawal policy for rape in the WACPS. While this perception and policy existed before the war, both have recently been more purposely crafted and promoted by NGOs, IOs, and the state. I found that in addition to cultural beliefs, officers' perception of rape has been shaped by training, the WACPS' policies, and the framing of the rape law. The WACPS' non-withdrawal policy has been structured primarily by IOs but also by women's activism for a tougher response to rape. Women's NGOs have, therefore, directly and indirectly influenced police behaviour.

Direct influence: The training sessions organized by women's NGOs, IOs, and the state have taught that prosecution is the only acceptable response to rape and have provided step-by-step guidance that aims to prevent the withdrawal of cases. AFELL's participation in the development of the 2009 Sexual Assault and Abuse Police Handbook, which reinforces this message, has further increased the organization's input into police training. This training aims to dispel any negative attitudes towards rape and to (re)form how officers think about the crime. The contrast in police officers' and the public's attitudes toward rape victims illustrates the effectiveness of this training. As discussed above, the public views some rapes as less serious and blames adult victims. The police officers interviewed did not express these views. Officers' references to their training when explaining their decision to refer cases to court underscores its importance. Officer No. 41 stated that, ‘we have been told time and time in all trainings that rape is not compromisable’.73

Even where training is not explicitly given as a reason for referring cases to court, it underlies officers' decisions, as other explanations such as the role of the state as prosecutor and the magnitude of the offence are conveyed to officers in training.74 This effect of training is expressed by Officer No. 36:

We have been receiving cases where the [victim's] parents just do not want to prosecute the person but then for the police, we look at the crime committed, it's a state crime. So we tell them that since you do not want to prosecute this person, but for us you will be serving as a state witness, even though you serve as a complainant in our document. Once we have evidence linking this man to the crime, whether you do not want to prosecute, we will still send this man to court and the state will prosecute him.75

In another example, Officer No. 32 stated that ‘with rape cases, everyone is aware … we don't take it lightly. It's not a minor offence. All major offences have to go to court. We don't compromise rape cases.’76 These statements illustrate the effect of training on officers' perception of rape and, ultimately, their response to the crime.

Women's NGOs also directly influence police officers when they aid survivors in navigating the criminal justice system and when they lobby the WACPS to regulate officers' actions. Their presence places pressure on and provides resources for the police to conduct investigations, to do so in a timely manner, and to eschew actions–including the demand of bribes–that lead complainants to withdraw cases. The pressure exerted on the police through the GBV Taskforce and other organizational platforms is also important in this regard. A staff member of WIPNET explained that when the local police fail to investigate adequately the reports of women who come to her organization for help, or refuse to make arrests, her organization makes a report to the central police headquarters and, if that fails, to the MoJ.77 She explained that women in the organization's sub-branches in the counties do not stop pressuring the police until the case proceeds to court. By taking these steps, women's NGOs lobby the WACPS to enforce its non-withdrawal policy, and hold it accountable. This accountability is important to ensure that policies on paper are effectively put into practice.

Indirect influence: AFELL spearheaded the campaign to strengthen the rape law and to eliminate all gaps that would lead to police withdrawal of cases. This amended law has signalled the gravity of the crime to police officers and conveyed the message that rapes should be prosecuted. It also provides clarity and direction and does not leave space for officers to use their discretion to determine if a case is ‘serious’ (and thus should be forwarded to court) or ‘not serious’ (and therefore should be withdrawn). In explaining their decisions to refer cases to court, some officers cited the provisions of the law. A female officer stationed in Monrovia stated that:

Rape cases, they [complainants] come in talking about withdrawal but there is nothing like withdrawal in rape cases … Because, it's a first degree felony in Liberia, it is viewed just as murder cases are viewed.78

A male officer stationed in Monrovia explained why he employs his discretion when responding to domestic violence (aggravated assault) but not to rape:

Rape cases are first degree felonies and aggravated assault is a second degree felony. For rape aspect we can't make our own decisions, we have to consult our superiors or the courts.79

Another officer, a male stationed in Monrovia, summarized the importance of the law in saying that he does not withdraw cases only because he fears disciplinary action from the WACPS but because ‘the laws says that anyone who rapes needs to be prosecuted’.80 These statements convey how the framing of the law has shaped officers' views and response to the crime.

