Keywords

Introduction

Bystander intervention is a broad phrase used to describe the action (or lack of) in an emergency when the bystander is not initially involved. Bystander intervention is encouraged as a prevention tool for health complications (helping if someone has a heart attack), bullying, sexual and interpersonal violence, drunk driving (friends don’t let friends drive drunk), and terrorism on transit (if you see something, say something). Interventions can include doing nothing, involving someone else (police, bartender, friends, other bystanders), or getting physically involved (administering CPR, stepping in between people fighting). Over the last couple of decades, bystander intervention training has been used as a tool for preventing sexual violence. This chapter briefly addresses prevention efforts that have paved the way for bystander intervention, discusses the efficacy and limitations of bystander intervention, and gives recommendations for moving bystander intervention training programs forward.

How Did We Get Here?

Rape prevention has a history that spans decades and some programs have been more successful than others. This section briefly examines prevention efforts that have led the field towards bystander intervention. While their research was not specific to sexual violence prevention, it is important to discuss the origins of the theory of bystander intervention. Latane and Darley (1970) were the first researchers to examine bystanders. Their research was prompted by the violent homicide of Kitty Genovese. Kitty was stabbed multiple times in the courtyard of her apartment building in Brooklyn and it took a half hour for her assailant to kill her. Multiple neighbors watched the incident unfold and none of them acted to help. Kitty Genovese’s murder was one of several similar stories preceding work by Latane and Darley which piqued their curiosity for why people witnessing a person in an emergency do not intervene to help. They offer a theoretical model for why people intervene or not. There are five different cognitive elements a bystander passes through when deciding whether to intervene. These elements are: noticing a victim or situation, interpreting the situation as an emergency, assuming responsibility to help, having knowledge on how to help in that situation, and then acting to help (Latane & Darley, 1970).

One theme among many prevention efforts is altering people’s social norms. One way to alter these norms is by examining attitudes and beliefs and then providing education and examples as to why some of these attitudes and beliefs are problematic and contribute to the norms that allow sexual violence to continue.

As men are overwhelmingly the perpetrators of sexual violence, logic suggests the power lies within them to stop engaging in this violence. One way in which prevention has targeted men is by addressing the peer level of the social-ecology. College men tend to care a great deal about what their peers think, so changing values and norms of peers may lead to behavior change among men (Fabiano et al., 2003; Schwartz, DeKeseredy, Tait, & Alvi, 2001). Some prevention programs encourage men to address their peers’ language (Barone et al., 2007; McMahon, 2010). In a program called The Men’s Project, men engaged with each other by dissecting gender roles and examining sexist language and rape myths in culture. The Men’s Project is a program which uses the ecological model by identifying and addressing risk factors at all four levels of the social-ecology. At the individual level, The Men’s Project dealt with rape myths and facts; the peer level taught men how to support a survivor and bystander intervention techniques; the community level looked at supportive environments; and the society level discussed privilege, gender socialization, identity intersection, and politics of oppression (Barone et al., 2007). The Men’s Project is conducted in 2-hour sessions over 10 weeks.

Barone et al. (2007) conducted focus groups to evaluate the program’s impact. Men reported observing change in peer behavior after challenging their sexist language. The men who were involved came to recognize how sexism and gender roles are very much ingrained in society. However, through speaking up (when they felt comfortable to do so), they noticed their peers began to change the language they used when describing women (Barone et al., 2007). The men who participated in this group were able to demonstrate that others’ behaviors can be changed through peer influence. They recognized the importance of their role in changing attitudes which led to behavioral change.

Fabiano et al. (2003) suggest that changing culture is an enormous task as “patriarchy and hypermasculine gender roles are deeply ingrained in individuals, families, social customs, laws, institutions…virtually every facet of living” (p. 106). They suggest utilizing a social norms approach as this has found some success in reducing college drinking, which they argue is another culturally entrenched phenomenon. They were interested in discovering how students’ actual norms compared with perceived norms about consent and bystander intervention. They found men’s perceived norms did not match with men’s actual norms regarding sexual behavior and encouraged prevention experts to engage in social norms approaches. By getting men’s perceived norms in closer alignment with actual norms, Fabiano et al. argue more men will become involved as social justice allies.

