Europeanization or National Specificity? Legal Approaches to Sexual Harassment in France, 2002–2012

Abstract

This paper examines whether—and if so how—a 2002 European Directive on sexual harassment has changed the practice and content of sexual harassment law in France. It finds that the European Directive shaped how French courts address sexual harassment and informed the content of a new sexual harassment law France passed in 2012. Yet, its influence has been mediated by dominant national attitudes about: (1) the nature of sexual harassment, (2) which legal institutions are best suited to address it, and (3) the character of women who claim to have been harassed. This paper further suggests that news reporting on a 2011 arrest of a French politician for sexual assault led to more positive attitudes about sexual harassment victims.

There is a lot of talk lately about the extent to which the European Union has eroded member states’ national autonomy. This concern came to a head in June 2016 when a majority of Great Britain's citizens voted to leave the European Union in a referendum known as “Brexit,” or British Exit. Proponents of the leave campaign argued, among other things, that the EU threatened British sovereignty (Lee 2016). Speaking in favor of Brexit, London Mayor Boris Johnson told a crowd of 200 people “This is not just the time to unshackle Britannia from her chains—though it certainly is—it's a time to speak up for freedom across the whole continent” (Wilkinson and Jamieson 2016). Yet, independent of the question of representation and control over European policy, it remains unclear the extent to which the European Union is bringing legal homogeneity to nation states. To shed light on this question, this paper examines the extent to which a 2002 European Directive shaped how sexual harassment law is practiced in France.

In September 2002, the European Union (EU) passed Directive 2002/73/EC (Zippel 2009), giving member states until October 5, 2005 to transpose the European guidelines into national law (Le Magueresse 2005). The directive defined sexual harassment as a form of “discrimination on the grounds of sex” that specifically includes “any form of unwanted verbal, non-verbal or physical conduct of a sexual nature [that] occurs, with the purpose or effect of violating the dignity of a person, in particular when creating an intimidating, hostile, degrading, humiliating or offensive environment.” It directed nation states to shift the burden of proof from the employee to the employer (Zippel 2009) and to establish institutions that would provide “independent assistance to victims of discrimination in pursuing their complaints about discrimination,” conduct “independent surveys concerning discrimination,” publish “independent reports,” and make “recommendations on any issue relating to such discrimination.” At the time the directive was passed, there was speculation about whether it would lead to national convergence in sexual harassment laws or whether national legal, political, and cultural differences would produce “a good deal of variation in how EU Directives on gender equality are transposed by national governments” (Zippel 2009: 144).

It is generally more difficult for member states to conform to EU directives when preexisting legal practices differ in important ways from European law (Zhelyazkova and Torenvlied 2011). This was the case in France, which has largely addressed sexual harassment via criminal—rather than civil—law and has defined it as a form of sexual violence, rather than discrimination. By examining the extent to which the 2002 European Directive has informed both the practice of French sexual harassment law since 2002 and a new French sexual harassment law passed in 2012, this paper contributes to a fuller understanding of how the European Union is shaping the practice and content of law in member states.

Background: French Law

When the European Directive was passed, French treatment of sexual harassment claims diverged from the Directive's mandates, both procedurally and substantively. First, France did not treat sexual harassment as solely a civil matter. Rather, its Criminal Code contained a specific sexual harassment statute, defining sexual harassment as “the act of harassing another with the goal of obtaining sexual favors.” The French criminal definition was narrower than the European definition, which evoked not only efforts to obtain sexual favors but also “any form of unwanted verbal, non-verbal, or physical conduct of a sexual nature” that has the purpose or effect of “creating an intimidating, hostile, degrading, humiliating or offensive environment.”

Another notable difference between French and European law was that, while the European Directive required that sexual harassment be defined as sex discrimination, the French Criminal Code defined sexual harassment as a form of sexual violence (Roy-Loustaunau 1995). This was evident in the sexual harassment statute's location—in the section on sexual violence, following the felony of rape and the misdemeanor of sexual assault (Dekeuwer 1993). Each of these crimes are technically mutually exclusive under French criminal law, with the difference between sexual harassment and sexual assault being that only the latter includes specific kinds of physical touching—such as forcibly kissing the victim or touching the victim's genital area, breasts, buttocks, or interior of the thigh. The French criminal sexual harassment statute has been located in the sexual violence section since the law was first introduced in 1991, when it defined sexual harassment more narrowly as: “the act of harassing another by using orders, threats or constraints, in the goal of obtaining sexual favors, by a person abusing the authority associated with his professional position.”

French lawmakers dropped the requirement of abuse of professional hierarchical authority in January 2002 to make the sexual harassment law consistent with a new law on moral harassment, which included peer harassment (Saguy 2003). This revision, however, left intact the framing of sexual harassment as sexual violence, not—as the European Directive would later require—sex discrimination. Finally, this, like all French criminal laws, was based on the presumption of innocence. But this hallmark of criminal procedure contradicts the terms of the European Directive's mandate, which presumes a civil law context and shifts the burden to the employer to prove the harassing conduct did not occur.

Since 1992, France has also provided some civil law recourse by prohibiting employers from dismissing, demoting, or otherwise professionally penalizing an employee for having “submitted or refused to submit to acts of sexual harassment” (Cromer 1995). The Labor Code, like the Criminal Code, has defined sexual harassment as sexual violence, not sex discrimination. As civil law, however, it is not constrained by the presumption of innocence and, as such, could respond to the Directive's mandate to shift the burden of proof to employers.

European Integration Versus French Cultural Exceptionalism

Work in political science suggests that the extent to which European nations fully endorse European law varies based on their political and legal traditions (Wind 2010), which differ across European nations (Gibson and Caldeira 1996). Scholars have predicted that cross-national differences in attitudes—such as the importance of individual liberty, support for rule of law, and perceptions of law's neutrality—would play a central role in how European law impacts national policies (Gibson and Caldeira 1996), but there have been few empirical tests of this proposition (Conant et al. 2017).

There were at least two specific legal and cultural obstacles to French adoption of European sexual harassment law. First, as discussed, French law historically defined sexual harassment as a form of sexual violence—not sex-based discrimination—and used criminal courts, rather than exclusively civil courts, to address it (Saguy 2000). Moreover, previous work suggests that French lawyers, activists, and others are committed to framing sexual harassment as a criminal act of sexual violence (Saguy 2003). Addressing the “same” infraction with different legal mechanisms—such as criminal versus civil law—creates challenges of institutional translation. Such “technical misfit” raises the costs of compliance with the EU directive (Zhelyazkova and Torenvlied 2011). For instance, civil law concepts of employer liability and burden of proof—enshrined in European Community law—are incompatible with the logic of criminal law.

