I. Evolving Digital Violence—The Origins

Digital violence, even in the twenty-first century’s more nuanced forms, is not a new phenomenon. It has existed since the origins of the internet, and especially since the first multi-user online spaces. The earliest reports of digital violence are not necessarily captured as digital sexualized violence, but nonetheless, the origins rest firmly in the male-dominated environments of the early online games. One of the earliest reports of digital sexualized violence is the rape of a game character in LambdaMOO in the early 1990s.Footnote ¹⁷ This, while one of the most high-profile, has tended to receive comment as an instance of internet regulation, rather than as a criminal act. Nonetheless, it is one of the examples of how early online violence adopted a sexualized element, albeit one which went unacknowledged and unchallenged by law enforcement and regulatory bodies. The manner in which virtual characters subject one another to violations—alleged to be virtual rapes—illustrates the difficulties in early discussions surrounding how to address such behavior. In 1993, this was addressed as a user issue within the game itself, rather than one involving a criminal complaint—although with some suggestion that, at best, the actions relate to simulation and nothing more harmful—relating to the virtual world in which the sexualized violence occurred. This is not the only, early, example of a situation where claims of digital sexual violence are conceptualized. In 2007, a further high-profile allegation relating to rape claims in a virtual environment was reported to the Belgian Police.Footnote ¹⁸

A narrow conceptualization of such alleged behavior may infer that there can be no rape of a virtual character which consists of little more than software code, but instead, the conduct complained of amounts to harassment with sexualized undertones in a digital environment. Making such narrow categorizations is tantamount to the dismissal of the broader context and root causes of such behaviors, but also the dismissal of the potential harms which may have been inflicted by virtue of the behavior in the online environment. More than this though, such an example highlights some of the difficulties that arise when incidents such as this are reported in the public domain—discussions which question whether it is even possible for digital violence, and digital sexual violence to occur.Footnote ¹⁹ Fortunately, debates, awareness, and understandings have developed since the reports of the 1990s and mid-2000s, albeit they remain far from adequate for addressing such issues, and the more developed forms of digital sexual violence. The criminal law was unprepared for instances of digital and virtual sexual violence in the 1990s, but the regulatory and enforcement challenges continue to grow as new forms of digital sexual violence emerge.

II. Virtual Black Eyes & Sexualized Threats: Normalizing Sexism Online

The shifting forms of digital sexual violence, moving towards OTFSV, and occurring in greater numbers across a multitude of platforms has been a matter of concern for the regulatory framework—both criminal, and internet—since the mid-2000s. Most notably, the modes of OTFSV have developed and become more sophisticated at the same time as technology has exposed greater numbers of people to such behaviors and harms.

Some early examples of digital sexual violence originate with the founders of one of what are now the ‘Internet Giants.’ One of the most notable is the ‘FaceMash’ website that was developed by Facebook creator Mark Zuckerberg in 2003. The website was designed to allow women to be pitted against one another, or farm animals, in a competition to see which would be voted as the most attractiveFootnote ²⁰ and serves as just one example, albeit one which developed into something more menacing, of the ways in which gender stereotypes and tropes play out in online spheres. This modified version of FaceMash’s inspiration, Hot or Not,Footnote ²¹ embodied within it a much nastier, vitriolic element of direct comparisons in the public domain. Originally suggested as more prankFootnote ²² than anything else, Zuckerberg claims to have not contemplated that this could give rise to and what is now a more broadly recognized phenomenon, online & digital bullying. In initiating FaceMash, this website arguably marks the emergence of image-based harassment, and highlights what is now one of the forms of digital sexualized harassment online. By creating the means to mine for images of women, and publicly share those images on a website, the predecessor to image based sexual abuse was conceived. It is not surprising that the successor to Zuckerberg’s FaceMash—Facebook—while being one of the leading social media platforms, is, more alarmingly, the one reporting the highest numbers of OTFSV incidents.Footnote ²³ The FaceMash example is followed by two much more severe, and extreme examples, both of which highlight on a global scale the graphic digital sexualized violence sent to women. First, the “Beat Up Anita Sarkeesian Game,” and second, “Zoe Quinn & Gamergate.”