Piccigallo, Lilley, and Miller (2012) conducted in-depth interviews with men who were involved in an all-male rape prevention campus group. Interested in the pathways by which these men came to be involved with these groups, the researchers conducted interviews to understand the decisions behind joining these groups. One of the main findings was most of the men had a personal connection to the issue of sexual violence against women which led to their group membership (Piccigallo et al., 2012). However, as argued by these authors, researchers and program creators must find a way to garner men’s interest in the issue without an established personal connection to it: the more men with personal connections to the issue means there is more sexual violence occurring to the women in these men’s lives.

Changing values towards safety and respect and engaging men in this change is an essential step in decreasing campus sexual assault (Fabiano et al., 2003). Gidycz, Orchowski, and Berkowitz (2011) conducted an evaluation of a program which combined bystander intervention approaches with social norms approaches. Follow-ups were conducted 4 months and 7 months after the program ended. They found men were less likely to intervene against sexually aggressive peers if they believed there was an absence of support from their friends. Similar to Fabiano et al. (2003), these researchers argued for a large-scale change of campus community norms. They state: “it may be necessary for the campus culture to provide continuous reinforcement of prosocial norms” (Gidycz et al., 2011, p. 735). If the campus culture values prosocial norms, males who are a part of this culture will value these norms, and peers will feel supported in situations which require intervention.

Providing education to increase knowledge about sexual violence, encouraging people to recognize the role their attitudes and beliefs play in perpetuating sexual violence, and making sexual violence everyone’s issue all contributed to the development and use of bystander intervention as a sexual violence prevention tool.

Where Are We Now?

The Bystander Decision-Making Process

Bystander intervention encourages people to be prosocial and prevent a negative event from happening. These events can range from harassment, crude jokes, threats of violence, and acts of violence. Despite whether someone intervenes, they are making a choice. There are three main categories of choices in bystander intervention; being a prosocial bystander, intervening in a further negative way, or choosing not to act. Beginning to seek out answers as to why people make any of these choices and the pathways which led them to those choices will be an important part of the development of bystander intervention prevention techniques. Researchers are beginning to answer these questions but there are many different situations and people interacting in terms of intervention behavior, which makes answering these questions complicated. Noticing a victim or situation and interpreting the situation as an emergency are the first two cognitive stages someone goes through when deciding whether to intervene. Some people may not intervene because they fail to recognize and/or interpret the situation as an incident needing their help. There are reasons at each of Latane & Darley’s five theoretical stages why someone may fail to intervene.

Fabiano et al. (2003) make suggestions for prevention educators in best utilizing a social norms approach. The authors suggest engaging men as allies and to stop “defining male culture as rape culture” (p. 110). Making these cultures synonymous causes men to be disengaged from rape prevention messages. Recognizing not all men are perpetrators of sexual violence may help nonperpetrators become active helpers in stopping the violence. Banyard et al. (2004) argue that participants in prevention programs may feel they are being addressed in a role they do not connect to: specifically women as victim and men as perpetrator. Being cast in these roles can cause individuals to tune out prevention messages because of a lack of identification with this role. Women often do not view themselves as potential victims. Men specifically tune out prevention information delivered in this fashion because they perceive it as being negative towards themselves. Schewe (2008) found that addressing people as potential bystanders, instead of dichotomizing them into victim/perpetrator roles was more effective than other types of prevention interventions.

As previously mentioned, bystanders must notice a victim and recognize the situation as an emergency (Latane & Darley, 1970). This can hinder the likelihood of intervening in situations of sexual violence because of a cultural tendency to blame the victim. Unfortunately blaming a rape victim is commonplace. Burn (2009) found when potential interveners engaged in victim blaming attitudes, they were less likely to intervene. This was found to be a stronger issue for men than women. Because of the strength victim blaming has in our society, bystander intervention prevention education efforts will need to include elements dispelling victim blaming. These efforts need to spread messages that regardless of the victim’s characteristics and actions, sexual assault is always the fault of the person committing the assault (Burn, 2009). Illuminating sources of victim bias should allow possible interveners to more easily recognize a potentially risky situation and thereby increase their likelihood of intervening.

