Why did she send it in the first place? Victim blame in the context of ‘revenge porn’

Abstract

‘Revenge porn’ or ‘cyber rape’ occurs when intimate images that were previously sent with permission are leaked to a wider audience without consent. This research investigated the perceptions that individuals form about ‘revenge porn’ victims, aiming to gain more understanding from a victimisation perspective as a first step towards improving victim outcomes. One hundred and twenty-two individuals were presented with a scenario depicting a leaked intimate image with a female victim. Two distinct nudity levels: low (lingerie) and high (bare-chest, breasts exposed) were included, and participants’ responses to the Sexual Double Standards Scale were analysed to determine whether acceptance of the traditional sexual double standard was correlated with victim perception. Results indicated that victims were perceived as more promiscuous and more blameworthy when they were more naked, and by participants with more traditional gender roles. There is a need for policy to address potential stigma directed at ‘revenge porn’ victims.

1. Introduction

Sexting, the sending and receiving of sexually explicit messages or images through electronic means, is a modern form of sexual expression that is growing rapidly in popularity, primarily among young adults (Dir, Cyders, & Coskunpinar, 2013; Klettke, Hallford, & Mellor, 2014). Coercive sexting occurs when a partner uses pressure or coercion to obtain the desired sexting messages and is also a growing problem (Kernsmith, Victor, & Smith-Darden, 2018). One of the potential consequences of sexting and coercive sexting is the distribution of these intimate images (often by an ex-lover, following a relationship breakdown) without the individual’s permission, commonly referred to as ‘revenge porn’ or ‘cyber rape’ (Kitchen, 2015). While the prevalence of sexting has been estimated at around 50% (Klettke et al., 2014), one in 10 Australians have reported having had a nude or semi-nude image of themselves dispersed online, or forwarded on to others without their permission, with the majority of victims appearing to be female (Powell & Henry, 2015). Negative outcomes for victims of leaked intimate images have been demonstrated to be widespread and potentially severe (Franklin, 2014). Yet, literature indicates that victims are likely to be blamed for their actions, and potentially even seen as sexually promiscuous for sharing such images (Feather, 1999).

Despite the documented high incidence of sexting in adult romantic relationships (Drouin, Vogel, Surbey, & Stills, 2013; Klettke et al., 2014), the perception persists that individuals who sext might be more promiscuous (Giroux, 2011). In fact, Henderson (2011) found that the most common reason for the sending of nude or semi-nude images was to appear sexy or to initiate sexual activity, with other reasons including the desire to gain attention, or to be seen as fun or flirtatious. Moreover, a considerable body of previous research has found participation in sexting to be correlated with other behaviours associated with promiscuity, including use of adult pornography and a lower age of first experience of sexual touching, sexual intercourse, oral sex partners, anal sex and unprotected sex (Crimmins & Seigfried-Spellar, 2014; Gong & Hoffman, 2012; Henderson, 2011). Since research also suggests that females are generally more likely to send intimate images and sexts than males (Klettke et al., 2014), females are therefore likely to be (comparatively) the target of negative perceptions about their promiscuity and sexual behaviour (Clayton & Trafimow, 2007; Lee & Crofts, 2015).

Most frequently, ‘revenge porn’ victims experience a loss of perceived or actual dignity and security, and lowered respect from family and friends (Franklin, 2014). These negative outcomes occur as a result of the victims being perceived as promiscuous, leading to negative subjective and social outcomes (Franklin, 2014). Traditionally, promiscuous women have been viewed as socially deviant (Clayton & Trafimow, 2007), and women tend to under-report their sexual partners, indicating an aversion towards being perceived as being promiscuous (Clayton & Trafimow, 2007). Conversely, men tend to over-report their sexual partners, which may reflect a desire to appear more sexually experienced, with research indicating that men may even have unwanted sexual intercourse due to perceived gender role expectations (Muehlenhard & Cook, 1988). This reflects sexual double standard theory where women are judged more harshly than men for comparable sexual behaviour (Milhausen & Herold, 1999). This standard has evolved into a conditional double standard, whereby men are permitted to have many sexual partners without penalty, while women are ‘allowed’ to engage in sexual relations only when they are in a committed loving relationship (Milhausen & Herold, 1999). Therefore, it is conceivable that those who more strongly endorse the traditional sexual double standard will judge female victims of ‘revenge porn’ more harshly than those who do not endorse the traditional sexual double standard.

