Legislative Approaches

Currently, 24 African nations where FGM/C is practiced have prohibited it by law or constitutional decree (Table 1). These include countries where FGM/C is widely practiced (e.g., Somalia, Guinea, Egypt, and Djibouti), countries where the prevalence of FGM/C is low (e.g., Uganda, Niger), as well as two countries that did not use the Demographic and Health Survey (DHS) module on FGM/C because the practice is believed to occur in less than one percent of the female population (South Africa, Zambia). The most recent ban has been introduced in Somalia, where the 2012 Provisional Constitution prohibits “circumcision of girls.” With the exception of the Central African Republic and Guinea, where bans on FGM/C were instituted in the 1960s, the process of legal reform began to take hold in African nations in the 1990s. This occurred in the context in which international opposition to FGM/C was being recast not only as a health issue, but as a human rights issue (Reference Boyle and PrevesBoyle & Preves 2000; Reference Shell-DuncanShell-Duncan 2008). As international consensus forged around framing FGM/C as a human rights violation, protection from FGM/C became understood as a right to be enforced, granted, and implemented by the state. Thus, legal strategies received new prominence, elevated over nonstate solutions (Reference Shell-DuncanShell-Duncan 2008). At the International Conference on Population and Development held in Cairo in 1994, strong statements were issued calling for governments to enact legislation specifically banning FGM/C, and to enforce such legislation (Reference Boyle and PrevesBoyle & Preves 2000). The World Health Organization (WHO) as well stated unequivocally that “protecting the rights of each and every citizen is the responsibility of national governments,” and specifically urged that “governments must enact and/or use anti-FGM laws to protect girls” (WHO 1999: 14–15).

Table 1. African Legislation on FGM: A Comparative Summary (as of August 2012)

DHS, Demographic and Health Survey; FGM, female genital mutilation; MICS, Multiple Indicator Cluster Survey; NA, not applicable.

In this context, legislative reform measures have been adopted by a growing number of nation states, but vary in terms of restriction. In Mauritania and Tanzania (as in some Western countries, including the Canada and the United States) FGM/C is illegal when performed on minors. By contrast, laws banning FGM/C at all ages have been passed in the majority of African countries. This reflects the view that adult women are not in a position to provide consent because of cultural pressures to undergo FGM/C, and the lack of economic and legal autonomy. The laws vary substantially in defining who is subject to punishment, and on the severity of penalties imposed. The criminal law enacted in Ghana imposes punishment on anyone who performs FGM/C, whereas in countries such as Ethiopia, Nigeria, Senegal, Togo, Uganda, and Zambia, those who participate in or commission the procedure are held criminally liable. In several countries (Benin, Burkina Faso, Côte d'Ivoire, Eritrea, Ethiopia, Kenya, Togo), fines can be levied not only against practitioners of FGM/C, but also anyone knowing about practice and failing to report it. Mandatory reporting requirements can be problematic from a legal perspective, particularly when health care workers have been sought to treat post-FGM/C complications, as they are also bound by duties to protect patient confidentiality (Reference KatziveKatzive 2003). In 2011, Kenya extended the 2001 ban on FGM/C among minors to apply to adult women, and also added an extraterritoriality clause, extending restrictions to citizens and residents who commit the crime outside the country's border. Some legal provision calls for increased penalties when FGM/C results in death (Benin, Burkina Faso, Côte d'Ivoire, Egypt, Ghana, Uganda), when performed by a medical professional (Burkina Faso, Côte d'Ivoire, Eritrea, Kenya, Mauritania, Uganda), or in cases of recidivism (Benin, Nigeria, Togo). FGM/C that results in death is punishable by a life sentence in prison in Kenya and Uganda, and by the death sentence in Ghana.

In some instances, criminal laws have been intended to be largely symbolic in nature, or to serve as a means of communicating a new norm that is backed by state support, with the goal of affecting behavior “expressively” by what it says, rather than what it does (sanctions) (Reference McAdamsMcAdams 2000). Educating the public about the new legal provision may serve this purpose. In most cases, however, the laws are intended to catalyze social change and to provide an “enabling environment” for the abandonment of FGM/C (UNICEF 2010). In such cases, proposed legislation must be accompanied by detailed plans for implementation, enforcement, and monitoring.