The first example is the game that was created by Bendilin Spurr, which allows users to virtually beat-up Anita Sarkeesian:

Sarkeesian became the object of Spurr’s game for campaigning to develop a video series deconstructing stereotypes of women and female game characters. The game description makes it abundantly clear that there is a significant sexualized element to it. The game was designed to allow users to click on an image of Sarkeesian and replicate virtually, punching her in the face, causing bruising, and lesions to appear on the image. The entire point of the game was to punch a woman in the face, “for having an opinion.”Footnote ²⁵ Sarkeesian herself notes that the culture of sexism is rewarded, with women being subject to increasing forms of silencing online:

The second example highlights a different element to OTFSV and focuses on the severity of threats and abuse sent to women, via digital and online means, in what is now known as Gamergate.Footnote ²⁷ The origins of Gamergate rest in behaviors which could generously, albeit dismissively, be described as online bullying. These behaviors are much more severe and include threats of physical harm including rape and murder, and are described by the woman at the center of Gamergate, as the destruction of her life.Footnote ²⁸ The trigger for Gamergate was game developer Zoe Quinn’s efforts in producing and developing the online game Depression Quest.Footnote ²⁹ The New Yorker infers that the reason for the volume of abuse and seriousness of the threats against Quinn—severe enough to make her leave her homeFootnote ³⁰—is related to the quality of the game she developed,Footnote ³¹ but others are less convinced and suggest that the trigger is the fact that she is a woman.Footnote ³²

Regardless of the reason for the motivation, the resulting behaviors and targeting of Quinn escalated when her ex-boyfriend wrote a series of blog posts indicating that she had cheated on him with other game industry workers. The backlash within gaming circles intensified and these allegations became almost unopposable truths. The ultimate result was not only intensified digital and sexualized threats in the form of rape, and assault, but also the public sharing of her address and phone number onlineFootnote ³³ so that these threats could potentially be carried out. In publicly sharing these details, the effect of the threats upon the victim is heightened, not because of the potential criminality, but rather because of the reach of online information which infers the potential of physical harm by persons unknown at times unknown. By making threats and sharing contact details, the stranger danger, and the unknown of whether these threats will lead to physical harm adds in itself greater psychological effects to the recipient of such communications. The emergence of violence in online environments has given rise to a new, and significant phenomenon of digital sexualized violence—one which manifests itself much more widely across the internet, but which emphasizes text-based sexual abuse (TBSA) and image-based sexual abuse (IBSA), and yet, such a phenomenon is not equally addressed by the criminal law. Specific legal provisions have been introduced to capture IBSA, and yet TBSA, despite its harms, remains at worse, almost unregulated, and at best, sporadically made to “fit” within non-specific legal provisions.

III. Sexual Abuses Online— IBSA & TBSA

The prominence of OTFSV is a hangover from a prolonged failure to address earlier behaviors and problems in online environments. The lack of criminal law responses to complaints of issues in online environments in the 1990s, early 2000s, and early 2010s demonstrates persistent failures to tackle the problem, which is now a phenomenon with wide-ranging impact.

In the last decade, digital sexual violence has developed still further, and evolved from the specific game-based violence seen in the 2000s. In particular, two examples indicate the evolution and scale of the challenge facing the criminal law. First, the development of image-based harassment into IBSA including other, associated forms of image harassment,Footnote ³⁴ such as upskirting, down-blousing, sextortion, and deepfakes.Footnote ³⁵ Second, the evolution of threats into text-based sexual abuses. These forms of OTFSV are all direct descendants from earlier problems, arguably facilitated and encouraged by the lack of appropriate responses by the legal system.

IBSA, also known as “revenge pornography,”Footnote ³⁶ has also seen a significant rise to prominence. The unauthorized and non-consensual dissemination of intimate images shares characteristics of behaviors seen earlier. In the non-consensual sharing element of for example, FaceMash, but with a much more sinister element that has also some shared origins in the allegations that triggered the backlash and online harassment of Zoe Quinn. The non-consensual sharing of intimate images has at its heart the capturing of the criminal sharing of nude or sexual photographs. Often these are taken when relationships are good, and disseminated in the pursuit of revenge, or destruction of reputation once things turn sour. The sexualized element is central to the harm, and to the sharing behavior that follows. Sharing images non-consensually that were never intended for public viewing is at the heart of the offence here. This form of OTSFV has unusually seen a relatively rapid rise in legislation from the criminal justice system,Footnote ³⁷ with criminalization and prosecutions being swiftly pursued. Such a response is, partially, because the ‘harmful’ aspect of images is easier to capture, and to categorize, particularly because in England & Wales a core element of the criminal offence is the causing of distress to the person depicted in the image.Footnote ³⁸ It is unfortunate that similar distress has not yet been captured in legislation for non-image-based forms of OTFSV.