Punishment for perceived promiscuity manifests as ridicule or personal attacks against a woman’s character (Papp et al., 2015), and consequences are similar to those reported by ‘revenge porn’ victims (Franklin, 2014) and by rape victims (Grubb & Turner, 2012). For example, ‘revenge porn’ victims often report a loss of personal dignity and lowered respect from family and friends (Franklin, 2014), while rape victims feel devalued by their peers (O’Donohue, Yeater, & Fanetti, 2003). This, combined with the considerable body of literature documenting the persistence of the sexual double standard, suggests that women may be viewed as socially deviant when an intimate image has been leaked, despite being a victim and not giving permission for the image to be viewed by a wider audience.

The concept of victim blame has been well established in the literature, occurring when victims of crime are judged by others as being responsible for the crime occurring (Grubb & Turner, 2012; Whatley, 1996). Victim blaming often occurs for rape victims and may be relevant in the context of ‘cyber rape’ victims (Kelley & Michela, 1980). The fundamental attribution error describes a bias whereby observers tend to overestimate personality or other intrinsic motivations and underestimate situational factors and other external forces that drive behaviour (Tetlock, 1985). Thus, when ‘revenge porn’ occurs, those viewing the intimate images are arguably likely to attribute the motivation to the individual (‘she should not have taken the image in the first place’, ‘she is promiscuous’, ‘she is deviant’) rather than situational factors (‘she was in a committed relationship when she sent the image’, ‘she did not give permission for the image to be leaked’), increasing the likelihood that judgments will be made about the individual in the image and the individual’s sexual promiscuity.

Negative perceptions formed about ‘revenge porn’ victims may be further explained by just world theory: the tendency for people to believe that the world is a just and orderly place where people get what they deserve (Lerner & Miller, 1978). This theory results in a tendency for observers to blame victims for their suffering, overestimating personal factors that influence behaviour (Lerner & Miller, 1978). In this context, individuals may negatively evaluate the morals of those who have sent intimate images, further placing blame onto them for their perceived immorality (e.g. ‘why did she take the picture in the first place?’). Indeed, just world theory has been commonly applied to explain victim blaming for rape, but has not yet been described as an explanatory mechanism for ‘revenge porn’ (Grubb & Turner, 2012).

Given that females have been shown to be more likely to engage in sexting, and taking into account the abovementioned sexual double standard, it is likely that women may be more impacted than men by ‘revenge porn’ in terms of social and emotional consequences. The limited research that has been conducted supports this notion. Powell and Henry (2015) found that women were more likely than males to be ‘very or extremely upset’ by digital harassment (including ‘revenge porn’), reflecting the blame, shame and humiliation that women are likely to face comparative to men (Powell, 2010). Franks (2015) found that 90% of ‘revenge porn’ victims were women, and the consequences for female revenge porn victims included loss of jobs, being subject to stalking and harassment, ‘doxxing’ and, on occasion, the requirement to change their identity due to prolonged harassment and ‘trolling’ online (Franklin, 2014; Salter & Crofts, 2015). Bates (2016) found that female ‘revenge porn’ victims may suffer from trust issues, posttraumatic stress disorder, anxiety and depression, and Franks (2015) found a heightened risk of suicide, with over half of ‘revenge porn’ victims reporting suicidal thoughts (Franks, 2015). There have also been cases of victims acting on these suicidal thoughts, with Celizic (2009) reporting a case where a female victim committed suicide in response to being taunted and bullied, after an image she sent to her boyfriend was leaked without her permission (see, e.g. Franks, 2015, for more examples).

Given the apparent relationship between sexting and behaviours traditionally associated with promiscuity, it is conceivable that individuals who have had intimate images leaked as a result of engaging in sexting may be perceived as promiscuous. In light of the severity of potential consequences for ‘revenge porn’ victims, a pressing justification emerges for establishing, through research, how individuals perceive these victims. Therefore, the aim of the present study was to investigate how female ‘revenge porn’ victims are perceived, whether they are viewed as sexually promiscuous and blameworthy, as a first step in considering how victims of ‘revenge porn’ might be stigmatised within the criminal justice system.