In Monrovia, the creation of Criminal Court E and the SGBV Crime Unit has also led officers to refer cases to court. These institutions convey to officers the seriousness of the crime, institute measures that make it difficult for officers to use their discretion to withdraw rape cases, and support officers to refer cases to court. Officer No. 26 summarized the importance of the unit:

If anybody comes to say I don't want to go to court, that is my relative, I don't want to go to court, we will tell you when you go to the SGBV [unit] you can tell them that but for now we have already turned the case over to SGBV.81

An officer discussing the role of the unit explained that,

We have constructive partners that we work with that help us a lot in the SGBV Unit. They provide the legal technical know-how about the process of handling cases and how to avoid the contamination of evidence.82

Another officer explained that ‘anytime cases of rape are reported, we contact the SGBV Unit and they advise us on what to do’.83

The support and supervision that the unit provides to investigating officers enhance investigations and contribute to their decision to refer rape cases to court. In the absence of an SGBV Unit in Gbarnga, officers relied on other less-resourced units within the MoJ. Officer No. 28 explained that ‘when a woman chooses not to prosecute, we contact the County Attorney to seek advice. If the office is closed, we get to the magisterial court.’84

Women's NGOs also provide support to rape survivors. Their presence gives confidence to women, many of whom are intimidated by the criminal justice system and who lack the financial resources to usher a case through the system. Their involvement not only empowers some women to report their rapes to the police, but also to persist until justice is served by the courts.85 Furthermore, the information that women's NGOs provide through radio programmes and other media encourages survivors to seek help from these NGOs and to cooperate with the police. In the absence of these programmes, many women would be ignorant of their rights under the law and, therefore, would continue to perceive customary courts and family meetings as the sole means of adjudication.

The decision of WACPS officers to refer cases to court has been shaped by their perception of the crime and by WACPS' policies. Officers' perception of the crime has been influenced by NGOs. AFELL's spearheading of the amendment of the rape law enabled them to play a central role in laying the foundation for Liberia's response to rape, as this law guides the behaviour of police officers. NGOs' participation in police training has also enabled them to influence police officers directly. Policing the police and holding them accountable is another avenue through which women's NGOs have shaped police response to rape.

Opportunities within the state for women's NGOs to participate in the design of policies and institutions have enabled them to influence implementation. IOs' focus on sexual violence and the political and financial support that they have provided to NGOs and the state have also bolstered the efforts of the women's movement. Furthermore, the strategies used and members' efforts at organizing across Monrovia and Gbarnga have allowed the women's movement to reach politicians, bureaucrats, implementers, and the public–all of whom matter to the implementation process.86 Building on the social movement literature, these findings show the conditions under which a social movement in a post-war African state has succeeded in influencing policy implementation.


Referring a rape case to court is no guarantee that it will be prosecuted, but it is a necessary step for victims to get justice. However, getting cases to court is an uphill battle for survivors of rape in Liberia and elsewhere. The case of Liberia demonstrates that police perception of the crime, and the organizational rules relating to it, shape the decision to refer cases to court. This article argues that women's NGOs in Liberia have contributed to shaping both police perceptions and the organizational rules, and in doing so have made police more likely to forward rape cases to court. Although some of their initiatives remain concentrated in Monrovia and resource constraints prevent them from reaching a significant proportion of women in other areas, they have set an important precedent and ensured that more women have their cases proceed to trial. Although police response to rape in Liberia remains deeply problematic, it is important to recognize the improvements that have been made, consider how these improvements have occurred, and use these lessons to think about how the implementation of rape and other GBV policies can be furthered across the continent.

This article has demonstrated that women's movements matter beyond policy adoption. Two conditions, a relatively open political environment and political and material support from IOs, have enabled women's NGOs to access and impact the implementation process. Future studies can build upon these findings to further advance our understanding of the influence of women's movements on policy implementation. There is a need to investigate implementation across policy issue areas and socio-economic and political contexts. The case of Liberia demonstrates the central role played by the women's movement in a state that is mostly open to the involvement of non-state actors in the policy process, in part because it is characterized by weak institutions. It will be important for theory building to probe the extent to which the presence of the movement matters in other contexts where the state actually has the resources to perform the services that are being provided by women's NGOs in Liberia, and so is less dependent on non-state actors. The relationship between the women's movement and IOs also needs to be probed. In Liberia, IOs have provided material and political support to women's NGOs but there is a gap in understanding the degree to which they set the movement's agenda and the tension that this causes on the ground. It is, therefore, important to examine how IOs, which have a strong presence in post-war states, are affecting women's movements.

The outputs and outcomes of the policy implementation process are determined by a complex interaction of factors. The women's movement is only one actor in this process and it faces many internal and external challenges. Nonetheless, this article demonstrates that despite significant constraints, the Liberian women's movement has succeeded in changing how police officers respond to victims of rape. This finding offers many opportunities to better understand and improve the delivery of rape policies in Africa and beyond.