Second, there is evidence that French sexual harassment law has been informed by a sense of French cultural exceptionalism in regards to gender. A good example of this is French reaction, in 1991, to news coverage of Anita Hill's testimony during the Senate confirmation hearings of Clarence Thomas to the Supreme Court. During those hearings, Hill testified that Thomas had sexually harassed her when she had worked for him at the Equal Employment Opportunity Commission (EEOC). In the United States, news reporting on the hearings raised consciousness, paved the way for better legal remedies against sexual harassment, increased the number of sexual harassment complaints filed with the EEOC, and encouraged companies to develop programs to deter sexual harassment (The New York Times Editorial Board 2011). French reporting on the very same incident, however, depicted Hill in an unsympathetic light and evoked a narrative of “American excesses” of feminism, Puritanism, and “Battle of the Sexes” (Saguy 2003). The French media contrasted a caricature of hysterical man-hating American women with an idealized representation of French women who appreciate a French brand of seduction (Badinter 1991; Ozouf 1995). Such media reports created a public relations challenge for feminist lawmakers working to pass a French sexual harassment law, who distanced themselves from perceived “American excesses” by limiting the sexual harassment definition to cases involving abuse of professional hierarchical power (Jenson and Sineau 1995). To the extent that French lawmakers, judges, lawyers, and other decision makers do not identify with the norms of the 2002 European Directive on Sexual harassment, this creates normative costs to implementation (Dimitrova and Rhinard 2005).

Yet dominant national cultural attitudes are not static, as demonstrated by a more recent scandal involving a prominent French politician. On May 15 2011, the New York Police Department (NYPD) arrested French Presidential hopeful Dominique Strauss Kahn (DSK) on charges of sexually assaulting a maid at the Sofitel Hotel. The initial French public response—including discussions of American “Puritanism” (Karlin 2011)—echoed French responses to the Hill-Thomas affair. Several leftist public intellectuals defended DSK, describing him as a charmer, seducer, and “friend of women,” but not a rapist (Levy 2011) and expressed indignation that one of their nation's elite men was being paraded handcuffed before cameras like a common criminal (Boulet-Gercourt 2011). Journalist Jean-Francois Kahn, trivialized the assault by saying that there had been no “violent rape attempt” but merely a “troussage de domestique,” literally lifting a female servant's skirts to have forced sex with her (Le Nouvel Observateur 2011). This initial reaction to DSK's arrest, however, ultimately offered France a long-awaited teaching moment on sexual violence and sexism. French journalists and the general public asked: What is sexual consent? What constitutes force? Some French commentators denounced the implication of Jean-Francois Kahn's remarks, that wealthy men should have sexual access to poor women employed in their service (Le Grand Barnum 2011).

This illustrates how dominant national cultural attitudes can shift over time, in response to the global circulation of new ideas and information, creating a novel context for the implementation and understanding of international and national laws. In this case, there is some evidence that French cultural attitudes about sexual consent and coercion shifted in response to news media reporting on the criminal charges brought against DSK in 2011, with potential implications for how sexual harassment cases are treated. For instance, greater awareness of the reality and severity of sexual violence and harassment could make judges—in civil and criminal courts alike—more likely to rule that sexual harassment did indeed occur.

Method

To evaluate the extent to which the European Directive has changed the practice of sexual harassment law in France since 2002, as well as the content of the 2012 sexual harassment law, I analyze relevant French and European laws. I further draw on 21 in-depth interviews, conducted between 2012 and 2014, with 18 French sexual harassment specialists. These include four members of the Association Européenne Contre les Violence Faites aux Femmes aux Travail (AVFT, European Association Against Violence Against Women at Work), which offers legal assistance and support to sexual harassment victims and is widely recognized as a leading French authority on sexual harassment.¹ I interviewed one AVFT member three times over three years and another twice in three years.

The sample also includes two members of the student organization CLASCHES (Collectif de Lutte Anti-Sexiste Contre le Harcèlement Sexuel dans l'Enseignement Supérieur or Collective for the Anti-Sexist Struggle Against Sexual Harassment in Higher Education), three lawyers specializing in sexual harassment, a legal scholar specializing in sexual violence, one judge, three employees of the Defender of Rights (two were interviewed together in a single interview), a government bureaucrat—interviewed in 2012 and 2014; two lawmakers, and one union leader who has developed a sexual harassment training program for union representatives. During the summer of 2012, I also observed legislative debates in the Senate and National Assembly over the new law, participated in AVFT and CLASCHES meetings, and took part in a rally. While it is too soon to provide a definitive assessment of the 2012 law, as the first decisions based on this law have only recently been delivered, I draw on the interviews, a 2016 National Assembly Report, and the secondary literature to analyze the content of the law and its early implementation.

Most of the respondents are advocates for effective sexual harassment law or for sexual harassment victims. As such, they are not representative of the French population, nor can they capture French employers’ attitudes about sexual harassment law. They are, however, ideally positioned to address the central concern of this paper—how the 2002 European Directive on sexual harassment has changed the practice of sexual harassment law in France, albeit from the perspective of sexual harassment victim advocates and legal professionals. To capture a broader range of views, I draw on a National Assembly report (2016) that includes various testimonies, including from MEDEF, an association representing French employers.

My status as a UCLA professor who had previously published a book comparing U.S. and French approaches to sexual harassment (Saguy 2003) gave me credibility and helped me gain access to interview respondents. The fact that I speak fluent French allowed me to conduct interviews in French without an interpreter and to consult French documents myself. My status as a woman seemed to facilitate rapport with women interviewees. When I began this research in 2011, I already knew several people at the AVFT and CLASCHES. In 2013, I was invited to present on a panel about French legal approaches to sexual harassment. It was there that I first learned—from another panelist who worked at the Defender of Rights and became my first contact there—about the Defender of Rights’ institutional incentive to advance a discrimination frame of sexual harassment. My existing contacts helped me make additional ones, while my established expertise made people interested in talking to me. I found that they were as interested in learning from me about how things worked differently in the United States as I was in learning about their work (for a discussion of information as a commodity of exchange between interviewer and interviewee, see Smith 2015). While my preexisting knowledge gave me credibility and entrée, my status as an American and nonlawyer allowed me to ask for additional explanation. My Jewishness—when known—further accentuated my outsider status in this heavily Catholic (but largely secular) country, although this may have been muted by my European descent, at least in interviews with white respondents (all but one).

I identify my interview respondents using a mix of real names and pseudonyms, based on whether or not the person interviewed is already a public figure or asked to be identified. I identify Senator Laurence Cohen, National Assembly Representative Catherine Coutelle, AVFT member Marilyn Baldeck, former AVFT President Catherine Le Magueresse, and legal scholar Claire Saas by name. I refer to the other respondents with descriptions or pseudonyms. I refer to a sexual harassment lawyer who works in both criminal and labor courts as “Ms. Fleury,” to a lawyer specializing in criminal law as “Ms. Collet,” and to a lawyer specializing in labor law as “Ms. Dubois.” I refer to three women who worked at the Defender of Rights as “Ms. Petit,” “Ms. Lefebvre,” and “Ms. Fournier,” respectively. I refer to the union leader as Mr. Bonnet.

Interviews lasted between 37 minutes and 1 hour 20 minutes, averaging 59 minutes. I audio recorded the interviews and had them fully transcribed. I used the qualitative data analysis software HyperRESEARCH to analyze the transcripts. Using HyperRESEARCH, I carefully read each of the interview transcripts, highlighted passages that spoke to a specific theme, and coded these passages with a name capturing that theme—e.g., burden of proof in Labor law cases, labor law procedures, labor law damages, is sexual harassment sex discrimination, sexism in legal proceedings, symbolic importance of criminal law, presumption of innocence in criminal law, dismissal of criminal cases, role of judges, Defender of Rights, or application of 2012 law. I then used the software to produce documents that showed all of the interview excerpts for each theme. The themes emerged inductively by reading the transcripts. I translated all interview excerpts—as well as excerpts from the French press cited above—myself and have edited them for clarity.

