1 Introduction
Unlike other crimes against the person, such as a physical assault or robbery, sexual assault is an ambiguous category because it involves a socially normative activity, sex, in a cultural climate where movies, television, social media, advertisements, video games and pornography regularly depict consensual sexual encounters that involve coercion or force, often reinforcing male aggression and female passivity (Heyes, 2016; Oliver, 2016).
For example, when I searched for ‘forced sex’ on youtube.com in September 2017, I found over two million videos that matched my search, some more graphic than others. Another phenomenon of the internet culture is the emergence of creepshot websites1 which are easily accessible and contain thousands of photographs of women’s body parts, unbeknownst to the women in the photos.2
In a cultural climate where forced or coerced sex is not always perceived as a criminal offence, messages via traditional and social media that sexually objectify women and girls and/or perpetuate rape myths are involved in maintaining rape-supportive beliefs (Ward, Merriwether, & Caruthers, 2006). Indeed, rape myths and sexual objectification are common in various media, gaming outlets, and advertisements (Driesmans, Vandenbosch, & Eggermont, 2015; Edwards, Turchik, Dardis, Reynolds, & Gidycz, 2011; Kahlor & Eastin, 2011; Stankiewicz & Rosselli, 2008) and have been shown to promote rape-supportive attitudes in newspaper readers (Franiuk, Seefelt, & Vandello, 2008), while Ferris, Smith, Greenberg, and Smith (2007: 500) reported that women are referred to as sexual objects almost six times per hour in reality dating shows. Overall, Oliver (2015: 9) considers that ‘contemporary mainstream youth culture values lack of consent’, with women conceived of as ‘game’ for hunting (Oliver, 2016), literally or figuratively. For many, sexual objectification is an everyday experience as Swim, Hyers, Cohen, and Ferguson (2001: 36) showed—diaries kept by 40 female college students for a fortnight revealed that 65% experienced at least one gender-based incident per week (‘comments or behaviours that reflected … traditional gender role prejudice and stereotyping’) while 23% experienced or witnessed sexual objectification.
There has never been any shortage of women getting raped in popular culture, but it seems to have reached a peak recently, from Broadchurch to Game of Thrones. … [T]he casual violence of rape in Game of Thrones becomes ever more ultra, and has such a playful quality that one almost forgets that there’s anything to object to. … What is it doing to us as viewers[?] … [I]f rape on telly is culturally necessary, we need to ask why. (Williams, 2017, n.p.)
when [sexually] objectified, women are treated as bodies that exist for the use or consumption of others, stripped of their individuality and personality … When this instrumental use of another person … involves a denial of human characteristics, it becomes denigrating for the objectified. (Vaes et al., 2011: 774–775)
people reserve full humanness to describe their own group, attributing a somewhat lesser humanity to the out‐group … . … [T]he out‐group is not necessarily completely deprived of humanness …; it suffices that the other is attributed less human qualities relative to the in‐group. (references omitted)
Referring to Haslam (2006), the authors go on to describe the distinction between different types of dehumanisation: ‘“animalistic” (i.e., the denial of uniquely human characteristics that distinguish us from animals such as civility and refinement) and “mechanistic dehumanization” (i.e., the denial of core human nature traits, such as emotionality, openness, and depth)’ (Vaes et al., 2011: 775).
Vaes et al. (2011: 775; 778) were the first to carry out a series of studies to ‘compare male and female objectified and non‐objectified targets and directly measure the extent to which they elicit human versus animal associations’. When heterosexual male and female participants viewed a range of male and female advertising images, classified as either objectified women (sportive women, mother figures and career women), or objectified men (father figures, sportive men and career men), ‘objectified women were the only targets that were less associated with human‐related words’. Interestingly, ‘both male and female participants dehumanized the objectified female targets to the same extent’.
In order to tease out the reasons for this latter finding, Vaes et al. (2011: 780–781) carried out a second study and found that female participants expressed affinity ‘judgments of vulgarity and superficiality’ in relation to images of objectified women compared to male participants who expressed affinity ‘judgments of sexiness’. In fact, the index used to measure affinity ‘moderated male and female participants’ dehumanization of objectified and personalized women’, revealing that the more female participants distanced themselves from sexualised images of women, and judged them to be vulgar and superficial, the more they tended to dehumanize them. By contrast, men’s sexual attraction to objectified female images explained their subsequent dehumanization of them, although ‘male participants who indicated feeling more distant from objectified female targets saw objectified and [non-objectified] women as equally (non) human’.
When male and female participants were primed with sex‐related or neutral words before viewing non-objectified images of attractive and non-attractive women in a third study, the authors reported that ‘good‐looking female targets do not need to be presented in a sexually objectified fashion per se to be dehumanized by men. Instead, an active sex goal in men is a sufficient condition to both objectify and dehumanize good‐looking female targets’ (Vaes et al., 2011: 781).
In the context of sexual assault, it is likely that processes of naming and shaming (such as, ‘slut’, ‘ho’, ‘whore’, and ‘prick-teaser’) result in a woman’s or girl’s dehumanisation, her rights to bodily integrity and justify her violation, as the phenomenon of gang rape of unconscious young women so graphically demonstrates, as discussed below.
From a theoretical perspective, the hunting of women and girls for sexual conquest might be associated with war-prone societies. Because ‘rape is endemic during war’, Zurbriggen (2010: 538) argues that the needs of societies involved in war include a type of masculinity that produces effective soldiers: toughness, aggression, restricted emotional display and the ability to dehumanise the enemy. Such dehumanisation is widely perpetuated through pornography, the use of which has been shown to be positively correlated with RMA, sexually aggressive behaviours, less progressive gender role attitudes and less empathy for rape victims (Allen, Emmers, Gebhardt, & Giery, 1995; Linz, Donnerstein, & Penrod, 1988; Malamuth, Addison, & Koss, 2000; Milburn, Mather, & Conrad, 2000; Peter & Valkenburg, 2009; Wright, Tokunaga, & Kraus, 2015; Zillmann & Bryant, 1982).
the valorization of non-consensual sex has reached the extreme where sex with unconscious girls … has become a goal of some boys and men. … In 2010 at Yale, fraternity brothers marched around the freshman dorms chanting, ‘No means yes, yes means anal’. (Oliver, 2015: 4)
Young men’s attitudes towards consent are formed by exposure to pornography … in which rape victims are depicted as enjoying sexual assault. … They have watched enough pornography to be convinced by the fantasy that whatever a woman says, and whether or not she is conscious, she enjoys it. For, within the pornutopia, the fantasy is that women enjoy violent sex, even abuse, conscious or not. (Oliver, 2015: 2; 3; see also Oliver, 2016)
Surveillance video showed Brandon Vandenburg helping to carry [his girlfriend] into his dorm room, where testimony showed he handed out condoms to three of his teammates and egged them on. The 23-year-old former player also took cellphone videos of the June 2013 dorm room rape and emailed them to friends in California while the crime was ongoing. (New York Post, 4 November 2016)3
Surveillance videos depicted the victim being dragged down a hallway, and dropped more than once while posted videos revealed that she was raped several times, including sodomy with a water bottle. None of the men who received Vandenburg’s videos of the rape reported it to the police.
In November 2016, Harvard University suspended its men’s soccer team from competing in the remainder of the soccer season after the discovery of a nine-page document which ranked the women soccer players according to their sex appeal, with vulgar comments being attached to their photos, including nicknames and sexual positions, such as ‘missionary’, ‘Doggy style’, ‘The Triple Lindy’ and ‘cowgirl’ (New York Post, 3 November 2016).4
We live in a privileged, advantaged society where women have access to education as a matter of right … . And yet girls and women are still subject to abuse or assault or violence of every kind. … Sexual assault is not a story of alcohol consumption or revealing fashions. Nor is it a story of bravado. It is a story of violence and crime and must be afforded its due gravity. (Allum, 2017)
But what causes educated, young men to film rape or rank women according to their sex appeal, or to chant, ‘“My name is Jack, I’m a necrophiliac, I fuck dead girls”’ or ‘to incapacitate “hot” girls using punch spiked with the drug Rohypnol’ (Oliver, 2015: 5)? Indeed, RMA is high among young groups of men who make up fraternity clubs and sporting groups (Bleecker & Murnen, 2005; Forbes, Adams-Curtis, Pakalka, & White, 2006; Foubert, Brosi, & Bannon, 2011; Hayes, Abbott, & Cook, 2016; Lutz-Zois, Moler, & Brown, 2015; Murnen & Kohlman, 2007).
Perhaps those fantasies are ultimately about accruing cultural power through the dehumanisation of women and girls. The extensive literature on the endorsement of men’s rape-supportive beliefs and their interest in, or propensity for, committing sexual assault (Abbey & McAuslan, 2004; Edwards, Bradshaw, & Hinsz, 2014; Lisak & Miller, 2002; Lutz-Zois et al., 2015) suggests that men who rape see their victims as objects or property to be used not only for their sexual pleasure but also for the power gained through the processes of dehumanisation.
Oliver (2015: 6) argues that ‘[r]aping an unconscious woman is the ultimate power trip, proving absolute dominance over another human being’ who has been utterly dehumanised during the process of rendering her unconscious and violating her. Nonetheless, that process of dehumanisation begins with a culture that tolerates, supports or condones rape-supportive beliefs and activities.
If ‘[p]ornography has penetrated mainstream culture … [into] our everyday lives’; if creepshot photos of women’s body parts and videos of gang rape are now circulated widely on social media so that ‘rape has become a spectator sport’ and ‘a form of public entertainment’ (Oliver, 2015: 5), then the criminal justice system is prosecuting rape in a rape-supportive culture with few procedures to guard against jurors’ adherence to rape-supportive beliefs.
Do the dual processes of sexual objectification and dehumanisation take place within the sexual assault trial, when rape myths are used to characterise a complainant’s behaviour before, during and after her alleged violation? Arguably, rape myths serve a particular function in a sexual assault trial—it is perhaps too simplistic to say that they distinguish ‘real’ rape from fabricated allegations, or distinguish stranger rape from acquaintance rape (McKimmie, Masser, & Bongiorno, 2014) since non-stranger rapes can and do result in convictions. Specifically, rape myths serve as a measure of the character or moral worth of the complainant at the same time as they constitute the means by which the female body is sexed according to the cultural values associated with femaleness (Cossins, 2015).
- i.
defence counsel and judges to objectify complainants so that they will be less likely to be associated with human-like qualities;
- ii.
female jurors to distance themselves from an objectified complainant (judging her to be vulgar and superficial); and
- iii.
male jurors to sexually objectify the complainant and judge her to be less than human.
Thus, rape myths within the sexual assault trial are the means by which complainants are dehumanised, thus confirming their low moral worthiness.
Rape law reformers have to deal with some serious questions, including why defence counsel are permitted to construct a narrative based on cultural fantasies about women and girls who dream of rough sex, or the beliefs that women and girls invite rape because of the clothing they wear or because they drink or take drugs or get into a man’s car or flirt. The cultural message that is sent via these beliefs is that once a woman or girl ceases to police her own body and behaviours then it is culturally acceptable for her to be sexually assaulted.
In fact, the criminal justice system is an institutional site which has, historically, encouraged beliefs that protect men from punishment for sexual assault as a result of judicial warnings about the ‘unreliable’ evidence of women and children; the historical requirement for corroboration of the evidence of sexual assault complainants; the historical relevance of a woman or girl’s sexual history/reputation to the issue of consent, and the largely uncontrolled cross-examination of complainants about their sexual histories, sexual behaviour, drug and alcohol consumption, delay in complaint, mental health history and so on. In other words, further evidence for the existence of a rape culture arises when the criminal justice system reinforces rape myths through its processes and procedures.