2. Method

2.1. Participants

A total of 122 participants completed the research online. The average age of participants was 27.60 years (SD = 10.48). The majority of participants were female (76%), Caucasian (94%) and currently in a relationship or married (64%). A majority of participants (59%; n = 72) reported that they had both sent and received an intimate image, 15% of participants reported having received an intimate image but not having sent one themselves (n = 18), and 2.5% (n = 3) reported sending an intimate image but not receiving one in return. Only 23% had neither sent nor received an intimate image. Just over 30% of participants reported having personal experience with the leaking of an intimate image, which could constitute ‘revenge porn’. There was no significant difference for the engagement in intimate image sharing behaviour between males and females, F(1, 120) = 0.009, p = .92, η_p² = .00.

Participants were generally able to identify who had uploaded the intimate image in their assigned scenario. Of those who answered the manipulation check question, 95% correctly identified the scenario to which they were assigned. It is notable, however, that there were many cases where participants chose not to answer this question (52%).

2.2. Procedure

Ethical approval for this research was granted by the [Flinders University] Social and Behavioural Research Ethics Committee. Scenarios were designed for this research and manipulated the motivation for the leaking of the image. Participants were randomly assigned either to read a scenario in which a female uploaded their own intimate image to Facebook (control condition), or to read a scenario in which a female experienced ‘revenge porn’: an ex-partner uploading an intimate image without permission (‘revenge porn’ condition). Each scenario included one of two levels of nudity: low nudity (depicting lingerie), or high nudity (depicting a bare chest with breasts exposed), with the order counterbalanced. The nudity levels were created pertaining to the literature (e.g. Beetles & Harris, 2005; LaTour, Pitts, & Snook-Luther, 1990), and a small pilot study (n = 3) confirmed that the nudity levels were classified as sufficiently distinct. Images were chosen from popular blog site Tumblr, and no faces were visible (Eck, 2003; Lunceford, 2012).

2.3. Measures

Two dependent variables were included in this research. First, a scale consisting of three items was constructed specifically for this experiment to measure the perceived promiscuity of the victim. Questions were measured on a 5-point Likert scale and reflected important themes in the promiscuity literature; for example, it has been found that people consider someone to be promiscuous if they have had a lot of sexual partners or engage in sexual activities with people they do not know well (Gong & Hoffman, 2012; Henderson, 2011). The literature translated to the items in the scale: How often does the victim have sex with people she doesn’t know?; How often do you think that the victim has sent other intimate images?; and How many sexual partners do you believe that the victim has had in her life?. Higher scores on the scale were intended to reflect higher promiscuity ratings.

The second dependent variable was a single question measuring the blameworthiness of the victim: ‘How much do you blame the victim for the leaking of the image?’. The item was measured using a 5-point scale ‘not at all’ to ‘extremely blameworthy’, with higher scores indicating higher blame endorsements. This question was included to ascertain whether participants blamed the victim for her image being leaked, despite not giving permission for the image to be viewed by a wider audience.

The Sexual Double Standard Scale (SDSS) was included to measure participant beliefs in traditional gender roles and ascertain whether endorsement of traditional gender roles impacted the attributions that individuals made regarding the promiscuity and blameworthiness of the victim (Muehlenhard & Quackenbush, 1998). The SDSS consists of 26 items, using a 4-point scale from strongly disagree to strongly agree. Scores range from 48 (reflecting greater sexual freedom for men than for women – the traditional double standard) to 0 (equal sexual freedom for both genders) to −30 (the reverse double standard).

3. Results

Results were analysed to determine whether there was a difference in the way participants viewed the images, the way they reacted to the images and the extent that the level of nudity affected attitudes towards the victim. To determine whether the level of nudity affected the perceptions of the victim’s promiscuity, a paired-samples t test was undertaken. As predicted, participants rated the victim as less promiscuous when she was wearing lingerie (M = 7.11, SD = 2.01) than when her breasts were exposed (M = 7.84, SD = 2.71). The higher nudity elicited a mean increase in promiscuity attribution of 0.73, 95% confidence interval, CI [0.409, 1.052], t(118) = 4.51, p < .001, d = 0.31. This indicates that when intimate images are leaked, victims may be attributed as more promiscuous and therefore potentially face more severe social consequences when the images leaked contain high degrees of nudity.

