Beyond ‘Revenge Porn’: The Continuum of Image-Based Sexual Abuse

In the last few years, many countries have introduced laws combating the phenomenon colloquially known as ‘revenge porn’. While new laws criminalising this practice represent a positive step forwards, the legislative response has been piecemeal and typically focuses only on the practices of vengeful ex-partners. Drawing on Liz Kelly’s (Surviving sexual violence. Polity Press, Cambridge, 1988) pioneering work, we suggest that ‘revenge porn’ should be understood as just one form of a range of gendered, sexualised forms of abuse which have common characteristics, forming what we are conceptualising as the ‘continuum of image-based sexual abuse’. Further, we argue that image-based sexual abuse is on a continuum with other forms of sexual violence. We suggest that this twin approach may enable a more comprehensive legislative and policy response that, in turn, will better reflect the harms to victim-survivors and lead to more appropriate and effective educative and preventative strategies.


Introduction
In the last few years, many countries have introduced laws combating the phenomenon colloquially known as 'revenge porn', with many more debating possible reform. 1 'Revenge porn' is typically understood as the non-consensual distribution of private, sexual images by a malicious ex-partner. It is a pernicious and gendered form of sexual abuse, the prevalence of which has increased exponentially with the ubiquity of the smart phone. While new laws criminalising this practice represent a positive step forward, the legislative response has been largely ad hoc. In particular, by focussing only, or mainly, on the paradigmatic case of the vengeful ex-partner, legislatures have failed to consider the range of abusive practices resulting from the non-consensual distribution and/or creation of private sexual images.
In this light, we propose that 'revenge porn' should be understood as just one form of a range of gendered, sexualised forms of abuse which have common characteristics forming what we have conceptualised as the 'continuum of imagebased sexual abuse'. 2 To develop this argument, in the first part of the article we analyse various forms of image-based sexual abuse, examining the variety and similarities of these practices. We then consider the extent to which this conceptualisation reflects the reality of sexual offending and the experiences of victim-survivors, including the varied nature of the harms suffered by differently situated victim-survivors which are more than the physical harms commonly associated with the criminal law: we argue that image-based sexual abuse forms part of the continuum of sexual violence. In essence, therefore, we are making two arguments: first, that there is a continuum of practices which together form our concept of image-based sexual abuse; and, secondly, that image-based sexual abuse is on a continuum with other forms of sexual violence. We conclude by suggesting that this twin approach may enable a more comprehensive legislative and policy response that, in turn, better reflects the harms to victim-survivors and lead to more effective educative and preventative strategies (McGlynn and Rackley 2017;Salter and Crofts 2015;Hampton 1998).

The Continuum of Image-Based Sexual Abuse
The concept of the continuum of sexual violence was first developed by Liz Kelly (1988) to explain the inter-relationships between different forms of sexual violence and to challenge the notion of a hierarchy of sexual offences. Her predominant concern was to provide the conceptual tools by which women's experiences of men's violence could be better understood as they were (and still are) not reflected in the 'legal codes or analytic categories' of existing research (Kelly 1988, 74). The concept also enables women to make sense of their own experiences 'by showing how ''typical'' and ''aberrant'' male behaviour shade into one another' (Kelly 1988, 75). In Kelly's work, a continuum is understood in two main ways. First, it is 'a continuous series of elements or events that pass into one another and which cannot be easily distinguished' (Kelly 1988, 76). The second meaning of continuum identifies a 'basic common character' underpinning and linking what might otherwise be seen as disparate phenomena (Kelly 1988, 76). The common character in sexual abuse is the 'abuse, intimidation, coercion, intrusion, threat and force' used to control (predominantly) women (Kelly 1988, 76). As Kelly (1988, 139) notes, identifying common characteristics is vital work for feminists in order to find the right 'names to describe women's experience[s]'. 3 Nonetheless, identifying 'common characteristics' must not be at the expense of recognising, and acting on, the differential experience of differently situated women. As Kelly herself has commented, the continuum concept is open to development, recognising the necessity of intersectional analysis (Kelly 2012). This understanding has been clearly articulated by Fiona Vera-Gray who comments that 'although all women and girls are in some way subject to gender discrimination, all women and girls are not discriminated against in the same way. Hierarchies of worth situate women and girls in relation to each other, as well as in relation to men and boys' (Vera-Gray 2017b). Further, we need to be attentive to how these 'hierarchical structures interact and intersect with gender inequality, and how its manifestation differs according to other markers of a woman's or a girl's social location' (Vera-Gray 2017b). This is vital not just in the development of conceptual analyses relating to violence against women (Choudhry 2016), but also for subsequent policy development (Strid et al. 2013).