Since rape myths justify sexual assault, minimise its harms and assign responsibility for the sexual violation to the victim herself, the denial of the validity of a complainant’s evidence in a sexual assault trial through the above methods reinforces any rape-supportive beliefs held by jurors which, in turn, has the potential to negatively impact their interpretation of the trial evidence, their assessments of complainant and defendant credibility, the assignment of blame and responsibility and ultimately their verdicts.
In summary, a rape culture exists where the sexual objectification of women and girls has become a widespread, endemic cultural phenomenon which leads to inaccurate beliefs about women’s and girls’ rights to bodily integrity and beliefs that permit male sexual access to their bodies irrespective of whether consent has been given. Together these beliefs are known as ‘rape myths’.
2 Rape Myth Acceptance (RMA) Studies
Traditional gender role beliefs define how female bodies are expected to behave in that the ‘degree to which an observer adheres to traditional gender role beliefs is … a key factor when assigning levels of responsibility blame’ to a victim in a rape scenario (Grubb & Turner, 2012: 447). It can also be expected that those with high levels of RMA will endorse the view that women who do not behave according to these gender role expectations invite male sexual aggression and are responsible for it.
In fact, RMA studies support but do not prove that English-speaking, Western countries are characterised by a ‘rape culture’ which has been defined as ‘a complex of beliefs that encourages male sexual aggression and supports violence against women’ (Buchwald, Fletcher, & Roth, 1993: vii). But questions remain: why is rape ‘the logical and psychological extension of a dominant-submissive, competitive, sex role stereotyped culture’ (Burt, 1980: 229)? Where does CSA fit into a rape culture? If some people consider that the use of drugs/alcohol or being sexually active is relevant to a complainant’s credibility, the question is why?
suggest that information about the victim affects rape attributions in terms of a ‘main effect’, i.e. it shapes perceptions irrespective of other variables. Several studies, however, provide evidence of an ‘interaction effect’, indicating that information about the victim affects some perceivers more than others. (Temkin & Krahé, 2008: 46)
Some of these misconceptions amount to long-standing, historical myths about the propensity of ‘immoral’ or ‘promiscuous’ women and girls to lie about being sexually assaulted. The question is, what impact do these misconceptions and myths have on jury decision-making? Are jurors able to put aside these beliefs when carrying out the important task of delivering a verdict in a sexual assault trial?
The summary in Chapter 4 of the influence of alcohol and drug consumption on perceptions of victim and perpetrator blame raises the broader issue of RMA in the general population from which jurors are drawn.
2.1 Prevalence of Rape Myths: General Population Surveys
In 2013, the Department of Health in Victoria, Australia conducted a large, cross-sectional community survey of Australians aged 16 years and older which involved more than 17,500 telephone interviews on a range of topics relating to violence against women (VicHealth, 2014). The survey followed two previous ones that had been conducted in 1995 and 2009. For example, in response to the statement, ‘Women are more likely to be raped by someone they know than a stranger’, a majority of respondents agreed in all three surveys: 76% (1995); 70% (2009) and 64% (2013). However, since 1995, there has been a statistically significant decrease in agreement that women are less likely to be sexually assaulted by a stranger (VicHealth, 2014: 10–11). This contrasts with recorded crime statistics for NSW that show that sexual assault victims are 6.4 times more likely to be sexually assaulted by someone known to them than a stranger (ABS, 2019a, Table 11).
As well, there was a statistically significant increase in agreement that ‘Rape results from men not able to control their need for sex’ between 2009 (35%) and 2013 (43%), although very little agreement that ‘A man is less responsible for rape if drunk/affected by drugs at the time’ (8%, 2009; 9%, 2013) (VicHealth, 2014: 14).5
Although a majority of survey respondents agreed that ‘Women rarely make false claims of rape’, a significant minority disagreed with this statement, a finding that was constant from 1995 to 2013 (41%, 1995; 40%, 2009; 41%, 2013).6 When the 2013 respondents were faced with the statement, ‘A lot of times women who say they were raped led the man on and later had regrets’, 38% agreed while only 10% agreed that ‘If a woman doesn’t physically resist— even if protesting verbally— then it isn’t really rape’. As well, 12% agreed that ‘If a woman goes to a room alone with a man at a party, it is her fault if she is raped’.7 Almost one-fifth of respondents agreed that ‘If a woman is raped while drunk/affected by drugs she is at least partly responsible’ (18%, 2009; 19%, 2013), while slightly fewer proportions of respondents agreed that ‘Women often say “no” when they mean “yes”’ (18%, 1995; 14%, 2009; 16%, 2013).
- i.
men;
- ii.
those aged 16–24 years, particularly males;
- iii.
people born overseas in non-English speaking countries.
Indeed, those ‘with a high level of support for gender equality have a higher level of understanding of violence against women and are less inclined to endorse violence-supportive attitudes’, leading to the supposition that the influence of gender and age may be due to ‘the rise of “raunch culture”, a greater emphasis on individuals looking after themselves and a belief that gender equality has been achieved’ (VicHealth, 2014: 20).
A third (33%) of respondents thought it is not ‘usually rape if a woman is pressured into having sex but there is no physical violence’;
A third of men and 21% of women believed if a woman flirted on a date it generally would not be rape, even if she did not consent to sex;
Almost a quarter (24%) did not believe that, in most cases, sex without consent in long-term relationships is rape;
Respondents over 65 had the most troubling attitudes to rape, with a third not believing that non-consensual sex in marriage or a relationship was rape, compared with 16% of people aged 16–24;
11% of respondents believed that the more sexual partners a woman has, the less harm she will experience from rape;
40% thought it is never or usually not rape to remove a condom without a partner’s consent;
11% were unsure or believed it is not usually not rape to have sex with a woman who is asleep or too drunk to consent.
2.2 Rape Myth Acceptance (RMA) Studies
- i.
Gender: men endorsed RMA at a higher rate than women;
- ii.
Race/ethnicity: Whites had lower RMA endorsement than those of other races/ethnicities;
- iii.
Education: those with higher education levels had lower levels of RMA endorsement (Suarez & Gadalla, 2010: 2019).
In terms of sociocultural attitudes, ‘[h]igher levels of RMA were strongly associated with higher levels of other oppressive beliefs’, including ageism, classism, racism, homophobia and religious intolerance. As well, higher endorsement of RMA was significantly more likely in those with ‘oppressive and adversarial attitudes against women’, such as acceptance of interpersonal violence and acceptance of rape and victim-blaming attitudes (Suarez & Gadalla, 2010: 2020).
In terms of behavioural/situational factors, there was a significant relationship between sexual aggression and RMA, as there was with elite athlete status, being a playboy and using degrading images of women (Suarez & Gadalla, 2010: 2023–2024).
Overall, this meta-analysis confirms that RMA continues to be as prevalent, at least in North America, as it was in 1990s (see Lonsway & Fitzgerald, 1994). Given that the internet and social media sites are borderless and the extent to which North American culture permeates other English-speaking, Western countries, this meta-analysis also has implications for the criminal justice systems in Britain and Australia.
Studies show that RMA is one of the most consistent predictors of victim blame in sexual assault scenarios (Grubb & Turner, 2012; Schuller & Wall, 1998; Suarez & Gadalla, 2010; Temkin & Krahé, 2008; Wenger & Bornstein, 2006). Mock jurors with high RMA scores are less likely to perceive the defendant to be guilty, compared to their low RMA-scoring counterparts (Stewart & Jacquin, 2010).
The attribution of blame to rape victims is considered to be influenced by an individual’s particular cognitive and emotional biases which are based on ‘widely shared, normative perceptions of role-conforming behaviour of men and women that restrict the range of what is considered rape and provide a basis for assigning responsibility to victims of rape’ (Temkin & Krahé, 2008: 48).
These biases influence an individual’s perception and interpretation of the context in which the alleged sexual assault occurred, as well as the victim’s and defendant’s credibility. Known as attribution theory (Weiner, 1980, 1985), this theory proposes that if individuals are able to attribute a victim’s situation to a cause within the victim’s control, this reduces sympathy for, and increases anger and blame towards, the victim. This means that ‘sympathy and anger mediate the relationship between controllability and helping’ (Sperry & Siegel, 2013: 17), although this process is influenced by an individual’s perceived similarity to, or empathy with, the victim. For example, Sperry and Siegel (2013: 21) found that sympathy for a victim ‘mediated the relationships between [victim] responsibility and helping, credibility, and to a lesser extent verdict’.
Indeed, victim sympathy may account for the documented differences between men and women in terms of levels of RMA.
commonly held beliefs about what a typical rape situation looks like which influence the decision-making process and guilty assessment of lay people as well as jurors. (van der Bruggen & Grubb, 2014: 524)
As ‘oversimplified and rigid cognitive schemas … [they] supply a structure around which information provided in a trial can be organized’ (McKimmie et al., 2014: 2275; references omitted). They explain the degree of blame attributed to complainants compared to defendants in various studies that have investigated the effect of complainant characteristics and behaviours using hypothetical rape scenarios.
Research has consistently shown that mock jurors who display relatively higher levels of RMA are significantly more likely to blame rape victims and less likely to blame offenders compared to those who display relatively lower levels (Basow & Minieri, 2011; Cohn, Dupuis, & Brown, 2009; Hammond, Berry, & Rodriguez, 2011; Hockett, Saucier, Hoffman, & Craig, 2009; Temkin & Krahé, 2008). While rape myths may vary across time and place, ‘they consistently follow a pattern whereby, they blame the victim for their rape, express a disbelief in claims of rape, exonerate the perpetrator and allude that only certain types of women are raped’ (Grubb & Turner, 2012: 445; emphasis in original).
[s]uch relationships display the important role that pre-trial bias appears to have upon juror decision making, with rape attitudes shown to be directly associated with juror beliefs in a defendant’s account of an alleged rape, though unsurprisingly, not with that of the complainant. Such findings directly support those reported in prior research in that, greater acceptance of sexually aggressive myths appears to reduce a juror’s propensity to believe a rape complainant’s testimony. (Willmott et al., 2018: ibid., 33)
Thus, complainant believability was significantly positively associated with guilty verdicts while defendant believability was significantly negatively associated with guilty verdict decisions (ibid.).
There are, however, inconsistencies in the definitions of rape myths and in the methodologies used to measure RMA in a certain cohort (Suarez & Gadalla, 2010: 2012), such as community members, university students and jurors. The Rape Myth Acceptance Scale (RMAS) developed by Burt (1980) has been used in several studies despite its documented limitations (Buhi, 2005; Lonsway & Fitzgerald, 1994, 1995) and is the most commonly used tool for measuring RMA (Suarez & Gadalla, 2010). In order to address the limitations of the RMAS, Payne et al. (1999) developed and validated the Illinois Rape Myth Acceptance Scale (IRMAS) as a standard measure of rape myth acceptance in a particular population, although other psychometric tools also exist for measuring rape myths (see Grubb & Turner, 2012 for a summary).