TBSA meanwhile refers specifically to the manner in which rape threats, and physical assault threats are communicated through online platforms. It is defined here as:

This form of OTFSV includes threats made to women in prominent positions and public life, especially elected officials.Footnote ⁴¹ It is particularly troubling that this form of OTFSV is often overlooked by the criminal law as being “just” a communications misuse offence,Footnote ⁴² rather than a form of digitalized sexualized violence, especially where studies suggest that thirty percent of women have experienced such forms of OTFSV at least once.Footnote ⁴³ The prevalence of this issue cannot be overlooked as it has tended to be previously. Often short messages on social media are not captured within the criminal law framework, and where they are, they are not prosecuted on the basis of having a sexualized element to the violence. In rare instances, consideration is given to the types of threat made—but this is unusual.Footnote ⁴⁴ It is particularly harmful for women choosing to express opinions on social media and online environments—high profile examples indicate this. For instance, the first black Member of Parliament in the United Kingdom, Diane Abbott, was targeted with more than 8000 tweets in the first six months of 2017 alone.Footnote ⁴⁵ This is not an isolated incident, with other prominent women also being subjected to similar levels of OTFSV communications which are threatening and abusive.Footnote ⁴⁶

IV. The Rise of OTFSV as a Challenge for the Criminal Law?

What the examples illustrate, beyond the evolution of digital sexual violence, is the vitriolic nature of OTSFV, and the significant volumes of it. The nature, form, and scale of OTFSV has not yet been adequately captured in the criminal law system. Significantly, the harmful nature and profound influence of OTFSV has been overlooked by the legal frameworks, not least when it comes to TBSA. The criminal law conceptions of harm tend to rest—because communications which have traditionally been placed within communications misuse rather than sexual violence offences—on notions of distress, and offensiveness.Footnote ⁴⁷ This conceptualization overlooks the personalized influence upon recipients of such violence.Footnote ⁴⁸ For instance, Zoe Quinn encountered harms that the criminal law framework does not necessarily recognize—she had to leave her home, and seek alternate accommodation. Similarly, in other instances, economic, social, as well as psychological harms have all been the result of profound and prolific OTFSV.Footnote ⁴⁹ These harms are not currently captured within the legal framework and by overlooking such harms, there is a significant void within the regulatory provisions which require ‘joined up thinking’Footnote ⁵⁰ from perspectives beyond criminal law.

It is nevertheless possible to identify three factors which mark out the damage that sexual violence in online environments can inflict that pose challenges for the criminal law. First, the rapidity with which digital sexual violence and OTFSV has risen to prominence; second, the lack of understanding and uniformity of approach in addressing digital crimes incorporating sexual violence; and third, the impact of such behaviors on the recipients and victims.

I. Criminal Law, OTFSV and Digital Harms

First and foremost, there are significant deficiencies in the way that criminal law conceptualizes OTFSV—and digital forms of violence more generally. Whilst harms arising from such forms of violence are gradually being acknowledged and discussed, both in the academic and law & policy making contexts,Footnote ⁵² the law lags behind the debates. Criminal law, and the criminal justice system have been slow in responding to the rise in OTFSV. That said, some forms of OTFSV, such as image-based sexual abuse (IBSA) and voyeurism, have been the subject of a speedy law reform in many countries across the world.Footnote ⁵³ In contrast, TBSA has not gained an equivalent level of recognition in the legal systems. As such, there is generally an absence of specific legal provisions dealing with these forms of sexual violence, with possibly a notable exception of IBSA provisions. This leaves non-image-based forms of TFSV largely undefined in law and, what follows, legally/criminally unaccounted for.

Other laws may serve as the basis for prosecuting online conduct involving TFSV—such as, for instance, provisions concerning threats, stalking or harassment, and communications offences.Footnote ⁵⁴ Although TFSV may indeed fall under these provisions, the act of TFSV would then need to satisfy the actus reus and mens rea for that particular offence, including different evidential thresholds. Given varying thresholds for these offences, this in turn leads to a highly fragmented and inconsistent approach to prosecuting TFSV which is likely to severely restrict the avenues of redress for the victims.