Aili Mari Tripp, Isabel Casimiro, Joy Kwesiga, and Alice Mungwa, African Women's Movements: Transforming political landscapes (Cambridge University Press, New York, NY, 2009).
Amnesty International, ‘“I can't afford justice”: violence against women in Uganda continues unchecked and unpunished’ (Amnesty International, London, 2010); Neil Andersson, Sharmila Mhatre, Nzwakie Mootsi, and Marina Penderis, ‘How to police sexual violence’, Crime and Conflict15 (1999), pp. 18–22; Bruce Baker and Helen Liebling-Kalifani, ‘Justice and health provision for survivors of sexual violence in Kitgum, northern Uganda’ (Coventry University, Coventry, 2010); Megan Bastick, Karin Grimm, and Rahel Kunz, ‘Sexual violence in armed conflict: global overview and implications for the security sector’ (Geneva Center for the Democratic Control of Armed Forces, Geneva, 2007); Convention on the Elimination of all Forms of Discrimination against Women, ‘NGO report on violence against women in Burundi’ (New York, NY, 2008); United Nations Population Fund, ‘Gender-based violence in Sierra Leone: a case study’ (United Nations Population Fund, New York, NY, 2005).
For data on the prevalence of GBV, see Liberia Institute of Statistics and Geo-Information Services [LISGIS], ‘Liberia demographic and health survey 2007’ (Ministry of Health and Social Welfare, Monrovia, Liberia, 2008); Women and Children Protection Section, Liberia National Police Force, Case Disposal Statistics, (2006–10), in author's possession.
Women and Children Protection Section, ‘Case Disposal Statistics’.
I am not arguing that police referral of cases to court is sufficient for the prosecution of the offender and for the complainant to receive justice. However, referring a case to court is a necessary condition for the prosecution of offenders.
Some exceptions are Gemma Lucy Burgess, ‘When the personal becomes political: using legal reform to combat violence against women in Ethiopia’, Gender, Place and Culture: A Journal of Feminist Geography19, 2 (2012), pp. 153–74; Shereen Usdin, Nicola Christofides, Lebo Malepe, and Aadielah Maker, ‘The value of advocacy in promoting social change: implementing the new domestic violence act in South Africa’, Reproductive Health Matters8, 16 (2000), pp. 55–65.
For a review of major methodological and theoretical themes and debates in the implementation literature, see B. Guy Peters and Jon Pierre (eds), Handbook of Public Administration (Sage, London, 2007), pp. 131–63.
For a discussion of the problem of over-determination and how it has hindered theory development in implementation studies see, Michael Goggin, ‘The “too few cases/too many variables” problem in implementation research’, Political Research Quarterly39, 2 (1986), pp. 328–47. The author also notes the failure of most studies to specify the causal patterns that correspond to the outcomes under observation, p. 329.
Peter J. May and Soren C. Winter, ‘Politicians, managers, and street-level bureaucrats: influences on policy implementation’, Journal of Public Administration Research and Theory19, 3 (2009), pp. 453–76; Marcia K. Meyers and Susan Vorsanger, ‘Street-level bureaucrats and the implementation of public policy’ in Peter and Pierre (eds), Handbook of Public Administration, pp. 153–63.
John Brehm and Scott Gates, Working, Shirking, and Sabotage: Bureaucratic response to a democratic public (University of Michigan Press, Ann Arbor, MI, 1997); Steven Williams Maynard-Moody and Michael Craig Musheno, Cops, Teachers, Counselors: Stories from the front lines of public service (University of Michigan Press, Ann Arbor, MI, 2003).
May, ‘Policy design and implementation’; Burgess, ‘When the personal becomes political’.
Herbert P. Kitschelt, ‘Political opportunity structures and political protest: anti nuclear movements in four democracies’, British Journal of Political Science16, 1 (1986), pp. 57–85.
Edwin Amenta, Neal Caren, Elizabeth Chiarello, and Yang Su, ‘The political consequences of social movements’, American Review of Sociology36 (2010), pp. 287–307.
Evelyn Z. Brodkin, ‘Implementation as policy politics’ in Dennis James Palumbo and Donald J. Calista (eds), Implementation and the Policy Process: Opening up the black box (Greenwood Press, Westport, CT, 1990), pp. 107–18; Eugene Bardach, The Implementation Game: What happens after a bill becomes law (MIT Press, Cambridge, MA, 1977).