In developing the idea of image-based sexual abuse, our aim is to provide both a new descriptive and a new conceptual tool by which to understand better the nature and extent of abuse experienced predominantly by women arising from the nonconsensual creation and/or distribution of private sexual images. 4 Image-based sexual abuse is a term to describe the variety and the nature of harms women experience. It provides a means of describing experiences 'as abuse without necessarily having to name it as a particular form of sexual violence' (Kelly 1988, 3 For a review of Kelly's continuum theory, including Liz Kelly's own re-assessment of her idea, see Brown and Walklate (2012). 4 The vast majority of victim-survivors are women (Whitmarsh 2015a, b;West Mercia Police 2016;Northumbria Police 2016;Wittes et al. 2016a, b), though men and boys are also targeted (Wolak and Finkelhor 2016). How victim-survivors experience these forms of abuse will obviously vary depending on their sexual identities, race and ethnicity, as well as socio-economic background, age and many other variables.
Beyond 'Revenge Porn': the Continuum of Image-Based… 27 157). It is also a valuable concept because it illuminates the abusive nature of the practices and the commonalities between seemingly disparate phenomena: it enables us to envision a continuum of image-based sexual abuse. Put differently, its breadth and flexibility creates a framework in which new experiences can be located and accurately understood as abusive-something which is especially important in this area where modes of perpetration rapidly change due to advances in technology. Drawing on Kelly's pioneering work, we make two arguments: first, that there is a continuum of practices that together form our concept of image-based sexual abuse; and, secondly, that image-based sexual abuse is on a continuum with other forms of sexual violence. 5 In relation to our first argument, the notion of the continuum underpins the development of our concept of image-based sexual abuse which encompasses all forms of the non-consensual creation and/or distribution of private sexual images. It includes, therefore, a range of abusive behaviours beyond the familiar example of 'revenge porn', such as 'sexualised photoshopping', sexual extortion (often labelled as 'sextortion'), 'upskirting', voyeurism and many other similar forms of sexualised abuse. We developed the concept in order to bring into focus the overlapping nature of various forms of abuse, including modes and motives of perpetration, and effects on victim-survivors. We recognised that there were 'no clearly defined and discrete analytic categories' (Kelly 1988, 76) into which these forms of abuse could suitably be placed, as existing legal and discursive categories fail to encompass or understand the range and extent of abuse being experienced. The idea of the continuum based around a 'basic common character' enables connections to be revealed between different forms of abuse with important discursive, policy and legal implications. In this way, we can move beyond the discrete categories of both law and public discourse, such as 'revenge porn' or 'upskirting', to recognise the 'commonality and connections' between different forms of abuse (Vera-Gray 2017a, 20-21).
Understood as a continuum, the concept of image-based sexual abuse is sufficiently broad and flexible both to embrace new ways of perpetrating, and experiencing, these forms of abuse. Thus far, law, policy and public discourse has tended to concentrate on specific categories of activity, harm, or particular motives, often focussing on one particular example, only to find other forms of abuse excluded and ignored. Further, the activities debated and/or prohibited are generally those already identified as legal categories and offences, or they are described and labelled according to media perceptions or perpetrators' motives. The forms of abuse, therefore, are not being defined or debated according to women's (and some men's) experiences. This piecemeal approach, which also largely ignores the differential experiences of women, enables legislatures and policy-makers to pay little attention to the sexual abuse faced predominantly by women.
In developing the concept of the continuum of image-based sexual abuse, we have identified the 'common character' of the various forms of abuse as follows: 5 The argument that image-based sexual abuse is on a continuum with other forms of sexual violence was first presented in a lecture given by Clare McGlynn, 'More than just 'revenge porn': challenging imagebased sexual abuse', at the conference 'More than Revenge: addressing the harms of 'revenge pornography', Monash University, 22 (i) the sexual nature of the imagery; (ii) the gendered nature of both perpetration and surviving the abuse (predominantly women as survivors of abuse and men as perpetrators); (iii) the sexualised nature of the harassment and abuse; (iv) the harms as breaches of fundamental rights to dignity, sexual autonomy and sexual expression; and, finally, (v) the minimisation of these forms of abuse in public discourse, law and policy. However, while all forms of image-based sexual abuse share a common sexual, sexualised and abusive essence or character, as we shall see, they are perpetrated in a wide variety and growing number of guises.

The Limits of 'Revenge Porn'
Public policy and media debate on this topic usually begins with the activity labelled as 'revenge porn'. This typically involves an ex-partner (usually a man) posting consensually created private sexual pictures or videos of their former partners (usually a woman) online and without consent, and in order to exact 'revenge' following the break-up of their relationship. The images are routinely reposted across the Internet, via social media, or on pornography websites including those specially dedicated to 'revenge porn'.