Next, to determine whether more blame was placed on individuals who had uploaded the intimate images themselves, as opposed to their image being leaked through revenge pornography, an independent samples t test was conducted. This revealed that the perceived blameworthiness of the victim was significantly different between the control and revenge conditions, t(120) = 5.00, p < .001, d = 0.90. More blame was attributed to the female who had uploaded the image herself, (M = 3.61, SD = 1.34) than to the female who had her image leaked through revenge pornography (M = 2.41, SD = 1.31). While the female who had self-uploaded her image received more blame, the above descriptive statistics indicate that blame was still apportioned to the female who did not consent to have her intimate images distributed to a wider audience (the revenge condition). As one in 10 Australians have been found to have had an image of themselves leaked without their permission (Powell & Henry, 2015), this indicates that many of these individuals may have blame apportioned to them and face potential social and emotional consequences.

Finally, to determine whether participants’ beliefs in traditional gender roles impacted the attributions made regarding victim blame and promiscuity attributions, Pearson’s product–moment correlations were undertaken. It was expected that individuals with stronger endorsements of traditional gender roles would perceive the female victim as more promiscuous and blameworthy. Scores on the SDSS were found to be correlated with blame attributions towards the victim, p < .001, r(120) = .37. This indicated that those with a high endorsement of the sexual double standard (reflecting more traditional gender roles) placed more blame onto victims than those with a lower endorsement of the sexual double standard. The same relationship was observed with promiscuity ratings, r(120) = .29, p ≤ .001. This indicates that overall, those who adhere to the traditional double standard are more likely to blame victims of leaked intimate images, and consider them to be promiscuous, than those with a lower adherence to the traditional double standard.

4. Discussion

For both the revenge condition and the self-uploaded control condition, victims were viewed as more promiscuous and more blameworthy when they were more naked. Those who strongly adhered to more traditional gender roles (evidenced by higher SDSS scores) attributed more blame to victims and perceived them as more promiscuous than those with greater sexual freedom (evidenced by lower SDSS scores). Therefore, it appears that ‘revenge porn’ victims are susceptible to being victim-blamed in a similar way to rape victims. Results also revealed that prevalence rates for the sharing of intimate images (e.g. engaging in sexting with visual content) was much higher in the current sample than in previous research, with over 80% of participants having sent or received an intimate image (Klettke et al., 2014). This suggests that either intimate image sharing is much more prevalent than previously thought, or that, in general, participation in intimate image sharing is increasing. In either case, this provides further justification for gaining a deeper understanding of how victims of ‘revenge porn’ are perceived, as it is likely that as sexting participation increases, so will the occurrence of ‘revenge porn’.

The current research aimed to investigate whether images with more nudity would elicit higher promiscuity perceptions than those with a lower degree of nudity. As predicted, when individuals were more naked (bare chest, breasts exposed) they were rated as more promiscuous than when they were less naked (lingerie). This aligns with previous research suggesting that level of dress is the most obvious marker of sexualisation and, by extension, of promiscuity (Reichert & Ramirez, 2000). This is consistent with literature pertaining to rape, where level of dress has also been found to impact perceptions of victims, potentially leading to further blame and stigmatisation (Peterson & Muehlenhard, 2004). Women are often blamed for their rape if they are perceived to be ‘asking for trouble’, for example by dressing in low-cut tops or short skirts (Peterson & Muehlenhard, 2004). Similarly, the current findings suggest that when an intimate image has been leaked by an ex-partner, the victim will suffer more social stigma and blame if the image is more naked. It is difficult to determine the prevalence of sexting that includes naked breasts compared with lingerie, given that sexting definitions vary between studies (Scott & Gavin, 2018). The current study establishes how perceptions can vary depending on variance in nudity level, and that level of dress can impact upon individuals’ perceptions of victims, transcending into more modern sex crimes like ‘revenge porn’.