Across Africa, even in the presence of specific laws against FGM/C, enforcement is variable. There have been scattered reports of prosecution or arrests in cases involving FGM/C in several African countries, including Burkina Faso, Egypt, Kenya, Ghana, Senegal, and Sierra Leone (UNICEF 2010). Reference ObioraObiora (1997: 357) highlights the fact that “diligent enforcement presupposes the availability of material resources and cooperative enforcement agents.” The 2004 Tanzania Human Rights report elaborates on the complex and varied problems that surround enforcement of the law: In some cases, police officers were unaware of the law; “victims” were often reluctant to testify against their own family members; bribes, in some instances, were purportedly given to leaders to assure freedom from prosecution; court cases often failed to offer an adequate standard of evidence despite having testimony from victims and confessions from parents and circumcisers (U.S. Department of State 2004). Hence, concerns about the broader negative ramifications of enforcement expressed by Reference GunningGunning (1992) over two decades ago remain salient: What is the psychological effect on a young, newly cut girl witnessing the arrest of her parents? What is the financial impact of imposing fines on people who are already poor? Given that women are the prime initiators and practitioners of FGM/C, would enforcement lead to the systematic imprisonment of women? Consequently, beyond the question of whether criminal laws can be used as a tool for ending FGM/C lies the question of whether they should be used to eliminate the practice. Rahman and Toubia, however, suggest that the debate over whether legislation should be used as a tool to combat FGM/C is moot, arguing that “the fact that many Africans and Western cultures have recently enacted laws prohibiting the practice creates a de facto role for legislation, the effects of which should be closely observed and documented in the coming years” (2000: xiv). The varied effects of enacting specific anti-FGM legislation at the community level are, however, poorly understood. We examine this, drawing on research on behavior change with respect to FGM/C in Senegal, to test predictions from two leading theories on social regulation.

Legal Theories on Social Regulation

Legal theories of social control are often described as falling predominantly into one of two schools of thought, law and economics and law and society, each of which examine the interplay between informal local social control and the formal legal system regulating human behavior, but differ with respect to the primacy given to each.

The law and economics movement rests on the assumption that social control is achieved primarily by the state through the legal system, and that governments are the chief source of rules and enforcement mechanisms (Reference PosnerPosner 2003). The law and economics approach emphasizes the role of incentives and their ability to influence behavior. It assumes that individuals are rational and forward looking, and alter their behavior in the face of legal incentives. Whether an individual commits a criminalized act is determined by an expected utility calculation. He or she would compare the costs and benefits of criminal activity, where the expected benefits include monetary and nonmonetary returns of the crime, discounted by the probability of realization, and the expected costs of the crime, which include formal and informal sanctions (e.g., social stigma), discounted by the probability of detection. This calculus is based on the assumption that the probability and magnitude of sanctions are known. If the expected benefits exceed the expected costs, a rational individual may commit the offense. This cost-benefit analysis is weighted against those of other legitimate options. Levying penalties against criminalized behavior elevates the potential cost, thereby altering the cost-benefit calculus and discouraging the behavior.

In his seminal article on economic approaches to criminal law, Reference BeckerBecker (1968) emphasized that enforcement is done by the state, rather than victims, and hence the state bears the cost of detecting, prosecuting, and punishing offenders. Limited resources constrain expenditure on enforcement, and are assumed to be allocated according to rational calculations about costs and benefits (Reference McAdams and UlenMcAdams & Ulen 2008). To maximize social welfare, microeconomic models determine optimum levels of expenditure and enforcement, and the magnitude and forms sanctions (fine, imprisonment, or a combination) required to optimally deter crime.

Law and society scholars note that while there are examples of successful deterrence-based strategies, there are also instances in which threat of criminal penalty has a limited impact. They emphasize that rules enforced at local levels often differ from formal law, but adhere to widely shared understandings of appropriate behaviors and practices. When laws are antithetical to the norms of a social group, the pecuniary costs may have limited power to deter behaviors (Reference EllicksonEllickson 1991; Reference TylerTyler 1990). Reference PosnerPosner (2000: 4) emphasizes that, “the law is always imposed against a background stream of nonlegal regulations—enforced by gossip, disapproval, ostracism and violence.” The influence of social groups may be powerful when there are strong social sanctions such as withholding or conferring signs of status and respect, or withholding access to material resources (Reference PosnerPosner 2000). Thus, the law and society movement holds that both law and social norms influence behavior, but often substantive norms supplement, and at times preempt, formal laws.

The law and society perspective further holds that laws may have limited power when they are poorly understood (Reference EllicksonEllickson 1991; Reference TylerTyler 1990). In his seminal study of closed range ordinances regulating cattle ranching in California, Reference EllicksonEllickson (1991) demonstrated that people often settle disputes in total ignorance of their legal rights, and that scant knowledge of the content of legal measures makes it improbable that fear of sanctions meaningfully regulates behavior. He argues, similar to Reference TylerTyler (1990), that beyond the threat of legal sanctions, other important factors influence compliance, including opinions of peers, informal sanctions from others for deviating from expected behavior, personal morality, and perceptions of legitimacy of the law and legal authorities (Reference EllicksonEllickson 1991). These describe a variety of norms that have been described by Reference Mackie and LeJeuneMackie and LeJeune (2009)Footnote ⁵ as falling into a typology shown in Table 2. Mackie originally described FGM/C as a social convention, locked in place predominantly by concerns over marriageability, and possibly as a peer convention (Reference Mackie, Shell-Duncan and HernlundMackie 2000). A convention is a social practice locked in place by interdependent expectations, and change requires coordinating a shift in the reciprocal expectations of interconnected social actors. More recently, Reference Mackie and LeJeuneMackie and LeJeune (2009) expanded the original application of the theory to also address other meanings associated with FGM/C that operate as various types of norms: social norms enforced by positive sanctions for compliance or negative sanctions for noncompliance; religious norms commanded and enforced by God; and moral norms, enforced by internalized values of right or wrong. These norms stand in contrast to legal norms, enforced by government sanctions formally stated in law. Social, moral, and religious norms may be intimately intertwined in complex cultural systems, and if at odds with legal norms, may generate resistance to complying with legal regulations.