Recent rape myth studies categorised according to the seven IRMASa categories
Categories of rape myths | Agreement* (%) | Studies and sample type |
|---|---|---|
1. ‘She asked for it’ (e.g. drunk) | ||
If a woman is raped while she is drunk, she is at least somewhat responsible for letting things get out of control | 41 | Aronowitz, Lambert, and Davidoff (2012) Sample: 237 college students |
20 | Vandiver and Dupalo (2012) Sample: 585 students, Texas State University | |
36.0 (M) 32.3 (F) | Carroll, Rosenstein, Foubert, Clark, and Korenman (2016) Sample: students from the US Military Academy (1169); the US Naval Academy (1916) & a mid-western, public university (393) | |
11.7 (M) 20.6 (F) | Sleath and Bull (2012)# Sample: 123 police officers from two UK Police Forces aged 27–54 years | |
A woman who “teases” men deserves anything that might happen | 22.9 (M) 15.7 (F) | Bendixen, Hendriksen, and Nøstdahl (2014) Random sample: 475 Norwegians, aged 20–60 years |
3.0 | Vandiver and Dupalo (2012) | |
When a woman allows petting to get to a certain point, she is implicitly agreeing to have sex | 18.3 (M) 13.3 (F) | Bendixen et al. (2014) |
If a woman makes out with a man, it is okay for him to push for sex | 63.0 | Aronowitz et al. (2012) |
When women are raped, it’s often because the way they said “no” was ambiguous | 6.7 (M) 3.1 (F) | Bendixen et al. (2014) |
14.2 (M) 8.6 (F) | Carroll et al. (2016) | |
5.0 (M) 7.9 (F) | Sleath and Bull (2012) | |
8.0 | Vandiver and Dupalo (2012) | |
When a woman is raped, she usually did something careless to put herself in that situation | 5.0 (M) 2.1 (F) | Bendixen et al. (2014) |
A woman who dresses in skimpy clothes should not be surprised if a man tries to force her to have sex | 22.2 (M) 17.8 (F) | Carroll et al. (2016) |
1.7 (M) 3.2 (F) | Sleath and Bull (2012) | |
21 | Vandiver and Dupalo (2012)b | |
When women go around wearing low-cut tops or short skirts, they’re just asking for trouble | 0.0 (M) 4.8 (F) | Sleath and Bull (2012) |
A woman who “teases” men deserves anything that might happen | 5.0 (M) 2.7 (F) | Carroll et al. (2016) |
1.7 (M) 1.6 (F) | Sleath and Bull (2012) | |
If a woman goes home with a man she doesn’t know, it is her own fault if she is raped | 0 (M) 7.9 (F) | Sleath and Bull (2012) |
When a woman is a sexual tease, eventually she is going to get into trouble | 18.3 (M) 20.6 (F) | Sleath and Bull (2012) |
A woman who goes to the home or apartment of a man on the first date is implying that she wants to have sex | 0.0 (M) 6.4 (F) | Sleath and Bull (2012) |
2. ‘It wasn’t really rape’ (e.g. no physical evidence) | ||
If a woman does not physically fight back, you can’t really say that it was rape | 10.0 (M) 8.2 (F) | Bendixen et al. (2014) |
7.4 (M) 5.7 (F) | Carroll et al. (2016) | |
0.0 (M) 1.6 (F) | Sleath and Bull (2012) | |
7.0 | Vandiver and Dupalo (2012) | |
A rape didn’t happen if the women has no bruises or marks | 5.0 (M) 3.4 (F) | Bendixen et al. (2014) |
0 (M) 1.6 (F) | Sleath and Bull (2012) | |
If the rapist doesn’t have a weapon, you really can’t call it rape | 0.7 (M) 0.4 (F) | Carroll et al. (2016) |
0 (M) 1.6 (F) | Sleath and Bull (2012) | |
1.0 | Vandiver and Dupalo (2012) | |
If a woman doesn’t physically resist sex—even when protesting verbally—it really can’t be considered rape | 3.3 (M) 1.6 (F) | Sleath and Bull (2012) |
If a woman claims to have been raped but has no bruises or scrapes, she probably shouldn’t be taken too seriously | 0 (M) 3.2 (F) | Sleath and Bull (2012) |
If someone came to me and claimed they were raped, my first reaction would be to not believe them | 5.0 | Vandiver and Dupalo (2012) |
3. ‘He didn’t mean to’ (e.g. he was sexually aroused) | ||
Men do not usually intend to force sex on women, but sometimes they get too sexually carried away | 13.3 (M) 13.8 (F) | Bendixen et al. (2014) |
31.8 (M) 18.6 (F) | Carroll et al. (2016) | |
16.7 (M) 9.5 (F) | Sleath and Bull (2012) | |
28.0 | Vandiver and Dupalo (2012) | |
Rape happens when a man’s sex drive gets out of control | 33.1 (M) 23.2 (F) | Carroll et al. (2016) |
8.3 (M) 4.8 (F) | Sleath and Bull (2012) | |
28.0 | Vandiver and Dupalo (2012) | |
When men rape, it is because of their strong desire for sex | 21.7 (M) 15.9 (F) | Sleath and Bull (2012) |
Rapists are usually sexually frustrated individuals | 11.7 (M) 12.7 (F) | Sleath and Bull (2012) |
When a man is very sexually aroused, he may not even realize that the woman is resisting | 5.0 (M) 14.3 (F) | Sleath and Bull (2012) |
4. ‘She wanted it’ (e.g. secret desire to be raped) | ||
Forced sex is a turn on | 23.0 | Aronowitz et al. (2012) |
Although most women wouldn’t admit it, they generally find being physically forced into sex a real “turn on” | 9.1 (M) 8.9 (F) | Carroll et al. (2016) |
0.0 (M) 3.2 (F) | Sleath and Bull (2012) | |
Although most women wouldn’t admit it, they generally like being physically forced to have sex | 8.0 | Vandiver and Dupalo (2012) |
Many women find being forced to have sex very arousing | 1.7 (M) 1.6 (F) | Sleath and Bull (2012) |
Even though the woman may call it rape, she probably enjoyed it | 8.3 (M) 5.2 (F) | Bendixen et al. (2014) |
Many women secretly desire to be raped | 2.0 (M) 2.6 (F) | Carroll et al. (2016) |
0 (M) 1.6 (F) | Sleath and Bull (2012) | |
2.0 | Vandiver and Dupalo (2012) | |
Some women prefer to have sex forced on them so they don’t have to feel guilty about it | 0.0 (M) 1.6 (F) | Sleath and Bull (2012) |
Many women actually enjoy sex after the guy uses a little force | 8.3 (M) 6.4 (F) | Sleath and Bull (2012) |
5. ‘She lied’ | ||
Rape accusations are often used as a way of getting back at men | 24.4 (M) 14.1 (F) | Carroll et al. (2016) |
31.7 (M) 28.6 (F) | Sleath and Bull (2012) | |
24.0 | Vandiver and Dupalo (2012) | |
A lot of women lead a man on and then they cry rape | 17.9 (M) 12.7 (F) | Carroll et al. (2016) |
18.3 (M) 20.6 (F) | Sleath and Bull (2012) | |
16.0 | Vandiver and Dupalo (2012) | |
Women who are caught having an illicit affair sometimes claim that it was rape | 60.0 (M) 65.1 (F) | Sleath and Bull (2012) |
Many so-called rape victims are actually women who had sex and “changed their minds” afterwards | 40.0 (M) 34.9 (F) | Sleath and Bull (2012) |
A lot of times, women who claim they were raped just have emotional problems | 6.7 (M) 19.1 (F) | Sleath and Bull (2012) |
6. ‘Rape is a trivial event’ (e.g. women tend to exaggerate) | ||
Women tend to exaggerate how much rape affects them | 3.3 (M) 2.8 (F) | Bendixen et al. (2014) |
5.2 (M) 3.9 (F) | Carroll et al. (2016) | |
0.0 (M) 3.2 (F) | Sleath and Bull (2012) | |
4.0 | Vandiver and Dupalo (2012) | |
If a woman is willing to “make out” with a guy, then it’s no big deal if he goes a little further and has sex | 3.4 (M) 2.2 (F) | Carroll et al. (2016) |
1.7 (M) 6.4 (F) | Sleath and Bull (2012) | |
3.0 | Vandiver and Dupalo (2012) | |
Rape isn’t as big a problem as some feminists would like people to believe | 1.7 (M) 4.8 (F) | Sleath and Bull (2012) |
Being raped isn’t as bad as being mugged and beaten | 0.0 (M) 1.6 (F) | Sleath and Bull (2012) |
If a woman isn’t a virgin, then it shouldn’t be a big deal if her date forces her to have sex | 0.0 (M) 1.6 (F) | Sleath and Bull (2012) |
7. ‘Rape is a deviant event’ (e.g. perpetrated by strangers) | ||
It is usually only women who dress suggestively that are raped | 9.2 (M) 4.9 (F) | Carroll et al. (2016) |
1.7 (M) 3.2 (F) | Sleath and Bull (2012) | |
Men from nice middle-class homes almost never rape | 4.3 (M) 2.7 (F) | Carroll et al. (2016) |
1.7 (M) 1.6 (F) | Sleath and Bull (2012) | |
2.0 | Vandiver and Dupalo (2012) | |
Rape is unlikely to happen in the woman’s own familiar neighbourhood | 4.0 (M) 2.2 (F) | Carroll et al. (2016) |
0.0 (M) 1.6 (F) | Sleath and Bull (2012) | |
1.0 | Vandiver and Dupalo (2012) | |
Rape mainly occurs on the “bad” side of town | 0.0 (M) 1.6 (F) | Sleath and Bull (2012) |
Usually, it is only women who do things like hang out in bars and sleep around that are raped | 1.7 (M) 4.8 (F) | Sleath and Bull (2012) |
In reality, women are almost never raped by their boyfriends | 3.3 (M) 1.6 (F) | Sleath and Bull (2012) |
Rape almost never happens in the woman’s own home | 1.7 (M) 4.8 (F) | Sleath and Bull (2012) |
The wide range of sample types of the five studies listed in Table 1 illustrates that RMA varies within different population groups and also between men and women. Overall, one-fifth or more of participants in each study agreed with, ‘If a woman is raped while she is drunk, she is at least somewhat responsible for letting things get out of control’ and ‘Rape accusations are often used as a way of getting back at men’.
- i.
‘If a woman is raped while she is drunk, she is at least somewhat responsible for letting things get out of control’ (11.7% (M); 20.6% (F));
- ii.
‘A woman who dresses in skimpy clothes should not be surprised if a man tries to force to have sex with her’ (1.7% (M); 3.2 (F));
- iii.
‘If a woman does not physically fight back, you can’t really say that it was rape’ (0.0% (M); 1.6% (F));
- iv.
‘Rape happens when a man’s sex drive gets out of control’ (8.3% (M); 4.8% (F)); and
- v.
‘Although most women wouldn’t admit it, they generally find being physically forced into sex a real “turn on”’ (0.0% (M); 3.2% (F)).
no significant effect was found for specialist rape victim training regarding levels of victim blaming. … This lack of effect is troubling, since it would be expected that specialist training to deal with sexual offenses would contain aspects that would address misperceptions about rape victims. (Sleath & Bull, 2012: 661)
Compared to other population groups listed in Table 1, British police officers exhibited the highest levels of RMA in relation to (vi) ‘Rape accusations are often used as a way of getting back at men’ (31.7% (M); 28.6% (F)), and (vii) ‘A lot of women lead a man on and then they cry rape’ (18.3% (M); 20.6% (F)). They also exhibited high levels of RMA in relation to (viii) ‘Women who are caught having an illicit affair sometimes claim that it was rape’ (60.0% (M); 65.1% (F)); and (ix) ‘Many so-called rape victims are actually women who had sex and “changed their minds” afterwards’ (40.0% (M); 34.9% (F)).
