Tactics against sexual harassment: the role of backfire

Journal of International Women's Studies, Vol. 7, No. 4, May 2006, pp. 111-125
pdf of published article


Greg Scott and Brian Martin

Greg Scott is an associate lecturer in Psychology at Charles Sturt University, Bathurst. He is interested in issues relating to gender equality, the role of culture in relation to psychosocial wellbeing, and the interface between socioeconomic and psychological change.

Brian Martin is associate professor in Science, Technology and Society at the University of Wollongong. He is the author of twelve books and many articles about dissent, nonviolence, scientific controversies, democracy and other topics. He was an active member of university sexual harassment committees for 15 years.

We thank Truda Gray and an anonymous referee for valuable comments. This research was supported by the Australian Research Council.

Go to

Materials on backfire

Brian Martin's publications on sexual harassment

Brian Martin's publications

Brian Martin's website


To oppose sexual harassment, it is useful to understand tactics commonly used by perpetrators. A useful approach to tactics is through the concept of backfire: if an action is perceived as unjust and information about it is communicated to receptive audiences, it has the capacity to cause outrage and consequently backfire on the perpetrator. Perpetrators regularly use five types of tactics to inhibit outrage: (1) cover-up of the action; (2) devaluation of the target; (3) reinterpretation of the events; (4) use of official channels to give the appearance of justice; and (5) intimidation and bribery of targets, witnesses and others. These tactics are regularly used against targets of sexual harassment. The deployment of these tactics is illustrated through the case of Anita Hill, who in 1991 accused US Supreme Court nominee Clarence Thomas of sexual harassment. An analysis using the backfire framework offers guidance on effective ways of deterring and countering sexual harassment.

Keywords: sexual harassment, backfire, Anita Hill, Clarence Thomas

Responses to the problem of sexual harassment have been dominated by procedural approaches, with priority given to laws, regulations and grievance committees. There has been considerable effort put into education about the problem, usually accompanied by emphasis on using formal procedures. Consequently, there has been a relative neglect of tactics used by perpetrators and targets of sexual harassment. Wise and Stanley (1987) argue that sexual harassment has been defined in a narrow fashion that leaves out the harassment of women in everyday life and also their practical means of resistance. Tactics used in practice can be gleaned from the numerous available accounts of sexual harassment. However, there is relatively little study of how to conceptualise tactics and to use such a conceptualisation to formulate anti-harassment strategies.

One of the few systematic treatments of tactics is by Langelan (1993), who draws on women's self-defence theory and nonviolent action theory to develop a framework for how individuals and groups can respond. Key insights from her analysis are that simply standing up and opposing harassment is remarkably effective, that group actions are immensely powerful, and that collecting information and building support are important. Langelan emphasises direct personal confrontation with harassers, relegating legal means to back-up at most.

In this paper, we present a different theoretical framework for analysing and developing tactics concerning sexual harassment, though one that is entirely compatible with Langelan's approach. Our framework builds on the phenomenon that attacks sometimes can backfire on perpetrators by generating greater support for the target. The classic form of backfire is when troops or police use heavy violence against peaceful protesters (Sharp 1973, 2005). Examples include the beating of nonviolent protesters as part of the salt satyagraha in India in 1930, which greatly increased support for independence in India and other countries, the shooting of black protesters at Sharpeville, South Africa, by white police in 1960, an event that punctured the international credibility of apartheid, and the shooting of protesters at a funeral procession in Dili, East Timor, by Indonesian troops in 1991, an atrocity that galvanised international support for East Timor's independence. Backfires can also occur during wartime; the 1968 massacre of Vietnamese civilians by US troops at My Lai is a dramatic example.

A closer examination of these and other examples shows that there are two key requirements for attacks to backfire: they must be perceived as unjust or otherwise inappropriate, namely violating social norms; and information about the attacks must be communicated to receptive audiences. In practice, there are many attacks that do not backfire; for example, there were other major massacres in East Timor before the 1991 Dili massacre. Therefore, it is worth looking at means used by perpetrators that inhibit the outrage necessary for backfire to occur.

Study of a wide range of cases (Martin, in press) suggests that it is convenient to classify tactics used by perpetrators to inhibit outrage into five categories: (1) cover-up of the action; (2) devaluation of the target; (3) reinterpretation of the events; (4) use of official channels to give the appearance of justice; and (5) intimidation and bribery of targets, witnesses and others. Ample evidence is available of every one of these tactics during the salt satyagraha, the Sharpeville massacre and the Dili massacre. Furthermore, the same dynamics can be found in cases well removed from violence-versus-nonviolence scenarios (Martin in press).

The backfire framework is a form of grounded theory (Glaser and Strauss 1967), namely theory developed out of close observation of empirical material. The robustness of the framework is shown by the existence of the five types of tactics in a wide variety of cases, including police beatings (Martin 2005a), censorship (Jansen and Martin 2004), defamation (Martin and Gray 2005), dismissals of academics (Martin 2005b), labour disputes (Smith and Martin in press), corporate disasters (Engel and Martin in press), war (Martin 2004) and torture (Martin and Wright 2003). These case studies do not reveal any significant category of tactics distinct from the five presented here. This conclusion is bolstered by a comparison with agenda management theory: nearly every one of the tactics used by governments to control the public agenda (Harding 1985; see also Eyestone 1978) fits into one of the five tactics for inhibiting outrage.