Table 2. Typology of Norms

Adapted from Mackie and LeJeune (Reference Mackie and LeJeune2009).

Thus, the law and economics and law and society perspectives differ with respect to the primacy given to legal norms, and point to different predictions about the legislation banning FGM/C in Senegal in several respects: knowledge of the law, perceived enforceability, dispute resolution, and behavior change (Table 3). We investigate these predictions drawing on data from a case study on behavior change with respect to FGM/C in Senegal. Specifically, we ask: Is there widespread knowledge of the existence of a law banning FGM in Senegal? Is the general public or are local leaders aware of the specifics of law? For those with knowledge of the legal provision, does this alter or influence their behaviors, or do local norms regarding FGM/C override legal restrictions? To what extent do normative factors influence compliance or noncompliance with the law? What are the indirect or unintended consequences of enacting anti-FGM legislation?

Table 3. Predictions of the Law and Economics and Law and Society Paradigms

FGC, female genital cutting; FGM, female genital mutilation.

FGM/C in Senegal

A three-year mixed-methods study on the dynamics of decisionmaking regarding FGM/C was conducted in The Gambia and Senegal. In order to examine responses to the law, we draw on data from the Senegalese portion of this study, as FGM/C is not illegal in The Gambia. At the national level, the prevalence of FGM/C in Senegal is 28 percent (according to the 2005 DHS; Reference Diop, Ndiaye and AyadDiop 2006),Footnote ⁶ although rates vary dramatically by ethnicity and by subregion. Our Senegalese study communities included mixed ethnicity borderland communities in Fatick and Kaolack District, where the estimated prevalences of FGM/C are 6 and 11 percent, respectively.

When discussing the significance of FGM/C, those in favor of the practice describe a number of interrelated social, moral, and religious norms that serve to uphold the practice. A number of informants linked FGM/C and Islam, arguing that FGM/C provides the cleanliness and purification required for religious participation and prayer. Many supporters of FGM/C also emphasize its moral virtues such as promoting chastity prior to marriage, and having controlled sexuality and fidelity once married. Greater emphasis, however, is placed on the importance of FGM/C as a crucial component of training and proper childrearing (Reference Shell-DuncanShell-Duncan et al. 2011). As girls are typically cut at a very early age, they undergo a transformation that is not only physical, but also prepares them to be recipients of this training. They are taught to “know the eye”—that is, they know how to display respect for elders, and to honor and uphold the traditions of their ancestors. As we describe elsewhere, FGM/C is widely seen as “women's business,” in that it is organized and practiced by older women on young girls (Reference Shell-DuncanShell-Duncan et al. 2011). It signals to other women who have undergone FGM/C that a girl has been trained to respect the authority of her elders who have undergone FGM/C, thereby creating an intergenerational hierarchy of power among women who have undergone FGM/C. In return for signaling their subordination, young women gain “insider” status, and hence access to the social connections and social support of elder women. We have described this as an intergenerational peer convention (Reference Shell-DuncanShell-Duncan et al. 2011). Those who are not cut are sanctioned by being denied access to the same level of social support. Uncut women may be insulted by being labeled solema, meaning not only “uncircumcised,” but also rude, ignorant, immature, uncivilized, and unclean. Solema are told they “know nothing,” and are often ostracized among women with FGM/C and excluded from attending ceremonial events such as weddings or FGM/C rituals.

Despite the strong cultural significance ascribed to the tradition of FGM/C by its proponents, increasingly intense debates center on whether it is an outmoded custom that should be discarded. For decades, Senegalese activists have been at the forefront of both local and global campaigns aimed at eliminating FGM/C. In February 1984, Senegal hosted a UN-sponsored conference that aimed to discuss anti-FGM policies. With delegates from 20 African nations in attendance, a pivotal outcome of this seminar was the formation of the Inter-African Committee on Traditional Practices Affecting the Health of Women and Children (IAC). One of the primary mandates of the IAC was to call for governments to use of existing national legislation to criminalize FGM/C, and to implement legislation specifically banning the practice.

In 1999, Senegal adopted a criminal law that prohibits the violation of “the integrity of the genital organs of a female person.” The law is applicable to anyone “who violates or attempts to violate” the prohibition; anyone who “provokes these sexual mutilations or gives instructions for their commission.” The penalty includes imprisonment for six months to five years, or where cutting results in death, hard labor for life.