Susan Franceschet, ‘Explaining domestic violence policy outcomes in Chile and Argentina’, Latin American Politics and Society52, 3 (2010), pp. 1–29.
Diane Mitsch Bush, ‘Women's movements and state policy reform aimed at domestic violence against women: a comparison of the consequences of women's mobilization in US and India’, Gender and Society6, 4 (1992), pp. 587–608.
Jelke Boesten, ‘Pushing back the boundaries: social policy, domestic violence and women's organizations in Peru’, Journal of Latin American Studies38, 2 (2006), pp. 355–378.
Amenta et al., ‘The political consequences of social movements’.
Ryken Grattet and Valerie Jenness, ‘Transforming symbolic law into organizational action: hate crime policy and law enforcement practice’, Social Forces87, 1 (2008), pp. 501–27.
Ibid., p. 516.
Kenneth T. Andrews, ‘Social movements and policy implementation: the Mississippi civil rights movement and the war on poverty, 1965–1971’, American Sociological Review66, 1 (2001), pp. 71–95.
Usdin et al., ‘The value of advocacy in promoting social change’.
Ibid., p. 60.
Amenta et al., ‘The political consequences of social movements’, p. 296.
Myra Marx Ferree, ‘Globalization and feminism: opportunities and obstacles for activism in the global era’ in Myra Marx Ferree and Aili Mari Tripp (eds), Global Feminism: Transnational women's activism, organizing, and human rights (New York University Press, New York, NY, 2006). pp. 3–23.
Filomena Chioma Steady, Women and Collective Action in Africa: Development, democratization and empowerment (Palgrave Macmillan, Gordonsville, VA, 2005).
Malehoko Tshoaedi, ‘(En)gendering the transition in South Africa: the role of COSATU women activists’, Transformation78, (2012), pp. 1–26, p. 7.
Aili Mari Tripp, ‘Women's movements and challenges to neopatrimonial rule: preliminary observations from Africa’, Development and Change32, 1 (2001), pp. 33–54; Sylvia Tamale, ‘“Point of order Mr Speaker”: African women claiming their place in parliament’, Gender and Development8, 3 (2000), pp. 8–15.
Studies have demonstrated, however, that the presence of a strong and autonomous women's movement is not a necessary condition for policy change. See this argument articulated in, Mala Htun, Sex and the State: Abortion, divorce, and the family under Latin American dictatorships and democracies (Cambridge University Press, New York, NY, 2003).
Tripp, ‘Women's movements and challenges to neopatrimonial rule’.
Akosua Adomako Ampofo, ‘Collective activism: the domestic violence bill becoming law in Ghana’, African and Asian Studies7, 4 (2008), pp. 395–421, p. 413.
Amnesty International, ‘I can't afford justice’; Andersson et al., ‘How to police sexual violence’; Baker and Liebling-Kalifani, ‘Justice and health provision’; Convention on the Elimination of all Forms of Discrimination against Women, ‘NGO Report on Violence against Women in Burundi’; Burgess, ‘When the personal becomes political’; United Nations Population Fund, ‘Gender-based violence in Sierra Leone’.
Burgess, ‘When the personal becomes political’, p. 162.
See Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences (MIT Press, Cambridge, MA, 2005, p. 67).
Human Rights Watch, ‘How to fight, how to kill: child soldiers in Liberia’, Human Rights Watch16, 2 (2004), pp. 1–43.
Interview, Sallimatu Kamara, Women Aid Incorporated, Monrovia, 28 June 2010.
Women and Children Protection Section, ‘Case Disposal Statistics’.
Deborah H. Isser, Stephen C. Lubkemann and Saah N'Tow ‘Looking for justice: Liberian experiences with and perceptions of local justice options’ (Peaceworks, 63, United States Institute of Peace, Washington, DC, 2009).
United Nations Mission in Liberia [UNMIL], ‘Research on prevalence and attitudes to rape in Liberia’ (UNMIL, Legal and Judicial Support Division, Monrovia, Liberia, 2008).
Interview, female member of staff, Women Aid Incorporated, Monrovia, 13 September 2010. Kofi Annan Institute for Conflict Transformation, ‘FAQs on rape and the revised rape law’ (Kofi Annan Institute for Conflict Transformation, Monrovia, Liberia, 2011).
LISGIS, ‘Demographic and health survey’, pp. 214–15.
UNMIL, ‘Research on prevalence’, p. 41.
Ibid., pp. 41–2.
Isser et al., ‘Looking for justice’, p. 6.