The coherence of the framework is suggested by its correspondence with leading psychologist Albert Bandura's (1990) "methods of moral disengagement," by which a perpetrator can psychologically minimise concern and guilt about an action: most of Bandura's categories correspond precisely to methods of inhibition in the backfire model. Similarly, Stanley Cohen's (2001) analysis of methods of denial of involvement or responsibility, by individuals and governments responsible for atrocities, corresponds nicely with backfire categories (Martin in press). These comparisons with other theories, designed for analysis in specific arenas, add confidence that the backfire framework captures the major tactics used in struggles over injustice.

We propose that this framework can be applied to interpersonal conflicts or situations, such as sexual harassment, not just broader sociopolitical or group processes, and, as such, can offer insight into ways to respond more effectively to it. We begin with some generalisations based on the sorts of cases reported in the sexual harassment literature (e.g. Farley 1978; Gomez-Preston 1993; Morris 1994; O'Donoghue 1997; Read 1982; Rutter 1996). Of course the varieties of sexual harassment are legion, and no general framework can capture the complexities and specificities of individual cases. Our generalisations should be seen in this light.

(1) Cover-up. In societies in which sexual harassment is stigmatised, perpetrators usually try to operate out of the public eye: generally, the more serious the harassment, the greater the secrecy. Much harassment is carried out one-to-one. To be observed by others is risky, unless they are perpetrators too; to have actions recorded on camera is even more risky. The importance of cover-up is suggested by imagining the reaction should a public figure grope a woman's body on television. The crucial role of cover-up is also suggested by routine advice to targets that they should collect evidence of harassment.

(2) Devaluation. It is common for targets of harassment to be labelled in various derogatory ways, for example as "sluts," "poor sports," "frigid," "humourless" and "hypersensitive." Sometimes they are accused of inviting attention. Rumours may be spread about their sexual behaviour or orientation. Devaluation is implicit in attempts to blame the target by saying she dressed or behaved provocatively or was otherwise responsible for the way men reacted to her. The subtext here is twofold: "she deserved it" and/or "it doesn't matter anyway." That is, why should we care that she was treated in this way?

(3) Reinterpretation. Perpetrators commonly say that their actions have been misunderstood: they are just being friendly, their comments were innocent, they were not really serious in what they said or did, and so forth. They may also claim that their actions were reciprocated or encouraged. The classic reinterpretation is to say that "no" means "yes." Such descriptions may or may not be sincere; the point is that they reinterpret a behaviour as being normal, acceptable or understandable.

(4) Official channels. Grievance procedures, courts and other formal channels for dealing with allegations of sexual harassment give the promise of moral justice, but the reality is often quite different: they may give only the appearance of justice (Rosenbaum 2004). There are several fundamental problems with official channels: they are procedural, often being driven by technicalities rather than substantive issues; they are slow, allowing outrage about events to decline with time; they rely on experts such as lawyers; they can be expensive in time, effort and money; and they often keep the issue behind closed doors rather than opening it to wider scrutiny. For these reasons, official channels give an advantage to those with more resources and staying power. The shortcomings of official channels are well known in relation to rape, for which trials often become a "second assault" (Stanko 1987).

(5) Intimidation and bribery. Targets of harassment often are threatened not to reveal or complain about what happened. Conversely, if they are compliant, they may be rewarded by continued employment or even promotion. Some rapists threaten to kill their targets should they reveal what happened. In less blatant forms of harassment, threats and bribes may be made in a more subtle, implied fashion.

On the basis of this brief overview, it seems that all five tactics that inhibit outrage may be used by sexual harassers and those who defend, excuse or cover up for them. The next stage in this analysis is to formulate counter-tactics, namely tactics to express and amplify outrage against sexual harassment. We address this in the conclusion. In the next section, we present a backfire analysis of an actual sexual harassment case: the 1991 case of Anita Hill versus Clarence Thomas, a prominent US case of sexual harassment. There are two important advantages in analysing this case. Firstly, because it is so prominent, extensive documentation from different perspectives is available. Secondly, Thomas's behaviour did backfire to a considerable extent and all five tactics of inhibiting and generating outrage can be observed. That is, the backfire dynamic is evident in this case. In many other cases that do not backfire, one or two tactics, such as cover-up, may be sufficient.

The Hill-Thomas case is complex, so we can only give a brief overview of the main events, enough so that we can highlight tactics. But this is not so different from many other cases of sexual harassment: they are also complex, just not played out so publicly and prominently. For our purposes, we do not need to judge the veracity of claims by either Hill or Thomas; rather, we look at the tactics used in the struggle between them.

Anita Hill-Clarence Thomas: an allegation of sexual harassment

On 1 July 1991, President George Bush, Sr., nominated Clarence Thomas to be a member of the US Supreme Court. The nomination had to be approved by the US Senate. During the months before approval, a key issue raised in objection to Thomas was an allegation that Thomas had sexually harassed Anita Hill, one of Thomas's former employees (Phelps and Winternitz 1993).