Following passage of the law, media reports claimed that the law was intended to be symbolic in nature. Senegalese parliamentarian Momar Lo, who introduced the bill in the parliament, was quoted claiming that “no one is really going to go to jail … The government will ensure the courts don't apply the law” (Reference HechtHecht 1999). There were no provisions in place, however, to ensure the purely symbolic nature of the law. In the first year following the passage of the law, there were two arrests, but following an emotional public outcry, the charges were dropped and no convictions resulted. In an effort to promote public acceptance of the law, the government supported a two-year program concentrated on awareness raising and educational outreach that disseminated information about the existence and content of the new law. Since that time there have been a number of high-profile arrests and convictions that have received a tremendous amount of media coverage, and have generated debate as to whether the law should be altered, ranging from calls to repeal the law to appeals to stiffen the penalties (UNFPA n.d.).

Senegal has been the site of a number of media campaigns and nongovernmental organization-sponsored initiatives aimed at ending the practice of FGM/C. It is the home of the massive Tostan program, whose goal is to enhance participants' active involvement in the social, economic, political, and cultural development of their communities (Tostan 1999). Beginning in 1997, the basic education program led to the organization of public declarations of community-wide abandonment of FGM/C. Tostan has now organized such declarations in over 5,000 communities in Senegal (as of March 2013, Tostan, pers. comm.).

Recent developments in Senegal make this site a unique setting for examining the effects and interaction of local nonlegal and legal factors influencing the norms surrounding practice of FGM/C. We seek to gain practical insights on the utility of law as a tool to eliminate the practice of FGM/C, and theoretical insights regarding the relationship between culturally mediated norms and formal law.

Methods

Data were collected over a 3-year period in Senegalese border villages in Fatick and Kaolack Districts beginning in 2004. This study site was selected after consulting with Tostan to determine a locale where the Tostan program had been implemented and where our research activities would not interfere with their ongoing work. Study sites did not provide a nationally representative sample, but were instead selected to capture variation in abandonment of FGC and the social dynamics surrounding the practice. Communities were ethnically diverse, included a mixture of families that did and did not practice FGC in the past, and varied as to whether they had or had not participated in the Tostan education program or public declaration.Footnote ⁷ Research in the borderland communities was of interest for several reasons. Movement across the border is very fluid, and communities on either side share very similar characteristics in terms of dominant religion (over 90 percent Muslim), subsistence activities, cultural practices, and climate. It is not uncommon for extended families to live divided on different sides of the political border, and some villages are literally bisected by the border. Therefore, this setting allowed us to examine the effects of differing national characteristics such as overall prevalence of FGM/C, as well as different intervention programs and national policies addressing FGM/C.

The study was conducted in two phases. The first phase involved qualitative research. In-depth interviews were conducted with men, women, community and religious leaders, health professionals, and former circumcisers. Interview were conducted by two local field-workers, one male and one female, who were trained in qualitative methods, the use of interview guidelines, and the overall goals of the research. In Senegal, a total of 98 interviews were taped, and later transcribed and translated into English. Six focus group discussions were also conducted, divided by gender, age (elder vs. younger men or women), and whether participants came from a village that had or had not participated in the Tostan program. At the University of Washington, the transcripts were then typed and coded using a qualitative data analysis program, Atlas ti (GmbH, Berlin, Germany). An initial set of codes was created by the senior investigators, and expanded while coding initial interviews. A subsample of 40 interviews was double coded and compared, revealing high consistency between investigators. The investigators met weekly to discuss emerging themes and constructs using a grounded theory approach (Reference Miles and HubermanMiles & Huberman 1994; Reference Strauss and CorbinStrauss & Corbin 1998). Focus group data were also coded, and themes and the degree of consensus or division were noted, and analyzed in a factorial design by constructing an overview grid organizing the focus group findings along break characteristics (gender, age, Tostan vs. non-Tostan) (following the methods of Reference Knodel and MorganKnodel 1993).

The themes and constructs identified in qualitative data analysis were then used to create an ethnographically grounded survey. This included creating a survey instrument to measure an important concept identified in our qualitative analysis, namely an individual's readiness to change. Categories of readiness to change combined two elements: an individual's preference regarding FGM/C and his actual or anticipated behavior (Reference Shell-Duncan and HernlundShell-Duncan & Hernlund 2006). During our pretest of the survey, we classified an individual's readiness to change using two methods: (1) a staging algorithm based on responses to six questions on preference and behavior, and (2) an individual's self-assigned readiness to change after having the categories explained (Reference Shell-DuncanShell-Duncan et al. 2010). The Senegalese version also included three questions on the law banning FGC. A developmental pretest of the survey was carried out in March and April 2006. This involved a conceptual review of the entire questionnaire with key consultants, evaluation of alternate forms of questions, testing the effect of question order, and assessing reliability using a test–retest process. The entire questionnaire was translated into Mandinka and Wolof, back translated into English, and discrepancies were reconciled. The concordance between self-described and assigned readiness to change was examined. However, since neither represents a “gold standard,” we contrasted the strengths and weaknesses of each categorization method. Because the self-described readiness to change stages resonated so clearly with pretest participants, we elected to use this measure in quantitative data analysis of the full survey. Further details of the pretest are found in Reference Shell-DuncanShell-Duncan et al. (2010).