The current research also aimed to investigate the perceptions of blame that participants placed onto victims. While more blame was placed onto the self-uploaded control condition, descriptive statistics revealed that blame was also placed onto those victims who had had their image leaked without permission. Conceivably, this blame was placed onto victims for initially capturing the image, despite not giving permission for the image to be viewed by a wider audience. This mentality of blame with regards to ‘revenge porn’ has been well documented within the media, with people often questioning: ‘Why did she take that image in the first place?’ (see, e.g. Foges, 2016). Responsibility to avoid falling victim to ‘revenge porn’ is often placed onto potential victims, absolving the (often male) perpetrators of any responsibility (Bates, 2016). For example, ‘revenge porn’ prevention risk management strategies have been published online, with one such article recommending ‘8 sexting rules’, one of which advises women to only send an intimate image if they have been in a relationship for more than one year (Brown-Warsham, 2012). This indicates that blame may be placed onto victims due to a perception that they have not been sensible in sending the images initially, much like the blame often placed onto rape victims (e.g. ‘Why was she drinking/walking alone/out late?’; Grubb & Turner, 2012). These risk management strategies are likely to be futile given that the initial images may have been coerced, taken without consent or fabricated through the creation of sexualised ‘deep fakes’ (Harris, 2018; Henry, Powell, & Flynn, 2018).

Attribution theory provides a potential explanation for the blame that may be experienced by ‘revenge porn’ victims. Individuals may be overestimating victims’ personality or other intrinsic motivations (e.g. ‘she is promiscuous’) while underestimating situational factors (e.g. ‘she was in a committed relationship’), a bias known as the fundamental attribution error. Just world theory may further explain the blame experienced by ‘revenge porn’ victims and describes the tendency for people to believe that the world is just and orderly, and people get what they deserve (e.g. ‘that wouldn’t have happened to her if she did not deserve it’). This blame may impact ‘revenge porn’ victims in ways similar to how rape victims are impacted; provision of good emotional support by others has been associated with better recovery for rape victims, while victim blaming has been associated with increased psychological distress and delayed recovery (Ullman, 1996). Indeed, both rape and ‘revenge porn’ involve the undermining of women’s sexual consent, the violation of fundamental rights to sexual autonomy, integrity and sexual expression and an invasion of women’s bodily privacy (McGlynn and Rackley, 2017), so the idea that victim-blaming may therefore transcend physical sexual assaults and also apply ‘revenge porn’ is a logical one. This is supported by McGlynn, Rackley, and Houghton (2017) who have argued that ‘revenge porn’ (albeit a more broad definition) be conceptualised as part of both a continuum with other forms of sexual violence and as part of a more specific continuum of image-based sexual abuse. Negative attributions made about ‘revenge porn’ victims could therefore impede emotional support networks, increasing psychological distress and delaying recovery for victims. Additionally, victim blaming may lead to victims under-reporting the crime and may reduce help-seeking behaviour and engagement with mental health services (Ullman & Filipas, 2001).

The blame that victims face may be compounded by individual beliefs in the traditional sexual double standard. Those with higher scores on the SDSS were more likely to place more blame onto victims than those with lower scores. This result is important given that the current sample was generally more liberal than expected in their beliefs about gender roles and sexual freedom, as evidenced in relatively low SDSS scores. Although the current sample was broad, the majority of participants were females aged in their twenties, both of which are markers for liberal beliefs (Inglehart & Norris, 2000). Therefore, ‘revenge porn’ victims are likely to face even more stigma in the real world, particularly within the criminal justice system where people they encounter are less likely to be females in their twenties. Victims may face individuals with more conservative views regarding gender roles, who, according to the present study, are likely to place even more blame and stigma onto victims. The same findings regarding gender role attitudes and an increased likelihood of victim blame have also been found specific to rape cases, with traditional gender role endorsement contributing to negative attitudes towards rape victims (Burt, 1980; Grubb & Turner, 2012). This is problematic for ‘revenge porn’ victims and may have numerous consequences, including the likelihood that the victims’ support network may be damaged if they feel they are not able to seek support from friends and family due to fear of blame and victimisation.