Sleath and Bull (2012: 659–660) reported that high levels of RMA among this group of police officers had a significant effect on the extent to which, individually, they blamed the victim, with both male and female offices displaying similar levels of victim blame. Specific types of rape myths were found to predict victim-blaming, that is, those that fell within the categories, ‘She wanted it’ and ‘He didn’t mean to’. Victim-blame was significantly higher when the perpetrator was an acquaintance compared to when he was a stranger.
By contrast, US students (a younger cohort) showed relatively higher levels of agreement with rape myths (i) to (v) and almost as high levels of agreement with myths (vi) and (vii). Generally, the community group of Norwegians displayed relatively low levels of agreement with all rape myths with their highest level of agreement being with ‘A woman who “teases” men deserves anything that might happen’ (5% (M); 2.7% (F)).
2.3 The Factors Influencing Victim Blame
One key question raised by the literature is what RMA scales are really measuring. Some consider RMA to be a measure of hostility towards women (Lonsway & Fitzgerald, 1994), or more broadly a measure of attitudes that deny the existence of rape and CSA (Forbes, Adam-Curtis, & White, 2004; Suarez & Gadalla, 2010; citing Cromer & Freyd, 2007). But what is the social purpose of rape myths about women and children? Before addressing this question, I discuss what is known about the factors influencing victim blame.
The literature reveals that ‘[t]here are a number of variables which have been found to influence the degree’ of victim blame, ‘including perceiver’s beliefs, victim characteristics and situational aspects … which are likely to influence every situation in a unique and unpredictable manner’ (Grubb & Turner, 2012: 444). Because it is ‘vital’ to understand the different types of observer or mock juror characteristics, and the way in which they influence jury decision-making (Grubb & Turner, 2012), the following discussion summarises the latest meta-analyses of RMA studies, as well as the results of a study of a large representative sample of laypeople in the UK.
- 1.
women who tease men;
- 2.
women who allow men to intimately touch them;
- 3.
women’s use of drugs or alcohol;
- 4.
women who dress sexy;
- 5.
women allowing the situation to get out of control; and
- 6.
women who do unsafe things (such as being out alone, hitch-hiking).
They also conducted an online survey of 2176 members of the UK general public (age range, 18–69 years10), using the 16-item Acceptance of Modern Myths about Sexual Aggression Scale (Gerger, Kley, Bohner, & Siebler, 2007), an instrument that is considered to measure subtler stereotypical beliefs about women and rape.
Rape scenarios were varied in terms of the use of force and complainant intoxication, as was the relationship between complainant and defendant (strangers, acquaintances with no prior sexual relationship, and ex-partners with a prior sexual relationship). All scenarios contained a verbal statement of non-consent by the complainant.
Participants’ judgements about defendants and complainants ‘were affected both by participants’ acceptance of rape myths and by information activating stereotypes about rape’ (Temkin & Krahé, 2008: 117). Ratings of defendant liability were directly influenced by the nature of the relationship in that the defendant was perceived as most liable when he was a stranger, compared to an acquaintance and ex-partner. Defendant liability was also greater when he used force than when he took advantage of the complainant’s intoxication (Temkin & Krahé, 2008: 103). In the scenarios where the complainant was intoxicated, defendant liability ratings did not vary according to the type of relationship. However, the more that participants agreed with female precipitation myths and modern rape myths, the more leniency was displayed towards the defendant, generally, but especially when he was an ex-partner.
The complainant was blamed significantly less when the defendant was a stranger and she was also accorded less blame when subject to force than when she was intoxicated. Overall, complainant blame was correlated with rape-supportive attitudes in that the more participants endorsed rape myths, the more blame was attributed to the complainant.
Temkin and Krahé (2008: 107–108) concluded that ‘differences in rape-supportive attitudes’ affected participants’ interpretations of rape scenarios. While, on average, men showed greater acceptance than women of female precipitation beliefs and modern myths about sexual aggression, ‘men and women did not differ substantially in their perceptions of defendant liability’ or in relation to complainant blame. Despite a verbal statement of non-consent being included in the scenarios so that the key element of the crime of sexual intercourse without consent was satisfied, use of heuristics in the form of rape myths influenced participants’ ‘appraisal of the evidence to the disadvantage of the complainant’ (Temkin & Krahé, 2008: 119). In other words, defendant liability and complainant blame were influenced by the relationship between the parties, the amount of force used by the defendant and the degree of intoxication of the complainant.
Temkin and Krahé (2008) concluded that schematic information processing11 is responsible for the different assessments of liability and blame so that the more sexual assault cases deviate from the stranger-rape stereotype, the more likely the case facts will be judged according to pre-existing RMA.
Bieneck and Krahé (2011) carried out a series of mock jury experiments with a group of undergraduate students to determine whether there is a special leniency bias in sexual assault cases compared to robbery cases by measuring victim and perpetrator blame. They found that perpetrator blame was significantly higher in the robbery scenarios compared to the rape scenarios. As expected, more blame was attributed to the rape victim than the robbery victim (Bieneck & Krahé, 2011: 1790).
When the perpetrator exploited the fact the victim was drunk … perceived perpetrator blame decreased in the rape cases but not in the robbery cases. … Similarly, if the victim was too drunk to resist … she was blamed more in the rape cases but not in the robbery cases. (Bieneck & Krahé, 2011: 1791)
These patterns can be seen as a reflection of schematic information processing because neither victim intoxication nor victim-perpetrator relationship are critical features of the legal definition of rape. (Bieneck & Krahé, 2011: 1793)
Because victim-perpetrator relationship and victim intoxication trigger stereotypical thinking about the veracity of a rape allegation, the authors concluded that a ‘special leniency bias’ exists in rape cases since background information about the victim ‘operated differently’ for different crimes (Bieneck & Krahé, 2011: 1794). Perpetrator and victim blame was constant in the robbery cases irrespective of the type of background information whereas perpetrator and victim blame varied as a function of victim intoxication and victim–perpetrator relationship.
It is likely that the reason for this variation was due to the legal issue of lack of consent which is an element of the offence in adult sexual assault cases but not in robbery cases. Thus, it appears that background information, that is, victim behaviour and victim–perpetrator relationship, was used by mock jurors as proxy measures to decide whether lack of consent existed when no other information was available.
2.4 The Independent Effects of Victim and Participant Characteristics on Victim Blaming
Because of the difficulties with differentiating between the independent effects of victim characteristics versus participant characteristics, van der Bruggen and Grubb (2014) carried out a literature review in order to summarise what is known about their separate effects on victim blaming in rape scenarios. These characteristics included gender, sexuality, victim carelessness, degree of resistance and the victim/perpetrator relationship.
In relation to victim resistance, the results have been consistent over a twenty-year period: ‘victims who do not resist tend to be blamed the most’, although non-resistant male victims are blamed more than their female counterparts (van der Bruggen & Grubb, 2014: 526). A number of studies also show that victims are blamed more in date-, acquaintance- and marital-rape scenarios than stranger-rape scenarios (van der Bruggen & Grubb, 2014).
By comparison, the influence of observer/participant characteristics on victim blame is not well known, partly due to the difficulties in accessing representative cohorts of laypeople or jurors. What does emerge from the literature is that male, heterosexual participants are more likely to endorse rape myths and blame rape victims compared to female, heterosexual, and male, homosexual participants (van der Bruggen & Grubb, 2014).
Thus, participant gender is widely accepted as having a significant influence on perceptions of victim blame (see studies cited in van der Bruggen & Grubb, 2014: 527), with men more likely than women to blame rape victims for their victimisation, irrespective of the gender of the victim and the type of rape (stranger, date, acquaintance or marital). If, as it appears, men are more likely to identify and sympathise with the offender and are more likely to judge the victim’s character, this has ‘immense implications for the criminal justice system’ (van der Bruggen & Grubb, 2014: 527) in terms of wrongful acquittals.
Indeed, RMA appears to act as a mediating variable between participant gender and victim blame, thus explaining men’s higher propensity for blaming rape victims (Suarez & Gadalla, 2010) and voting guilty in sexual assault trials (Devine & Caughlin, 2014). The various stereotypes that make up RMA are believed to be used by mock jurors when assessing ambiguous evidence, blame and responsibility (McKimmie et al., 2014) since ‘offense-related and victim-related schemas influence impressions about the blameworthiness of the defendant and complainant’ (Stuart, McKimmie, & Masser, 2016: 2).
As well, participant gender role attitudes or levels of benevolent sexism have been shown to influence RMA, in that the higher the degree of benevolent sexism the more a rape victim is blamed (Grubb & Turner, 2012; van der Bruggen & Grubb, 2014), especially in date-rape scenarios compared to stranger-rape scenarios (Grubb & Turner, 2012). Other studies have found an indirect relationship between stereotypical gender role attitudes and victim blame, with such attitudes ‘act[ing] as mediating variables between observer gender and degree of [victim] blame’ (van der Bruggen & Grubb, 2014: 528).
2.5 Investigating the Independent Effects of Victim Stereotypicality and Gender Stereotypicality
While ‘[t]he exact relationship between gender role conformity, rape myth acceptance and victim blaming is unclear’, Grubb and Turner (2012: 447) suggest that ‘gender role conformity enhances RMA and ergo victim blaming by reinforcing cognitive schema that support the traditional stereotypical notion that rape victims are deserving of their misfortune’. The studies below provide some answers.
When Masser, Lee, and McKimmie (2010) investigated the independent effects of victim stereotypicality (verbalised non-consent, some physical resistance and assisting the police investigation) compared to gender stereotypicality (‘bad’ mother versus ‘good’ mother behaviour), they found that it was only when victims behaved in counter-stereotypical ways that the gender stereotypicality of the complainant affected mock jurors’ perceptions of blame.
In a later study, McKimmie et al. (2014: 2278) investigated the impact of victim stereotypicality independently of gender stereotypicality ‘in the context of stereotypical [stranger-rape] and counter-stereotypical [acquaintance-rape] offenses’. Victim stereotypicality amounted to the victim pushing the perpetrator away, trying to stop the removal of her clothes and cooperation with the police. Counter-stereotypicality involved no resistance and unwillingness to cooperate with police. Like other studies, RMA was a significant predictor of verdict by community participants, with higher endorsement of rape myths predicting acquittals and higher ratings of victim blame. The participants were also more likely to vote guilty and had more positive perceptions of the victim when the offence was stereotypical (stranger-rape) and when the victim was stereotypical (actively resisted).
While victim stereotypicality ‘had no effect or a lesser effect on participants’ perceptions in the stranger rape condition’ (McKimmie et al., 2014: 2287), in the acquaintance-rape condition, a counter-stereotypical victim resulted in more victim blame, less defendant blame and fewer guilty verdicts. While most participants voted guilty in both conditions, not guilty verdicts were rarer in the stranger-rape condition compared to the acquaintance-rape condition. Reasons for guilty verdicts focused on the defendant’s menacing behaviour in the stranger-rape scenario from which participants probably implied that the victim was not consenting, such that the victim’s behaviour had less salience.
suggest that perceivers generally start with the ‘real rape’ crime prototype … in attempting to interpret what happened and how blame should be apportioned … . [T]his classification may motivate different questions to be asked of the evidence, leading to a ‘step-down’ through the hierarchy of schemas and, subsequently, the use of victim stereotypes to evaluate the complainant in counter-stereotypical cases. (McKimmie et al., 2014: 2294)
Ellison and Munro (2013) conducted a qualitative study with 160 jury-eligible mock jurors who were presented with four mini rape trials that included a number of elements common to acquaintance rape cases, and played in real time by professional actors and barristers. The ambiguous facts involved a complainant and defendant who had been in an eight-month relationship which had ended two months previously. During a meeting at the complainant’s home, when they drank coffee, wine and talked, they kissed as the defendant was leaving. Although the defendant tried to initiate sexual intercourse, the complainant asked him to stop and pushed his hands away. The defendant admitted that sexual intercourse had taken place but maintained that it had been consensual. Evidence from a forensic specialist showed that the complainant had suffered bruises and scratches consistent with considerable force but had sustained no internal bruising. The mock jurors were told that this evidence neither confirmed nor contradicted the complainant’s rape allegation.