Findings

I find that the 2002 European Directive on sexual harassment has shaped how French courts address sexual harassment. This influence, however, has been mediated by a persistent commitment to framing sexual harassment as sexual violence, rather than sex discrimination, and a continued belief in the importance of criminal—rather than civil—law. Entrenched sexism has also stymied the influence of the 2002 European Directive, but news reporting on the criminal charges brought against DSK seems to have created more empathy and understanding of sexual harassment victims. After discussing each of these findings below, I turn to how the European Directive shaped the content of the 2012 French sexual harassment law and examine the law's early implementation. Here too, we see how European influence has been mediated by French commitment to an understanding of sexual harassment as a criminal act of sexual violence.

French Application (or Lack Thereof) of European Law

France was supposed to transpose the European Directive into national law by October 5, 2005. Not all of the provisions pertaining to sexual harassment—including the definition, employer prevention, dissuasive sanctions, and adequate compensation for victims—however, were inscribed into French law by this deadline (Le Magueresse 2005). The definition of sexual harassment as sex discrimination was not inscribed into French law until May 27, 2008 and, only then, in response to European pressure. This law revised the sections in the Labor and Penal Code on discrimination but not those sections on sexual harassment specifically, highlighting French resistance to conceptualizing sexual harassment as sex discrimination. French civil judges did not begin to shift the burden of proof onto employers until September 2008, even though French law had adopted this principle from the directive. Judges only began to change their approach to burden of proof after four separate decisions from the Cour de Cassation, ruling that judges had to “evaluate the facts [as presented by the employee] as a whole and determine if they allowed for the presumption of the existence of the alleged harassment.” If so, it was incumbent on the employer to “establish that they [these facts] did not characterize a situation of harassment.”² According to AVFT's Marilyn Baldeck, “even the most specialized labor lawyers” and judges incorrectly thought that the 2008 Cassation Court rulings had created new rules, whereas they, in fact, merely clarified a rule that had been in effect for six years. Since cases often last four years, the AVFT only began to “reap the benefits” in terms of favorable decisions in 2012. Baldeck reported that the AVFT had won all of its 15 cases between January 2012 and June 2013, whereas winning was “rather exceptional” only a few years earlier.³

In practice, the burden of proof is shared between the employee who has to provide evidence of the harassment and the employer, who “has to prove that those facts do not constitute harassment,” according to Ms. Fleury, a sexual harassment lawyer who works in both criminal and labor courts. Moreover, providing evidence of harassment is difficult. Ms. Lefebvre said that this is her “daily work” at the Defender of Rights and yet, “it's not always very obvious.” Compared to other discrimination cases, it is difficult to get other employees to testify because it often happens “behind closed doors, in an office, in the bathroom.”

Because there is less direct evidence and because one can—since 2002—show a pattern of evidence, rather than having to point to a single definitive moment, lawyers make their case by collecting “a lot of different elements: medical expertise, testimony, emails, correspondence” (Ms. Fleury). This is labor intensive. Ms. Dubois—a lawyer specializing in criminal law—said that only 10 percent of her files are about sexual harassment but that these take up half of her time. Pointing to a thick binder on her desk, she said, “that's a sexual harassment case that I have not even argued yet, compared to two others [pointing to much thinner binders] that are not sexual harassment. Just in terms of documents, there are many more things.” Sexual harassment cases also require more emotional support, compared to, in the words of Ms. Dubois, “someone who lost his or her job for an economic problem and has not been traumatized.” Still, the lawyers I interviewed agreed that the new rules about burden of proof have made it easier to win their cases on behalf of sexual harassment victims (see also Assemblée nationale 2016).

The Defender of Rights and Use of European Law and the Discrimination Frame in France

Also in response to the 2002 European Directive's mandate to establish independent institutions for investigative purposes, in late December 2004, France established the French Equal Opportunities and Anti-Discrimination Commission (Haute autorité de lutte contre les discriminations et pour l'égalité or HALDE),⁴ later subsumed into the Defender of Rights (Défenseur des Droits).⁵ The Defender of Rights cannot initiate legal proceedings nor directly sanction employers or employees. However, sexual harassment victims and their lawyers can ask the Defender of Rights to conduct an investigation and use the investigation's findings in a trial; it is a state crime not to cooperate with a Defender of Rights investigation. Ms. Petit, who used to work at the Defender of Rights, said that, compared to similar institutions elsewhere in Europe, the Defender “has a very large power […] of investigation.”

The Defender of Rights’ powers of investigation are important in a context in which French civil lawyers do not themselves have these powers. Ms. Dubois explained that, unlike U.S. Lawyers who can question witnesses during the discovery period, French lawyers are not empowered by law to “question people, make site visits, or conduct audits,” so that, without the Defender of Rights, they can only build their case with information received from their clients or the police report, when there is a police investigation. In criminal cases, the prosecutor conducts his or her own investigation, but there are no prosecutors in civil cases. The Defender of Rights investigation is especially important in state employment, which—unlike private employment—does not have a Work Inspector, who also has some powers of investigation. Thus, an investigation from the Defender of Rights can meaningfully strengthen a plaintiff's case.

In delivering their conclusions, Defender of Rights jurists invoke the applicable law, from the “highest to lowest source,” in the words of Ms. Lefebvre. Thus, they begin with the European Directives, which state that sexual harassment is contrary to the principle of equal treatment between men and women and constitutes discrimination on grounds of sex. They then cite the French Labor Code, including the combination of Labor Law 132-1, prohibiting workplace discrimination on the basis of several protected classes including sex, and the law of May 27, 2008, which transposes the European Directives, specifying that “any act of a sexual nature suffered by a person and having as the goal or effect of undermining his or her dignity or creating an intimidating, hostile, degrading, humiliating or offensive environment” constitutes sex discrimination. Finally, they cite relevant jurisprudence.

The Defender of Rights has an institutional imperative to frame sexual harassment as sex discrimination. Because the Defender of Rights—and the HALDE before it—is empowered to combat cases of discrimination, it can only intervene in cases of sexual harassment if sexual harassment is understood as a form of sex discrimination. As Ms. Lefebvre said, “We are qualified on matters of discrimination […] If it is not considered discrimination, we would not be qualified to treat the files!” When asked about the work she did at the Defender of Rights on sexual harassment, Ms. Petit spoke about how she had to teach sexual harassment victims and their advocates that “sexual harassment constitutes discrimination and, as discrimination, is within the expertise of the Defender of Rights.” She said that an important part of her work consisted of partnering with associations working on sexual harassment to convince them to ask the Defender of Rights to conduct an investigation. It also involves explaining to lawyers and associations that, rather than do their job in their stead, the Defender of Rights can assist them by conducting investigations that they are not empowered to conduct themselves. The Defender of Rights has made flyers and brochures to inform the public that sexual harassment is a form of discrimination and, therefore, sexual harassment victims can seek its assistance.

The interviews suggest that the idea—championed by the Defender of Rights—that sexual harassment is a form of sex discrimination has had some limited resonance in France. A minority of the French lawyers I interviewed were receptive to the idea that sexual harassment constitutes a form of sex discrimination. For instance, when I asked, Ms. Collet said that sexual harassment was a form of sex discrimination because, for her, all sexual violence is sex discrimination in that women are disproportionately victimized. Similarly, Ms. Fleury agreed that sexual harassment constitutes sex discrimination because “it is because one is a woman that one is treated like that” and “it trips her up [in her career trajectory].” She acknowledged that while French law does not define sexual harassment as sex discrimination, it is defined this way “at the European level… in the directive.” A judge who has ruled on several cases of sexual harassment evoked a principle that he learned in the 1980s in a training session run by the European Court of Justice in which “a national judge is the common law judge of the European standard.” That implies, he said, that “when a European norm exists,” he must “privilege it even if the French law is more restrictive.”