Belief in rape myths influences intervention behavior. One study found rape myths specific to victim blaming correlated positively with a lack of perceived need for intervention (McMahon, 2010). Further, regardless of the specific type of myth, higher rates of any rape myth acceptance led to lower rates of intentions to intervene, therefore sexual assault prevention educators must continue to include rape myths in their curricula. McMahon (2010) found further support which showed students are more likely to intervene in situations of blatant sexual violence, such as an actual assault, rather than intervening when another person makes a sexist joke or uses sexist language. While students are more likely to intervene in blatant sexually violent situations, increasing student awareness of the spectrum of sexual violence and how it all contributes to rape may help encourage students to also intervene in cases of sexist jokes or language.

Banyard et al. (2004) believe bystander intervention can help facilitate new norms, both at the individual and community level, which will better set up communities for intervening in sexual violence. They believe bystander intervention can model prosocial bystander behavior as well as teach bystanders valuable skills. By not labeling individuals as perpetrators or victims, but as bystanders, friends, witnesses, or allies, it places the onus of responsibility on all parties who are willing and able to help. Lastly, when looking at ecological theory, the authors posit if community-level norms are shifted in this way, then societal-level norms will eventually begin to reflect those of the community (Banyard et al., 2004). By empowering all individuals to be active bystanders, the more community members become involved in the fight against sexual violence. This empowerment will hopefully lead to prevention efforts resting on everyone’s shoulders.

Casey and Ohler (2012) gathered a group of male participants who were involved with groups committed to ending violence against women. Through conversations with these men, they found only about a quarter of their participants reported intervening every time they were “confronted with exploitive, offensive, or inappropriate behavior by other men” (p. 77). They were surprised that even men who were actively engaged and trained in techniques to address behavior on the sexual violence spectrum still sometimes had difficulty in confronting others’ use of sexist language, jokes, or coercive/forceful behavior. Some of the men reported difficulty intervening because of a norm in male culture to not interfere with another male’s sexual exploits (i.e., cock-blocking). These men’s willingness to intervene was affected by their perceived relationship and status to the offender as well as their perceived norms of group members in the area at the time of the offending situation (Casey & Ohler, 2012). While there are individual men trying to help end violence against women through addressing their peer groups, challenges still exist even to those who are trained in intervention techniques. This finding suggests the need for more work at addressing the community and society levels of the social-ecology. Men may become more comfortable with intervening as societal norms change.

When Bystander Intervention Works

Shifting the prevention focus to engaging all community members to be prosocial bystanders has had some positive evaluation results. There is a myriad of bystander training programs, some with more evaluation research than others. One of these programs is Bringing in the Bystander®(BITB). BITB is offered as either a one- or multisession training which includes prevalence, causes, and consequences of sexual violence and discussions and role playing surrounding prosocial bystander behavior and safe intervention techniques. Banyard, Moynihan, and Crossman (2009) evaluated the BITB program and found it to be successful at changing beliefs about bystander behavior and increasing behavioral intention to intervene among student leaders, who are already more engaged in helping other students than their nonleader peers. A different evaluation of this same program found post-tests given anywhere from 2 to 12 months after participation continued to yield results which showed attitude and behavior changes (Banyard, Moynihan, & Plante, 2007). The long-term positive change in attitudes and behaviors is significant because many evaluation studies show these changes directly after program participation, but rarely are attitudes and behaviors tested with any amount of time passing in between the program and the post-test.

In another evaluation, Cares et al. (2015) looked at BITB on two separate college campuses; one rural with most students residing on campus and the other urban with a mix of residential and commuter students. Testing this program between two different college populations was important because often campuses want a prevention program which is not expensive, does not take a long time, and applies to diverse settings and students. Despite some differences between genders and campuses, the researchers were able to demonstrate through pre- and post-tests as well as a 12 month follow-up that “initial changes in attitude lasted at least as long as twelve months post program” (Cares et al., 2015, p.180).