Thomas is an African-American who was well known for his conservative ideology and opposition to affirmative action as a means of addressing racial and gender-related inequality. Thomas' nomination split the civil rights community (Brock 2002, p. 97), thus increasing the nomination's chance of success. On the one hand, Thomas would become only the second African-American to sit on the US Supreme Court and thus could be viewed as a minority group representative. On the other, Thomas clearly conformed to the conservative ideology of the Bush administration. Thomas' nomination was strongly supported by conservatives such as Republican Senator John C. Danforth.

At the time of the nomination, Anita Hill was a law professor at Oral Roberts University in Oklahoma. She was asked her thoughts on the nomination by a friend, Karolyne Murdock. For the first time in nearly ten years, Hill disclosed to Murdock that while employed by Thomas, first at the Office for Civil Rights, and later at the Equal Employment Opportunity Commission, she had been sexually harassed by Thomas. It was the first time anyone had suggested that she should raise a complaint about the behaviour (Hill 1997).

Having completed a law degree at Yale University, Hill was employed by Thomas in the Office for Civil Rights in August 1981. According to Hill, "the atmosphere of mutual respect soon began to erode" (Hill 1997, p. 68). Thomas' conversation gradually became less work focussed and more personally oriented with Thomas, despite Hill's rebuffs and protestations, suggesting that they should share a closer relationship. According to Hill, "his confessions about his life became more personal, more graphic, and more vulgar" (Hill 1997, p. 69), including, among other things, vivid discussions of Thomas' sexual interests and of acts he had seen in pornographic movies, including scenes of bestiality, group sex and rape. According to Hill, "I told him repeatedly that I didn't want to talk about those kind of things." (Hill, pp. 69-70).

Despite these alleged incidents, when Thomas took up a position at the Equal Employment Opportunity Commission (EEOC) in 1982, Hill accepted Thomas' offer to continue working for him, a point often raised as inconsistent for someone who had been sexually harassed. According to Hill, she was faced with a choice between certain employment at the EEOC compared with uncertainty at the Office for Civil Rights. Thus, "Despite my concerns, I knew that Thomas in an odd way was offering me job security. Putting my misgivings aside in hindsight, even foolishly so, I decided to take the job at the EEOC as Thomas' assistant" (Hill 1997, p. 73). However, several months after starting in this new position, Hill alleges that the harassment began again. As a result, when offered a position teaching law at Oral Roberts University in May 1983, she gladly accepted.

At the time of the alleged harassment, Hill confided in only two people, friends Susan Hoerchner and Ellen Wells. Having now disclosed the harassment to Murdock some ten years later (and several others subsequent to this disclosure) and for the first time told that she should make a complaint, Hill was faced with the question of what exactly she should do. She says she was reluctant to make a formal complaint and was even warned by friends not to become involved given the highly controversial and divisive nature of Thomas' nomination. Thus, rather than coming forward with the complaint, Hill decided that she would wait until contacted by government investigators during the course of Thomas' background checks and nomination assessment for the Senate Judiciary Committee (Hill 1997, p. 103). Eventually she was contacted, prepared a four-page written statement, and was interviewed by agents from the Federal Bureau of Investigation. After the FBI report cleared Thomas, Senator Danforth took the nomination from the Judiciary Committee to the full Senate. But after the media reported Hill's allegation on 5 October, public pressure led the Senate Judiciary Committee to call a rushed hearing to examine the matter. Following the hearings, the Senate narrowly confirmed Thomas' nomination to the Supreme Court.


Cover-up is a critical tactic for inhibiting outrage, because an event cannot backfire unless it is exposed. Cover-up of Hill's alleged harassment, and of the forthcoming allegation, was attempted at several stages. On the last day of Hill's employment with Thomas at the EEOC, Hill accepted a dinner invitation from Thomas, said to be purely a professional courtesy. Hill alleges that during the course of this dinner, Thomas stated that, "if I ever told anyone about his behavior toward me, it would ruin his career" (Hill 1997, p. 81). While linked also to intimidation, Thomas' statement can be treated as an attempt to cover up his behaviour.

Attempts to cover up Hill's allegation of sexual harassment are also evident in the Senate Judiciary Committee's handling of the allegation during the month of September. For example, Hill was informed on 13 and 15 September that the Senate Judiciary Committee would not investigate her allegation, in large part because of her request for confidentiality (Hill 1997, p. 112). In other words, Senators knew about Hill's claims but were prepared to do nothing about them, at least until media stories forced their hand, about which they were greatly upset (Phelps and Winternitz 1993, pp. 233, 240). Later, during the committee hearings of 11-13 October, Thomas and his advocates projected an image of propriety and asexual professionalism within the workplace. To emphasise this point, eight female character witnesses were called upon to testify on Thomas' behalf (Miller 1994, pp. 461-468), despite the committee having previously declared that character witnesses would not be permitted (Danforth 1994, p. 149). Collectively, these witnesses stated that neither had they witnessed Thomas engaging in acts of sexual harassment, nor did they believe him capable of such acts (Miller 1994, pp. 461-468). In contrast, Hill was not afforded the opportunity to present witnesses with evidence to the contrary.