[n]otwithstanding the fact that our jurors received an extended judicial direction which … emphasised that rape by an acquaintance or intimate partner is no less serious than rape by a stranger, our findings suggest that jurors continued to struggle to convict because of engrained expectations regarding resistance and sexual miscommunication. (Ellison & Munro, 2013: 303)
A number of jurors observed that the complainant had ‘egged on’ or encouraged the defendant by allowing him into her home, offering him a glass of wine, engaging him in friendly conversation over the course of a few hours, kissing him, and generally creating a comfortable ambience … with jurors—male and female—frequently censuring the complainant for giving out ‘mixed messages’ … or … ‘being a prick-tease’. (Ellison & Munro, 2013: 311)
This belief in the responsibility of women to control intimate encounters then allows jurors to determine the ‘reasonableness’ of a defendant’s mistaken belief as to consent, or the reasonableness of the steps he took to make sure there was consent, depending on the jurisdiction (see further, Chapter 7).
As Ellison and Munro (2013: 311; 312) observe, their findings are consonant with other studies that have shown people’s greater preoccupation with the complainant’s conduct rather than the defendant’s, as discussed above, thus ‘testif[ying] to the strength and reach of these socio-sexual “scripts” … [to] afford a more lenient treatment to defendants’. Jurors with these views thus excluded the possibility that a woman could have a social occasion with an ex-partner in the home they used to share, without sex being the inevitable outcome if the man so desired: ‘jurors clearly relied upon a series of expectations regarding “appropriate” forms of behaviour towards ex-partners … [and] reflected on their own experiences with relationship breakdown to determine what (sexual) meaning might reasonably be imposed’, such that the complainant was seen as being far too welcoming, ‘needy, clingy or still in love with the defendant’ while the defendant took the opportunity to ‘“try his luck”’ (Ellison & Munro, 2013: 312–313), with the ability of the complainant to withdraw consent becoming an unreasonable possibility after a consensual kiss.
the onus was typically placed upon the complainant to correct any potential misreading of her behaviour by communicating her lack of consent unequivocally … [T]his stretched to … a de facto requirement, that the complainant physically resist and/or use force where it was clear that her verbal protestations were not being taken on board by a defendant whom many perceived as having diminished deciphering skills on account of his aroused condition. (Ellison & Munro, 2013: 314; emphasis in original)
Jurors compared the complainant’s lack of resistance with that of a woman confronted by a stranger, where the risk of serious violence would leave a woman ‘“petrified”’, ‘“like jelly”’, rather than the less traumatised complainant who, with no history of domestic violence, would be ‘more inclined to “fight tooth and nail”’ (Ellison & Munro, 2013: 314–315).
Although the jurors in this study were presented with evidence from a forensic medical examiner that it is common for rape victims to display no vaginal injury as a result of sexual assault, ‘several jurors continued to voice the (unfounded) counterview that genuine resistance to an unwanted sexual attack would inevitably result in “some kind of tearing” or “inside trauma”’ (Ellison & Munro, 2013: 317–318). Overwhelmingly, jurors believed that there would be evidence of internal damage and the lack of it equated with sexual arousal and, therefore, consent.
Even though the complainant sustained bruises to her chest, wrist and upper thigh, jurors advanced unsubstantiated explanations for her ‘“superficial-ish”’ injuries, with the most common suggestions being that ‘the complainant had self-inflicted the injuries to bolster the credibility of her false rape allegation, that she had sustained them accidentally in … a separate sexual or other physical activity, or that she was simply prone to bruising easily’, that the parties had previously enjoyed ‘“rough”’, ‘“raunchy”’ or violent sex or that the encounter was a particularly passionate one as they ‘“reignit[ed] an old flame”’. The eccentricity of these views suggested they were drawn from ‘(media-informed) scripts about the intensity and urgency of heterosexual “reunion sex”’. A suggestion by some jurors that ‘a woman could freeze during a sexual attack out of fear or shock’ was rejected by others who considered such a response was only plausible in a stranger rape attack where there was a risk of additional violence (Ellison & Munro, 2013: 315).
As a result of their doubts, 80% of the 20 juries returned not guilty verdicts while 71% of jurors in post-deliberation questionnaires said that the previous relationship between the complainant and defendant was an important or a highly important factor in arriving at their verdicts (Ellison & Munro, 2013: 318–322).
However, in an ambiguous fact situation, jurors’ own experiences or media-informed beliefs constitute a limited source of information for assessing the behaviour of two people whom they had never met and creating a narrative in which to understand an alleged sexual assault, such as the assertion that ‘“love makes people do crazy things”’ (Ellison & Munro, 2013: 314). The authors concluded: ‘on the whole, jurors displayed a striking willingness to construct a range of unsupported hypotheses regarding the parties’ prior sexual relationship’, even though no evidence of it had been raised in the simulated trial, although some assumed the defence had been prevented from doing so (Ellison & Munro, 2013: 319; 320).
In summary, these deliberations contain gendered assumptions about women’s and men’s behaviour (clingy, needy complainant, her need for revenge when rejected), media-driven views of heterosexual relations (to account for injuries) and the sympathy displayed to, and/or excuses made for, the defendant who may have merely misunderstood the complainant’s behaviour, and, sexually aroused, could not control himself.
alone will be insufficient to overcome some of the obstacles that render jurors reluctant to convict in such cases. Instead, a thorough interrogation of prevailing expectations regarding socio-(hetero)sexual behaviour and communication is required, alongside renewed efforts to disavow jurors of the assumption that a victim’s ‘normal’ response to rape would be to fight back aggressively, receiving and causing injury (including internal injury) as a result. (Ellison & Munro, 2013: 321)
In Chapters 11 and 12, I discuss the reforms that would be necessary to overcome the obstacles to prosecuting (acquaintance) rape cases.
The key limitation of Ellison and Munro’s study (2013) is that they did not undertake empirical analyses to examine the relationships between jurors’ RMA, misconceptions, credibility determinations and verdicts, nor was any coding data for the deliberations reported. Nor did the authors determine the extent to which educational/judicial guidance changed mock jurors’ misconceptions, or the relationship between mock jurors’ exposure to educational/judicial guidance, their credibility perceptions and verdicts.
rape myths were used extensively in every trial and were a routine way for the defence to undermine prosecution witnesses’ credibility. … Rape myths were used to portray victim/survivor behaviour as either normal or ‘abnormal’. For example, juries were told to evaluate whether the demeanour of victim/survivors was ‘consistent’ with expectations; namely, that they were visibly distressed immediately after the offence, when initially reporting, and at trial. (Smith & Skinner, 2017: 449)
the normality of physical resistance (e.g. ‘Squirming is the word [Victim/Survivor] uses throughout. She doesn’t do anything else … What would you do?’ (Smith & Skinner, 2017: 449));
the abnormality of delayed complaints to police;
revenge as a motive for complaint;
the abnormality of maintaining contact with the accused; and
inconsistencies by a complainant as evidence of dishonesty.
comments made by prosecution barristers and judges aimed to challenge these [myths]. In fact, prosecution barristers resisted stereotypes at some point in every full trial, often by outlining the possible reasons for the victim/survivor’s ‘suspicious’ behaviour. (Smith & Skinner, 2017: 453)
were not particularly complementary about victim/survivors. For example, the T1 prosecution barrister called the three victim/survivors ‘stupid’ and ‘odd’ instead of asserting that their actions are common after sexual victimisation. … [O]ne barrister, who was observed in both prosecution and defence roles, used myth-busters when prosecuting, yet invoked myths when defending. (Smith & Skinner, 2017: 453)
any link between the use of myth-busters and conviction rates. This is not surprising since any challenge to the relevance of myths was strongly undermined by defence barristers and several factors meant they were kept relevant for juries.
[Victim/Survivor] didn’t bite, kick or scream … You don’t have to … but it is good evidence in cases where that does happen and so you may think it’s useful to know that she didn’t bite, kick or scream. (Smith & Skinner, 2017: 455)
2.6 Summary
The above discussion reveals the existence of engrained, cultural stereotypes that influence people’s perceptions of the responses of female sexual assault victims, including the cultural expectations that a truthful victim will act in particular ways and will not have ‘encouraged’ the defendant through her ‘risky’ behaviours. A complainant’s behaviours are, therefore, used as a proxy for determining lack of consent, thus raising the question, what are the implications of victim behaviours that do not match expectations associated with the stranger-rape stereotype?
- i.
men have a need for, and a right to, have sex;
- ii.
it is the responsibility of women to control this need; and
- iii.
women encourage men’s need for sex through their dress and flirtatious behaviours which constitutes implicit consent.
Because the ‘consensual sexual script is generally agreed upon within cultures and across genders’, it is common for defence counsel to adopt this script when challenging a complainant’s account by pointing to stereotypical behaviours that ‘characterize consensual sex’ in order to undermine the complainant’s credibility (McKimmie et al., 2014: 2296), with counter-intuitive behaviours well known to be the focus of defence counsel strategies in court (Cossins, 2009; Smith & Skinner, 2017; Zajac, Gross, & Hayne, 2003; Zajac & Hayne, 2003).
Suarez and Gadalla (2010: 2027) have argued that ‘rape prevention programs must be broadened to incorporate strategies’ that address the beliefs associated with RMA. But how are we to deal with RMA within sexual assault trials to address the use of rape myths, not only by jurors but also by judges and lawyers? How are we to eliminate the acceptance of styles of cross-examination which feed on rape myths and victim blame by turning everyday behaviours by both young men and women, such as dancing, drinking, flirting and staying out late at night, into a narrative that ‘she was asking for it’? How are we to educate jurors about the fact that many of the symptoms of sexual victimisation, such as alcohol and drug use (see Table 1, Chapter 4), are misused by the defence to undermine a complainant’s credibility?
- i.
that stranger-rape represents the model for judging the behaviours of victims who are sexually assaulted by relatives, partners, ex-partners, friends or acquaintances;
- ii.
‘inaccurate portrayal[s] and perception of the number of false rape allegations’ (Grubb & Turner, 2012: 445), thus denying the true prevalence of sexual assault;
- iii.
victims’ beliefs in their own victimisation, their ability to recognise what happened to them as sexual assault and, hence, reporting rates. For example, LeMaire, Oswald, and Russell (2016: 332) found that victims with higher RMA endorsement ‘were less likely to label their past sexual assault experience as rape’;
- iv.
police investigations and prosecutorial decisions (Frohmann, 1991; Spears & Spohn, 2006; Spohn & Holleran, 2001; Temkin & Krahe, 2008) since prosecutors filter out those cases ‘unlikely to result in a conviction because of questions about the victim’s character, … [and] behavior’ (Spears & Spohn, 2006: 501); and
- v.
jury decision-making.
