Abstract
Considerable scholarly attention has been paid to a range of criminal behaviours that are perpetrated with the aid of digital technologies. Much of this focus, however, has been on high-tech computer crimes, such as hacking, online fraud and identity theft, or child exploitation material and cyberbullying. Less attention has been paid to ‘technology-facilitated sexual violence’, where new technologies are used as tools to perpetrate or extend the harm of a sexual assault, extend control and abuse in a domestic violence situation, or distribute sexual or intimate images of another without their consent. In this article, we focus on the scope and limitations of criminal legislation for responding to these varied but interconnected gendered harms. We argue that although there have been some developments in a range of international jurisdictions, particularly relating to the phenomenon of ‘revenge pornography’, much more needs to be done both within and beyond the law. Whilst we support the intervention of the criminal law, we argue that equal attention must be given to policies and practices of educators, law enforcement agencies, service providers, online communities and social media networks to fulfil the promise of equal and ethical digital citizenship.
Introduction
There is growing media, activist, practitioner, legal and scholarly attention to the phenomenon of technology-facilitated sexual violence (TFSV), where mobile and online technologies are used as tools to blackmail, control, coerce, harass, humiliate, objectify or violate another person. Part of the challenge is to devise appropriate terminology to describe a vast array of different gender-based online harms such as ‘revenge pornography’, ‘virtual rape’, ‘cyberstalking’ and ‘online gender-based hate speech’ as well as the use of new technologies to perpetrate more traditional or conventional crimes, such as domestic violence or sexual assault. Existing terminology and the laws that govern such offences in many jurisdictions internationally do not adequately capture the scope, nature or intersection of such harms. Legislative and policy responses frequently treat existing and new technologies merely as ‘tools’ of abuse and as such elide the unique ways in which victim survivors experience harms (Henry and Powell, 2015b). These behaviours are frequently framed in public discourse using euphemistic, titillating or narrow language that produces a paradigmatic conceptualization of the behaviour (e.g. ‘revenge porn’) and in the process excludes other related behaviours or leads to both victim blaming and perpetrator exoneration. On the other hand, focusing more broadly on the phenomenon of TFSV provides an opportunity to reflect on the similarities among diverse behaviours, impacts and responses as well as the problematic language surrounding them.
In this article, we examine the scope and limitations of criminal laws for responding to TFSV. We draw predominantly on Australian law as a means to illustrate the ways in which a common law country is grappling with these issues and the adequacy or inadequacy of current approaches within this context. We examine the emerging literature on TFSV and the criminal law (both in Australia and internationally) and critically examine legislation as well as case law. We argue that although perpetrators can be prosecuted under existing criminal laws, these laws are often ill-equipped to capture the gendered harms resulting from these behaviours. Part of the issue, we argue, concerns law’s pacing problem. Law has typically been slow to respond to the emergence of new technologies as tools of abuse. Existing laws are often inconsistent, outdated and poorly enforced across state, territory and federal jurisdictions in Australia (Henry and Powell, 2015a). Moreover, law is often treated as the most effective, if not sole, remedy for such harms. Although we focus specifically on the role of criminal law, we conclude the article by arguing that attention must be given to broader measures both within and beyond law to tackle the problem of digital abuse and violence.
In the article, we first conceptualize the phenomenon of TFSV as a form of gender-based violence. We then explore a range of different behaviours that are encompassed under this term and the array of existing criminal laws for responding to each of these behaviours.
TFSV and Harassment: A Continuum of Gendered Violence
Although there is currently a lack of empirical data on the prevalence of online forms of sexual violence and harassment, and as such little is known about the gender of victims and perpetrators, or the causes of these behaviours, we hypothesize that TFSV is a gendered phenomenon for three interconnected reasons. First, research to date supports the theory that although not the exclusive victims of violence, women and girls are the main targets of online digital sexualized violence. Studies on ‘sexting’, for instance, show that young women are more likely than young men to send explicit images of themselves (via mobile phone or other means) as a result of pressure or coercion from their male partners or peers (Drouin et al., 2015; Ringrose et al., 2012). Other studies suggest that women and girls are the primary targets of the non-consensual distribution of intimate images online (revenge porn) (Citron and Franks, 2014; CCRI, 2014). Research also consistently points to the prevalence of sexual violence in ‘offline’ contexts, where women and girls are disproportionately the victims of sexual harassment, domestic violence and sexual violence (see e.g. ABS, 2006, 2013; Heenan and Murray, 2006).
Second, the impacts of these behaviours are gendered because women and girls may experience adverse impacts due the persistence of outdated myths and expectations surrounding sexual norms and expectations for women specifically. 1 And third, regardless of the gender of the perpetrator (or indeed the victim), a key factor underlying the perpetration of sexual violence is the social and structural context of gender hierarchization – a ‘historically constructed pattern of power relations between men and women and definitions of femininity and masculinity’ (Connell, 1987: 98–99).