Two other women, Angela Wright and Sukari Hardnett, contacted the committee with details concerning Thomas' inappropriate workplace behaviour. In a telephone interview with committee staffers, Wright stated that Thomas had consistently pressured her to date him, had commented on the size of her breasts and had made derogatory comments about other women in his employ (Hill 1997, p. 232; Miller 1994, pp. 376-407). In an affidavit sent to the committee, Hardnett stated that "If you were young, black, female, reasonably attractive and worked directly for Clarence Thomas, you knew full well you were being inspected and auditioned as a female ... You knew when you were in favor [and] ... You knew when you had ceased to be an object of sexual interest ... To maintain that Clarence Thomas' office was untainted by any sexuality and permeated by loving, nurturing but asexual concern is simply a lie" (Miller 1994, p. 472). While neither woman claimed that they had themselves been sexually harassed by Thomas, their evidence would have served to rebut statements made by Thomas and his advocates that he would not engage in behaviour that could be perceived as sexual harassment. However, neither Wright nor Hardnett were called upon to testify at the hearing (Hill 1997, pp. 231-233).

A final example is Thomas' reluctance to take a polygraph (lie-detector) test. Hill's lawyer announced to the media on 13 October that she had taken and passed a polygraph test, increasing for many the credibility of her claims. In contrast, Thomas refused to take a polygraph test, saying something like, "that it would be a 'spectacle' for a sitting federal judge to take a lie detector test [and] I'm not going to be further humiliated" (Danforth 1994, pp. 182-183).

For a decade, Hill's allegations of harassment were known to only a few people: neither Hill nor her confidantes pushed for greater publicity. But after Thomas's nomination, Hill gradually revealed the story to a few others. Enough information was circulated to arouse the interest of journalists, in particular Timothy Phelps, who pursued and broke the story, leading to massive media exposure (Phelps and Winternitz 1993, pp. 227-245). Cover-up failed, albeit ten years after the incidents.


Devaluation is an effective tactic in inhibiting outrage as it effectively reduces the credibility of an individual or group. As the credibility of the target decreases, so too does the propensity for outside observers to believe that an injustice has been done.

Hill was labelled quite simply as a liar and as a tool for special interest groups (Hill 1997, p. 2). Within the conservative ranks of politics, Hill's allegation was regarded as little more than an eleventh-hour smear campaign lacking any substance whatsoever (Brock 2002; Danforth 1994). During the publicly televised Senate Judiciary Committee hearing, personal attacks from several key senators were particularly notable, both during the hearing itself, and via the media covering the hearing (Miller 1994). Further, committee members made use of questionable testimony in their attempts to discredit Hill. John Doggett, a friend of Thomas and acquaintance of Hill, said that Hill had fantasised about Doggett's romantic interest in her: "It was my opinion at that time, and is my opinion now, that Ms. Hill's fantasies about my sexual interest in her were an indication of the fact that she was having a problem with being rejected by men she was attracted to" (Miller 1994, p. 428), suggesting that this could explain Hill's "fabricated" allegation. Doggett's affidavit was released to the press, and thus the public, before Hill or the entire Senate Committee had seen it (Hill 1997, p. 194).

Capitalising on Doggett's description of Hill, Senator Danforth obtained an affidavit from psychiatrist Park Deitz. Deitz had published a newspaper article following Hill's testimony describing the condition of erotomania: "a rare delusion of some women that particular men in positions of power ... have romantic interests in them" (Danforth 1994, p. 155). Believing that the 'erotomania hypothesis' could discredit Hill, Senator Danforth attempted to release publicly Deitz's affidavit prior to the confirmation vote. In Senator Danforth's own words, "I thought it my job to show that Anita Hill was not telling the truth. I wanted to present what I thought was relevant evidence to the public - evidence that was not presented to the committee ... My intention was to present documents that refuted her credibility in a contest where credibility was the sole issue" (Danforth 1994, pp.187-188). Although the affidavit was not actually released, Deitz was called upon to provide an interview with the media.

Attempts at devaluation did not stop with the end of the committee hearings, nor even with the confirmation and swearing in of Thomas to the Supreme Court. Shortly after the hearings, the conservative newspaper The Washington Times published an article giving weight to a fabricated story claiming that Hill admitted that the sexual harassment claims were false (Hill 1997, p. 248). David Brock, in his book The Real Anita Hill, deliberately set about destroying Hill's character and credibility by providing an incredibly biased and vicious account of the Hill allegation and subsequent events (Brock 1993). A decade later, Brock wrote another book, Blinded by the Right, in which he admitted that he had been hired by conservatives to write The Real Anita Hill as a hatchet job (Brock 2002): "no respectable publication ... had ever seen the likes of the sexist imagery and sexual innuendo I confected to discredit Anita Hill. These were but two ingredients in a witches' brew of fact, allegation, hearsay, speculation, opinion, and invective labelled by my editors as 'investigative journalism'" (Brock 2002, pp. 107-108). The biography of Clarence Thomas contains many derogatory comments about Hill (Thomas 2001, e.g. pp. 392-393).

Devaluation can effectively inhibit outrage as it reduces the credibility of an individual or group. Those who wish to counter devaluation can provide evidence, present testimonials from others, and use appearance and demeanour to create a good impression.