Limits of European Law and of the Discrimination Frame in France

While Defender of Rights jurists and some judges and lawyers rely heavily on the European Directives, many French lawyers do not, even when the Directive would support their cases. As Ms. Lefebvre explained, if the Defender of Rights cites the European Directive but the plaintiff's lawyer does not evoke it in their conclusions, the judge will not take it into account. Ms. Lefebvre, who practiced French law for four years before working at the Defender of Rights, explained that most French lawyers and judges are very “Franco-French”: “It's the [French] Labor Code, the [French] Criminal Code, and a little [French] jurisprudence.” Likewise, the judge cited above said that for many of his colleagues “the European norm is a little bit of an intellectual luxury; it's nice but it's not a priority. It's not essential. The essential is French law.”

Even for many specialists familiar with the European definition, however, the discrimination frame did not resonate in the case of sexual harassment. When asked whether sexual harassment is a form of sex discrimination, labor lawyer Ms. Dubois initially responded, “sexual harassment as such is a discrimination: it's true that the European directives say so.” A moment later, however, she noted that French law does not define it as such and admitted that she herself was not fully convinced that the definition would fit all cases of same-sex harassment. After discussing scenarios in which a discrimination frame did not seem to fit, she concluded: “For me, it's not a problem of sex discrimination; it's more a problem of social discrimination. It's an exercise of power, sexual harassment. That's why I am not comfortable with this notion of sex discrimination.” Ms. Fournier, who worked at the Defender of Rights, said that it is even more “complicated in labor courts to recognize sexual harassment [as discrimination] because there are so many texts,” while the judge said that integrating multiple legal sources complicates the work of French judges who are terribly overworked.⁶

Just because some AVFT jurists use European law and work with the Defender of Rights does not mean that all AVFT jurists personally conceptualize sexual harassment as a form of sex discrimination. A longtime AVFT jurist said about sexual harassment: “We, at the AVFT, we consider that it is a violence […] an attack to integrity.” In the summer of 2013, Marilyn Baldeck said that the AVFT would like to define sexual harassment as both sexual violence and sex discrimination—to have their cake and eat it too—so that they could use all available legal avenues to combat sexual harassment. Likewise, Catherine Le Magueresse has called France's legal approach to sexual harassment a “hybrid model,” in that the Criminal Code treats it as sexual violence, Labor Law treats discriminatory actions following sexual harassment as a violation of the right to work, and law #2008-496—adapting various provisions of Community law in the combat against discrimination⁷—treats it as discrimination (Le Magueresse 2008). Another AVFT jurist, however, said in the summer of 2014: “The AVFT opposes anything that could dilute or hide sexual violence behind other concepts, whether that be the concept of discrimination or the concept of moral harassment or the concept of suffering at work. […] For us, sexual harassment is part of sexual violence.”

Likewise, union leader Mr. Bonnet said in 2013 that sexual harassment “should remain its own subject. It should not be put with discriminations; it mustn't be put with gender inequality. […] It is more of a separate subject and it's better like that.” Indeed, at Mr. Bonnet's union, separate people are in charge of professional equality and sexual harassment, although both report to an officer in charge of women's issues more broadly. Another branch addresses all violence except sexual violence and a fourth deals with all discrimination except sex discrimination. When asked why sex discrimination was not part of discrimination more broadly, Mr. Bonnet said he thought that sex discrimination, which “affects half of the population,” should not be lumped together with other forms of discrimination.

It seems that few sexual harassment victims think that they have suffered sex discrimination either, which helps explain why they do not typically appeal to the Defender of Rights. Only ten out of almost 360 cases at the Defender of Rights in the summer of 2014 were sexual harassment cases. Defender of Rights employee Ms. Lefebvre mused: “I wonder if people even know that sexual harassment is discrimination.” Ms. Fournier added that almost all of the women who have come to the Defender of Rights have come via the AVFT, after the organization informed them that they could appeal to the Defender of Rights.

In fact, some argued that France labor law jurisprudence is moving farther away from a discrimination frame. An AVFT jurist said that, since 2002, in response to European law, employers are now held strictly liable for “a safety obligation,” meaning that “even if [the employer] did everything to ensure that all went well, if the work has the consequence of negatively affecting the workers’ health, he is held responsible” (see also Assemblée nationale 2016). Catherine Le Magueresse said that French judges are increasingly framing sexual harassment as a “psychosocial risk.” Doing so allows them to invoke the employer's obligation to “take the measures necessary to assure safety and protect the physical and mental health of workers” (article L4121 of the Labor law). However, this formulation does not acknowledge structural inequalities. Drawing on her experience legally representing sexual harassment victims, Ms. Fleury said that evoking psychosocial risk often leads to favorable judgments for the victim and an invocation of employer responsibility without “recognizing sexual harassment” per se. Rather, “they are identifying a degradation of the worker's health.”

French Commitment to Criminal Law

French activists, lawyers, and legal experts continue to defend the importance of criminal sexual harassment laws, despite the difficulty victims have in winning such cases or even getting their day in court (Assemblée nationale 2016). Even though the Defender of Rights works almost exclusively in civil courts, Defender of Rights employee Ms. Petit nonetheless said she believes “in the symbolic importance of the criminal conviction.” Criminal lawyer Ms. Collet similarly said: “it's important to name it [as a crime].” A longtime AVFT jurist said “sexual harassment is a misdemeanor, it is an attack on a person's integrity and must be sanctioned as such.” Baldeck said that only the Criminal Code can declare, on behalf of the “whole society,” that a behavior “troubles the social order.” When asked if it would not be better to only bring sexual harassment charges in labor courts given that victims were more likely to prevail there, National Assembly Representative Catherine Coutelle (Socialist) said “No. I do not think that it should only be a question of the Labor Code. It is truly something about which one must raise awareness, that this is a misdemeanor and that harassment must not happen.”

The other side of the symbolic importance of criminal law in France is the devaluation of civil law. As Le Magueresse explained, “If it is not a penal infraction, it means that it is not serious.” Le Magueresse further noted, “civil law is not our culture,” explaining: “If you walk around in France, you buy the newspaper, you'll always have a page for the criminal cases […] you have nothing for the civil [ones].” According to a longtime AVFT jurist, addressing sexual harassment exclusively in civil court would imply that “this behavior is not so serious.” She noted that, in seeking monetary damages through a civil trial, some victims feel as if they are “like a prostitute; he touched me and he pays.” Both Le Magueresse and Baldeck noted that women often say they want to go to criminal court to prevent the harasser from harassing other women. While the employer has some responsibility, explained Le Magueresse, for firing the victim, not believing her, and not investigating her complaint, ultimately it is the harasser who is responsible and who must be charged: “he must be stopped and to stop him, that's [the role of] the penal [system].” Baldeck admitted that the Criminal Code serves this function in a context in which the monetary damages accorded by civil courts are so low as to have little to any dissuasive impact on employers or potential harassers. “We need the penal as long as the civil is so ungenerous,” she concluded.