'Revenge porn' is disturbingly big business. Before it was shut down in 2012, Hunter Moore's notorious 'Is Anyone Up?' website, which regularly featured 'revenge porn', was said to receive over 300,000 unique visitors a day (Lee 2012). Though figures vary, and date quickly, it is estimated that there are around 3000 dedicated 'revenge porn' websites (DeKeseredy and Schwartz 2016), and more than 30 sites operating in the UK (Ridley 2015). These sites host images of women and sometimes men, but it the images of women that are more frequently viewed and commented on (Whitmarsh 2015a, b). Not surprisingly in this light, the evidence on prevalence suggests high levels of abuse and a gendered impact. In the first year following the criminalization of 'revenge porn' in England and Wales, over 200 cases were prosecuted (Crown Prosecution Service 2016, 11; BBC News 2016a). This is likely to be the tip of the iceberg, with earlier data revealing over 1000 incidents reported to police in a nine-month period (Sherlock 2016). Further, organisations supporting survivors of domestic abuse report the increasing use of technology, and specifically distribution and threats to distribute private sexual images, as part and parcel of the process of coercive control and domestic abuse. 6  support via the UK's Revenge Porn Helpline, the vast majority are women (Government Equalities Office 2015). 8 The 'revenge porn' moniker is also used to describe the growing problem of hacked private sexual images being shared online. In 2014, for example, over 200 sexual images of women, many well-known celebrities, were hacked or stolen and went viral (Farrell 2014) 9 ; and similar hacks occur with regularity (Hyland 2016;Daley 2016). While often labelled 'revenge porn', these incidents have also often been granted new monikers by the media-'Celebgate' or 'The Fappening'-terms which invariably serve to minimise the harms experienced by victim-survivors.
The harms experienced as a result of all of these forms of abuse are considerable (McGlynn and Rackley 2017; Bates 2017) and are evidence of the common character of different forms of image-based sexual abuse. As well as the abuse of trust and privacy, and the breach of the victim-survivor's fundamental rights to dignity, sexual autonomy and sexual expression, there is often vicious harassment and abuse which is highly gendered and sexualised (Citron and Franks 2014, 353). The sexual double standard prevails as the abuse meted out typically castigates women for traversing expected norms of femininity and sexuality, and often involves sexual and sexualised language and threats, including rape threats (Citron and Franks 2014, 353;Jane 2014). The harassment becomes 'a team sport' as 'cyber-mobs' 'compete to be the most offensive, the most abusive' (Citron 2014, 5). Often social media profiles, names and addresses are revealed online, resulting in the very real threat of physical violence.
In recounting their experiences of harassment and abuse following 'revenge porn', victim-survivors have described the harms suffered as a form of sexual assault. Actor Jennifer Lawrence, whose private sexual photos were hacked and distributed worldwide, described what happened as a 'sex crime' (Rifkind 2014;Arthur 2014;Rickman 2014). Other victim-survivors have made similar arguments in public debates, with YouTuber Chrissy Chambers describing her experiences as 'sexual assault' 10 and another victim-survivor stating that '[h]ow anyone can fail to see revenge porn as a sexual crime is beyond me' (The Yorkshire Post 2016).
The term 'revenge porn', therefore, neither adequately describes the activities for which it is used, nor does it convey the nature and extent of the harms perpetrated. Further, while many of the new laws enacted in this area cover some aspects of 'revenge porn', most remain restricted in scope and offer only limited redress. 11 As discussed below, the term is at once too narrow and applied too widely. The continuum of image-based sexual abuse goes beyond this and it includes the nonconsensual creation and distribution of private sexual images. 8 Statistics from the US Cyber Civil Rights Initiative (2013) found that 90% of the victim-survivors of 'revenge porn' were women. See also (Lenhart et al. 2016). 9 The perpetrator, Ryan Collins was sentenced to 18 months in prison in October 2016 (France-Presse 2016). 10 Chambers states ''the reality is that what happened to me was a form of sexual assault, and it should be treated that way by the law'', see https://www.change.org/p/protect-victims-end-revenge-porn-now Accessed 16 November 2016. 11 Such as the requirement in English law, for example, that the offender has a direct, not reckless, intention to cause the specific victim-survivor distress (Gillespie 2015).

The Confines of Conventional Voyeurism
There are many ways in which private sexual images are created without consent, some of which are covered by current criminal laws, an example being voyeurism. Conventionally understood, voyeurism is the surreptitious viewing, and/or photographing or recording of images, of sexual or 'private acts' for the purposes of sexual gratification, where the perpetrator knows the other person does not consent to being observed for sexual gratification, with perpetrators colloquially described as 'peeping toms'. 12 In England and Wales, this conduct appears to be increasing in prevalence with reported incidents rising steeply (Office for National Statistics 2015, Appendix tables). Nonetheless, the laws on voyeurism provide a paradigmatic example of existing social and legal categories focussing only on narrowly defined conduct and from a perpetrator's perspective, thereby failing to reflect the experiences of victimsurvivors. The commonplace label and reference to 'peeping' also serves to minimise the impact and significance of the conduct, as with so many other labels in this area.