In this article, we claim that TFSV is fundamentally an issue of gender. We acknowledge that men and boys are also subjected to TFSV and that these harms are likewise significant. Nonetheless, it is important to note that women and girls may be targeted for particular forms of digital abuse (e.g. sexual assault or revenge porn) precisely because of their gender and the perpetrator’s intention to ‘slut-shame’. We also acknowledge the adverse impacts on already marginalized members of the community on the basis of gender, sexuality, race and religion. In perhaps the majority of cases of TFSV, whether against male or females, or members of particular ethnicities, races, religions, or gender identity or sexual orientation, the motivation for the behaviour stems essentially from the objectification and debasement of the ‘other’ as many of our examples throughout the article show.
In our discussion, we categorize these different behaviours in turn and give an explanation as to how criminal law has addressed these harms, taking into account the scope of law (what is/is not covered); the extent to which legislation captures the unique harms of digital abuse; and any outstanding issues. We acknowledge that there are significant overlaps between these different categories. 2
Dimensions of Digital Sexualized Violence
Revenge Porn: The Distribution of Sexual or Intimate Images Without Consent
The terms revenge porn, ‘non-consensual sexting’, ‘involuntary porn’ (Burns, 2015) and ‘non-consensual pornography’ (Citron and Franks, 2014; Franks, 2015) are used interchangeably to refer to the distribution of sexually explicit or intimate images (photos or videos) without the consent of the subject. 3 In some instances, the victim has taken the image herself/himself (a ‘selfie’), or consented to someone else taking the image of them, but has not consented to its distribution. In other cases, images have been doctored by superimposing the face or identity of a victim onto an existing pornographic image, or intimate images have been hacked from the victim’s computer or mobile device. In further cases, nude or semi-nude images have been taken when a victim is asleep, unconscious, alcohol or drug affected and/or during a sexual assault and then distributed among peer networks or shared online (see e.g. Powell, 2015, and discussion below). 4
In one Australian case, a man threatened to distribute a topless photo of his victim in order to coerce her to have sex with him after she told him she wanted to end their casual sexual relationship (Bryan, 2015). Other instances involve the more ‘conventional’ or paradigmatic revenge porn cases, including a 2010 New Zealand case where the accused posted a nude photo of his ex-girlfriend on Facebook after having logged into her account, uploaded the image, unblocked her privacy settings and then changed her password (Stuff, 2010). Similarly, in the first social networking-related conviction in Australia in 2012, a Sydney man received a 6-month jail sentence after posting six nude images of his former girlfriend on Facebook (Police v. Ravshan Usmanov, 2011, NSWLC 40). 5 In the former case, the accused was charged with an indecency offence (morality and decency), and in the latter case, the accused was also charged with an indecency offence (publishing indecent articles). By way of comparison, in the United Kingdom in July 2015, a 21-year-old man became the first person to be convicted under the new revenge porn legislation after sending intimate images of a woman to her family and sharing them on Facebook (BBC, 2015).
The term revenge porn is a misnomer since not all perpetrators are motivated by revenge (Franks, 2015) and not all content constitutes or serves the purpose of ‘pornography’. First, in relation to motivations, whilst the paradigmatic revenge porn scenario might be of a scorned and spiteful ex-lover posting images of his or her former partner onto mainstream or ‘ex-girlfriend’ pornography sites, or imageboard and social media sites, not all perpetrators distribute images out of vengeful motivations. The term revenge porn is inadequate to describe many situations where intimate or explicit images are being distributed for reasons such as coercion, blackmail, fun, sexual gratification, social status or monetary gain. For individuals who solicit images on dedicated revenge porn threads within online communities (such as reddit, 4Chan or 8Chan), or for those who provide online platforms for the distribution of explicit images, revenge might not be a motive at all. 6 The term is furthermore inadequate to capture the harm caused where the distribution or threat of distribution of an explicit image is deployed as a means of intimidating, silencing or otherwise extending power and control over victims of domestic and sexual violence (Henry and Powell, 2015a).
The second issue with the terminology – both revenge porn and non-consensual pornography – is that the focus is on the content of the image, which is framed as pornography, regardless of the image or the intent of the original creator of that image. Whilst this may fit with Dworkin and MacKinnon’s (1984: 321) definition of pornography as ‘the graphic sexually explicit subordination of women’, it fails to recognize that many images are not per se ‘sexually graphic’ (e.g. mid-dressed or shower shots). Moreover, labelling the distribution of non-consensual images, pornography has the potential both to minimize the harm done to victims and to liken the images to an acceptable and/or desirable subgenre within commercially produced online pornography. As Rackley and McGlynn (2014) also note: [t]here is a danger that the framing of this form of harassment and abuse as “pornography” shifts attention away from the motivations and actions of the perpetrators of the abuse and onto the content of the image and actions of the victim.