The fact that men are more likely to endorse RMA than women suggests that a jury dominated by men would be more likely to return a not guilty verdict in a sexual assault trial, particularly in cases where the evidence is ambiguous and/or the complainant’s conduct before, during or after the assault is perceived as ‘risky’ or blame-worthy behaviour. Indeed the ‘strong positive association between RMA and sexual aggression and other hostile attitudes’ (Suarez & Gadalla, 2010: 2025) towards women suggests that a complainant’s conduct and credibility will be central to a jury that is dominated by those with hostile attitudes towards women.
3 Interactions Between Offence, Victim and Perpetrator Stereotypicality
It goes without saying that the effect of extra-legal factors on jury decision-making in sexual assault trials is complex so that it is important to understand the relationship between the key aspects of a sexual assault trial: the type of offence (stranger vs acquaintance); the victim’s characteristics and behaviours; and the perpetrator’s characteristics and behaviours. A number of mock jury studies have been able to provide further insights into the interactive effects of various stereotypes on credibility assessments and verdicts.
Schuller et al. (2010) examined the difference between mock jurors’ endorsement of gender stereotypes compared to victim stereotypes in determining the influence of myths and misconceptions on jurors’ decision-making. Based on existing research that suggested that ‘emotionality [of the complainant] is … congruent with the feminine stereotype’, Schuller et al. (2010: 768; 770) explored ‘if, and how, perceptions of emotionality interact with victim and gender stereotypes to influence judgments of sexual assault victims’; that is, whether or not both ‘“good” women and “good” victims should be emotionally expressive’.
In their study, the calm and unemotional victim was perceived as being less representative of both women and rape victims in general, while the complainant who physically and verbally resisted the defendant was perceived to be more like a ‘real’ victim. Compared to the victim who only resisted verbally, the physically resistant complainant’s claim of sexual assault was rated as more believable. Believability was also higher for the emotionally distressed complainant, compared to the calm/controlled victim. However, the influence of emotional distress was only found when the complainant was gender stereotypic, that is, when she was portrayed as a ‘good’ mother compared to the complainant who was portrayed as a ‘bad’ mother. Significantly more blame for the sexual assault was apportioned to the ‘bad’ mother (counter gender stereotypic behaviour) compared to the ‘good’ mother while the validity of the claims made by the ‘bad’ mother and the non-emotional, ‘good’ mother were both perceived to be low.
Schuller et al. (2010) concluded that the validity of a sexual assault claim is viewed through two stereotypical prisms, in that a complainant’s allegation and credibility is measured according to the expectations associated with her gender and with a ‘typical’ victim; that is, a victim who conforms to gender expectations and acts like a ‘typical’ victim (high emotionality and physically resistant) will be viewed more positively. But this raises the question, isn’t the concept of the typical victim, itself, a gendered concept? Has the literature ignored this relationship?
Masser et al. (2010) also manipulated gender and victim stereotypes in order to determine the independent effects of good/bad gender stereotypes versus real/not real victim stereotypes on mock juror attributions of blame in sexual assault scenarios. They found that gender stereotypicality (good mother versus bad mother behaviour) only influenced blame attributions when the complainant acted like a counter-stereotypical victim, that is, by not physically resisting the perpetrator, although it was not clear from their research why victim stereotypes were not considered to be based on gendered stereotypes.
In other words, the concept of a stereotypical victim is not gender neutral because of the cultural climate in which sex offences are prosecuted and the historical origins of victim stereotypes (as discussed above). Thus, blame attributions appear to depend on gendered expectations of female victim behaviour so that the good mother/bad mother stereotype reinforces the concept of stereotypical and counter-stereotypical victim behaviour in particular circumstances.
McKimmie et al. (2014) carried out a further study to examine the independent effects of offence prototypicality and victim stereotypicality on mock jurors’ perceptions in simulated sexual assault scenarios which involved acquaintance and stranger rape scenarios. The study was based on the concept of cognitive schemas (stereotypes and crime prototypes) for interpreting case evidence and witness credibility.
Generally speaking, it appears that the less a complainant matches the qualities of a stereotypical rape victim, the more she will be blamed for the defendant’s sexual behaviour (Masser et al., 2010; McKimmie et al., 2014) which supports the view that victim stereotypicality is a gendered concept. However, because there are several variables in each different sexual assault case, such as the type of relationship between complainant and defendant, degree of resistance, and the social context in which the sexual act occurred, understanding the precise schemas or stereotypes that operate, and the point at which they operate in a trial, is ‘critically important … in identifying how to understand and overcome their potentially negative influence’ (McKimmie et al., 2014: 2277). It is also important to understand how offence prototypes interact with victim and gender stereotypes.
- i.
the complainant physically resisted the removal of her clothing and cooperated with police;
- ii.
the complainant did not physically resist (only asking the defendant to stop) and did not fully cooperate with police.
The authors found that RMA was a significant predictor on outcomes, ‘with stronger endorsement of rape myths associated with a lower likelihood of a guilty verdict’, as well as lower ratings of complainant credibility and defendant blame, and higher ratings of victim blame (McKimmie et al., 2014: 2283; 2285). However, guilty verdicts were more likely ‘when the offense was stereotypic [stranger rape] and also when the victim was stereotypic [physically resistant]’. In this context, ‘[p]articipants had more positive perceptions of the complainant and the prosecution case and less positive evaluations of the defendant and the defense case’ (McKimmie et al., 2014: 2287).
It was only in the acquaintance rape scenario that victim stereotypicality influenced participants’ views of the complainant which were more positive than when the victim’s behaviour was counter-stereotypical. With the latter behaviour, participants were less likely to find the defendant guilty and the complainant was perceived as less credible and more to blame.
Comments provided by participants who read the acquaintance rape/counter-stereotypical scenario were even more illuminating, such as ‘“[c]omplainant’s decisions or behaviour on the night put her at risk of rape”’ which was the most frequent reason for a not guilty verdict (McKimmie et al., 2014: 2289; 2293), confirming the views expressed in other studies that women’s behaviours incite rape. This study shows that mock jurors focused on the complainant’s behaviour when the defendant was an acquaintance, but focused on the defendant’s behaviour in the stranger rape situation, such that a woman’s ‘risky’ behaviour is used to measure her degree of responsibility, indicating that jurors judge victim behaviour through the lens of gender expectations.
Thus, perceptions of sexual assault are gendered and contextual so that when the parties are known to each other, these perceptions appear to depend on the female complainant’s behaviour which will be measured through the lens of stereotypical expectations about female victims’ behaviours such as physical resistance, and not leading a man on. Thus, the main comments made by participants in the acquaintance rape scenario ‘focused on establishing whether consent to the sexual act was given or implied’ (McKimmie et al., 2014: 2294), results which assist in considering reform options for the vast majority of sexual assaults, that is, where the defendant is known to the complainant.
RMA functions in terms of a cognitive schema that actively guides information search when individuals are confronted with a rape case, resulting in a focus away from the perpetrator and toward the victim, whose characteristics and behavior are at the heart of most rape myths. (ibid., 2332)
More recently, Bongiorno, McKimmie, and Masser (2016) found that when presented with counter-stereotypic victim behaviour, mock jurors who were culturally similar to perpetrators considered them to be less guilty and less deserving of punishment, compared to those who were culturally dissimilar. This increased leniency shown towards culturally similar perpetrators was explained by increases in victim blame and decreases in perpetrator blame.
the ‘complainant’s decisions or behaviour on the night put her at risk of rape’ was the most frequently given reason for [a not guilty] verdict. This theme was most dominant in the acquaintance rape/counter-stereotypical victim condition.
This suggests that in acquaintance rape cases, jurors begin with an assessment of the female complainant’s behaviour in order to determine whether her behaviour amounts to implied consent.
Finally, Stuart et al. (2016: 9) found that when they manipulated perpetrator and victim characteristics, ‘typical rape victims’ were seen by mock jurors as significantly more weak or vulnerable, frightened, helpless, scared, sad and depressed, ashamed and embarrassed and to have lower self-esteem than women in general. Typical rapists were perceived to be significantly more aggressive, angry towards women, insensitive, impulsive and significantly less socially skilled than men in general. When they examined the influence of offence, victim and perpetrator stereotypes on perceptions of victim and defendant behaviour, they found that mock jurors perceived the victim more positively and the perpetrator more negatively when both the offence and victim were described as stereotypical (stranger rape, aggressive force, victim who physically resisted).
However, perpetrator stereotypicality had no effect on perceptions of the victim’s behaviour while the effect of victim stereotypicality did not differ as a function of offence stereotypicality. Instead, victim stereotypicality predicted how the perpetrator was perceived: the perpetrator was perceived as more stereotypical when the victim was stereotypical than when she was counter-stereotypical, thus confirming the finding from other studies that a female victim’s behaviour influences perceptions of responsibility and that victim stereotypes are gendered concepts.
Because the authors also found that victim stereotypicality affected perceptions of offence stereotypicality, this suggests that gendered assumptions about women are the default stereotype that guides mock jurors’ perceptions of blame, responsibility and consent, including perceptions of the defendant’s behaviour. Since particular victim behaviours are not elements of a sexual assault offence, the complainant’s behaviour should not logically affect perceptions of a perpetrator’s conduct. Nonetheless, her behaviour appears to be the filter through which judgements about the alleged sexual conduct are made, suggesting that, for example, a resistant victim colours observers’ perceptions of the defendant’s conduct compared with a non-resistant victim.
The saliency of this approach by jurors makes sense when the cultural context in which these perceptions are made is taken into account—a context which is populated by images of sexually ‘provocative’ women and girls who enjoy forcible sex together with images of sexually entitled men, as discussed above. In other words, if predatory sexual relations are culturally normative, then a female victim’s response may be the only measure for determining lack of consent.
- i.
the less that a complainant matches the qualities of a stereotypical rape victim, the more she will be blamed for the defendant’s sexual behaviour, suggesting that victim stereotypes are gendered concepts;
- ii.
gender stereotypicality only influences blame attributions when the complainant acts like a counter-stereotypical victim;
- iii.
mock jurors perceive a victim more positively and the perpetrator more negatively when both the offence and victim are described as stereotypical, (stranger rape, aggressive force, victim who physically resisted);
- iv.
guilty verdicts are more likely in stranger rape cases and also when the victim acts in stereotypic ways, such as being physically resistant;
- v.
when the victim’s behaviour is counter-stereotypical, the defendant is less likely to be found guilty and the complainant is perceived as less credible and more to blame;
- vi.
perpetrator stereotypicality has no effect on perceptions of a victim’s behaviour;
- vii.
victim stereotypicality predicts how the perpetrator is perceived—the perpetrator is perceived as more stereotypical when the victim is stereotypical than when she is counter-stereotypical;
- viii.
victim stereotypicality affects perceptions of offence stereotypicality;
- ix.
the effect of victim stereotypicality did not differ as a function of offence stereotypicality;
- x.
mock jurors focus on the complainant’s behaviour when the defendant is an acquaintance, but focus on the defendant’s behaviour in stranger rape scenarios;
- xi.
mock jurors who are culturally similar to perpetrators consider them to be less guilty and less deserving of punishment, as a result of increases in victim blame and decreases in perpetrator blame;
- xii.
overall, a female victim’s behaviour influences perceptions of responsibility for both parties, leading to the conclusion that gendered perceptions of victim stereotypicality are the driving force behind credibility assessments and verdicts in sexual assault trials.