A further complicating factor is posed by the contextual aspect of the TFSV. Namely, that they largely take place in the online sphere. Although, as we argue elsewhere,Footnote ⁵⁵ the online nature of the offences per se is not an obstacle in ensuring criminal accountability for TFSV, it is proving problematic when it comes to consideration of TFSV due to the way in which underlying offences are characterized in law. For instance, some of the existing criminal law provisions in England & Wales—which, in principle could apply to cases involving online TFSV —tend to focus on concepts such as ‘proximity’ or ‘hearing’ which prove redundant in the context of acts taking place in the online sphere, especially on social media. This can be illustrated by section 4(1) of the Public Order Act 1986, amended in 2006, which makes it an offence to use threatening, abusive, or insulting words or behavior, or to display to another person any writing or visible representation which itself is threatening, abusive, or insulting with the intention of causing the person to whom threats are made to believe that, “immediate unlawful violence will be used against him” or “to provoke the immediate use of unlawful violence by that person.” The overreliance on the notions of physical proximity, immediacy or provocation of violence, as well as requirement that the threat must be heard by the person to whom the threat is directed, make the provision realistically redundant in the context of potential prosecution of online TFSV. Furthermore, in the UK, the key laws governing the communications offences—the Malicious Communications Act 1988Footnote ⁵⁶ (MCA 1988) and the Communications Act 2003 (CA 2003)Footnote ⁵⁷ are only partially suited for prosecution of modern day communication offences, including those involving TFSV. Primarily, the offences enshrined in the MCA 1988 and the CA 2003 focus on the misuse of a public network rather than the nature of the act committed or the effect it has on the recipient/victim. Furthermore, a very high threshold for both offences combined with the need to meet the public interest threshold for the prosecution make them unlikely avenues of redress in cases involving online TFSV. For instance, the threshold for satisfying a Communications Act offence under § 127, requires that the communication be “grossly offensive”—a threshold which is difficult to satisfy, and which has been the subject of criticism as a result.Footnote ⁵⁸ This evidential threshold, even where satisfied, does not mean that there will be a prosecution—the public interest test also must be satisfied separately. Together, these thresholds mean that even though a criminal offence exists, it is not necessarily frequently utilized.

II. Is the Criminal Justice System Fit for Dealing with OTFSV Cases?

Aside from the necessary legislative reform, broader questions arise about the capacity and suitability of the criminal justice system to deal with cases involving OTFSV. As the example of England & Wales shows, whilst there exists an overall commitment of the justice system to tackle cybercrimeFootnote ⁵⁹ —which, in theory, should include its gendered forms, such as OTFSV —in practice it is rather selective as to which cybercrimes are indeed tackled, largely to the exclusion of online forms of sexual violence, particularly those which are not image-based. For instance, the CPS Guidance on Social Media VAWG Offences states that “communications that contain images or videos of women with very serious injuries, or of women being raped, or of women being subjected to sadistic acts of violence, accompanied by text that suggests that such assaults/rape/acts are acceptable or desirable may well, depending on the context and circumstances, be considered grossly offensive.”Footnote ⁶⁰ As such, whilst the Guidance acknowledges the possible gross offensiveness—and, what follows, harmfulness—of text-based dimensions of OTFSV, it sees it as merely accompanying feature of IBSA rather than a stand-alone type of OTFSV which may warrant prosecution. Furthermore, the test for public interest in prosecuting social media offences, which may include OTFSV, is relatively high and, following CPS Guidelines, it is likely to be met relatively rarely.Footnote ⁶¹

Added to this is another complexity, the question of attitudes of the police and other actors in the justice system towards policing and accountability for acts of OTFSV. The misperceptions of the harmfulness of online abuse, especially where it takes sexual and/or gender-based forms, continue to permeate the justice system. Often seen as not ‘really’ violent—in the sense that offline acts of violence, such as knife crime or offline sexual offences, are—reports of online sexual abuse are often seen as a stretch on the limited police resources, even by the police chiefs.Footnote ⁶² The issue of resources is a valid one, albeit one that needs addressing through meaningful reform rather than being used as an excuse not to tackle online criminality.Footnote ⁶³ The volume of various forms of online violence, including OTFSV, frequently paired with the anonymity of the perpetrators and the lack of appropriate and efficient reporting mechanisms pose a barrier to the effective processing of such cases. It is therefore difficult to envisage meaningful change with regard to OTFSV given the overwhelmingly negative attitude towards regulating and establishing accountability for these forms of violence on the one hand, and the practical barriers on the other.