For a discussion of the causes and consequences of women's movements' focus on sexual violence see Peace A. Medie, Police Behavior in Post-Conflict States: Explaining variation in responses to domestic violence, internal human trafficking, and rape (University of Pittsburgh, unpublished PhD dissertation, 2012), pp. 193–200; Benjamin de Carvalho and Niels Nagelhus Schia, ‘Sexual and gender-based violence in Liberia and the case for a comprehensive approach to the rule of law’, Journal of International Relations and Development14, 1 (2011), pp. 134–41.
Mary Moran, Liberia: The violence of democracy (University of Pennsylvania Press, Philadelphia, PA, 2006).
For example, in March 1988 female residents of the Matadi Estates in Monrovia criticized the police for poor handling of the investigation into the rape and murder of one of the estate's female residents. They appealed to several government agencies that were involved in the case to make it a priority. See Daily Observer [Monrovia], ‘Estate women’, 1 March 1989, pp. 1 and 6.
Abigial E. Disney and Gina Reticker, Pray the Devil Back to Hell [film] (Fork Films, USA, 2008); African Women and Peace Support Group, Liberian Women Fighting for the Right to Be Seen, Heard and Counted (Africa World Press, Trenton, NJ, 2004); Daily Observer [Monrovia], ‘It is time for peace’, 8 June 1990, pp. 1 and 6; S. Togba Slewion, ‘Christian women want an end to Nimba crisis’, Daily Observer [Monrovia], 16 May 1990, p. 6.; Daily Observer [Monrovia], ‘Women call for ceasefire’, 30 May 1990, pp. 1 and 6.
African Women and Peace Support Group, ‘Liberian women fighting for the right to be seen’.
All organizations are underfunded and face resource constraints.
The political opportunity structure is defined as the allies and structures within state institutions that determine movement involvement in the policy process.
Interview, Ruth Caesar, former Chairperson of the Committee on Gender Equity, Women, and Child Development in the National Transitional Legislative Assembly, Monrovia, 5 June 2011.
Medie, Police Behavior in Post-Conflict States.
United Nations Population Fund, ‘Liberia's Gender Based Violence National Action Plan’ (United Nations Population Fund, 2006).
Interview, female member of staff of Liberian NGO, Monrovia, 29 June 2010; interview, female member of staff of Liberian NGO, Monrovia, 15 July 2010.
Interview, Zeor Bernard, Interim President, AFELL, Monrovia, 7 September 2010.
Interview, Deweh Gray, Commissioner, Law Reform Commission, Monrovia, 19 August 2010.
The problem persists.
Interview, female member of staff of Liberian NGO, Monrovia, 15 July 2010.
Interview, Anita Rennie, Bong County Gender Coordinator, Gbarnga, 5 May 2011; observation of GBV Taskforce meetings, Monrovia, 27 August 2010 and 30 September 2010.
Interview, Felicia Coleman, Chief Prosecutor, SGBV Crime Unit, Monrovia, 14 July 2010.
Daily Observer [Monrovia], ‘Two get ten year jail each’, 10 March 2011, p. 20; Front Page [Monrovia], ‘Two men held for rape’, 30 July 2010, p. 8b.
Meyers and Vorsanger, ‘Street-level bureaucrats and the implementation of public policy’.
Interview, female international NGO staff member, Monrovia, 1 September 2010.
Carvalho and Schia, ‘Sexual and gender-based violence in Liberia’.
Ministry of Justice, ‘Annual Report of the Attorney General of the Republic of Liberia–October 1,1967 to September 30, 1968’ (Ministry of Justice, Monrovia).
Ministry of Justice Annual Reports covering the periods: 1 October 1971–30 September 1972; 1 October 1972–31 December 1972; 1 January 1977–31 December 1977. Many of the reports were destroyed during the war.
Interview, male police officer, Monrovia, 10 June 2010.
Ministry of Justice, ‘Sexual assault and abuse handbook: for the women and children protection section of the Liberian National Police Force’ (Ministry of Justice, Liberia, 2009).
Interview, male police officer, Monrovia, 15 September 2010.
Interview, male police officer, Monrovia, 14 September 2010.
Interview, member of staff, WIPNET, Monrovia, 30 June 2010.
Interview, Officer No. 19, Monrovia, 29 September 2010.
Interview, Officer No. 50, Monrovia, 10 July 2010.
Interview, Officer No. 16, Monrovia, 10 October 2010.
Interview, female police officer, Monrovia, 16 September 2010.
Interview, Officer No. 2, male, Monrovia, 12 July 2010.
Interview, Officer No. 4, male, Monrovia, 20 September 2010.
Interview, male police officer, Gbarnga, 24 May 2011.
Interview, Zoe Bernard, Interim President, AFELL, Monrovia, 7 September 2010.
Meyers and Vorsanger, ‘Street-level bureaucrats and the implementation of public policy’.