There are some advantages to criminal over civil law, namely the fact that the state prosecutor is empowered to investigate and prove that the felony or misdemeanor occurred and that—compared to civil courts—criminal courts accept a wider range of evidence. Yet, as in U.S. criminal law, French criminal law holds that a person is innocent until proven guilty. The European Directive presupposes a civil context, putting the burden on the harasser to prove the conduct at issue did not occur. But this would undermine a central tenet of criminal law and is thus not adopted within criminal procedures. And in cases of sexual harassment—which French criminal law defines as falling short of sexual assault—proving guilt is further complicated by the fact that, unlike with rape or sexual assault, there is no physical or forensic evidence. Most cases are dismissed before they even make it to trial, typically on the grounds that “the offense is insufficiently characterized,” according to several of my interviews.⁸ According to former Minister of Justice, Christiane Taubira, “out of the approximately 1,000 complaints filed annually on the basis of sexual harassment, a large number of complaints are dismissed before a trial—classements sans suite; 80 only lead to a conviction, usually suspended, sometimes a fine. And the rate of appeals is 25 percent” (Assemblée nationale 2016). The cases that make it to trial generally involve complaints from multiple women about the same person (Assemblée nationale 2016) and often involve not only accusations of harassment but also of the more serious crime of sexual assault (Le Magueresse 2011, 2014b, Saas, Forthcoming).

Prosecuting a crime under a less serious crime, called déqualification, is common in France and not limited to sexual violence. Prosecutors similarly routinely bring charges of robbery when they could have brought charges of armed robbery, according to legal scholar Claire Saas. The practice of déqualification was controversial among my respondents. Some argued that it can increase the prosecutor's odds of winning the case and, when potential felonies are treated as misdemeanors, save time and money. Others pointed out that it minimizes the seriousness of the crime, leaves sexism intact, and does not address the problem of the underfinancing of the justice system (Le Magueresse 2014a). Indeed, some legal scholars bemoan that the first sexual harassment law created a legal category below that of sexual assault, enabling aggressors to circumvent the more serious charge (Assemblée nationale 2016).

Two lawyers I interviewed (Ms. Collet and Ms. Petit) said that the police are especially likely to categorize sexual assault—or even rape—as “sexual harassment” when it takes place at work because they associate sexual harassment with employment. Catherine Le Magueresse said that victims fall prey to the same logic, reporting to the police that they were sexually harassed when they were in fact sexually assaulted (e.g., forcibly kissed or touched on the breasts, buttocks, genital area, or interior of the thigh). Once categorized as harassment, Le Magueresse said it is difficult to convince a prosecutor to reclassify an incident as assault. Those who overcome these various obstacles and denounce sexual harassment or assault but lose their cases risk being found guilty of slanderous defamation, incurring fines and even jail sentences (Assemblée nationale 2016; Le Magueresse 2011, 2014b).

Sexism

French attitudes about gender and sexuality further limit the protection offered by existing French laws, according to my interview respondents. Claire Saas said that most judges do not take sexual harassment seriously. They may think a sexual harasser “went a bit far,” but they are not truly shocked. This comes through, said Saas, in the euphemisms used to describe the harassment—a “stolen kiss,” “caress,” or a “(misplaced) gesture of tenderness.” Ms. Fleury spoke of a case of an older female judge who responded to the account that Ms. Fleury's client's boss had put his hands on her buttocks, touched her breast, and incessantly commented on her buttocks by asking: “Don't you like it?” Le Magueresse said that defense lawyers still openly suggest that women who wear certain clothes are asking to be sexual harassed or assaulted. Defender of Rights employee Ms. Lefebvre said it was very common for the colleagues of women who had been sexually harassed to say, “she showed cleavage, she was seductive, she wore short skirts, she was a tease.” Ms. Lefebvre said there is work to be done before sexual harassment is taken seriously and that people comprehend the gravity of sexual comments or pornographic images on a computer. At the moment, she said, “it has to go very far before it is considered sexual harassment.” She said it had to involve sexual touching, which technically should move it into the category of sexual assault.

Sexism leads judges and jurors to assume that victims of sexual violence are lying, according to the people with whom I spoke. In the first stage of a French labor court case, four judges—including two who represent the employer and two who represent the employees—hear the case. Ms. Dubois explain that, in typical labor law cases that do not involve sexual harassment, she knows from the start that she can convince the two judges representing the employees; the challenge is to also convince the two judges representing the employer. In contrast, when representing a victim of sexual harassment, “I know it is going to be difficult [to convince] all four.” Ms. Fleury—who represents victims of sexual harassment in both criminal and labor court—also said that the labor court trial is emotionally “more difficult” than criminal proceedings because the employer-judges are next-to-impossible to convince (see also Assemblée nationale 2016). Moreover, because none of the judges are professionally trained, they are more likely to ask the victim inappropriate and hostile questions. Ms. Fleury said she has never won in this first stage; when she wins, it is after the case is sent to a professional judge (départage) or before an appeals court. In the meantime, her clients have to go through this harrowing first stage before four nonprofessional judges. She said that judges often assume sexual violence victims are making up the entire story “to get money,” even though compensatory damages in French sexual harassment cases are typically no more than a few thousand dollars. Based on her experience representing sexual harassment victims in criminal court, Ms. Collet said that defense lawyers there try to discredit the victim by casting aspersions on her psychological state and personal integrity, so that she is “institutionally harassed one last time to find out if her word is true or false.”

AVFT members and lawyers said that, within this general context of sexism, the attitude of the judge and the skills of a victim's lawyer in managing sexist hostility can make all the difference in the verdict. Ms. Collet explained that a bad judge will look at the victim in a way that is “cynical, cold, distant and very judging,” making her feel uncomfortable and undermining her ability to give her best testimony about how things unfolded, what she thought at the time, and what she thought later. In contrast, a good judge will alleviate the victim's discomfort, allowing her to provide better testimony. Ms. Collet said it was crucial to have a competent lawyer who can take control of the situation and insist that people speak differently to the victim. She explained that she herself role-plays with her clients and works with psychologists to prepare clients for hostile and sexist questions because “if the victim is not prepared for these bad questions, she will not know how to respond.” Ms. Collet explained that, in sexual violence cases, the defense lawyers often insult, humiliate, and try to discredit the lawyers representing the victim so that the victim will lose confidence in her lawyer, fire her, and hire a new one, thereby weakening her case.

A longtime AVFT jurist noted that things are even worse in the French overseas departments, such as Guadeloupe and Martinique, where she and another colleague provide training to feminist associations. There, she said, “there have never been trials for sexual harassment, never, never, never, never. When we go there to do trainings, the women say, ‘but why do you want us to denounce [such behavior]? There is no work. We are obligated [to go along with it].” She said that one woman told her that, at the end of a job interview, the employer told her, “you see all the women working in my office? You have seen this room here? They have all passed through here. Your call.” She said other women said that, “even if I was harassed or assaulted, I cannot return to the police because the last time I was assaulted and I went to the police station, the policeman wanted to sleep with me.” She said that one woman who sought out the AVFT had been harassed by a leading figure in Martinique and could not find a lawyer willing to represent her. “She came to Paris, and it was a Parisian lawyer who went [to Martinique to represent her],” said the AVFT jurist. AVFT participated in the trial as a civil party and the accused was found guilty (TGI de Fort de France, 8 April 2002).