Hill succeeded in all these areas. Despite furious attempts to find dirt from her past, Hill's detractors could find nothing damning. Her friends and colleagues spoke glowingly of her, with only a few detractors to be found. Finally, and crucially, Hill was impressive in her personal presence. Dressing stylishly, she undercut suggestions that she was an extreme feminist, a label being used in a derogatory manner by her conservative opponents. Presenting her views calmly and in a measured and dignified manner, she added immensely to her credibility. At her first public appearance, before journalists, colleagues and students in Oklahoma on 7 October, all this came together: she was greeted with a standing ovation (Phelps and Winternitz 1993, p. 244). Nina Totenberg, the Supreme Court correspondent for National Public Radio, and first to air Hill's allegations, commented: "Who can forget Anita Hill, trembling almost imperceptibly, as she walked into the ornate Senate Caucus Room, and seeming to gain strength as she testified. Political operatives at the White House initially were so impressed with her steady performance as a witness that they privately concluded on Day One that Clarence Thomas would not be confirmed" (Totenberg 1994, p.6).


Reinterpretation serves to inhibit outrage by providing alternative explanations for the event, in this case the act of sexual harassment, and of the motivations behind the exposure of such an event (Hill's allegation). If an event can be reinterpreted to promote the perception that the event was not what it appeared to be, or the exposure of such an event is maliciously motivated, then outrage is likely to be reduced.

The most creative and masterful example of reinterpretation was evident in Thomas' testimony in the evening session of the Senate Judiciary Committee hearing on 11 October. Having already read a prepared statement that morning, Thomas' testimony followed Hill's. Thomas opened by "unequivocally, uncategorically [sic]" denying all of the allegations made against him by Hill. However, Thomas quickly shifted his focus away from the subject of sexual harassment to the actual function and validity of the hearings. Thomas stated, "this today is a travesty. I think that it is disgusting. I think that this hearing should never occur in America. This is a case in which this sleaze, this dirt, was searched for by staffers of members of this committee, was then leaked to the media, and this committee and this body validated it and displayed it in prime time over our entire nation." He went on to suggest that, "something is dreadfully wrong with this country, when any person, any person in this free country would be subjected to this. ... from my standpoint, as a black American, as far as I am concerned, it is a high-tech lynching for uppity blacks who in any way deign to think for themselves ... it is a message that, unless you kow-tow to an old order, this is what will happen to you, you will be lynched, destroyed, caricatured by a committee of the U.S. Senate, rather than hung from a tree" (Miller 1994, pp. 117-118). With this statement, Thomas in effect sought to reinterpret the hearings as a racial assault as opposed to an investigation into allegations of sexual harassment, casting himself as the victim of a racially motivated attack as opposed to a sexual harasser.

Other examples of reinterpretation were evident both during the hearings and in media releases by senators involved in the hearing. For example, during questioning, Senator Dennis DeConcini asked if Hill was perhaps in some way responsible not only for the sexual harassment, particularly in her second period of employment with Thomas, but also for the media spectacle that the current hearing had become, given that she had not reported her allegations of sexual harassment for ten years (Hill 1997, p. 136; Miller 1994, p. 103). In his questioning of Hill, Senator Alan Simpson was incredulous that, following such "repugnant" behaviour, Hill had visited with Thomas twice, had taken him to the airport, and had telephoned him 11 times (Miller 1994, p. 107). Indeed, prior to the hearing, Simpson appeared on ABC Television's Nightline presenting to the public a telephone log of 11 calls made by Hill to Thomas in the ten years following the alleged harassment, implying that a woman who had been sexually harassed would not continue to contact her harasser. As Hill states, "The campaign that began with DeConcini's 'blame the victim' remark continued with Simpson's labeling me the aggressor in my relationship with Thomas" (Hill 1997, p. 137).

Simpson made a revealing comment: "It's a harsh thing, a very sad and harsh thing, and Anita Hill will be sucked right into the - the very thing she wanted to avoid most. She will be injured and destroyed and belittled and hounded and harassed, real harassment, different from the sexual kind, just plain old Washington variety harassment which is pretty unique in itself" (Hill 1997, p. 139; Phelps and Winternitz 1993, p. 276). As Hill suggests, such a statement can be interpreted as suggesting that sexual harassment is quite distinct from real harassment. That is, sexual harassment can be reinterpreted as "tolerable, if not excusable - that is mild or harmless, or at least less harmful than the 'real' kind he had apparently experienced or inflicted as part of the politics of Washington" (Hill 1997, p. 139).

Official channels

The use of official bodies and processes can inhibit outrage, as many observers may believe that such bodies provide justice. For example, many people would have believed that the Senate Judiciary Committee hearing provided a fair and just platform for assessing claims. This was not the case in the Hill-Thomas hearings. According to Phelps and Winternitz (1993, p. 274), the Senate, in giving only one additional week for hearings before Thomas' nomination was put to a vote, made a serious investigation of Hill's claims impossible.

With regard to knowledge of the processes involved in first raising her allegations to the committee, and subsequently in the hearing itself, Hill was at a distinct disadvantage. Despite her requests about how her statement would be handled by the committee, very little information was forthcoming. Concerning questions such as "who would make the enquiries? Whom would they contact? How would the information be processed, and who would get it?" Hill was simply told, "I cannot give you information about how your statement will be handled ... That is up to the committee" (Hill 1997, p. 113). Hill was later informed that she would be interviewed by the FBI, and that they would talk to her corroborating witness, to Thomas, and to anyone he named, though nothing of how the investigation would proceed following this (Hill 1997, p. 114). Similarly, prior to the Senate Judiciary Committee hearing, Hill was formally notified on 8 October, leaving her just three days to seek legal counsel and to notify and prepare witnesses.