Some have instead suggested that revenge porn is a form of sexual assault or ‘cyber rape’. 7 Whilst there are issues defining revenge porn in these terms (as either pornography or sexual assault/rape), no other term to date has proved satisfactory to capture the types of behaviours that emerge under this category. One alternative term is ‘image-based sexual exploitation’ (see Powell, 2009, 2010). This mirrors debates regarding child exploitation material, which uses this term to distinguish it from pornography but yet at the same time highlights the harmful circumstances of its production and the continued harm associated with its dissemination. First, image-based sexual exploitation captures a wide variety of motivations driving the non-consensual distribution of intimate images, including that of revenge. Second, such images themselves need not be pornographic per se but rather are used as a form of sexual exploitation (e.g. for sexual gratification, coercion, humiliation, revenge and other reasons). And third, this term captures a broader range of contexts where the image was originally produced, thus including situations where the victim has taken a selfie and has shared it with someone else, but has not consented to any broader forms of distribution. Such a term can also encapsulate situations where the perpetrator or someone else takes the image or when a victim does not know they are being filmed or photographed. Although we prefer this term for the above-mentioned reasons, we nonetheless continue to use ‘revenge pornography’ in this article due to its popular use in public and academic discourses. 8
In response to the increasing prevalence of revenge pornography, as well the growing appreciation of the significant harms and impacts on victims, a number of jurisdictions have introduced specific legislation to criminalize the non-consensual distribution of intimate images. These include the Philippines (2009; maximum (max.) sentence 7 years), Israel (2014; max. sentence 5 years), Japan (2014; max. sentence 3 years), Canada (2014; max. sentence 5 years) 9 , the United Kingdom (in England and Wales) (2014; max. sentence 2 years) 10 and New Zealand (2015; max. sentence 2 years). 11 At the time of writing, in the United States, 25 states had passed some form of revenge porn legislation. 12
In Australia, Victoria is the only state or territory in Australia to have a specific offence that makes it illegal to maliciously distribute, or threaten to distribute, intimate images of another person without their consent (2014; max. sentence 2 years; section 41DA and section 41DB, Crimes Amendment (Sexual Offences and Other Matters) Act 2014 (Vic)). South Australia has broader legislation making it a criminal offence to distribute without consent an ‘invasive’ image (2013; max. sentence 2 years; section 26C, Summary Offences (Filming Offences) Amendment Act 2013 (SA)). At the Commonwealth (federal) level, whilst there is no specific criminal offence in place, broader telecommunications offences can and have been used in relation to using a ‘carriage service to menace, harass or cause offence’ (section 474.17, Crimes Legislation Amendment (Telecommunications Offences and Other Measures) Act (No. 2) 2004). 13 In all other Australian states and territories without specific legislation, options for criminal charges include stalking (where a course of conduct needs to be established), blackmail, voyeurism or indecency. These existing criminal offences are ill-suited to the diverse types of behaviours that are captured under the label of revenge pornography. Moreover, prosecuting offenders under indecency or obscenity offences implies problematically that the images themselves are offensive rather than the perpetrator’s behaviour.
The introduction of specific revenge pornography criminal legislation regardless of jurisdiction raises four key issues. The first concerns the nature and content of the image. Different jurisdictions use a range of terms, including ‘intimate images’, ‘private sexual material’, ‘nude or semi-nude images’ and ‘sexually explicit images’. However, whilst in many cases, the law specifies that private, sexual or intimate images must be those ‘not of the kind ordinarily seen in public’ and must be what a reasonable person would consider ‘sexual’, in practice it is not clear what would constitute a sexual or ‘intimate’ image (McGlynn and Rackley, 2015). The Victorian legislation, for example, defines ‘intimate image’ as ‘a moving or still image that depicts (a) a person engaged in sexual activity; (b) a person in a manner or content that is sexual; or (c) the genital or anal region of a person, or, in the case of a female, the breasts’.
It is unclear whether images that a ‘reasonable person’ in the community might not consider either intimate or sexual would actually be covered by these divergent laws. For instance, would existing laws criminalize the non-consensual distribution of an intimate image of a Muslim woman in her underwear without her hijab on (Yosufzai, 2015)? Similarly, it is unclear whether ‘female breasts’ include cleavage shots and whether transgender or intersex persons are also protected under such legislation. We contend that the law should take into account the nature and content of the image, the degree to which the distribution affects the privacy of the person, as well as the degree to which the distribution of images violates that person’s community’s standards of acceptable conduct.