Several of the people with whom I spoke, however, said that news reporting on the initial criminal charges brought against DSK alleviated some of the mistrust of sexual violence victims in metropolitan France. Ms. Dubois said that, since media reporting on this incident, “we don't feel like we are automatically assumed to be crazy.” The case and the discussion that followed it created, she said, “a kind of upheaval in French society,” where people are better able to openly consider sexual harassment accusations. “That doesn't mean we always win our cases” or that “it will be easy to win them” but “it is possible to hear what is going on,” she said. Ms. Fleury said that the charges brought against DSK helped raise awareness that powerful men can also harass, assault, and rape. Likewise, Le Magueresse said the DSK case taught that “even if you are very powerful… you can get caught.” Ms. Fleury added that the fact that the charges brought by the Sofitel maid against DSK were credible enough to prompt an arrest—even if ultimately the prosecutor decided not to pursue the case—demonstrated that one can prove the occurrence of sexual violence without a witness by establishing “a pattern of evidence,” including “how the victim told someone about it” or “was troubled afterwards.” One AVFT jurist said that news media coverage of the charges against DSK increased “very considerably the number of victims that appeal to us.” While she said she was pleased to see more women becoming aware of their rights, she lamented that the AVFT does not have the resources to help so many women and that, while the scandal unfolded, news media demands competed for scarce AVFT staff time.

A Priority Question of Constitutionality

In May 2012, France's Constitutional Council (Conseil Constitutionel)—the highest constitutional authority in France—ruled that the existing French sexual harassment law in the Criminal Code was unconstitutional, immediately repealing it and creating a “juridical vacuum” for sexual harassment victims who no longer had legal recourse in their previously constitutionally viable and ongoing cases of criminal sexual harassment. The ruling was a response to a priority question of constitutionality (QPC)—brought by Gérard Ducray, a high-ranking French politician who had been convicted of sexually harassing one of his employees. The QPC was brought before the Cassation Court and forwarded by the Cassation Court to the Constitutional Council. The Council ruled that the legal definition of sexual harassment, revised in 2002 as “the act of harassing another with the goal of obtaining sexual favors,” was too vague (AFP & Reuters 2012; Rigaud 2013).

Marilyn Baldeck learned by chance that the QPC was in motion. Usually there is a lot of publicity surrounding a QPC, according to Baldeck, since QPCs raise questions about fundamental rights and are argued before the highest constitutional authority, conferring prestige on the lawyer bringing them. Typically, specialized and general news outlets publish articles about a new QPC. In contrast, about Ducray's QPC, there was “nothing, nothing at all,” according to Baldeck. Ducray's lawyer deposed the QPC on December 15 2011. The AVFT, however, only learned about it in late February 2012 from one of Baldeck's friends. The friend was an immigration lawyer who was bringing her own QPC and saw the sexual harassment QPC on the list when looking on the Council's website to see when her own was scheduled. The friend immediately called Baldeck to tell her—on the off chance that she did not already know—that a QPC on sexual harassment was scheduled for a hearing.

After learning of the QPC, the AVFT made the controversial decision to join the QPC—as a third party with a legitimate interest in the case—in order to, in the words of Marilyn Baldeck, “put up barriers to [Ducray's] action.” Baldeck explained that the AVFT's objectives were twofold. The first goal was to postpone the repeal of the law until a new one could be passed. The second was to counterbalance Ducray's discourse in the media. While Ducray's lawyer argued that the law should be repealed because it attacked fundamental liberties and penalized attitudes of seduction, the AVFT criticized the law for not sufficiently protecting sexual harassment victims. Indeed, the AVFT had criticized the law since its revision in 2002, lobbying the government to put a better one in place, to no avail (AVFT 2006).

The AVFT failed at the first goal. The association was not able to present its own report, since no AVFT employees are licensed lawyers and the Council only permits licensed lawyers to present a QPC. The association thus hired an outside lawyer—who was not involved in the report's preparation and was thus less familiar with its content than the AVFT members—to present its report. Baldeck said that this was frustrating. Moreover, Baldeck noted with disgust that the members of the highest court—almost all white men of advanced years—were disrespectful during the hearing, talking, winking, smirking, and chuckling, during the two lawyers’ argumentations. This provides additional evidence of how persistent sexism stymies sexual harassment victims’ pursuit of justice. While the hearing was officially open to the public, Baldeck said that two other AVFT members were turned away despite arriving on time. In all, she said, the hearing was “in complete violation of the rules” and “a certain form of dignity and solemnity that one expects for this type of jurisdiction.” The Council ruled to immediately repeal the law without replacing it, leaving thousands of sexual harassment victims with court cases in progress without the original legal basis for their cases.⁹

While unable to defer the repeal, the AVFT succeeded in changing the dominant discourse. News media reports echoed the AVFT analysis about how the previous law did not protect victims and, as I observed in the legislative debates that were to follow, lawmakers would repeat this theme like a refrain.

The 2012 Law

In the wake of the sexual assault charges against DSK, the Constitutional Council's ruling elicited public outrage (Saguy 2012), prompting socialist candidate François Hollande to vow to pass a new law if elected. Two months of intense discussions between the new government, lawmakers, feminist activists, lawyers, and others about a new law followed Hollande's election. According to Ms. Collet, who took part in these debates, these discussions drew heavily on the European Directive's definition. Ms. Collet explained that the European definition was crucial in opening the minds of the senators—typically older men.

Indeed, the 2012 French law was strikingly similar to the 2002 European Directive in its wording. The law—passed on August 6, 2012—defined sexual harassment as “the act of imposing on a person, in a repeated fashion, words or behavior of a sexual connotation that either undermines his or her dignity because of its degrading or humiliated nature or creates an intimidating, hostile or offensive situation for him or her.” In the summer of 2013, Marilyn Baldeck underscored the significance of the reference to “his or her dignity,” explaining that she has never seen the use of possessive pronouns in penal law and that this brings in the specific victim's subjectivity: “It's the dignity of this specific woman, not dignity in general. Thus, [it is] a definition of dignity that can be specific to this woman, depending on her personal history, her level of vulnerability, the context in which it happened.”¹⁰ Similarly, by speaking of a hostile environment “for him or her,” Baldeck argued that the law implicitly recognized that the same situation can create a hostile environment for one person but not another. Of course, she concluded at that time, “we will have to see if the judges go as far as the letter of the law does.”

The 2012 French law also included a second section, describing behavior that is not technically sexual harassment but which is prohibited under the sexual harassment law. This behavior need not be repeated and involves “any form of serious pressure in the real or apparent goal of obtaining an act of a sexual nature, whether this is sought to benefit the author of the acts or a third party.”

Despite important similarities between the 2002 European Directive and the 2012 French law, there were also notable differences. Unlike the Directive, the French Criminal Code did not define sexual harassment as sex discrimination. Rather the new sexual harassment statute was placed in the same section of the French Criminal Code as previous ones: Sexual Violence. This was never discussed during the Parliamentary debates, nor did the lawmakers or activists I interviewed see it as a problem or even notable. When, in late July 2012, I asked Senator Laurence Cohen of the French Communist Party—one of the supporters of the bill—whether sexual harassment was a form of sex discrimination, she said that she had “never particularly asked [herself] that question in those terms.” She explained: “I've always seen it as an act of violence. It is, of course, also discrimination… a form, yes. [pause]… I still think we must maintain sexual harassment as a violence because that allows us to consider it as a very serious act.” Catherine Coutelle (Socialist)—National Assembly representative and president of the delegation of women's rights and equal rights between men and women—conceded that sexual harassment is also a form of sex discrimination “in the sense that one attacks a woman because of her sex.” Still, she said, “for me, [sexual harassment] is really an act of violence. It is not because there is a physical act but because it destroys people; it destroys them.”