The second phase of the study involved administering the questionnaire, data entry, and quantitative data analysis. A multistage sampling procedure was used in which villages where FGM/C was practiced by at least some residents were identified, and then stratified into Tostan and non-Tostan villages, and whether they were large or small. Cluster sampling of villages was done in each stratum, and compounds were randomly selected in each village. The number of compounds sampled was proportional to the total number of compounds. Residents were asked to list all female residents between ages 18 and 40 who had given birth to at least one girl. Using a random number table, one eligible woman was recruited from each compound. Women were asked if they had a husband, and if he was alive. If they responded positively, a random number table was used to select one out of every five husbands to be invited to be interviewed. The overall response rate was 96.4 percent. In Senegal, survey data were obtained from 265 women and 82 husbands. Quantitative data from the survey questionnaire were analyzed using Stata software (Statacorp, College Station, TX, USA).

We evaluated associations between readiness to change and perceived power of the law among Senegalese respondents from families for which FGM/C was a tradition using logistic regression (LR). As the outcome of interest for LR models, self-described readiness to change was characterized in two binary variables: willing abandoners versus willing adherents, and contemplators versus willing adherents.

As the predictor of interest, a law score was created by combining responses to three questions on enforceability and power of the law: “agree” responses were given three points, “unsure” two points, and “disagree” responses one point. The total number of points was then averaged, for a variable ranging from one to three, with a higher value indicating stronger expressed belief in the power and enforceability of the law banning FGM/C in Senegal.

Confounding in LR models was defined as a change in the effect of the predictor of interest greater than 10 percent with the inclusion of a control variable. Age, sex, and ethnicity were evaluated for confounding. Only age met this criterion in one model, and thus was controlled in both LR models.

We present results of this LR as both odds ratios (ORs) and predicted probabilities; ORs tend to overstate associations for outcomes that are not rare (e.g., >5 percent prevalence; Davies, Crombie, & Reference Davies, Crombie and TavakoliTavakoli 1998), while predicted probabilities avoid this distortion. Predicted probabilities were calculated by calculating values for logit X (where X is the binary readiness to change outcome) across the range of law scores, holding age constant at the observed mean; the probability π(X) was then calculated as exp(logit X)/[1 + exp(logit X)] and graphed to represent the association between readiness to change and law scores.

Results

Characteristics of survey respondents are shown in Table 4. Women and the subsample of husbands represent multiple ethnic groups. Overall, 49 percent of the women reported having undergone FGM/C. Interethnic marriage has become increasingly common, as the tradition of arranged marriage is giving way to “love marriage.” This has led to an increasing number of FGM/C-incongruent marriages, where one spouse is from a family that traditionally practiced FGC and the other did not. We draw on analyses of the qualitative and survey data to evaluate predictions of the law and economics and law and society paradigms.

Table 4. Sample Characteristics

^* Arabic school only: the respondent's only formal schooling was religious education.

^† Husband's circumcision tradition does not match wife's tradition.

FGC, female genital cutting.

Enforceability

The law and society paradigm asserts that because behavior is guided primarily by local norms, and with minimal knowledge of or concerns regarding formal legal provisions, community members should not regard the law as enforceable. In the law and economics paradigm, a key element to social regulation through law is that it must be seen by the public as both legitimate and enforceable. However, rather than striving for large-scale enforcement, deterrence theory rests on a psychological process whereby individuals are dissuaded from committing a crime if they perceive legal sanctions as certain, swift, and/or severe (Reference Williams and HawkinsWilliams & Hawkins 1986). Highly publicizing a few cases of prosecution should dissuade at least some community members from practicing FGC (Reference ObioraObiora 1997). Reference Williams and HawkinsWilliams and Hawkins (1986), however, draw a distinction between perceived legal sanctions and objective legal sanctions (the actual likelihood of arrest), and argue that objective measures of legal sanctions do not in all instances equate with perceived legal sanctions. This appears to be true in our study communities. Informants, when asked, were unaware of any cases of enforcement of Senegal's anti-FGM law in their own communities, or elsewhere in Senegal. Many people, nonetheless, viewed the law as potentially enforceable. Simply rumors or imaginings of enforceability generated fear of prosecution. Indeed, lack of clarity on the details of the legal restriction contributed to the impression that the law could be broadly applied. This finding was supported by the survey questionnaire data as well: 84 percent of respondents agreed with the statement, “The law banning female circumcision is more powerful than we are, so we must change the practice,” and 80 percent agreed with the statement, “Someone who openly breaks the law banning female circumcision needs to worry about being punished” (Table 5).