Specifically, voyeurism laws tackle a very particular form of conduct, namely spying on 'private' acts (conventionally understood) for the purposes of sexual gratification. It does not include, therefore, the many ways in which voyeuristic abuse is now perpetrated. First, the 'sexual motive' requirement precludes those who perpetrate this form of abuse for a myriad of other purposes including causing distress to the victim-survivor, to secure notoriety or bond with a friendship group, or for financial gain (Minting 2016;Newsletter 2015). 13 In recent years, there have been a number of high profile cases involving the non-consensual recording and distribution of private sexual images in which the perpetrators, while not necessarily involving the necessary mens rea for the offence of voyeurism, have caused their victims significant harms. For example, in 2010, US university student, Tyler Clementi committed suicide after his roommate covertly filmed, and later streamed, him having a 'sexual encounter' with a man (Pilkington 2010). In Australia, a woman was left, in the words of the judge sentencing the perpetrators, with 'her whole world … shattered, her dignity stolen … her self-worth destroyed' 14 after a film of her having consensual sex was live-streamed, without her knowledge, to six others watching next door (Richardson 2012, 151). While in these notorious cases the police and prosecutors have shoe-horned the circumstances into other offences and secured convictions, it is unlikely to be happening in all such cases. Victim-survivors will not know there are other offences that might encompass their experiences, and many police, prosecutors and perpetrators are likely to be similarly unaware. Furthermore, shoe-horning practices into conventional privacy-related offences risks obscuring the nature of the abuse and reducing any potential expressive effect of the criminal law.
Further, technological advances have created means of harassment and abuse, such as the taking of private sexual images in public, which do not fit within the 12 For example, Sexual Offences Act 2003 section 67-68. 13 See for example, R v D where a defendant claimed the filming was part of a photographic project: http://devonchambers.co.uk/component/content/index.php?option=com_content&view= article&id=36&Itemid=149. Accessed 16 November 2016. 14 Acting Justice Nield quoted in Knaus and Inman (2013). traditional confines of voyeurism. 'Private', in the context of voyeurism, means a specific place conventionally understood as a distinct location, such as the home, toilet or changing room. Such provisions are a clear demonstration of how legal categories based on supposedly isolated forms of conduct and for specific purposes leave many victim-survivors unprotected. Laws, as currently interpreted and enacted, therefore largely fail to cover the range of experiences of abuse. If we focus instead on the harms experienced by victim-survivors, and acknowledge a range of activities on a continuum, we would recognise the links between different forms of abuse and act accordingly. The nature and harms of voyeurism discussed above help to identify the 'common character' of image-based sexual abuse, as further evidenced by examining the phenomenon of 'upskirting'.

The Invisibility of 'Upskirting'
The most glaring omission from conventional voyeurism laws is the failure to encompass the practice of 'upskirting'. This involves the non-consensual taking of images of an individual's pubic area underneath their outer clothing in public places (Gillespie 2008, 376;Spencer 2008;McGlynn and Downes 2015). As with other media-friendly monikers, the term 'upskirting' is problematic. While accurately describing the practices involved, it minimises the nature and impact of the abuse. Of course, the belittling of sexual harm is not new. However, in this case, the terminology itself works to hinder the understanding of these activities as harmful. This is easily demonstrated by a front page headline of the UK's largest selling daily tabloid newspaper in October 2016 reporting on a case of 'upskirting' of a flight attendant by a passenger on board a British Airways (BA) plane which read: 'Jet perv films BA girl's undercarriage' (Moyes 2016).
And, unsurprisingly, the practice is on the rise. One reason for this is that the ubiquity of smartphones and internet access make it easier to take and distribute the images. There is also an alarming array of devices specially designed for these purposes such as a camera hidden in the perpetrator's shoes (see e.g. Chong 2015; EJInsight 2015). Women-and the images are almost exclusively of women (Perrie 2016; BBC News 2016b)-have been furtively photographed or filmed on university campuses (Siegel et al. 2006, 11-12), trains (MacLaughlin 2016) and in supermarkets (Yensi 2015). The images regularly appear online on the many pornography websites dedicated to this genre. One such website exposed in a national UK newspaper reportedly receives 70,000 views a day and is valued at £130 million (Perring 2015).
Such forms of covert creation of images highlight a gap in the current ad hoc legal regimes. 15 As 'upskirt' images are taken in public, they generally fall outside 15 This is despite voyeurism provisions being enacted by the Government in 2002 stating that they wanted 'cases where a photographer takes indecent photographs of someone without their consent and for example posts them on the Internet or in a pornographic magazine to be treated particularly seriously by the courts' (Home Office 2002, 32). Some countries have adopted specific laws covering 'upskirting', usually following a public scandal. For example, Crimes Act 1900 A1900-40 Republication No 107, section 61(B)(5) in the Australian Capital Territory and Summary Offences Amendment (Upskirting) Act 2007 in Victoria, Australia. of conventional voyeurism offences. Most pertinently, the person is not undertaking a 'private act'. Indeed, as Alisdair Gillespie has noted, the nature of the act they are doing-for example walking around a supermarket-is typically the very opposite of a 'private act' (Gillespie 2008, 376-377). And yet such images are archetypally 'private'. They are a recording of the genital area and/or underwear of the victimsurvivor who had no intention of revealing themselves in such a way. 16 Therefore, as well as current laws failing to encompass the practices known as 'upskirting', we can see that the nature and harms share commonalities with other forms of imagebased sexual abuse. The abuse is focused on sexual and sexualised images, the victimisation and perpetration is gendered, 17 the harms are often minimised and the harms are various and extend beyond any immediate risk to physical safety.