BITB has been widely evaluated. Moynihan et al., (2015) conducted an evaluation of BITB and the Know Your Power® social marketing campaign. Utilizing separate campuses, researchers tested the effectiveness of BITB by using Know Your Power® as a control. Both campuses were exposed to the Know Your Power® campaign and one campus participated in BITB training. While both the treatment and control groups had a decrease in reported bystander behavior after 1 year, those who also received the in-person BITB program and the social marketing campaign reported higher levels of bystander behavior after 1 year. Further, students who reported low opportunity for interventions on pre-test measures, reported an increase in helping strangers at the 1 year follow-up, leading the authors to believe that the in-person training potentially increased students’ awareness and identification of intervention opportunities (Moynihan et al., 2015).

Green Dot is a bystander training method used on college campuses. Green Dot attempts to decrease barriers to engaging in bystander behaviors and teaches students to select an intervention they feel safe to carry out. One evaluation of Green Dot comes from the University of Kentucky (Coker et al., 2011). Green Dot consists of a motivational speech encouraging students to connect with the issue of sexual violence and present intervention as something manageable for students to do. Green Dot then trains smaller student groups to recognize risky situations and implement bystander behavior. Green Dot encourages diffusion of behaviors using peer opinion leaders (POL). POLs were recruited from many different student subgroups to increase student exposure to the trained POLs. Green Dot is different than BITB because it teaches students to look for “high-risk potential perpetrator behavior” as opposed to identifying victims. Students who only participated in the motivational speech reported more active bystander behaviors than students who received neither of the Green Dot components. Participating in both the motivational speech and the specialized training resulted in the most reported intervention behaviors as well as the lowest rates of rape myth acceptance (Coker et al., 2011).

Evaluations of these programs found both intent to intervene and actual intervention behavior increased and there was a decrease in attitudes which have been linked to low intervention intention such as rape myth acceptance. Further, most of the evaluations of these programs were able to conduct follow-ups after lengths of time had passed and found that intervention intentions and behaviors were still greater than prior to training.

When Bystander Intervention Does Not Work

Several limitations hinder the ability of bystander intervention to prevent sexual violence. One of these limitations involves social norms. An infrequently addressed limitation of bystander intervention is that it involves conflicting social norms. Bystander intervention encourages prosocial behavior such as helping others’ in need and intervening with peers’ inappropriate language. However, this is in direct conflict with the social norm of minding one’s own business. Further, there is a social norm among men to not interfere with other men’s sexual activity. In the moment where intervention occurs, the issue requiring intervention must be so great as to overcome the norm to not interfere too much with others’ personal matters. This may be the reason behind why students are much more likely to intervene in situations of sexual assault than with peers’ sexually problematic language. Further, if sexual violence is occurring between people in a relationship, bystanders may feel less responsibility for intervention because of a norm to respect others’ privacy (Banyard, 2011).

A second limitation is that sexual violence is often committed in isolation. The very definition of isolation is a lack of bystanders. Therefore, bystander intervention training must focus on teaching people to recognize potentially violent situations before the victim is isolated. Lastly, situations where interventions could occur are very specific and full of nuanced details specific to each scenario. Developing measures to attempt to gauge all the different contexts under which someone chooses (or not) to intervene would be a daunting task. Using interviews with students would allow researchers to determine commonalities among all the different types of situations college students could find themselves in as potential interveners. Such research could help to create measures that could give more insight into situationally specific barriers and facilitators of bystander intervention.

Where Are We Going?

Over 20 years have passed since Schewe and O’Donohue (1993) reviewed rape prevention programs and determined prevention programming suffers from a lack of knowledge about the cause(s) of rape. Unfortunately, this is still the case for many sexual violence prevention programs. In her CDC report to the White House Task Force to Protect Students from Sexual Assault, DeGue (2014) states one best practice for sexual assault prevention is to be strongly based in theory. Until recently, sexual assault prevention has not had a strong base in any theory of sexual violence; likely caused by a limited understanding of the etiology of rape.