Table 5. Respondent's Views on Enforceability and Power of the Law

The Role of the Law in Promoting Behavior Change

According to the law and economics rationale, people ought to look to the law to determine the appropriate course of action. The law must be accepted as legitimate, and fear of punishment should, at least for some, motivate abandonment of FGC. Many law and society scholars, by contrast, contend that formal law has been overrated as an instrument of change, and point to a growing body of evidence that reveals that segments of social life are located and shaped beyond the reach of the law. Hence, moral, religious, and social norms and concerns over rewards and sanctions from the social group exert considerable influence on individual actions, above and often beyond externally imposed criminal punishment.

While examining responses to the law, it is important to note that in our study communities, there were active, and at time heated, debates as to whether FGM/C has become an obsolete custom that should be discarded, or whether it is a deeply valued tradition handed down from ancestors that must be retained. Our data show that there exist a wide range of opinions on FGM/C not just across communities, but even within families. In previous work (Reference Shell-Duncan and HernlundShell-Duncan & Hernlund 2006), we have described the process of behavior change with respect to FGM/C in terms of stages of readiness to change that combine information on individuals' (1) actual behavior (abandon or not abandon FGM/C), and (2) preference (supports or opposes FGM/C) (Figure 1). Willing adherents are those who favor continuation of FGM/C and are continuing the practice in their family.Footnote ⁹ Reluctant adherents are those who personally favor abandonment, but continue the practice because others who participate in the decision have not yet become persuaded to abandon the practice. Contemplators are those who continue the practice, but are experiencing ambivalence and question aspects of the practice. Reluctant abandoners are those who personally support the practice, but have abandoned it because other people or social pressures have forced them to stop. Willing abandoners describe individuals who personally favor abandonment of FGM/C, and have been able to act on this preference.

Figure 1. Dimensions of Readiness to Change Female Genital Cutting. Individual preference for female genital cutting (FGC) and actual behavior regarding FGC combine to form five stages of change for FGC: willing adherent, contemplator, reluctant adherent, willing abandoner, and reluctant abandoner (after Shell-Duncan & Reference Shell-Duncan and HernlundHernlund 2006).

We examine readiness to change in Senegalese women who come from a circumcising tradition, as well as in the subsample of husbands (Table 6). With greatest frequency, women in this group describe themselves as willing adherents, while husbands are most commonly self-described reluctant abandoners. Pairing husbands and wives, complete information is available for 48 couples. Stage of change concordance within couples was 52.63 percent. This demonstrates substantial potential discord within couples: Almost half of husbands were in a different stage of readiness to change than their wives, with husbands being less likely to be willing adherents of FGM/C than their wives, and more likely to self-identify with “none” of the categories of readiness for change.

Table 6. Distribution of Self-Described Stage of Readiness to Change among Senegalese Women and a Subsample of Husbands

^* Restricted to husbands of women for whom FGC was a tradition before the law banning FGC.

We evaluated associations between stage of readiness to change (willing adherents vs. willing abandoners and willing adherents vs. contemplators) and demographic characteristics (age, sex, education, ethnicity, and marital status) among the full sample of female respondents from families for whom FGM/C was a tradition using univariate LR. The only significant association observed was between willing adherents (vs. willing abandoners) and age (OR: 1.07; 95% CI: 1.02, 1.14); no other demographic characteristics were significantly associated with stage of readiness to change (data not shown).

Tables 7 and 8 describe the results of LR for law scores and stage of readiness to change, controlling for age: In Table 7, willing abandoners are compared to willing adherents; in Table 8, contemplators are compared to willing adherents. Predicted probabilities calculated from this LR are represented graphically in Figure 2, and show that increasing law score (stronger agreement with assertions that the law banning FGM/C is powerful and readily enforced) was significantly positively associated with the willing abandoner or contemplator (vs. willing adherent) stage of readiness to change. In comparison to those who support the continuation of FGM/C, those who have abandoned the practice or are actively questioning the practice have significantly greater fear of prosecution. This is consistent with the prediction of the law and economics theory that the potential pecuniary costs help dissuade the practice. However, among ardent supporters of FGM/C, fear of prosecution and belief in the power of the law is significantly lower. This is also consistent with the prediction of the law and society theory that when legal norms run counter to social norms, threat of legal sanctions holds little sway. Consequently, readiness to change is strongly associated with the influence of threats of punishment and beliefs in power of the law. This suggests that imposing legal regulations in communities where there is unanimous support for the practice of FGM/C will have little effect. However, where there are active debates and divergent opinions about the continuation of FGM/C, legislation can give added strength to those in favor of abandonment.