Victims may also feel afraid to engage in the criminal justice system and come forward to report their image being leaked, fearing negative judgments, a lack of action or perhaps even their own prosecution. This fear may not be unwarranted, and in some jurisdictions victims may be charged as perpetrators due to obscenity laws, and underage victims have been accused of creating child pornography when reporting revenge porn to the police (Folderauer, 2014). Given the assertions of Stewart and Maddren (1997), who found a correlation between level of victim blame and the likelihood of police charging the offender there is an indication that the perceptions police officers have about victims may impact the likelihood of prosecution, which may translate to other crimes such as ‘revenge porn’. Grubb and Turner (2012) argue that internal gender role biases may skew decision-making processes within the criminal justice system and recommend that those involved in the criminal justice system (particularly jurors) be made aware of such biases. It is therefore suggested that the same disclosures be made within the criminal justice system and ‘revenge porn’, in an effort to prevent the attributional biases revealed in the present study.

While the present research has provided insight regarding the perceptions that individuals form about victims of ‘revenge porn’, ‘revenge porn’ research is in its infancy, and there is scope for extending the current research beyond a simple scenario. Incorporating other variables into the analysis could provide a deeper understanding of the mechanisms underlying victim attributions – for example, including leaked images depicting sexual acts, which may elicit different perceptions (Citron & Franks, 2014). Instances have also been documented where an ex-partner has leaked a naked image of a victim that was taken without her permission (Citron & Franks, 2014). As the current research only looked at photos that were initially captured with consent and then distributed to a further audience without consent, including images that were initially taken without consent or taken under coercion may provide further insight. This is a particularly relevant avenue for future research as sharing intimate images due to pressure or coercion is twice as prominent in females than males (Englander, 2012). While the current study did include two levels of nudity, female genitalia were not visible in the highest nudity condition, and this too may have impacted perceptions. It is also important to note that only one side of the attribution process was looked at in the current research: the attributions formed about ‘revenge porn’ victims. However, much information could be gained by investigating the perceptions formed about the perpetrator – for example, investigating the level of blame attributed to the perpetrator and the level of promiscuity attributed to the perpetrator for also initially engaging in the sexting behaviour. The current research also only included heterosexual couples, female victims and male perpetrators, and including gay and lesbian couples, male victims and female perpetrators in further research may provide for further insight.

As ‘revenge porn’ is a relatively new phenomenon, the law is still ‘catching up’, and many lawyers may not appreciate the potential for severe impact due to this type of victimisation. Currently, laws vary from state to state; however, the Enhancing Online Safety (Non-consensual Sharing of Intimate Images) Bill 2018 (Cth) is currently before the Australian House of Representatives and will make ‘revenge porn’ a federal offence, with penalties of over $100,000 for perpetrators. All facets of the criminal justice system should understand the severe impacts that ‘revenge porn’ may have on victims. Sexual assault education programmes aimed at adolescents have been developed as a prevention strategy and implemented in many secondary schools and colleges, particularly in the United States (Anderson & Whiston, 2005). Similar programmes have been developed in Australia, such as the Sexual Assault Prevention Program for Secondary Schools (SAPPSS; Imbesi, 2008). SAPPS has shown promising results, with evaluations indicating an increase in understanding of sexual assault issues and ability to discuss issues in an open, respectful and appropriate way, which are best sustained through multiple and ongoing initiatives to address sexual assault (Imbesi, 2008). Given the similarities between rape and ‘revenge porn’ victims, similar initiatives may also provide benefits and improve outcomes for ‘revenge porn’ victims.

5. Conclusions

The current research has provided an insight into how people perceive the victims of ‘revenge porn’. The results reflect a tendency for individuals to blame ‘revenge porn’ victims, despite them not having given permission for the image to be viewed by anyone other than the intended recipient. Those with higher endorsement of traditional gender roles, as reflected by scores on the Sexual Double Standard Scale, were found to perceive the victim as more blameworthy and promiscuous. These findings, across a range of measures, contribute to the growing body of literature on sexting and revenge pornography, and give reason to continue exploring these perceptions and attitudes. This is especially prominent given the high prevalence rates for engaging in the sending of intimate images found in the present study. With future research, policies and programmes may be able to be developed and successfully implemented to reduce the stigma and blame faced by victims, which may assist in them engaging more in the criminal justice system.

Ethical standards

Declaration of conflicts of interest

Tahlee Mckinlay has declared no conflicts of interest.

Tiffany Lavis has declared no conflicts of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of Flinders University and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

The data are not publicly available due to ongoing research with the dataset.