Moreover, whereas the European Directive spoke of “unwelcome verbal, non-verbal or physical conduct of a sexual nature” the 2012 French law spoke of a harasser “imposing on a person, in a repeated fashion, words or behavior of a sexual connotation.” Ms. Collet noted that the European Directive's definition seemed to follow the “Anglo-Saxon definition,” requiring some translation to work in the context of French criminal law. For instance, the term “unwelcome”—used in the European Directive—assumes the perspective of the victim. This is consistent with a civil law approach but contradicts the logic of French penal law, which takes the perspective of the accused. Thus, the concept of “unwelcome verbal, non-verbal or physical conduct of a sexual nature” was translated into the idea of a harasser “imposing on a person, in a repeated fashion, words or behavior of a sexual connotation.”

In an official notification instructing judges how to apply the 2012 law, however, then Minister of Justice Christiane Taubira clarified that “non-consent of the victim is thus one of the constituent elements of the misdemeanor, which presupposes acts imposed by their perpetrator, and thus suffered and unwanted by the victim” (La Garde des Sceaux 2012, emphasis added). The Minister of Justice thus instructed judges to treat the term impose as synonymous with unwelcome, thereby directing them to “look from the perspective of the victim,” according to Marilyn Baldeck. The official notification went even farther, by saying that lack of affirmative consent—say, through silence—should be interpreted as evidence that the behavior was unwelcome, thereby inverting the common assumption that a woman who does not explicitly object to sex consents. Specifically, the document stated that the law “does not in any way require that the victim expressly and explicitly inform the perpetrator that s/he was not consenting” and that a judge could interpret a “cluster of clues”—including “permanent silence” or a “request for intervention addressed to colleagues or a hierarchical superior”—as evidence of lack of consent.

Baldeck interpreted this as placing the burden of proof on the harasser to show that his advances were welcome, consistent with the European Directive but potentially violating the principle of the presumption of innocence. She speculated that the minister “put into the official notification all that she had been unable to put into the law.” Whereas “normally an official notification helps interpret the law; this official notification provides new rights,” said Baldeck. Yet, Baldeck pointed out, the official justification does not have the power of law: “judges can do as they please afterwards.” Catherine Le Magueresse said she doubted that many judges would even read this official notification.

The 2012 law increased the maximum penalty for a convicted sexual harasser from one year in prison and a €15,000 fine to two years in prison and a €30,000 fine. In certain aggravated cases, the law allowed for up to three years of jail and €45,000 fine. These included: (1) the harasser abused hierarchical professional authority; (2) the victim was 15 years old or younger; (3) the harasser knew of a victim's particular vulnerability due to age, illness, infirmity, physical or psychological deficiency, or pregnancy; (4) the harasser knew of a victim's particular economic or social vulnerability; or (5) several people harassed or assisted in the harassment of the victim. The 2012 law also introduced a new Article 225-1 into the Criminal Code prohibiting retaliation against sexual harassment victims or witnesses.

In the civil context, the 2012 law revised the Labor Code to use the modified criminal definition. It added language saying that not only should employers “take all necessary measures to prevent” sexual harassment, as was stated in the previous law, but also that they should take all necessary measures to stop and punish sexual harassment that does occur.

During the summer of 2012, as the law was being crafted and debated, AVFT members expressed concern about whether their perspective would sufficiently shape the new law. While one of the early bills was identical to an AVFT proposal, subsequent versions deviated from it in ways AVFT members found troubling. Yet, the government—and especially then Minister of Justice Christiane Taubira—subsequently repeatedly solicited their feedback, and AVFT members expressed enthusiasm about the final version (see also Assemblée nationale 2016), which incorporated many of the AVFT's suggestions (Le Magueresse 2014b).

Members of the student group CLASCHES expressed frustration that the 2012 law was not able to address problems with sexual harassment in education and specifically higher education. In cases of sexual harassment in higher education, victims have to appeal directly to the university president who decides whether or not to conduct an investigation (Assemblée nationale 2016). Any such investigation is then conducted by the alleged harasser's colleagues. CLASCHES members had hoped that the 2012 law could reform this procedure, creating disciplinary bodies comprised of faculty from universities other than that of the harasser. They said that they were told, however, that changing this procedure was a regulatory rather than a legislative change and thus could not be addressed in the 2012 law. Since then, an official notification published on November 25 2012¹¹ reminded university presidents that they had the obligation to prevent sexual harassment, and a 2013 law relative to higher education and research specified that disciplinary bodies must have equal numbers of men and women and include the representative of the mission on gender equality (Assemblée nationale 2016). A 2015 decree¹² further specified that, when there is an “objective reason” to doubt impartiality, disciplinary bodies may be delocalized (Assemblée nationale 2016).

While CLASCHES regretted that the 2012 law did not do more to protect sexual harassment victims, the French association MEDEF—representing French employers—criticized it for doing too much. It asserted that “over half of conversations at the coffee machine” could be characterized as “sexual words or behavior” and that the 2012 sexual harassment law was therefore too broad (Assemblée nationale 2016).

While it is still too early to provide a definitive assessment of the 2012 law, it does not seem to have increased the number of criminal prosecutions or convictions, which remain tiny, according to a National Assembly report presented on November 16 2016 (Assemblée nationale 2016). This report finds that, whereas surveys show that 1 in 5 French women have been sexual harassed at some point in their careers, in 2014, there were only 1048 complaints filed nationwide in criminal court and of those 65 convictions—a number that is somewhat higher than the period between 1994 and 2003 (30–40) and somewhat lower than the period between 2006 and 2010 (70–85). Between 1994 and 2010, only 3–4 cases each year involved actual jail time (Assemblée nationale 2016), and in those cases—AVFT employees say—there was actually evidence of sexual assault. In only 10–12 percent of these cases was there a fine, usually of about €1000, and cases took on average 27 months. The National Assembly report (2016) does not suggest that this has changed since 2012. This same report notes that cases of sexual assault continue to be prosecuted as the lesser crime of sexual harassment, despite the fact that the official notification from the Minister of Justice (La Garde des Sceaux 2012) explicitly condemned this practice.

The new law does not change the underlying practice of criminal proceedings—such as cases being dismissed before getting a hearing and the presumption of innocence—leading the AVFT to continue to focus its energies on the labor courts, where it is increasingly successful. Referring to labor law cases, Baldeck said: “One must not paint an overly positive picture, but we win more and more and better and better.” Baldeck does not attribute these positive developments in labor cases to the 2012 law. Indeed, the cases about which she spoke were based on the old labor law, which was never repealed. Rather, she explained, the better decisions are due to a “more careful application of rules about proof in discrimination cases,” thanks in large part to two important 2013 court decisions instructing judges to take into consideration the evidence as a whole, rather than rejecting pieces of evidence that do not stand on their own.¹³

I found no evidence that the new law is changing much how unions address sexual harassment either. In the summer of 2013, Mr. Bonnet praised the new law but said that it would change nothing for the work of the unions. He explained that the previous law already provided the unions with plenty to work with: “the work that we do not do, that our union sections do not do, was not linked to the problem of the law. It is not because the law was badly written for the victims that we did not take charge of the victims. That wasn't the problem.” The problem with the unions, according to Mr. Bonnet, is that lawyers do not want to use the existing laws. Rather than using sexual harassment laws in the Labor Code, they argue their cases under other labor laws, such as wrongful discharge. This is consistent with what previous work has shown for the late 1990s (Saguy 2003). The National Assembly report further notes that “while information and training policies have been developed in ministries, prevention policies remain poorly developed in companies” (Assemblée nationale 2016: 26).