The second issue concerns the intention of the perpetrator. Some jurisdictions require that the accused must have distributed the image with the specific intent to cause emotional distress, whilst others require ‘malicious intention’ or, alternatively, proof of harm to the victim. As noted above, perpetrators may have diverse motivations for distributing private sexual images without consent, including revenge, coercion, humiliation, blackmail, sexual gratification, social notoriety and financial gain. Many perpetrators then do not necessarily have the intention to cause harm or emotional distress. This poses a challenge to legislators: Should laws apply to third parties who do not know that the image was originally distributed without consent but who then goes on to distribute the image? Or should the laws only apply if the person knows, or has reason to know, that the other person did not consent to the distribution of the image? As legal scholar Citron (2014: 15) claims, ‘[i]t should not be a crime, for instance, to repost a stranger’s nude photos having no idea that person intended them to be kept private’. That does not mean it is ethical to do so however, the law should only capture those who knowingly betray another person’s privacy or those who were reckless to whether or not the person consented to the distribution of the image (e.g. they did not give any thought to it). Moreover, whilst proving intent to cause harm or distress might be relatively straightforward in some cases, in others, the difficulty of proving intent of the original distributor of the image, as well as others who repost and distribute the images, may serve as an obstacle to successful revenge porn convictions (McGlynn and Rackley, 2015).
A third issue concerns the harms of revenge pornography and whether this could replace or supplement the mental element of the accused in specific legislation. It is important to acknowledge that in some instances, there may be little or no harm to the subject of the image. Conversely, in other situations, there may be adverse and far-reaching physical, psychological, social and financial consequences for victims, including risks to personal safety due to stalking and/or further domestic violence; shame and humiliation; altered relationships with others; reputational damage; loss of employment prospects; victim blaming; withdrawal from social life and low self-esteem and paranoia among a whole range of other afflictions (see Citron and Franks, 2014). Therefore, we argue that the introduction of specific criminal legislation is important to acknowledge the harms associated with the non-consensual distribution of intimate images. In the absence of legal frameworks to address this serious and emerging issue, victims, perpetrators and the community more generally will continue to place the blame on women, and in the process, exacerbate existing psychological and social harms.
Finally, the fourth issue surrounding the criminalization of revenge pornography concerns the platform or medium in which such images are shared and distributed and how they are shared. For instance, does revenge pornography only include images distributed in online form? Does ‘distribution’ also encapsulate the ‘showing’ or ‘sharing’ of these images (e.g. a person showing a video or photo on their mobile phone to another person)? In some jurisdictions, it is made clear that images include photograph or film sent as text messages, distributed on social networking sites or distributed in offline form. This indicates that although revenge pornography is an emerging issue due to the development of new technologies, it is not actually a new phenomenon as people have been distributing intimate images without consent prior to the advent of smart phones and social media.
Furthermore, there is some debate about whether website operators should have immunity from prosecution (Cecil, 2014; Franklin, 2014). In Australia, under proposed federal legislation, website operators would be liable for up to 5 years if they possess, control, produce, supply or obtain for commercial purposes or some kind of benefit ‘private sexual material for use through a carriage service’ (Criminal Code Amendment (Private Sexual Material) Bill 2015). In other words, the onus is on the website operator to take all steps to ensure that another person providing the image has verified that they own the image and/or that the image is being distributed with the subject’s consent. It should not be enough for websites to simply provide victims with an option of requesting their images be removed from their websites. In October 2015, for instance, Pornhub, the largest pornography site on the Internet, announced a reporting option for victims of revenge pornography but did not take the extra step of requiring uploaders to verify images are consensual (Brown, 2015).
Although there are a range of existing civil avenues available to victims (including under civil laws such as anti-discrimination, copyright, breach of confidence and defamation), neither civil or criminal laws adequately protect or provide a remedy for victims of revenge pornography. This is for three key reasons. The first is that the costs associated with civil litigation may be overly burdensome for the average victim who may not have the financial means to bring civil action under existing laws. The second reason is that existing criminal offences that are not specific to revenge pornography behaviours do not capture the harms of revenge porn or provide an effective deterrent against these behaviours since in the absence of legislation, perpetrators may not know that the non-consensual distribution of intimate images is a wrong. And third, neither new nor existing civil or criminal laws are capable of stopping the spread of the image once it has been posted online (see Citron and Franks, 2014).
Although some legal experts in the United States have argued that existing criminal offences are sufficient to capture revenge porn behaviours (see e.g. Budde, 2014; Stokes, 2014), there has been increased pressure on lawmakers to introduce specific criminal legislation – in Australia and elsewhere. Citron and Franks (2014: 349), for example, argue that specific legislation is important ‘to convey the proper level of social condemnation for this behaviour’. The benefit of specific legislation is that it captures behaviours in the context of intimate partner violence (where the behaviour is part of an overall pattern of intimidation and abuse) and those outside the domestic violence frame. Specific legislation also has expressive value, sending a signal to the community that such behaviours are abhorrent and deserve to be punished accordingly. The potential outcome may be a move away from victim blaming.