With regard to the hearing itself, procedural fairness for both parties was not a high priority. According to Senator Danforth (1994, p. 149), he convinced the committee to allow female character witnesses for Thomas to be called, despite the fact that the agreed rules of the hearing precluded them. Similarly, Danforth's attempts to obtain and publicly release damaging affidavits from some of Hill's former students demonstrated inequity in the processes of the hearings. In Danforth's own words, "I was far more concerned about procedural fairness for Clarence than about procedural fairness itself. I wanted Clarence to be represented by counsel who would cross-examine Anita Hill and all other witnesses. But I wanted to obtain damaging affidavits from her former law students and release them to the public without any cross-examination at all. ... In my quest for affidavits, I was showing no concern at all about fairness to Anita Hill" (Danforth 1994, pp. 162-163).

Official channels did much more to dampen than amplify outrage over Thomas' harassing behaviour. The FBI investigation found in favour of Thomas, while the Senate hearings were biased against Hill. What broke open the case was publicity, especially media coverage. Without this, the Senate hearings would never have been held. Nor, without media interest, would the hearings have been publicised.


Intimidation can also be an effective tool in inhibiting outrage as threats of reprisals can decrease the likelihood that victims and witnesses will be willing to expose an unjust attack. One factor in Hill's reluctance to confront Thomas at the time of the harassment was that she feared his power, even after she stopped working for him (Phelps and Winternitz 1993, pp. 236, 318).

Following the release of her allegations to the public, Hill was subjected to threatening phone calls. "I went home alone to a ringing telephone and an answering machine full of messages, some worse than I had anticipated - death threats and threats of rape or sodomy. People felt free to leave the most cruel and revolting messages imaginable" (Hill 1997, pp. 128-129). Even following the October hearings, Hill continued to receive hate mail and bomb threats both at home and at work (Hill 1997, p.244).

Although subject to threats and abuse, Hill did not succumb: she wrote up her allegations, testified before the Senate committee and later wrote a book about the events. Thus, in her case, the tactic of intimidation did not succeed.

Intimidation was also used against the journalists who exposed Hill's story. Timothy Phelps, who broke the story in Newsday, and Nina Totenberg, who broadcast the first interview with Hill, were subpoenaed to appear at a Senate investigation into how the story reached the public domain; later their telephone records were subpoenaed. The investigation also targeted Senate staff and advocacy groups (Phelps and Winternitz 1993, pp. 437-440).

We have presented evidence that all five tactics of inhibiting outrage from sexual harassment were deployed in the Hill-Thomas case. But in this particular case, these tactics did not fully succeed. Cover-up failed because Hill was willing to speak out and the media broke the story. Hill, by her competent and brave behaviour, partially countered attempts to devalue her. Her original interpretation that Thomas sexually harassed her was contested but not totally defeated by the alternative interpretation that criticisms of Thomas were racially motivated. (That Hill herself is an African-American was important in the interpretation struggles, because her claims could not be dismissed as racially inspired.) The formal investigations of Hill's allegations served to give the appearance of fair treatment, though behind the scenes there was bias in favour of Thomas. Finally, Hill persisted despite intimidation. The result was that Thomas' alleged behaviour in sexually harassing Hill ended up backfiring on him - a decade later. This is evident in the final confirmation vote, 52-48 in favour of Thomas, the smallest margin in US history. Further, although public opinion initially favoured Thomas, this eroded over time. According to journalist Nina Totenberg (1994, p. 7), "When it was over, public opinion polls showed that people believed Clarence Thomas by a margin of two to one, a ratio that would reverse itself in less than two years."

The Hill-Thomas case greatly increased attention to sexual harassment generally. "The hearings ripped open the subject of sexual harassment like some sort of long-festering sore. It oozed over every workplace, creating everything from heated discussions to an avalanche of lawsuits" (Totenberg 1994, p.7). Donations to female political candidates dramatically increased. In the 1992 federal elections, more women were elected than ever before. Many women enthusiastically supported Hill, and their energising anger was called "the Anita Hill effect." Suddenly Congress started taking claims about sexual harassment much more seriously (Phelps and Winternitz 1993, pp. 419-427).


Sexual harassment, like other actions perceived as unjust, has the potential to backfire. The Hill-Thomas case is perhaps the most prominent case in which sexual harassment ended up being very counterproductive for the harasser. But most cases of sexual harassment do not backfire: the perpetrator is able to use a variety of tactics to counter the outrage that can result from sexual harassment. The Hill-Thomas case reveals the use of all five types of tactics: cover-up, devaluation, reinterpretation, use of official channels, and intimidation and bribery. It is plausible to expect that these tactics will be used in many other cases of sexual harassment. Therefore, it is worth outlining ideas on how to counter these tactics.