Cultural expectations of victims’ responses to rape compared to documented responses of victims to trauma
Expected features of a rape scenario | Expected (masculine) responses to trauma/threats | Documented responses of victims to trauma/threats |
|---|---|---|
Location: public space | Scream, shout to alert passers-by | Silence due to tonic immobility or freeze response |
Resistance | Physical resistance, fight | Freeze response |
Injury | Demonstrable injuries | Lack of resistance minimises injuries Medically unusual for genital injuries |
Complaint | Immediate complaint | Delayed response due to shame, psychological trauma, self-blame |
As Table 2 demonstrates, there is a huge gap between ‘commonsense’ expectations about how sexual assault victims should respond and the reality of a sexual assault victim’s psychological and physiological responses to trauma. How to bridge this gap will be discussed in Chapters 11 and 12.
4 Mock Jurors’ Beliefs About Victims of Child Sexual Abuse
The type and extent of the misconceptions held by laypeople about children’s reliability as witnesses and their responses to CSA have been established in a number of studies (Cossins, 2008; Cossins, Goodman-Delahunty, & O’Brien, 2009; Goodman-Delahunty, Martschuk, & Cossins, 2017a) so that the influence of jurors’ pre-existing misconceptions on their perceptions of credibility and verdict cannot be discounted in filling in the gaps in evidence in CSA trials.
Like the studies involving adult sexual assault trials, these studies also demonstrate the centrality of a child’s credibility to jurors’ decision-making. As with adult complainants, the believability of a child’s story and her/his credibility are predictors of defendant guilt, although laypeople’s perceptions of credibility are mediated by their age, gender, education and misconceptions. Nonetheless, CSA trials feature an added complicating factor due to the range of misconceptions that laypeople hold about children’s ability to give reliable evidence.
Juror gender exerts a significant main effect in that men are more sceptical than women about children’s claims of sexual abuse and rate children as less believable, less credible and less competent and more culpable for the abuse than do women (Bottoms & Goodman, 1994; Clark & Nightingale, 1997; Crowley, O’Callaghan, & Ball, 1994; Devine & Caughlin, 2014; Gabora, Spanos, & Joab, 1993; Golding, Bradshaw, Dunlap, & Hodell, 2007; Goodman-Delahunty et al., 2010; McCauley & Parker, 2001; Orcutt, Goodman, Tobey, Batterman-Faunce, & Thomas, 2001; Quas, Bottoms, Haegerich, & Nysse-Carris, 2002).
Compared to women, men give higher estimates of false reports by children (Quas et al., 2002) while ‘[w]omen tend to be more positive about children’s competence as witnesses, their ability to identify assailants, and their memory for trauma’ (McCoy & Gray, 2007). As well, juror characteristics are related to the type of misconceptions jurors hold about CSA and children’s reliability as witnesses in that men are significantly more likely to endorse misconceptions about CSA than women (Cossins, 2008; Goodman-Delahunty et al., 2010).
The more positive views held by women about children’s claims of sexual abuse are evident in the results that have shown that female jurors and jury-eligible citizens are more likely to convict in a CSA trial, compared to their male counterparts (Devine & Caughlin, 2014).
An Australian study on the misconceptions held by a representative sample of jury-eligible citizens investigated their knowledge about children’s memory, reliability, suggestibility and responses to sexual abuse (Cossins et al., 2009). The results showed that gender, age and educational attainment were associated with the degree of endorsement of CSA misconceptions. On average, participants correctly answered only 4% of the 20 items contained in a questionnaire created for the study although they ‘held the most erroneous views about children’s ability to provide reliable and truthful accounts of their experiences’ (Cossins et al., 2009: 441).
A majority of participants were unsure about 35% of the questionnaire items and ‘reported high rates of uncertainty … [about] a number of statements that might affect a juror’s perception of a child’s credibility’. These included uncertainty about: whether a child would avoid the abuser, that abused children show strong emotional reactions, that a physical examination rarely provides evidence of sexual abuse and that non-leading, open-ended questions are unlikely to lead children to make false claims of sexual abuse (Cossins et al., 2009: 440; 442).
This lack of understanding of children’s reactions to sexual abuse and children’s memory and reliability as witnesses influences mock jurors’ credibility assessments and verdicts. For example, Gabora et al. (1993) found that jurors’ endorsements of misconceptions were associated with low ratings of complainant credibility, fewer guilty verdicts and higher ratings of defendant credibility. Similar results were reported by Goodman-Delahunty et al. (2011) in that endorsement of CSA misconceptions by mock jurors negatively impacted their ratings of complainant credibility and their verdicts in simulated CSA trials, with the impact of jurors’ misconceptions on verdict being mediated by the perceived credibility of the child complainant.
- i.
A medical examination almost always shows whether or not a child was sexually abused;
- ii.
Children are easily coached to make false claims of sexual abuse;
- iii.
Children aged 7–8 years are easily manipulated to give false reports of sexual abuse;
- iv.
Children sometimes make false claims of sexual abuse to get back at an adult;
- v.
A child victim of sexual abuse will avoid their abuser; and
- vi.
Children who are sexually abused display strong emotional reactions afterwards.
Generally, we know that jurors with greater knowledge and fewer misconceptions about CSA are more likely to believe a complainant and to convict, than jurors with less knowledge and more misconceptions about CSA; and that the perceived credibility of the child complainant mediates verdict (Goodman-Delahunty et al., 2011; Goodman-Delahunty, Cossins, & Martschuk, 2016).
Laypeople’s and jurors’ misconceptions about children’s responses to sexual abuse and suggestibility as witnessesa
Beliefs | Agreement(%)* | Study |
|---|---|---|
1. Abuse claims made by 4- or 8-year- olds are false 1a. Abuse claims often prove to be false | 77b >33 | Quas et al. (2005) Morison and Greene (1992) |
2. A psychologist can tell whether a child’s description of an event has been influenced by another adult | 68 | Quas et al. (2005) |
3. A physical examination by a doctor will almost always show whether a child has been sexually abused | 57 21.4 (46.7 uncertain) 29.4 (25.3 uncertain) | Quas et al. (2005) Kovera and Borgida (1997) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
4. Children are easily manipulated into giving false reports of sexual abuse 4a. Children are sometimes led by an adult to report they have been abused when they have not 4b. Children can be persuaded by an adult to say something untruthful | 53 >50 >50 8.8 (51.1 uncertain) 23.6 (38.1 uncertain) 29.7 (61.2 uncertain) 62.4 (31.5 uncertain) | Kovera and Borgida (1997) Quas et al. (2005) Morison and Greene (1992) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
5. Child sexual abuse will result in physical damage and physical evidence | >50 >50 | Kovera and Borgida (1997) Morison and Greene (1992) |
6. Repeatedly asking children general, open-ended questions (e.g. what happened?) leads them to make false claims of sexual abuse | 46 4.1 (49.8 uncertain) | Quas et al. (2005) Cossins et al. (2009) |
7. Most sexually abused children exhibit bizarre sexualised behaviour 7a. Children who are abused display strong emotional reactions | 41 26.6 (51.7 uncertain) 47.8 (31.1 uncertain) | Quas et al. (2005) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
8. A victim of sexual abuse will avoid the abuser | 34.1 (42.3 uncertain) 43.2 (29.6 uncertain) | Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
9. A typical victim would resist, cry out for help or escape the offender 9a. A sexually abused child typically cries out for help and tries to escape 9b. Child victims of sexual abuse respond in a similar way to the abuse | >33 20.8 (30.2 uncertain) 19.5 (26.1 uncertain) 16.8 (38.7 uncertain) | Morison and Greene (1992) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) Goodman-Delahunty et al. (2017a) |
10. Children sometimes make up false claims of sexual abuse for revenge 10a. Children sometimes make up stories about being sexually abused when they actually have not 10b. Children sometimes make false claims of sexual abuse to get back at an adult | 33 7.5 (60.5 uncertain) 5.9 (63.1 uncertain) 25.7 (42.7 uncertain) | Quas et al. (2005) Cossins et al. (2009) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
11. Children cannot remember events well enough to be reliable witnesses 11a. Children can be trusted to give reliable statements 11b. Children can accurately remember past events 11c. Memories of children aged 7–8 years about traumatic events are not as accurate as those of adults | 31 66.5 (31.7 uncertain) 68.9 (27.6 uncertain) 13.4 (22.3 uncertain) | Quas et al. (2005) Goodman-Delahunty et al. (2017a) Goodman-Delahunty et al. (2017a) Goodman-Delahunty et al. (2017a) |
12. Inconsistencies in a child’s report is evidence of lying | 29 29 2.1 (35.5 uncertain) 8.1 (20.2 uncertain) | Quas et al. (2005) Morison and Greene (1992) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
13. When a child discloses more details over time, this indicates the report is false | 25 6.9 (20.7 uncertain) | Quas et al. (2005) Goodman-Delahunty et al. (2017a) |
14. Few child sex abuse cases are based on physical evidence | 23.1 [64.6 uncertain] | Cossins et al. (2009) |
15. Children who retract their reports have fabricated 15a. Children who change their reports of sexual abuse were probably lying in the first place | 21 21 1.7 (24.7 uncertain) 9.7 (25.9 uncertain) | Quas et al. (2005) Morison and Greene (1992) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
16. Delay in complaint is evidence of lying | 9 1.8 (13.8 uncertain) | Quas et al. (2005) Cossins et al. (2009) |
17. Stepfathers are more likely to abuse children than natural fathers 17a. The perpetrator of CSA is normally a stranger to the child 17b. A friend or family member is less likely to sexually abuse a child than a stranger | 16.5 (51.6 uncertain) 4.1 (19.7 uncertain) 15.8 (14.8 uncertain) | Cossins et al. (2009) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
18. Repeatedly asking children specific questions (e.g. did he touch your private parts?) often leads them to make false claims of sexual abuse 18a. Repeating questions such as ‘What happened? What else happened?’ leads children to make false abuse claims | 58 17.3 (67.7 uncertain) 16.8 (39.8 uncertain) | Quas et al. (2005) Cossins et al. (2009) Goodman-Delahunty et al. (2017a) |
The top five misconceptions focus on the falsity of a child’s allegation and the belief that a medical examination will show evidence of the abuse, with studies showing relatively high levels of agreement that children’s abuse claims are often false and that children are easily manipulated into making false claims of abuse.
Even for misconceptions with relatively low levels of agreement, Cossins et al. (2009) and Goodman-Delahunty et al. (2017a) reported relatively high levels of uncertainty about: ‘Children are sometimes led by an adult to report they have been abused when they have not’; ‘Repeatedly asking children general, open-ended questions (e.g. what happened?) leads them to make false claims of sexual abuse’; ‘Children who are abused display strong emotional reactions’ and ‘A victim of abuse will avoid their abuser’. The extent to which jurors incorrectly agree with, or are uncertain about, particular misconceptions suggests that jurors may be vulnerable to defence counsel suggestions that reinforce these misconceptions, the most common being allegations of a false complaint.
5 The Heuristic-Systematic Processing Model of Juror Decision-Making
The above summary of the literature invites the difficult question as to whether the decision-making capacities of laypeople as jurors are inadequate for the complex tasks required of them in sexual assault trials, since almost every aspect of the realities of sexual assault are in conflict with ‘commonsense’ beliefs about victims and perpetrators.