Recognising Sexualised Photoshopping
The ad hoc nature of legal regimes can also be noted when we consider the implications of photoshopping (digital manipulation of images). The concern arises where, without consent, a pornographic image is superimposed onto an individual's head/body part, such that it looks as if that individual is engaged in the pornographic activity: what we have called 'sexualised photoshopping'. 18 Advances in technology now mean that it is often impossible to tell that edits have been made to an original image and indications suggest that a not inconsiderable proportion of nonconsensually distributed images are photoshopped. 19 Indeed, a market has been created around this practice, with websites offering to produce such images. 20 While these generated images are not 'real', the harms are very real. Many victimsurvivors experience a breach of trust as well as a loss of dignity, harassment and/or abuse (e.g. Jeffreys 2014; Blott and Martin 2016). For them it matters little whether the images were 'originally' sexual, or photoshopped. As the organisation Australian Women Against Violence Alliance (2016, 6) note: 'the fact that an image has been altered, or is even composed of images taken of different women, does not diminish the potential harm resulting from its dissemination'. 16 This situation is in distinct contrast to the individual who has voluntarily exposed themselves and relinquished control over who sees the image, such as the 'streaker' at a sports event. Such situations can easily be excluded from the remit of any criminal offence by excluding 'voluntary exposure' as we recommended in our submission to the Scottish Government consultation on the Abusive Behaviour and Sexual Harm (Scotland) Bill (McGlynn and Rackley 2015a, para 4.2.2) following the provision in the Illinois Criminal Code 2012, section 11-23.5(c)(3). 17 Harms will be differentially experienced, including on the basis of sexuality, race, ethnicity, age and social class. The emphasis here is also on the commonality of gender as a locus for these forms of abuse. 18 In definitional terms, we suggest that even if the original photo was public, it becomes a private image when it has been edited in this way because the person has lost any agency and freedom to choose the portrayal of their sexual self. Further, the harms are no less than in cases of other forms of image-based sexual abuse. 19 In an investigation carried about by Charlotte Laws, who has campaigned in the US for laws covering 'revenge porn', an estimated twelve percent of 'non-consensual pornography' has been photoshopped, or otherwise edited and manipulated (Gladstone and Laws 2013). 20 For example, the website 4Chan. For a discussion, see Gander (2016).
Sexualised photoshopping challenges legal regimes to be more reflective of the harms which victim-survivors suffer and of changing modes of perpetration. We can see here abusive practices 'shading' into one another (Kelly 1988). In terms of legal categories in English law, however, the non-consensual creation and distribution of the photoshopped image is not criminalised. 21 Indeed, in responding to a request by one of us to extend the law to cover such manipulated images, a Government minister claimed that manipulated images, while 'still distressing', do 'not have the potential to cause the same degree of harm as the disclosure of images that record real private, sexual events' (Ministry of Justice 2016). The experiences of victimsurvivors suggest otherwise. In particular, we can see similarities between the nature and harms of other forms of image-based sexual abuse and sexualised photoshopping. It is a sexualised and gendered phenomenon.

The Growth of Sexual Extortion
Sexual extortion (often colloquially known as 'sextortion') generally describes practices whereby perpetrators coerce individuals, often young people, into creating and/or sharing private sexual images, as well as deploying threats to force further image-creation. Alternatively, webcams, phones or data storage areas such as the iCloud are hacked to obtain consensually taken images or videos without the person's consent, with perpetrators using threats and blackmail to solicit further images and/or sexual practices and, in some cases, money (Wolak and Finkelhor 2016;Wittes et al. 2016a;Powell 2015, 2016b).
Perpetrators of sexual extortion are generally men, 22 with children and young adults (both women and men) the common target. 23 Where the victim-survivors are adults, 'women are the primary targets', leading researchers to conclude that 'adult sextortion therefore appears to be a species of violence against women' (Wittes et al. 2016a, 4). 24 In a recent study, one perpetrator was found to have '15,000 webcam-video captures, 900 audio recordings, and 13,000 screen captures' that were predominately of women (Wittes et al. 2016a, 2). Harms experienced by victim-survivors are similar to those who have spoken out following acts of 'revenge porn', including adverse impacts on their mental health as well as fear of physical assault (Wolak and Finkelhor 2016, 31-33). Indeed, the authors of the aforementioned study described sexual extortion as 'akin to sexual assault' and a form of 'remote sexual violence' (Wittes et al. 2016b, 4). One victim-survivor described feeling 'like [they were] being raped through a phone' (Wolak and Finkelhor 2016, 31). 21 While English law fails in this regard, the Scottish Government included digital manipulation of images in its recent provisions [Abusive Behaviour and Sexual Harm (Scotland) Act 2016, sections 2 and 3] and the Irish Law Commission recommends such a provision (Irish Law Reform Commission 2016, paras 2.202). 22 Wittes et al. (2016a, 12) found that all the perpetrators were male. Wolak and Finkelhor (2016, 9) found that there were female perpetuators as well. 23 Wittes et al. (2016a, 12) found 71% involved only victim-survivors who were minors. See also the news report of the thousands of images of children found by the police in Scotland (Forster 2016). 24 See also Segall (2016).