Cover-up, the first tactic of inhibiting outrage, can be countered by exposure: revealing information about the harassment to receptive audiences. This need not be media coverage: telling friends and colleagues can be quite effective. Of course the very first step is to tell the harasser. That is widely recommended (except when there is a danger of serious reprisals) and also fits with Langelan's (1993) recommendation to confront harassers with a description of their actions and a demand that they stop.

Harassers hardly ever seek publicity. Targets of harassment often allow cover-up to occur because they are embarrassed, frightened or humiliated, and do not want attention. Their feelings must be respected. However, it is important to emphasise that exposure of harassment by at least some women is essential for generating outrage and making sexual harassment backfire. If enough targets of harassment speak out, then harassers will feel less secure that their own actions can remain hidden.

Devaluation of the target, the second tactic of inhibiting outrage, can be countered by taking steps to increase the target's reputation, for example, by presenting evidence of work performance and honesty. Testimonials by others are valuable, including stories of good deeds and sensible judgement. So is exemplary behaviour by the target: continued good work, suitable demeanour, lack of vindictiveness. Of course it can be quite unfair to expect the target of harassment to behave impeccably. Nevertheless, in terms of effective tactics, it is unwise to be seen to conform to any of the derogatory labels that might be used to discredit the person. The reality is that victims of injustice are regularly scrutinised for any flaw or weakness, which is then used to discredit them. Being prepared for such tactics and being prepared to counter them is essential.

Reinterpretation, the third tactic of inhibiting outrage, can be countered by showing flaws in the alternative perspectives and by continuing to assert the original interpretation: that what happened was indeed sexual harassment and that the harasser is responsible.

The use of official channels, the fourth tactic of inhibiting outrage, can be countered by either avoiding these channels or discrediting them. In many cases, sexual harassment grievance procedures and laws serve to give only the appearance of fair treatment, while in practice doing little to penalise harassers or discourage potential ones. In the Hill-Thomas case, the Senate Judiciary Committee hearings were an official channel, but on their own did not break open the case: it was media coverage that both forced the hearings to occur and ensured attention to Hill's allegations. This is quite different from most sexual harassment cases that involve grievance procedures or court trials, for which these official processes do little to foster wider outrage: rather, through their technicalities, slowness, dependence on experts and imbalance in favour of institutions over individuals, official channels usually dampen outrage.

To recommend avoiding official channels goes against the orientation implicit in the institutionalisation of sexual harassment laws and procedures. It is certainly true that official channels sometimes lead to moral justice. That is not the issue. The question is whether official channels are the most effective tactic to mobilise support through generating outrage. The evidence from a wide range of other injustices, outlined in the introduction, suggests not. The recommendation to avoid official channels is compatible with Langelan's perspective on effective tactics. It can also be said that if using official channels, it is wise to use them in conjunction with publicity.

The recommendation to avoid official channels may seem to put too much of the onus for action on the harassed individual, who is already vulnerable. However, the alternative to official channels is not personal heroism but rather mobilisation and solidarity, namely gaining support from other women and perhaps sympathetic men, including activists, to take action. Some of the most powerful responses to harassers, as described by Langelan, are by groups of women. In order to mobilise support, the harassed individual has to take some action, if only to tell a friend who is willing to help. But it should be remembered that using official channels also requires the harassed individual to take action, usually with less sympathy and support than in a collective response.

What does this imply about sexual harassment agencies, laws and procedures? Would it be better to abolish them, to reduce the illusion that they provide a reliable path to justice? These are big questions. Our analysis here says only that official procedures are commonly used in a way that reduces outrage over the unjust actions of those with more power. So the key to making official channels work better is to increase the power of subordinate groups. In workplaces where feminist consciousness and organisation are high, it is more likely that sexual harassment procedures will be effective. In other workplaces, procedures will be more effective when there is a parallel process of mobilisation, with publicity and education about harassment and support for targets.

Official channels are more likely to be responsive when under the spotlight: they cannot so easily hide weaknesses in their performance. At the very least, it would be useful for statistics to be published about the number of formal complaints to an agency, how long it took for them to be processed, and what the outcomes were. With this information, potential complainants are in a better position to judge whether to make a formal complaint. At the moment, victims are at a terrible disadvantage because they have no information about the success rate of various courses of action. Collecting and publicising information about what happens when different paths are chosen - including time and effort required, emotional impact, the likely outcomes - is essential to help targets make informed decisions.

Our comments here on what to do about formal sexual harassment processes are tentative. The backfire model focuses on tactics, and points to the common role of official channels, conceived as a tactic, in dampening outrage. The implication is that, for the purpose of increasing outrage and making harassment backfire, it is better to aim at mobilisation. The backfire framework, in focusing on tactics, does not directly address the issue of how institutional processes should be reformed; more investigation is needed into this crucial area.

Intimidation and bribery, which together constitute the fifth tactic for inhibiting outrage, can be countered by refusing to be intimidated or bribed and by exposing attempts to do so. This is easier said than done. It is important that at least some women stand up to these tactics.

The analysis here thus leads to a clear set of recommendations for tactics that are likely to increase outrage over sexual harassment. Using these tactics increases the chance that sexual harassment will backfire on the perpetrator. When this happens, it sends a loud signal to the perpetrator and to any other potential perpetrators. Backfire is thus a powerful tool in deterring sexual harassment.