Figure 2. Effect of Fear of Prosecution on Stage of Change among Senegalese Respondents. Significant associations between increasing law score (reflecting agreement with statements about the enforceability of laws banning FGC) and the predicted probability (calculated from results of logistic regression) of a willing abandoner (compared to willing adherent; N = 99; middle), or a contemplator (compared to willing adherent; N = 68; below), controlling for respondents' age.

Table 7. Association between Law Scores and Odds of being a Willing Abandoner versus being a Willing Adherent among Senegalese WomenFootnote * by Logistic Regression (N = 99)

^* Restricted to women who identified as coming from a family that had traditionally practiced FGM/C.

CI, confidence interval; OR, odds ratio.

Table 8. Association between Law Scores and Odds of being a Contemplator versus being a Willing Adherent among Senegalese WomenFootnote * by Logistic Regression (N = 68)

^* Restricted to women who identified as coming from a family that had traditionally practiced FGM/C.

CI, confidence interval; OR, odds ratio.

Knowing that there are strongly divergent views on whether the practice of FGM/C should be abandoned, as well as varied views on the enforceability and power of the law, it is understandable that our qualitative data reveal a wide range of responses to the law. A small number of respondents report abandoning FGM/C following media broadcasts informing the general public about the anti-FGM law, although emotional reactions to this change varied tremendously. Some reported being sad about being forced to leave a valued cultural practice, and viewed themselves as powerless against the force of the law:

In my opinion, I feel sad about the law because it is more powerful than us and we can not disagree with the law. The law forced us to stop it, but it should not be stopped because it is our tradition that we found our grandmothers practicing. (Young Mandinka woman)

For others, knowledge of the legal ban did not on its own motivate abandonment of FGM/C. Instead, FGM/C continued, but generated deep anxiety about the prospect of criminal sanctions. When the Tostan intervention was later introduced in the village, however, their anti-FGM program was welcomed as a means of ending the ongoing fear of punishment. As one Sereer woman commented: “The campaign was after the law. But after the law people were doing it. The campaign made the law stronger and many people have agreed that they have stopped.” This lends support to the assertion that the law may facilitate the change process by giving additional leverage to those who have become persuaded about the need to abandon FGM/C.

At the same time, we found that those who abandon the practice were forced to grapple with the social repercussions of not having circumcised their daughters, and often experienced misgivings about their decision. One Mandinka mother, for instance, reported sleeplessness from worrying about the fate of her two daughters who have fallen into what she described as the “law trap.” She explained: “the law was passed seven years ago and my daughters were seven and five years old at the time. Now the eldest is fourteen years old. If I look at them I get scared.”

Williams and Hawkins (Reference Williams and Hawkins1986) distinguish between the direct and indirect consequences of legal sanctions that promote deterrence, a distinction that is important in the case of FGM/C as well. The Senegalese law, for instance, is aimed at deterring not only parents of young girls, but also circumcisers. Interviews with some former circumcisers revealed that they had stopped practicing following the passage of the law. Additionally, while the skills of circumcisers were previously passed down from one generation to the next, most claimed that they did not train an apprentice before “putting down the knife.” Some informants in our study communities noted that, in the absence of trusted and experienced local circumcisers, the demand has been met by “traveling circumcisers,” who reportedly circulate throughout communities in The Gambia, Guinea Bissau, and elsewhere. Many informants were suspicious of the training and qualifications of these circumcisers, and suspected that they are motivated by economic gain:

1. Q What is your opinion about the law banning FGM?

2. A My opinion towards it is that when I heard of law banning FGM, I felt happy about it because in the past, our ancestors did it accordingly and correctly, but now everybody wants to be a circumciser because of money and will not be qualified. (Elderly Sereer woman)

One indirect effect, then, of deterrence of circumcisers is that those favoring continuation of the practice must weigh the risks of employing a traveling circumciser with questionable skills. A number of people reported this as a reason to favor legal bans on FGM/C.

While direct or indirect effects of threats of legal sanctions resulted in abandonment in some cases, this was not universally true. Some people ascribe strong cultural value to the practice, and, as predicted by the law and society paradigm, rewards and sanctions from their peers and elders exerted considerable influence, above and often beyond externally imposed rewards or punishments. Some described legislative reform efforts as an attempt to “break culture” and force assimilation with dominant ethnic groups that do not practice FGM/C (what some called “Wolofization”). One Mandinka woman remarked, “If they (the people who passed the law) did not leave their tradition, we will not leave ours. They want us to be like them.” In some of our study communities, reactance simmered below the surface, and it took little to incite fervent vocal opposition to the law. In one conservative community, following a tense focus group discussion, participants stood and chanted phrases such as, “We will not let our culture be destroyed!”