The practices known as sexual extortion demonstrate the changing ways in which sexual abuse is perpetrated, here without contact between individuals and using new technologies and modes of operation not envisaged even just a few years ago. We can see the links between sexual extortion and other forms of image-based sexual abuse in terms of the impacts on victim-survivors and their experience of the harms as forms of sexual abuse, as well as the harms being generated through the nonconsensual creation and/or distribution of private sexual images. In terms of legal redress, in many jurisdictions there are a range of existing criminal offences which cover some aspects of 'sexual extortion', such as blackmail or causing another to engage in sexual activity. Thus, the means of redress is often available, though great efforts are often required to convince prosecutorial authorities that existing laws can be used in these circumstances. There similarly remains work to be done to challenge long-standing views of sexual offending as only involving contact, and abuse perpetrated remotely being just as significant and long-lasting as physical harms.

Images of Sexual Assault and Rape
One of the most disturbing examples of non-consensually created private sexual images involves the recording of rapes or other forms of sexual assault. 25 In a notorious US case from 2013, two high school footballers were found guilty of raping an incapacitated young woman after pictures and films of the crime were distributed across social media. The images were used to further harass and humiliate the victim-survivor, blaming her for the assaults and including death threats against her (BBC News 2013;Carpentier 2013). 26 The legal response in some jurisdictions to the filming and recording of sexual assault and rape shows that the divide between the real world and the virtual world is no longer tenable; both the physical assault and the film causes harm. However, there is often no legal response to the harm caused by threats to (further) distribute the images. It is not just the recording of the sexual assault that causes harm, but also the subsequent and continuing coercive use of these images. Often victimsurvivors are threatened with the exposure of the video should they report the assault to the police. Too often, what is missing is the recognition of the harm caused by such a threat, notwithstanding the perpetrator's ability or otherwise to make good on it.
We have already noted that the focus on 'revenge' skews reality. In this context, while some perpetrators who record sexual assaults might be motivated by sexual gratification, more typically such actions and images are part of a form of (male) bonding or initiation ritual or as a misguided attempt to gain notoriety (DeKeseredy and Schwartz 2016). Further, a law that focuses on the motivations of the perpetrator, risks overlooking the experiences of the victim-survivors and the harms caused.

Locating Image-Based Sexual Abuse on the Continuum of Sexual Violence
So where does this leave us? When Kelly (1988, 76) developed the concept of the continuum of sexual violence, she explained its value as enabling us to document and name the range of abuse, intimidation, threat and force that women experience, particularly where there are no legal or other discrete categories already in common usage. Further, it enables us to see the interconnecting nature of many experiences. In this context, as well as arguing that there is a continuum of image-based sexual abuse, we are suggesting that image-based sexual abuse is on a continuum with other forms of sexual violence. In this way, we are drawing the connections between different forms of sexual violence, enabling a more holistic legal and policy response.
This approach is central to understanding image-based sexual abuse as sexual abuse; but it also has potentially significant discursive, legislative and policy consequences. Image-based sexual abuse is on the continuum of sexual violence because it shares common characteristics with other forms of sexual violence. As identified above, the abuse is sexualised, sexual imagery is the focus of the abuse, and women experience these phenomena as a form of sexual assault. The harms experienced include those identified by Kelly (1988, 76) in describing her sexual violence continuum including 'abuse, intimidation, coercion, intrusion, threat and force'. In addition, two of the authors have argued elsewhere that the harms of image-based sexual abuse include the violation of fundamental rights to sexual autonomy, integrity and sexual expression (McGlynn and Rackley 2017). Taken together, these elements would suggest to us that image-based sexual abuse is sufficiently similar to other experiences of sexual violence that it sits on a continuum with them.
However, in the main, legislators and policy-makers have yet to recognise imagebased sexual abuse as a form of sexual violence. While Israel currently understands forms of image-based sexual abuse as a sexual offence 27 and the Australian Legal and Constitutional Affairs References Committee (2016, para 5.37) discussed this phenomenon as a 'sex crime', there is no general acceptance of this argument. The Irish Law Commission (2016, para 1.07) recently rejected the idea that forms of abuse including 'revenge porn' and 'upskirting' are sexual offences, emphasising instead the breach of privacy involved in such activities. While these acts of abuse are without doubt egregious breaches of privacy, conceptualising the harm in this way inhibits recognition of the gendered, sexualised and abusive nature of the practices of image-based sexual abuse (McGlynn 2017). The UK Government has similarly rejected the idea of image-based sexual abuse as a form of sexual offending on the basis that the offence 'requires no element of sexual contact, sexual intent or gratification' (Ministry of Justice 2016; The Yorkshire Post 2016). Worryingly, this entirely misunderstands the nature of sexual offending. Sexual offences are so because of the mode of perpetration (sexual acts), rather than the motive, and are designed to protect sexual autonomy. While some sexual offences do require a 'sexual gratification' motive, this is not exclusively so. 28 While the impetus to sexually offend is clearly complex, we do know that offences are not carried out only for the purpose of sexual gratification. One study found that 'the most common type of rapist is one who is motivated by power and control' (Robertiello and Terry 2007, 511). Others identified the motives as including revenge and punishment, sexual access of unwilling/unavailable women, recreation and adventure (Scully and Marolla 1985;Mann and Hollin 2007, 3-9). While a United Nations study found that 'sexual entitlement' was an important motivation, it also found that prominent motivations included entertainment, as well as anger and punishment (Fulu et al. 2013). Therefore, while sexual gratification can be part of a motive for sexual offending, it is only part and, moreover, even where the intent may include sexual access, this is closely associated with beliefs relating to entitlement to sex, disregard for consent and seeing sexual assault as a means of collective punishment of women.