Our analysis is of tactics to increase outrage over perceived injustice. What is perceived as unjust, excessive or inappropriate is neither static nor the same for all people, but rather a continually negotiated matter. The naming and stigmatising of sexual harassment was and is an achievement of the women's movement. Without the ongoing efforts to make people understand and oppose sexual harassment, no outrage would occur. But given that sexual harassment is perceived as a wrong in many societies, backfires can occur. Furthermore, by using the tactics to magnify outrage, sexual harassment is more likely to backfire and to thus continue to be perceived as an injustice.


Bandura, A. Mechanisms of moral disengagement. Pp. 161-191 in Origins of Terrorism: Psychologies, Ideologies, Theologies, States of Mind, edited by W. Reich. Cambridge: Cambridge University Press, 1990.

Brock, D. The Real Anita Hill. New York: Free Press, 1993.

Brock, D. Blinded by the Right: The Conscience of an Ex-Conservative. New York: Three Rivers Press, 2002.

Cohen, S. States of Denial: Knowing about Atrocities and Suffering. Cambridge: Polity Press, 2001.

Danforth, J. C. Resurrection: The Confirmation of Clarence Thomas. New York: Viking, 1994.

Engel, S. and Martin, B. Union Carbide and James Hardie: lessons in politics and power. Global Society, in press.

Eyestone, R. From Social Issues to Public Policy. New York: Wiley, 1978.

Farley, L. Sexual Shakedown: The Sexual Harassment of Women on the Job. New York: McGraw-Hill, 1978.

Glaser, B. G. and Strauss, A. L. The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine-Atherton, 1967.

Gomez-Preston, C. with Reisfeld, R. When No Means No: A Guide to Sexual Harassment. New York: Carol, 1993.

Harding, A. Unemployment policy: A case study in agenda management. Australian Journal of Public Administration, Vol. 44, September 1985, pp. 224-246,

Hill, A. Speaking Truth to Power. New York: Doubleday, 1997.

Jansen, S. C. and Martin, B. Exposing and opposing censorship: backfire dynamics in freedom-of-speech struggles. Pacific Journalism Review, Vol. 10, No. 1, April 2004, pp. 29-45.

Langelan, M. J. Back Off! How to Confront and Stop Sexual Harassment and Harassers. New York: Simon & Schuster, 1993.

Martin, B. Iraq attack backfire. Economic and Political Weekly, Vol. 39, No. 16, 17-23 April 2004, pp. 1577-1583

Martin, B. The beating of Rodney King: the dynamics of backfire. Critical Criminology, Vol. 13, No. 3, 2005a, pp. 307-326.

Martin, B. Boomerangs of academic freedom. Workplace: A Journal for Academic Labor, Vol. 6, No. 2, June 2005b.

Martin, B. Justice Ignited: The Dynamics of Backfire. Lanham, MD: Rowman & Littlefield, in press [2007].

Martin, B. and Gray, T. How to make defamation threats and actions backfire. Australian Journalism Review, Vol. 27, No. 1, July 2005, pp. 157-166.

Martin, B. and Wright, S. Countershock: mobilizing resistance to electroshock weapons. Medicine, Conflict and Survival, Vol. 19, No. 3, July-September 2003, pp. 205-222.

Miller, A., ed. The Complete Transcripts of the Clarence Thomas - Anita Hill Hearings. October 11, 12, 13, 1991. Chicago: Academy Chicago Publishers, 1994.

Morris, C. Bearing Witness: Sexual Harassment and Beyond - Everyone's Story. Boston: Little, Brown, 1994.

O'Donoghue, W., ed. Sexual Harassment: Theory, Research, and Treatment. Boston: Allyn and Bacon, 1997.

Phelps, T. M. and Winternitz, H. Capitol Games: The Inside Story of Clarence Thomas, Anita Hill, and a Supreme Court Nomination. New York: HarperPerennial, 1993.

Read, S. Sexual Harassment at Work. Feltham, Middlesex: Hamlyn, 1982.

Rosenbaum, T. The Myth of Moral Justice: Why Our Legal System Fails to Do What's Right. New York: HarperCollins, 2004.

Rutter, P. Sex, Power, and Boundaries: Understanding and Preventing Sexual Harassment. New York: Bantam, 1996.

Sharp, G. The Politics of Nonviolent Action. Boston: Beacon Press, 1973.

Sharp, G. Waging Nonviolent Struggle: 20th Century Practice and 21st Century Potential. Boston: Porter Sargent, 2005.

Smith, K. and Martin, B. Tactics of labor struggles. Employee Responsibilities and Rights Journal, in press.

Stanko, E. A. Intimate Intrusions: Women's Experience of Male Violence. London: Routledge & Kegan Paul, 1985.

Thomas, A. P. Clarence Thomas: A Biography. San Francisco: Encounter Books, 2001.

Totenberg, N. "Introduction." Pp. 5-8 in The Complete Transcripts of the Clarence Thomas - Anita Hill Hearings. October 11, 12, 13, 1991, edited by A. Miller. Chicago: Academy Chicago Publishers, 1994.

Wise, S. and Stanley, L. Georgie Porgie: Sexual Harassment in Everyday Life. London: Pandora, 1987.