Training and education surrounding consent has become more predominant in the prevention landscape, but few theories argue that rape is caused because of a misunderstanding of consent. Further, much programming has focused on rape myths and traditional gender roles; and while feminist theory speaks to both, the larger context of feminist theory points to many different etiologies for rape. McPhail (2015) argues that while feminist theory is widely believed to only deal with power and control, this ignores other feminist theorizing of rape. While power and control can certainly be pointed to as a motive for rape in society (i.e. rape culture) and for individuals (man wants to be dominant over woman), this explanation alone does not count for the sexual nature of rape, the intersectionality of rape and other forms of oppression, and motivations in which men are attempting to demonstrate masculinity (McPhail, 2015). Moreover, feminist scholars discuss the importance of examining gender-based violence through a lens of intersectionality. Kimberle Crenshaw argues that race and gender cannot be viewed as mutually exclusive (Crenshaw, 1997). Angela Davis has argued that attempting to eliminate rape by only examining it as a gender issue and ignoring the role race and sexuality play will result in failure (Davis, 1990). Most prevention programming lacks an intersectionality lens and thereby neglects recognizing the complex ways in which gender, race, sexual orientation, and disability status interacts with and upon potential perpetrators, victims, and bystanders.

While Latane and Darley (1970) outlined a theory of bystander intervention, their theory accounts for the steps someone takes in the intervention decision-making process and is not specific to violence prevention. Banyard (2011) and McMahon (2015) have conceptually addressed this need for a more thorough and comprehensive theoretical approach to bystander intervention, however these pieces focus on increasing prevention efforts to community and society contexts of the social ecology, rather than developing a prevention-based theoretical model. One theory which may help bridge the gap between the social ecological model and bystander intervention that has some empirical support is routine activities theory. This theory could possibly adequately address the need for bystander intervention training. Indeed, sexual violence has all three elements of routine activities: motivated offenders, suitable targets, and lack of capable guardianship. Bystander intervention training could increase the amount of capable guardianship, thereby resulting in prevention as all three elements of routine activities theory would no longer be met (Cohen & Felson, 1979). Further, increasing capable guardianship effectively applies to the peer and community contexts of the social ecology; which has been called for by multiple scholars (Banyard, 2011; DeGue, 2014; McMahon, 2015). Prosocial bystanders and capable guardians should be considered synonymous terms.

Lack of capable guardianship can refer to several elements when viewed in the context of sexual violence. It could be a lack of bystanders to intervene or victim incapacitation. Schwartz and Pitts (1995) argue that there are an increased number of likely-or motivated- offenders on college campuses due to all male peer groups who encourage sexual abuse of women, such as fraternities and athletic teams. When looking at victims, specifically women, Schwartz & Pitts (1995) discuss suitable targets. They argue the college environment is one which encourages alcohol consumption, thereby making women who voluntarily drink large amounts of alcohol “suitable” for sexual victimization. Lastly, college campuses specifically can be lacking in capable guardianship, the third element of routine activities theory. Traditionally, campus administration does not take sexually violent crimes seriously and offenders receive little to no punishment for their actions (Schwartz & Pitts, 1995). Coupled with all male campus groups who are supportive of sexual violence towards women, the pool of guardians who are likely to intervene becomes even shallower.

Schwartz, DeKeseredy, Tait, and Alvi (2001) take capable guardianship one step further. Using the idea of rape culture from feminist theory, they argue North America has a culture which supports male sexual violence and allows for women to internalize an expectation for sexual violence and to believe they deserve what happened to them. This internalized expectation and belief of deservingness work together to decrease effective guardianship from society as a whole (Schwartz et al., 2001). A culture which supports male sexual violence not only decreases effective guardianship, but likely increases motivated offending. When sexually violent crimes are not prosecuted, not heavily punished, and victims are blamed or not believed, this may lead to offenders perceiving a lack of societal concern for this type of crime and increase motivation to commit sexual violence because the perceived risk is less than the perceived reward.

Theoretical models regarding how people view victims and offenders of sexual violence could further aid in the development of a specific prevention approach through bystander intervention theory. The ideal victim is a concept in which society believes some people are more deserving of victim status or than others. People who appear weak and attempt to defend themselves from an attacker are more likely to be given a victim status. Christie (1986) further discusses how the concept of ideal victims creates ideal offenders. Ideal offenders are explicitly dangerous and bring about visions of scary strangers lurking in shadows. Criminologists know that these rigid and explicit definitions of victims and offenders are simply not reality. Those who study sexual violence recognize how these notions affect all aspects of this crime. Rape victims can be blamed and seen as less of a victim because of previous sexual encounters, what they were wearing, or what activities they were engaged in prior to being raped. On the contrary, men of higher status who commit rape are rarely viewed as offenders; or men are not viewed as able to be victims of rape.