Further, sexual assault does not require contact. The approach of the UK Government entirely misunderstands the existing law which includes a number of offences, already labelled as sexual, that do not require contact, including preparatory offences such as grooming, causing individuals to perform sexual acts (without contact with the perpetrator), coercing individuals to watch sexual acts and indecent exposure, to name a few (Sexual Offences Act 2003). 29 In sexual offences such as the Scottish offence of coercing a person to look at a sexual image, the harm to the victim-survivor is the 'invasion of his or her sexual autonomy'. 30 Moreover, a focus on physical contact fails to recognise the blurring between online and offline worlds both for experiencing and perpetrating abuse (Bluett-Boyd et al. 2013). One commonality of the forms of image-based sexual abuse is the blurring of the real and virtual worlds. For example, for the victim-survivors of sexual extortion, who felt 'raped through a phone' (Wolak and Finkelhor 2016, 31), the strict dichotomy (or hierarchy) between physical acts of sexual offending and online abuse is not the reality and is no longer tenable.
Accordingly, we are suggesting that image-based sexual abuse is a form of sexual violence, and therefore on a continuum with other forms of sexual violence. This approach aids understanding of the phenomenon as it better reflects the experiences of victim-survivors and helps to ensure a broad approach to legislative and policy responses, including those which reflect what Elizabeth Stanko describes as a 'woman-defined understanding of what is threatening and … potentially violent' (Stanko 1985, 10; discussed in Vera-Gray 2017a, 6).

Beyond 'Revenge Porn': Image-Based Sexual Abuse
When examples of abuse, from so-called 'revenge porn' to recording of sexual assaults, are examined alongside each other, it becomes possible to see the continuum of practices and harms. When we then place these on a continuum of sexual violence, we see that these abusive practices are not only part of a larger, gendered phenomenon, but that currently public discourse, law and policy is not dealing with the seriousness of the harms. Our analysis has demonstrated the gendered, sexualised and abusive extent and nature of these practices and how victim-survivors experience them as a form of sexual assault. It is for these reasons that we have developed the term and concept 'image-based sexual abuse'.
But what's in a name? As already noted, there is general agreement that the term 'revenge porn' is problematic. The term 'pornography' wrongly focuses attention on the 'perceived actions by the victim' rather than on the perpetrator (Australian Legal and Constitutional Affairs Reference Committee 2016, para 5.4). At the same time, the language of 'pornography' has led to some legislatures, including (at least initially) England and Wales, to suggest that regulation is dependent on an image being 'pornographic'. 31 More generally, while some would describe any image featuring the 'graphical sexually explicit subordination of women' as pornographic, and in turn as abusive (MacKinnon 1984, 321), the term 'pornography' lends a sense of choice and legitimacy and does not sufficiently capture the non-consensual nature of the practices. Equally problematic is the label 'revenge' which spotlights the motive of the perpetrator and overlooks alternative reasons for the sharing of images, such as group bonding, notoriety or financial gain. For these reasons, the terms 'non-consensual pornography' (Franks 2015) and 'involuntary porn' (Barmore 2015) while having the benefit of describing a broader range of practices than 'revenge porn', still raise concerns regarding the use of the language of 'pornography'. In particular, 'non-consensual pornography' could be misunderstood as referring to specific genres of pornography featuring non-consent and abuse, which would limit its use as a tool for referring to broader practices. It also raises questions as to whether placing 'non-consensual' in front of 'pornography' assumes and thus legitimates all other forms of pornography as 'consensual' (Tyler 2016). Further, the language of porn risks eroticising the harms of image-based sexual abuse, and may encourage the media's salacious interest and reporting of these phenomena.