These laws may also have a positive impact through better articulating community standards about digital exchanges. One example is corporate levels of responsibility, including website policies on unsolicited images. In 2014, for instance, reddit announced a ban on the posting of sexually explicit images of a person without their consent after it was heavily criticized for allowing the distribution of hacked nude photos of Hollywood actors. In March 2015, Twitter followed suit, banning the posting of unauthorized sexual images (individuals who do so will have their accounts locked and users will be required to delete the content before they can return to using the site). In June 2015, Google announced a new reporting mechanism for victims who can now request images of them be excluded from Google Internet searches. In July 2015, Microsoft announced a similar function that will enable victims to have content removed from its Bing search engine and on its OneDrive and Xbox Live cloud services. Finally, as mentioned above, Pornhub, in October 2015, also introduced a process that allows victims to request their images be taken down from the site.
These developments, we contend, are further reminders of the importance of action beyond the law to address the growing problem of TFSV. But most pertinently, together changes in laws and policies at different levels demonstrate the importance of a multifaceted approach to this issue. Yet one of the most pressing issues concerns the persistence of sexually explicit or intimate images in cyberspace post-distribution and the fact that many existing laws have very little impact on whether the content is removed, particularly since many sites where the images are hosted are based outside the victim’s country. As the Internet readily enables re-blogging and reposting, it may be impossible to retract the image once it has been distributed. This is another reason why it is important to look beyond law and to focus energies on primary prevention measures as well as the development of codes of corporate and user responsibility and accountability for tackling revenge pornography and other forms of digital violence.
The Recording and/or Distribution of Sexual Assault Images
Whilst revenge pornography is currently receiving much attention internationally, it is only one form of technology-facilitated image-based abuse. Another emerging behaviour concerns the recording (‘creation’) and/or distribution of sexual assault images. The 2012 Steubenville Ohio case is one recent example. In that case, an incapacitated 16-year-old girl was raped by two high school football players over a 6-hour period at various locations. The rapes were recorded and then distributed via mobile phones and websites, such as You Tube and Instagram. Two of the perpetrators (Ma’lik Richmond and Trent Mays) were convicted in juvenile court for the rape of a minor and distributing child pornography. Three adults were indicted for tampering with evidence and the obstruction of justice and two young women pled guilty to aggravated menacing charges in relation to tweets they sent threatening the victim (Welsh-Huggins, 2013).
The Steubenville case garnered enormous media attention, in part because of the biased and victim blaming media reporting of the case but also because of the unprecedented role that social media played in both the collection of evidence and the collective outrage towards the event itself and the reaction to it. Although the dozens of individuals who recorded and/or distributed the images onto social media sites were not prosecuted, in court the mobile phone images and text messages helped to initiate the prosecution’s case as well as provide evidence. Moreover, these images revealed the role of ‘bystanders’ who stood by and callously recorded the violence and then distributed the images via mobile phone and social media. In addition to the behaviour of the two accused and the authorities indicted for obstructing the course of justice, this was further ‘proof’ of the problematic culture of rape that allowed not only such an incident to happen in the first place but also allowed the abuse to be captured, disseminated and voraciously consumed (Powell, 2015).
A surprisingly similar example concerns both the creation and distribution of images of simulated sex acts by US soldiers (sodomy, oral sex and bondage) forced upon Iraqi prisoners at Abu Ghraib during the war in Iraq in 2003. Although the acts themselves have been widely condemned as homophobic, racist, misogynist and imperialist, as Puar (2004: 531) suggests: … what is exceptional here is not the actual violence itself, but rather the capture of these acts on film, the photographic qualities which are reminiscent of vacation snapshots, mementos of a good time, victory at last, or even the trophy won at summer camp. (emphasis added)
Another issue concerns whether the covert filming and distribution of consensual sex constitutes a sexual assault or a rape under existing criminal law legislation. Consider, for example, the Australian Defence Force Academy (ADFA) ‘Skype Scandal’, where a male cadet secretly broadcast otherwise consensual sex with a female cadet via Skype so that five of his peers could watch from another room without her knowledge or consent. One of the two accused, Daniel McDonald, was found guilty in October 2013 under the Australian Commonwealth telecommunications legislation, including sending offensive material over the Internet and a related indecency offence (see Byrne, 2013). Whilst the convictions recognize that the broadcasting of sex without consent was ‘offensive’ and ‘indecent’, they do not adequately acknowledge the harm experienced by the victim. It is arguable that the victim’s consent was vitiated by the deception involved in broadcasting the sexual encounter without consent. This is not dissimilar to other forms of sexual activity by fraud or deception cases in Australia and internationally, where convictions for rape have been successful despite the original consent of the victim to the act itself (see e.g. Crowe, 2011, 2014; Syrota, 1995). However, such an approach is not without limitations. For instance, some legal scholars suggest there is the potential to inadvertently minimize the harms of rape as a physical, sexual violation, if the psychological harms of sexual activity by fraud or deception are included under the same offence category (see Roffee, 2015). This suggests, perhaps, that a separate sexual offence may be needed to address the harms to victims in such instances.