Lastly, there remain significant questions over whether a meaningful criminal law reform addressing OTFSV is even possible. It is unlikely that new criminal laws—however modern and adequate—will be capable of addressing the underpinning causes of OTFSV, including patriarchy, structural gender-based violence, pervasive gender stereotyping, and gender-based inequality which permeates the societal structures. Feminist critique of criminal law has long highlighted the gender-bias of the criminal law and the justice system which ignores experiences of women and fails to conceptualize, prosecute, punish and redress harms suffered by women—especially those arising in relation to sexual offences.Footnote ⁶⁴ Victim blaming, stereotypes surrounding gender roles, and trivialization of experiences of the victims of sexual violence are likely to be reproduced at trials of cases involving OTFSV due to the deeply entrenched biases of the justice system and actors involved in handling as well as adjudicating such cases. This, in turn raises doubts as to whether harms suffered by the victim of OTFSV would be legally recognized and redressed.

III. Is Internet Law Fit to Respond?

Beyond questions surrounding the suitability of the criminal justice system to respond to deal with cases of OTSFV, there is the caveat that a substantial portion of the discussion inevitably must also fall within the sphere of online regulation and platform regulation more specifically. Historically, where regulation has been part of the legislative agenda, the emphasis has fallen on attempting to ensure that the internet remains workable,Footnote ⁶⁵ but also protects the platform and website operators from burdensome responsibilities.Footnote ⁶⁶ In so doing, the legendary e-Commerce DirectiveFootnote ⁶⁷ is one of the first regional attempts to offer some element of regulation to internet platforms. It does not however actually regulate content—the content is something which is addressed through the introduction of the liability shield provisionsFootnote ⁶⁸ which have operated to ensure that platform operators are not responsible for the content that is posted and shared on their platforms. Criticisms of these provisions have abounded in recent years,Footnote ⁶⁹ not least from the platforms themselves who have sought to refrain from becoming involved in so-called editorial roles and therefore have been able to remain within the protections of the liability shields.

That said, given the rise of OTFSV, and other online harms, there has been a growing appetite for addressing online content through legal regulation. This has resulted in the Online Harms White PaperFootnote ⁷⁰ in the UK for example, which is intended to lead to legislation introducing a statutory duty of care.Footnote ⁷¹ This duty, which still requires some clarification in terms of its proposed operation, has been watered down from initial suggestions that executives of online platforms would be criminally prosecutable for non-compliance with the Online Harm regime.Footnote ⁷² Other examples of initiatives designed to address regulation include the German Network Enforcement ActFootnote ⁷³ which focuses on hate speech and hateful content online and takedown penalties for online platforms which are non-compliant within specific timescales. These are not the only examples—Austria,Footnote ⁷⁴ and FranceFootnote ⁷⁵ have also introduced proposed legislation to address elements of the internet regulation. While these legislative developments are not specifically designed to address the OTFSV phenomenon, they do represent a shift in attitudes towards internet regulation.

That said, the difficulty in all of these remains one of enforcement. Moreover, the proposed Digital Services Act, tabled by the European Parliament in December 2020Footnote ⁷⁶ indicates that there will be a greater obligation imposed on platform operators to ensure that online content is not illegal. This changes significantly the position operators have enjoyed under the current eCommerce Directive and its regime. There are nevertheless stark signals that this will change, and there is appetite from platform operatorsFootnote ⁷⁷ for this to be the case. What remains unseen is how these various provisions are supposed to be enforced and utilized. In all situations, especially for those in continental Europe, there remains the question of how varying national regimes fit into the framework proposed by the DSA.

If platforms and their operators are subjected to greater responsibilities and have to now take an active role in content control, as has been suggested through the DSA proposal, this is likely to cause potential problems for users. Where liability is to be imposed on the platform, this could lead to over-zealous platform actions to protect their own interests. They are after all, commercial and profit-driven entities who will be seeking to mitigate their potential liabilities. More importantly with no specific thought given to how the criminal law may be dealing with OTFSV, there seems to be little joined-up thinking being considered to ensure that OTFSV is a key consideration in changes to the legal frameworks address internet regulation. Similarly, no matter how well thought out proposed changes to legal regimes are, it is incredibly unlikely that any internet law will be able to tackle the underlying causes of OTFSV. In sum, while internet regulation may prove capable of responding, the issue remains as to how effective that response is likely to be.

The recent proposals for reform in the context of online harms in UK do not raise much hope for a meaningful and gender-sensitive capturing of such harms.Footnote ⁷⁸ Disappointingly, although not entirely unexpectedly, the emphasis of the reform seems to fall from considering harms associated with TBSA leaving, once again, little hope for paving the way to justice, however imperfect, for the victims of OTFSV.