The fact that civil damages—when awarded at all—remain low in France does little to incentivize French employers to develop sexual harassment prevention or training programs. Mr. Bonnet has led an effort to convince his union to pressure employers to take preventive action and has had some success. He noted, however, that some employers fear that if they develop a prevention program, people will think there has been a case of sexual harassment in the company. Moreover, while union representatives are also potentially well situated to support sexual harassment victims, they rarely play this role. Mr. Bonnet said that sexual harassment victims typically do not turn to their union representative and that “without victims” it is hard for the union to prioritize the issue among all the other issues it confronts.

Conclusion

In sum, the 2002 European Directive on sexual harassment has shaped how French courts—especially civil courts—address sexual harassment. It has also informed the content of the sexual harassment law France passed in 2012. European law's influence has been mediated, however, by several national factors, including: (1) French commitment to the idea of sexual harassment as sexual violence, rather than sex discrimination; (2) French belief in the importance of criminal law for affirming right and wrong; and (3) sexism.

This paper makes several contributions. Empirically, it extends earlier studies of sexual harassment in France (Saguy 2003) and in the European Union (Zippel 2006, 2009) by documenting how the legal treatment of sexual harassment in France has changed in the past decade and a half in response to growing European influence and specifically the 2002 European Directive on sexual harassment. It has shown that the directive has shifted the burden of proof onto employers in civil cases, led to the establishment of the HALDE/Defender of Rights, and shaped the content of the 2012 law.

The influence of the European Directive and of the Defender of Rights, however, has been mediated by national attitudes about the nature of sexual harassment (sexual violence versus sex discrimination), about which legal institutions are best suited to address instances of sexual harassment (criminal versus civil courts), and about the nature of women who claim to have been harassed (Can they be trusted? Did they ask for it by how they dressed?). Of course, France has no national monopoly on sexism. Sexism informs court deliberations elsewhere as well, including in the United States (Estrich 1988; Schulhofer 1998; Schultz 1990). This paper has suggested, moreover, that—in the French case—news reporting on sexual assault charges brought against DSK have produced more empathy for victims of sexual assault and harassment. Likewise, while there are particular challenges to bringing sexual harassment claims in French criminal court, potential plaintiffs in the United States also face well-documented challenges to filing complaints in civil courts. These include potential plaintiff's social psychology and lack of financial, emotional, and social resources (Bumiller 1988). When they do pursue civil remedies, U.S. plaintiffs often pay huge personal and financial costs that make them feel as if they have lost even when they win their legal case (Berrey et al. 2017).

These findings have implications for our understanding of globalization and European integration. The European Directive has shaped how sexual harassment cases have been addressed in France. Yet, its influence has not been immediate, inevitable, direct, or complete. It took six years and four Cassation Court rulings before French civil judges began shifting the burden of proof onto employers, as ordered in the directive. This underscores the essential role played by French courts in making the European Directive a reality in France. More generally, it points to the crucial role European member states’ courts play in either developing or “containing” new rights emanating from the European Union (Conant 2002). Likewise, while the European Directive provided the impetus for the creation of HALDE/Defender of Rights, it would require the hard work of this institution's employees to promote an understanding of sexual harassment as a form of sex discrimination. This underscores the central role of government agency employees in enacting European law.

Yet, this paper has also documented how the dominant French perspective—reaffirmed by the 2012 law—that sexual harassment constitutes a form of sexual violence, rather than sex discrimination, has hindered the implementation of the 2002 European Directive in France. This misalignment between how France and the European Union frame sexual harassment not only constituted an important cultural-cognitive barrier but also—in so far that it favored a criminal versus civil law context in French and European law, respectively—generated a “technical misfit” that has raised French costs of complying with the EU directive (Zhelyazkova and Torenvlied 2011). Finally, this case underscores how dominant national cultural attitudes—such as sexist ideas about sexual harassment victims—shape how European law is implemented in nation-states but can also shift in response to the global circulation of information and ideas.

The continued importance the French respondents place on criminal law, despite innumerable practical limitations, may puzzle American readers, who live in a country where sexual harassment per se is not a crime but rather violates state and federal civil laws. While AVFT employees are increasingly encouraging sexual harassment victims to use civil law procedures for practical reasons, they expressed misgivings about abandoning criminal charges. It is not only that they say criminal convictions dissuade future harassers. It is also that, in their view, only criminal court has the power to reinstate social order by declaring an act to be a violation of the collectivity. AVFT jurists report that harassment plaintiffs feel that accepting monetary damages for sexual harassment is akin to prostitution. Future work should further investigate the origin and impact of such national attitudes about different kinds of legal proceedings, that make it seem natural for Americans—but distressing for French—that sexual harassment be addressed solely in civil courts.

One unanticipated empirical finding was that French courts are increasingly treating sexual harassment as a psychosocial risk. This increases employer liability since employers are legally obligated to protect their employees from psychosocial risks. Yet it further obscures the sex-discrimination component of sexual harassment by lumping sexual harassment with various other psychosocial risks that affect employees regardless of sex. Sociologist Paige Sweet has similarly shown how, in the context of U.S. domestic violence advocacy, medicalized risk—especially as instantiated via screening technologies—obscures gender inequality and gendered violence while promoting gender essentialism (Sweet 2015). Future work should examine how new understandings of sexual harassment as a psychosocial risk shape how employers and employees think about the causes and effects of sexual harassment.

Future work should also examine how the 2012 law is applied in the years to come, as it is still too soon to definitively evaluate its implementation. The National Assembly has proposed to conduct its own follow-up report in 2018 (Assemblée nationale 2016), which would provide a valuable resource for social scientists investigating this question.

This paper has focused on sexual harassment law as it applies to private employment. Several interview respondents and a 2016 National Assembly report (Assemblée nationale 2016) acknowledged that legal protections in education, sports, the military, and the public sector continue to be lacking. Future research should examine the persistent challenges faced in these domains, as well as the legislative and regulatory changes that could be made to better remedy and prevent sexual harassment in these arenas. It would also be useful to interview sexual harassment victims, employers, and others not captured in the current interview sample, about their experiences with sexual harassment in France.

At the time of this writing, the future of Europe as a political, legal, and economic entity is uncertain. There is much talk about the harms and benefits of Europeanization and, more broadly, globalization. Too often, however, such talk is not grounded in careful research. This article has endeavored to rigorously examine how the European Union, and specifically the 2002 European Directive on sexual harassment, has shaped French sexual harassment law. It found that European law has shaped French jurisprudence and legislation in ways that have benefited sexual harassment victims, even though its influence has been mediated by French attitudes and practices. I hope that this article will inspire additional social scientific research investigating the concrete effects European law and politics is having on member states.