In summary, the term revenge porn is inherently flawed to capture the motivations of the accused persons who create and distribute images of sexual assault. These recordings and their distribution may well be for human rights or evidence-gathering purposes (e.g. in wartime), yet little attention has been given to the ethics or criminality of reproducing torture images in this field. It is thus important to scrutinize these acts and to consider the harms done to victims when violent images are distributed, regardless of the intention of the distributor. Likewise, it is important to explore whether amendments are required to existing sexual offences legislation to specifically criminalize the recording and/or distribution of sexual assault (see Powell, 2010).
Technology-Facilitated Sexual Assault and Coercion
Online technologies such as mobile phones, email, social networking sites, chat rooms and online dating sites (among a range of other platforms) are also being used by sexual predators as a means of facilitating a rape or sexual assault, representing a further example of TFSV. Indeed, there are at least three forms of online sexual coercion or sexual assault via digital technologies that are emerging in media reports and case examples. First, where a perpetrator befriends the victim online through a mobile phone app, social network or online dating site before then meeting them in person and sexually assaulting them. For example, in June 2012, popular social networking site Skout suspended accounts of under-18s after a number of sexual assaults were carried out by adult perpetrators against underage victims (e.g. Perlroth, 2012). In July 2014, a 22-year-old man was arrested on a rape charge in Louisiana after allegedly raping a woman he had met on the dating and ‘hook up’ app Tinder (Hodges, 2014). In September 2014, an Irish woman in her 30s reported having been raped after meeting a man, also in his 30s, on the Tinder app (McMenamy, 2014). In 2015, Tinder was once again under the spotlight after users alleged that the app is helping sexual predators to contact and groom underage youths (Huynh, 2015) as well as numerous adult assaults (Shadwell, 2015).
A further example concerns sexual coercion and ‘sextortion’, a form of non-physical coercion where a person procures ‘sexual cooperation by putting some kind of pressure on a victim’ (Barak, 2005 : 80). This can take the form of eliciting private information or a sexual image from a victim and using this material to blackmail, bribe or threaten the victim to engage in either virtual or in-person sex acts. A third example concerns a form of ‘rape by proxy’ (see Frosh and Dumais, 2014; O’Connor, 2013), where communications technologies are employed to solicit a third party to sexually assault a person, whether through deception, including false or mimicked identity, or more direct means. Such has been the context of a number of cases reported in the media where a perpetrator, often a male ex-intimate partner, has placed advertisements in online classifieds or community forums inviting others to rape a victim, either by posing as the victim or through direct requests. For instance, in December 2009, a Wyoming (US) man placed an advertisement on Internet classifieds site Craigslist posing as his ex-girlfriend and asking for ‘a real aggressive man with no concern for women’ (Correll, 2010). According to media reports, one week later, a man who had responded to the advertisement forced entry to the victim’s home and raped her at knifepoint. Whilst an extreme example, the use of dating apps, online classifieds and community forums to facilitate sexual crimes is an issue regularly featured in media reports (see Furness, 2012; Meyer, 2012; Noonan, 2011), though currently there is little empirical data on the prevalence of attempted and/or completed rape via such means.
Sexual assault, whether facilitated by communications technology or otherwise, is already subject to criminal legislation in all common law countries. As such, the extent to which these behaviours can be understood as ‘new harms’ requiring amendment to existing laws, or are simply sexual assault by a different means, is worthy of interrogation. To some extent the manner through which a perpetrator accesses and ‘grooms’ both young and adult victims does not change the criminal nature of the contact offence itself. Some jurisdictions are responding to the disturbing trend of rape by proxy by considering criminal offences addressing communications that solicit others to commit a sexual offence. For instance, in Maryland (US), a Senate Bill was passed (SB50, 14 April 2014), which prohibits a person from using the identity of an individual without consent to invite, encourage or solicit another to commit a sexual offence against another person (see also Frosh and Dumais, 2014). The sexual felony offence, with a 20-year maximum penalty, recognizes the harm of such an act, regardless of whether a rape was ultimately committed (at which point existing offences including aiding and abetting a rape would apply). Furthermore, the mode of facilitation itself raises consideration of whether third party service providers, such as dating applications, online classifieds and community forums, ought to be required to take more proactive measures to address the risk that their service is used to facilitate a sexual assault. For example, in many media reports, perpetrators have used fake profiles either to misrepresent their age for the purposes of targeting young victims and/or in an attempt to avoid detection (e.g. Inman, 2014; Portelli, 2015). Such cases suggest that service providers should consider more rigorous approaches to confirming individuals’ identities linked to their profiles on such apps and sites.
Cyberstalking and Criminal Harassment
The final example of TFSV concerns cyberstalking. Although there is no single and universally agreed-upon definition of cyberstalking in the international research literature, Reyns et al. (2012: 1153) describe it as ‘the repeated pursuit of an individual using electronic or Internet-capable devices’. Such a definition is useful as it is inclusive of a wide range of behaviours, whether perpetrated via mobile phone, email, instant messages, chat services, online discussion or bulletin boards as well as social media or other digital technologies. Cyberstalking includes repeated unwanted communications; repeated unwanted sexual advances or requests; repeated threats of violence; as well as surveillance and monitoring of a victim’s location, daily activities and/or communications, whether facilitated by cameras, listening devices, computer software and mobile phone applications, or global positioning system (GPS) location information (see Reyns et al., 2012; Spitzberg and Hoobler, 2002).