Other supporters of FGM/C responded to threats of legal sanctions by taking the practice underground. As Gunning predicted, “if people support the practice, then the law will not actually prevent the surgeries from being performed: people will just hide the fact that they are doing it” (1991-92: 229). For ethical reasons we did not in our study directly ask people if, following the ban, they had continued to practice FGM/C. Nonetheless informants often offered general comments:

1. Q: What changes have you seen in the practice over the years?

2. A: People are not practicing it as how it was in the past. Even if they are doing it, then it is done secretly but not openly done. And that is a change I have noticed in our community here. (Sereer woman)

Secrecy, some informants explained, was achieved partially by eliminating the public celebration that had formerly accompanied the practice. Additionally, it was considered more difficult to detect that FGM/C performed on infants than on older girls, as infants are not visibly absent from village activities during recovery. Consequently, as predicted by the law and society paradigm, among ardent supporters of FGM/C, legal norms ran counter to the social norms upholding the practice, and had little influence in deterring the practice. This reveals that when legislative reform is not accompanied by other effective means of addressing the broader cultural context, threat of criminal sanctions does not appear to be an effective means of catalyzing behavior change. However, when the practice of FGM/C was contested, and where the process of change was underway, legal sanctions strengthened the stance of those in favor of abandonment of FGM/C.

Discussion

We have analyzed our data on responses to the law banning FGM in Senegal with respect to two leading theories on social regulation: the law and economics and law and society paradigms. Each theory leads to divergent predictions on individuals' knowledge of the law, views on enforceability, use of the formal legal system in dispute resolution, and ability of the law to change behavior. Our research reveals support for certain elements of each theory, shedding light on how legal norms are variably weighed against social, moral, and religious norms that serve to uphold the practice of FGM/C. We find that threat of criminal sanctions is weighed against the effects of defying local norms, and this calculus is strongly influenced by the degree to which the cultural value has been called into question.

Key findings of our research include the fact that mass enforcement of the law is not required to generate widespread fear of prosecution. However, fear of prosecution, while contributing to motivation to abandon FGM/C for some individuals, results in multiple responses. While some people viewed the ban as a reason to reluctantly abandon FGM/C, others defied the ban and continued the practice underground. Reactance in our study communities did not spark visible public protests, although in some communities deep resentment over the “criminalization of culture” simmered just below the surface and boiled over when the topic was raised. It is important to note as well that there exist some individuals not only in elite urban echelons of Senegalese society, but in a relatively poor, rural context as well, who expressed having serious reservations about the prospects of having FGM/C performed on their daughters and who welcomed the passage of the law and intervention efforts. Surprisingly, these multiple responses coexisted not just within a single region or village, but even within an extended family. Importantly, where the practice of FGM/C was being contested, legislation served to give added strength to those contemplating or favoring abandonment. This lends support to the assertion of Rahman and Toubia (Reference Rahman and Toubia2000: 13) that “law can be a useful tool for change, giving NGOs and individuals greater leverage in persuading communities to abandon the practice.”

Many scholars have argued that law alone seldom changes behavior, and that legal reform must be accompanied by other change efforts that address the culturally mediated social, moral, and religious norms that may uphold the practice (e.g., Obiora Reference Obiora1997; Rahman & Toubia Reference Rahman and Toubia2000). In our study communities, some respondents who continued to secretly practice FGM/C expressed deep anxiety over not complying with the law, and feared being caught by the authorities. This fear did not, in and of itself, lead to abandonment of FGM/C in most instances, but instead led to residents to welcome the Tostan program to their community. The Tostan intervention was positively viewed as a means of organizing change that would lead to compliance with the anti-FGM law, and other Tostan programs, such as literacy, preventive health care, and sanitation, were welcomed as well. Our findings support the view expressed by UNICEF (2005: 29), who describe the role of legislation in terms of “creating an enabling environment”; they emphasize that, beyond serving as a deterrent, legal bans provide a support mechanism for those who have or wish to abandon FGM/C. Consequently, as Lewis and Gunning (Reference Lewis and Gunning1998) anticipated, a legislative approach can work in a complementary fashion with an integrated community-based intervention approach when addressing FGM/C.

At the same time, it is important to emphasize that in our study communities, those who practiced FGM/C were keenly aware that they are in the minority in Senegal, and that the practice has been targeted for elimination. There were widely divergent views on whether the practice of FGM/C should be discarded or retained, and readiness to change was found to be strongly associated with the perceived influence of threats of punishment and beliefs in the power of the law. Where the practice of FGM/C was being contested, legislation was able to give added strength to the opinion of those who favor abandonment. Even in the context of customary dispute resolution, the weight of force of the new legal norm was able to be co-opted and used to strengthen the stance of those whose norms most closely aligned with the new legal rule. Our findings are, however, specific to the border region of Senegal, a setting in which the minority of residents in the region practice FGM/C, the nation's dominant ethnic group does not practice FGM/C, and decades of campaigns against the practice have sparked active debates. The successful elements of legislative reform seen here may not be found in other regions of Senegal, or in other countries, in which a majority of the population practices FGM/C. Further research is needed to more fully understand responses to legislative reform across the range of social contexts in which it is being implemented, and should investigate how it can best complement other reform strategies aimed at eliminating FGC.