In order to understand why the disputed evidence in a sexual assault trial invites the use of myths, prejudices and misconceptions, the theoretical framework proposed by Chaiken and colleagues which describes a dual-process model of jurors’ decision-making—the heuristic-systematic processing model—is described below (Chaiken, 1980; Chaiken & Trope, 1990; Chen, Shechter, & Chaiken, 1996; Eagly & Chaiken, 1993; see also Gigerenzer, Todd, & the ABC Research Group, 1999).
- i.
‘jurors’ decisions involve a complex and nuanced set of cognitive processes that transform difficult choices into easier ones by amplifying one alternative perspective on the evidence and deflating competing perspectives’ (Winter & Greene, 2007: 741);
- ii.
individuals, generally, hold particular beliefs and attitudes that affect their reasoning processes and ‘bias reasoning performance’; and
- iii.
when jurors are ‘faced with complicated cognitive tasks’ and lack the motivation or ability to understand and interpret the evidence, they will rely on heuristic cues to determine guilt (Chen & Chaiken, 1999; Winter & Greene, 2007).
a strategy that ignores part of the information, with the goal of making decisions more quickly, frugally, and/or accurately than more complex methods.
The heuristic-systematic processing model posits that two particular processes mediate individual decision-making. Heuristic processing, sometimes referred to as the conflict ‘between the “head” and the “heart”’ (Süssenbach, Albrecht, & Bohner, 2017: 292; citing Gawronski & Bodenhausen, 2014: 448), involves little or no scrutiny of evidence and low cognitive effort because the individual uses (persuasive) heuristic cues or generalisations about human behaviour, such as ‘there’s no smoke without fire’ or ‘children always lie’, or other stereotypes to arrive at a decision. This is also known as implicit evaluation (as opposed to explicit evaluation: Süssenbach, Albrecht et al., 2017) and is an inherently unreliable method since heuristics are derived from individual beliefs and experience which may not accurately reflect the human behaviour that is the subject of a criminal trial.
Heuristic processing tends to be relied on when individuals are required to process ambiguous or incongruent evidence (which typically arises in a circumstantial evidence case, such as a sexual assault trial) but lack the ability or motivation to engage in systematic processing. This latter type of reasoning involves greater scrutiny of the evidence and higher cognitive effort, particularly as the complexity of the evidence increases. The higher a person’s motivation and the less his or her emotional involvement in a case, the more likely he or she will engage in systematic processing (Maheswaran & Chaiken, 1991).
The use of heuristics will depend on the degree of emotional involvement of the fact-finder which is likely to be influenced by levels of RMA, identification with the complainant or the defendant and individual experiences of similar crimes, along with factors such as gender, age and education.
magistrates based their bail decisions on one cue [either defendant or crime-control related] and they differed as to the information they used. … Finally, magistrates’ explicit statements of cue use did not match that proposed by the models of their decision making policies. (ibid.)
In fact, Dhami (2003) found that ‘the fast and frugal model’ correctly predicted 92% and 85% of all magistrates’ decisions in two London courts, respectively.
decisions made in the legal context are potentially susceptible to the same biases and limitations that characterise social information processing in general, such as the tendency to interpret information selectively by concentrating on what is consistent with the perceiver’s pre-existing attitudes. (ibid.)
The stereotypes that make up various rape myths, for example, represent ‘generalized knowledge structures’ (Bieneck & Krahé, 2011: 1787) which are used to attribute blame and responsibility, assess witness credibility and arrive at a verdict. The use of such generalisations presents a major problem for the criminal justice system since ‘legal definitions of sexual assault typically do not contain references to’ factors, such as ‘the type of victim-perpetrator relationship’ or the victim’s behaviour (ibid.).
Within a criminal justice setting, the use of heuristics is also problematic because this form of decision-making ‘does not search through all available information, does not weigh information in an optimal way, does not integrate all relevant information and bases its decision on only one cue’ (Dhami & Ayton, 2001: 161).
For example, a study of sentencing decisions by prosecutors found that they ‘considered only a limited number of factors and neglected factors that are legally relevant and rated as highly important’ as a result of relying on ‘simple decision heuristics’ (von Helversen & Rieskamp, 2009: 387; 388). While the use of heuristics depends on the legal task, some evidence shows that ‘that the position of the decision-maker in the decision hierarchy has an important influence on the strategies’ applied so that the inability to rely on another’s decision might result in a decision-maker using compensatory measures such as heuristics (ibid., 389). The authors conclude that ‘[t]o change the decision process, it appears necessary to additionally formulate “procedural” rules (in addition to “factual” rules) that specify how factors should be used in arriving at a decision’ (ibid.).
A number of factors may influence jurors’ interpretations, motivations and emotional involvement in the evidence and facts of a case, such as pre-trial publicity (which may establish a pro-prosecution bias), activation statements by trial counsel and judges (discussed in Chapter 6), individuals’ own experiences of similar crimes or events, as well as their own skills, competencies and individual differences such as gender, age and education (Devine & Caughlin, 2014) and their beliefs in the constancy of human behaviour.
Heuristic cues may arise from any one of these influences or other persuasive factors to do with the trial such as the impressive reputation of an expert witness, ‘the heinous character of the crime [or] the horrific nature of certain evidence’ (Winter & Greene, 2007: 746). An individual’s processing style influences his/her understanding of evidence and, therefore, the quality of his/her reasoning. In a complex trial or one relying solely on circumstantial evidence, ‘difficulties in comprehension may spark different processing strategies. For example, confusion may lead to a recourse to heuristics’ (Honess & Charman, 2002: 73) and incomplete or inadequate interpretation of the evidence.
Charman, Honess, and Levi (2001) found that when jurors’ comprehension difficulties triggered heuristic processing, this was associated with poorer evidence recall. This type of reasoning may also influence jurors to make up their minds early in the trial (also known as pre-decisional distortion: Carlson & Russo, 2001) and, once that occurs, ‘they stop thinking about the evidence too hard’ (Honess & Charman, 2002: 74). Both judges and jurors are susceptible to heuristic reasoning when gaps in the evidence give rise to confusion.
In order to determine whether or not jurors in a sexual assault trial use heuristic cues based on their degree of RMA, it is necessary to assess the extent to which they engage in systematic (or explicit) processing of the evidence compared with less effortful, heuristic (or implicit) reasoning.
situational (i.e. specifics of the rape case) and personal factors (i.e. RMA) influence implicit judgments of a rape case (just as they influence explicit judgments). … [T]he results imply that people’s gut responses to a case add to the explanation of their final explicit judgment – particularly for people with a low NFC [need for cognition], who do not like to engage in effortful processing, this effect was pronounced. (ibid., 299)12
Specifically, those with higher RMA levels, men, and those in the ‘non-real-rape’ case were more likely to be ‘strongly pro-perpetrator and anti-victim’. As well, ‘participants’ implicit associations regarding the rape case predicted their explicit case judgments’ (ibid., 299). Thus, implicit evaluations are influenced by situational and individual factors which, in turn, influence explicit judgements, such as verdict and blame. However, the authors noted that, ‘[a]lthough people with a low NFC showed a particularly pronounced relationship between implicit and explicit judgments’, this effect was also evident in those with high NFC, although weaker. Thus, as suggested by the large literature on RMA, the participants in this study ‘seem to have relied to varying degrees on their first gut response to the case. This gut response— just like the later explicit verdict— was already influenced by a situational variation, the manipulation of the rape case, and attitudinal differences’ (ibid., 300).
In summary, this study shows, for the first time, clear evidence of the association between implicit evaluations and explicit judgements (verdict, blame). Even though ‘a certain amount of influence of one’s gut response on explicit case judgments is to be expected and acceptable’, it is important to ask under which conditions the influence of implicit evaluations or heuristics (such as RMA) can be reduced in order to increase the likelihood of more effortful, systematic reasoning in sexual assault trials (ibid., 301).
Indeed, Eyssel and Bohner (2011) showed that when participants read a vignette of a rape case, those who had high RMA and received more irrelevant evidence were less likely to blame the defendant, suggesting that the irrelevant information increased the impact of RMA, possibly through heuristic processing.
considering the opposite is assumed to be an effective corrective strategy, as it counters biases such as biased information search … . Future research should thus examine whether such an instruction reduces the impact of first implicit case evaluations on individuals’ final verdict. (ibid., 301; references omitted)
6 Summary: Jury Decision-Making in Sexual Assault Trials
The jury is inevitably open to criticism when it comes to their role as fact-finders in criminal cases since they comprise a group of laypeople with little or no legal training. A number of studies have shown that juries are not impartial fact-finders since they are prone to taking into account extra-legal or legally irrelevant information in delivering a verdict. In sexual assault trials, this chapter has shown that jurors’ decisions are influenced by both the strength of the evidence against the defendant and their biases and misconceptions about adult and child victims of sexual assault. In particular, both jurors and mock jurors have been shown to focus primarily on the complainant’s behaviour in acquaintance rape scenarios, and primarily on the defendant’s behaviour in stranger rape scenarios.
Overall, the summary of the literature in this chapter reveals that observers in simulated sexual assault trials make judgements in relation to blame, credibility and verdict based on the case facts (also known as situational factors), including not only the strength of the prosecution’s evidence, but also based on variables that are unrelated to the elements of the offence. These include: the relationship between complainant and defendant, the complainant’s behaviour prior to, during and after the offence (such as, consumption of drugs/alcohol, clothing revealingness, lack of physical resistance, delay in reporting), as well as observers’ attitudes and adherence to rape myths and misconceptions about children and CSA, that is, factors personal to the decision-maker.
Generally speaking, the more a sexual assault case contains elements that accord with negative victim and gender stereotypes (or with stereotypes about lying children) and the more unlike the ‘real rape’ stereotype, the more likely it is that blame will be attributed to the complainant and observers will be more likely to find the defendant not guilty. Meta-analyses show that the higher an individual’s level of RMA, the more likely it is that an individual will make negative judgements about the complainant and the prosecution’s case and acquit the defendant. Similarly, jurors who are more likely to endorse misconceptions about children and CSA are less likely to believe the complainant and more likely to acquit.
Because decision-making is based on an interaction between case factors (legal and extra-legal) and an individual’s personal beliefs, it would be inaccurate to conclude that only stranger or ‘real’ rape scenarios will result in convictions. For example, we know that women are more likely than men to vote guilty in both adult and child sexual assault cases. However, it can be expected that the more ambiguous the prosecution evidence, the more doubts will be created in observers’ and jurors’ minds and the more they will rely on extra-legal factors to deliver a verdict.
The effects of evidence strength decrease when extra-legal factors are taken into account in statistical analyses, such that omitting the effects of these factors leads to inflated measures of the effects of legally relevant evidence. However, it is worth asking, what do we expect jurors to do when a sexual assault case is word against word, there is little or no evidence to support the complainant’s allegations and defence counsel are permitted to undermine the complainant’s evidence by relying on myths, prejudice and stereotypes in opening and closing addresses, and when testing the complainant’s evidence during cross-examination?
- i.
when jurors are faced with complicated cognitive tasks, such as a case involving ambiguous evidence, and lack the motivation or ability to understand and interpret the evidence, they will rely on heuristic cues to determine guilt, rather than systematic, more effortful processing of the evidence;
- ii.
there is clear evidence of the association between implicit (heuristic) evaluations (that is, ‘gut response’) and explicit judgements (verdict, blame).
What can be done to reduce the influence of implicit evaluations or heuristics (such as RMA and misconceptions about CSA) in order to increase the likelihood of more effortful, systematic reasoning in sexual assault trials?
These issues are discussed in the following chapters.