Two theories which hail from psychology which could be beneficial to include in a larger theory of prevention through bystander intervention are self-categorization theory and theories surrounding implicit bias (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; Greenwald & Krieger, 2006). Self-categorization theory looks at how people classify themselves in relation to different groups. Levine, Cassidy, Brazier, and Reicher (2002) found bystanders are more likely to intervene when victims are in-group members. They argued that recognition of common group membership can increase helping. Further empirical support for self-categorization theory showed categorical relation to others affected whether people viewed situations relevant to themselves (Levine, et al., 2002). If self-categorization can provide some explanation for the bystander intent to intervene with a victim, it may also provide some level of explanation for if a bystander does not intervene because of who the perpetrator is. Examining the relationship between helping behavior and identification with victim or perpetrator “may deepen our understanding of and strengthen our predictions about bystander behavior” (Levine et al., 2002, p. 1461).

Self-categorization theory may be able to provide some explanation for implicit bias. As constructs, implicit bias refers to our individual psychology and self-categorization refers to our collective psychologies. These collective psychologies are what we use to define ourselves relationally to groups similar and different from ourselves (Reynolds, Turner, Haslam, & Ryan, 2001). Combining self-categorization theory with implicit bias may hold a key into unlocking how people identify and how those identifications help or hinder bystander intervention behaviors. The ability to better predict bystander behavior will lead to more specific and tailored training programs. If people are more likely to help those with whom they identify (or, conversely, less likely to help those with whom they do not identify), program developers must figure out a way to address this subconscious identification with others. This could potentially be done by strongly encouraging situational awareness, using examples to increase empathy for all types of victims, or even including implicit association tests and education about how self-categorization may affect decision-making during bystander training to increase participant awareness of this possibility.

Theories about implicit bias suggest that people do not always have control over their perceptions, judgments, and impressions (Greenwald & Krieger, 2006). Implicit biases are based on discriminatory attitudes or stereotypes. Empirical testing of theories of implicit bias comes from the Implicit Association Test (IAT) and other computer-generated tests. Analysis of results from the IAT show that two-thirds of people show preference for whichever group is the advantaged group for that test (i.e., black vs. white or young vs. old; Greenwald & Krieger, 2006). Interestingly, those groups which are typically disadvantaged mirror these preferences, often preferring the dominant group and not the group to which they belong (Jolls & Sunstein, 2006).

Implicit bias and ideal victim/offender bias may play a role in whether someone notices a situation as worthy of intervention or classifies a situation as an emergency which requires them to act in some way. If, due to some subconscious bias, whether that be related to race, gender, or what a victim or perpetrator is supposed to look/act like (or a combination of these), a bystander does not perceive a person to be a victim, they will not take any action to intervene. Implicit bias is believed to be automatic, such that people have no time to engage in advance cognitive processes in situations of quick decision-making (Jolls & Sunstein, 2006). This ties in with the theoretical argument that the literature and prevention efforts neglect how racism, gender, and violence are linked.

Using theories from multiple disciplines to assess barriers behind intervention behavior is a big step in the right direction towards improving bystander intervention training. Researchers and prevention educators must continue to examine the varying reasons why people do and do not intervene with a specific focus on the role of the intersectionality of race, class, gender identity, ability, and sexual orientation. Identifying how our bias and self-categorization into varying groups affects our perceptions of other people and the role this plays in whether we believe someone to be a victim or an offender will be tantamount in prevention programs which encourage prosocial bystander behavior in situations of sexual violence.

For example, practitioners need to be aware of how intersecting roles affect not only intervention behavior but also the earliest stages in the intervention decision-making process. Men are rarely viewed as potential victims; white women are more likely to be viewed as victims than women of color; women of color are more likely to blamed for their victimization (Katz, Merrilees, Hoxmeier, & Motisi, 2017). Addressing these issues and encouraging training participants to recognize their own internal bias and how bias may contribute to their own intervention decision-making should improve the efficacy of bystander intervention sexual assault prevention programs.