Whilst cyberstalking encompasses a range of digitally abusive behaviours that may be perpetrated by intimate partners, sexual or dating partners, acquaintances and strangers alike, research nonetheless suggests that perpetration and victimization are gendered in particular ways. For example, whilst men also report stalking victimization, research and crime data suggest that such victimization may be less common, is more likely to be perpetrated by a stranger or acquaintance rather than by a partner or former partner and causes less fear for male victims compared to female victims (see e.g. Logan, 2010; Wigman, 2009). 14 For women in particular, there appears to be much overlap between cyberstalking and more conventional forms of stalking in terrestrial or offline spaces, though there are also some apparent differences (Nobles et al., 2014 ). Cyberstalking, for instance, when occurring in intimate partner violence contexts, might represent one component of a broader pattern of abuse where male perpetrators use multiple strategies to monitor, control, harass or threaten their often female victims (Diette et al., 2014; Southworth et al., 2007; Woodlock, forthcoming). One notable difference made by electronic communications and other technologies is that a stalking perpetrator can potentially have constant access to their victims, extending the feelings of exposure, vulnerability and fear of victims (Diette et al., 2014). Indeed, technologies give perpetrators easy access to their victims, explaining why, in a recent survey of domestic violence advocacy workers (DVRC, 2015), 98% of survey respondents said their clients had experienced some form of technology-facilitated stalking and abuse. It is important not to overlook the qualitative differences between offline and online forms of stalking in terms of harms or impacts on victims. As stated by the US National Network to End Domestic Violence (NNEDV) (2015) Safety Net Project, ‘Intimidation, threats, and access of information about victims aren’t new tactics … However, the use of technology as a tool … means that the harassment and abuse can be much more invasive, intensive, and traumatising’ (emphasis added).
Laws criminalizing stalking, whether in physical space or by electronic means, often require three core elements: a ‘course of conduct’ (or repeated behaviours over time) that is intrusive and/or unwanted and that threatens to harm or cause fear of harm to victims (Spitzberg and Hoobler, 2002). For example, in Victoria (Australia) under the Crimes Act 1958 (Vic), stalking is defined as a person ‘engaging in a course of conduct which causes apprehension and fear’ and includes any of the following behaviours: contacting the victim by any means; publishing on the Internet or by email, or other electronic communication to any person or statement or other material about the victim; tracing the victim through electronic communication; keeping the victim under surveillance; and a number of other acts. Whilst such legislation would appear to capture the harms of stalking in the specific context of intimate partner violence, there are some limitations with the framing of both a course of conduct and ‘causes apprehension and fear’ in the context of emerging forms of digital harassment and abuse. Foremost of these is that some single or one-off actions (such as posting a revenge pornography image alongside identifying information and an invitation to others to contact, harass or rape the victim) may in and of themselves cause considerable apprehension or fear in a victim. Second, continuing with this particular example, the one-off action may in turn encourage and result in subsequent harassment by third parties, or the extensive distribution of the harassing content, such that the impact of the single act results in a series of actions by others. Alternately, a repeated course of invasive conduct, such as repeatedly posting offensive, malicious, or personal information about a person, might be humiliating, shameful or harassing, but not cause fear or apprehension. In the United Kingdom, for instance, a summary offence of criminal harassment exists as a separate offence to stalking to address such instances where a course of conduct is harassing but does not cause fear of violence, distress or alarm (see section 2 and section 2A, Protection from Harassment Act 1997 (UK)).
Further limitations of conventional stalking offences are brought into even sharper focus when considering other, non-partner stalking or harassment-type behaviours that have emerged in cyberspace. In online forums, games environments, email and social media, for example, women and men appear to be increasingly targeted with threatening language, threatening images (sometimes simulating a violent assault) and the posting of identifying and/or private information (also known as ‘doxing’). Some scholars have gone further to identify the online harassment and abuse directed at women in particular as ‘gendertrolling’ (Mantilla, 2013), misogynist ‘e-bile’ (Jane, 2014) and sexist or gender-based ‘hate speech’ (Lillian, 2007).
Whilst much sexist trolling or hate speech might be more readily addressed in civil laws (and is thus beyond the scope of this article), there are further offences of criminal harassment that may apply in some instances where the hateful speech includes threats directed at an individual. Many jurisdictions also include specific criminal offences regarding communicating threats to kill or inflict bodily harm, such as in Australia at the federal level (e.g. section 474.15 of the Criminal Code Act 1995 (Cth)). In Victoria, an offence regarding threats to commit a sexual offence (section 43, Crimes Act 1958 (Vic)), includes ‘where a person (A): ‘makes to another person (B) a threat to rape or sexually assault B or a third person (C); and … A intends that B will believe, or believes that B will probably believe, that A will carry out the threat’ (emphasis added).
One of the problems with online threats to kill, rape or inflict bodily harm is that despite causing subjective fear, alarm or distress for a victim, such threats are not always taken seriously as ‘credible’ or ‘real’ threats. For example, in New South Wales (Australia), in order to meet the requirements of the threat to kill offence, the threat must be communicated such that a reasonable person would take it to represent an actual proposal to kill or harm and that a ‘merely hypothetical proposal will not suffice’ (R v. Leece, 1995, 78 A Crim R 531, Higgins J, at para. 536; emphasis added). Similar reasoning has been applied to cases in other jurisdictions internationally, including a significant recent US Supreme Court decision, in which it remains unclear whether threats to harm made online (such as via Facebook), may constitute ‘real threats’ that a reasonable person would expect to cause fear and are therefore not protected free speech. 15 Whilst the assumed distance of a harasser making threats to kill or rape on social media might lessen the risk of actioning violence, 16 for targets of such threats, the anonymity or diffused identities of online harassers means that their proximity, connection to the victim, and/or capacity to act on the threat are all unknowns. This in turn has the potential to cause heightened fear.
As in the other behaviours explored here, cyberstalking is an emerging and increasingly common phenomenon that the law has struggled to keep pace with as the technology has evolved. Many jurisdictions are yet to decide how to best capture the harms of criminal harassment and cyberstalking in online spaces. The challenges of cross-jurisdictional detection and punishment continue to loom large, as do the issues surrounding the distinctions and crossovers between conventional forms of criminal behaviour, and those emerging behaviours where technology not only serves as a tool of abuse but can cause new harms and present real threats to psychical and bodily autonomy, integrity and freedom. Given the gendered nature of these various behaviours, it is important to look to the criminal law as an important means of responding. However, attention must also be given to a range of measures both within and beyond law, including acknowledgement and action of the pervasive othering and objectification that occurs in both offline and online spaces.
Conclusion
Whilst this article has focused on criminal law responses to various forms of TFSV, it is important to emphatically state that law should not be seen as the sole remedy for these offences. This is not only because increased criminalization poses a real challenge to dealing with such harms but also because law is inherently modelled on a model of individualized, depoliticized justice that fails to address the deeply embedded and underlying (structural and individual) causes of gendered violence. As such, there should be equal attention given to the policies and practices of service providers of online communities and social media networks and their responsibility to proactively address this problem by providing mechanisms for users to report offensive, hateful and/or harassing content and dedicate sufficient resources towards monitoring and removing this content. There are a range of other non-legal measures that can be developed to ensure the ethical use of communications technology, including police and sector training around the impacts of digital abuse as well as evidence gathering methods; clear community guidelines; clear, consequential and enforceable terms of use on Internet sites; greater care from website providers to ensure customer safety (e.g. on dating sites); agreements between police and service providers to facilitate the timely gathering of evidence; victim hotlines and other mechanisms to provide free and confidential advice; and educational initiatives and public awareness campaigns designed to foster ethical digital citizenship. Above all, the onus should not be on victims or potential victims to ensure their own safety. Instead, attention needs to clearly shift to perpetrator behaviours and the development of community standards around ethical digital relationships.
There are some key and perplexing challenges that cut across both conventional and technology-facilitated forms of sexual violence and harassment. The first is how to prevent violence before it occurs. Such ‘primary prevention’ measures include educational programmes around respectful relationships or gender or sexuality which must, in addition to tackling gender dichotomies and inequalities, also focus on ethical digital interactions (Henry and Powell, 2014). The second most important issue is that we must think very carefully about how to respond to violence and injustice after it happens: what kind of criminal laws are appropriate and do they capture the harms associated with online digital abuse? What role do non-criminal laws play and non-legal mechanisms too? We would argue that a multipronged approach is essential to tackling gender-based digital violence.
As Butler (2004) asks, ‘how should justice be done?’ and ‘what just do we owe to others?’. In relation to both primary prevention and justice responses, we suggest that justice must be thought about beyond a perpetrator/victim paradigm. In other words, it is vital to think also about the causes of violence, the measures we implement to prevent this violence and the ways in which we respond when violence occurs – as both a problem of individual and collective or societal dimensions. These two key challenges demand we implement fundamental structural changes to the ways we think about gender, sexuality and violence and ‘the other’ in an ever expanding and shifting digital age full of peril and promise.
Footnotes
Acknowledgements
We would like to thank Brent Collett for his research assistance, as well as Martha Piper, Alex Davis, the journal editors and the two anonymous reviewers for offering extremely helpful suggestions on how to improve the article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by an Australian Research Council (ARC) Discovery Grant (DP130103094).