In avoiding the language of pornography, others have used terms such as 'intimate' and 'sexually explicit' (Henry and Powell 2016a, 7;Henry and Powell 2015, 113). These too have their drawbacks. The term 'sexually explicit', for example, is likely to limit the range of images covered to those which include considerable nudity and/or sexual acts, perhaps overlooking images that are equally harmful but where the woman is, for example, wearing underwear. The term 'intimate'-adopted by a number of countries in their legislation and policy documents-is potentially broader than sexual as it could be used to include nonsexual, but still private, material. However, while the non-consensual disclosure of private, non-sexual images can be no less distressing to the individual depicted, 32 we focus on 'sexual' images in order to highlight the need to respond to the specific sexual harms perpetrated against women through the attack on their sexual self, sexual autonomy and agency. A further option is 'image-based sexual exploitation' used by Anastasia Powell and Nicola Henry (2016b) which draws on the terminology of child exploitation material (Henry and Powell 2016b), and others have referred to 'intimate image exploitation' (Legal Voice 2015; see also Anderson and Prasad 2015). The language of exploitation helpfully mirrors the approach to material involving children, and could encompass a wider range of practices which might not be characterised as 'abusive'. Nonetheless, we argue that the term 'abuse' better explains the harms experienced by victim-survivors and emphasises the similarities with other forms of sexual abuse. 33 For these reasons, terminology matters because when definitions are too narrow, this works to restrict redress and support. It matters because media-friendly monikers, though attention grabbing, at best downplay and at worst erase the everyday reality of abuse, harassment and violence of all forms of image-based sexual abuse. Terms such as 'upskirt' and 'revenge porn' have the effect of minimising the harms. As a result, many see image-based sexual abuse as a 'harmless prank' undertaken by 'misunderstood tricksters' (Jane 2014, 539). In this culture, perpetrators are 'rarely reprimanded', while victim-survivors are 'frequently chastised as hypersensitive or humourless' (Jane 2014, 539). We can also see parallels with this approach in Fiona Vera-Gray's (2017a) work on street harassment which also draws on Kelly's continuum concept. In drawing attention to, and problematising, men's everyday and routine intrusions, which are so often minimised both by society and women themselves, Vera-Gray demands that we recognise their significance. In a similar vein, just as routine street harassment can have a cumulative effect on women, so too can the normalised harms of imagebased sexual abuse. When we see forms of image-based sexual abuse, or online harassment, being defined as 'normal', this is a societal justification for (predominantly) men's behaviour enabling women's complaints to be dismissed 32 For example, the distribution of the image of a Muslim woman without her usually worn headscarf (Scottish Government 2015, para 37). 33 The term 'image-based sexual violence' was suggested by a representative of Dublin Women's Aid at the 2016 Irish Association of Criminal Justice Research and Development conference. Such a term can draw parallels with other forms of sexual violence. However, the risk is that the term 'violence' signals physical violence to many. We hope the term 'abuse' has greater breadth and resonance in public, policy and legislative debates. (Kelly 1988, 104). The value, therefore, of the concept of image-based sexual abuse, and in particular the naming of this as sexual abuse, is to reflect women's experiences and to resist the minimisation of these forms of harm.
For these reasons, we argue that the term 'image-based sexual abuse' immediately and accurately conveys both the focus on the non-consensual creation and/or distribution of private sexual images and on the seriousness of the harms. To this end, it has a valuable descriptive role. We have also developed 'image-based sexual abuse' as a concept that explains the common character of these forms of abuse, and enables us to examine together this 'series of elements or events that pass into each other and cannot be easily distinguished' (Kelly 1988, 76). In addition, we argue that the value of this term is that it also identifies the practices under consideration as a form of sexual abuse. And so, while we acknowledge that different cultures and political contexts may often demand alternative terminology and concepts, 'image-based sexual abuse', as defined and developed here, more accurately captures the extent and nature of the harm.
Accordingly, as we have argued elsewhere, our concept of image-based sexual abuse demands legislative and policy responses which recognise image-based sexual abuse for what it really is-a form of sexual violence (McGlynn and Rackley 2017). As such it demands a response that is both preventative and punitive; that provides both criminal and civil law options for victim-survivors; and which informs a culture-both within and outwith the legal system-in which the actual and potential harms of these practices are addressed.

Conclusions
Writing in the late 1980s, Kelly discussed how women were encouraged to deny or minimise the impact of obscene phone calls, flashing, 'peeping' and many other forms of abuse (Kelly 1988, 102). Fast forward to today, and women are still being told to 'just ignore' such conduct, as well as now being advised not to take online abuse or various forms of image-based sexual abuse so seriously; to get a sense of humour, to take it easy. Indeed, public discourse minimises women's experiences through the use of media-friendly terminology and the legislative focus on particular activities excludes other experiences and fails to appreciate the scale of the problem. Kelly argued that we needed the right 'names' (Kelly 1988, 139). We suggest that it is time to 'name' these practices as image-based sexual abuse.
Image-based sexual abuse exposes the variety of practices that shade into one another, and it encompasses creation, distribution and threats. It thus highlights the overlapping characteristics of these practices, exposing the common denominator as harassment and abuse, predominantly against women: these practices form the continuum of image-based sexual abuse. Our concept has the added benefit of being more flexible and future-proof. The examples of image-based sexual abuse discussed here are necessarily indicative and are not exhaustive. Unlike its media-friendly forerunners, naming the many and varied practices of image-based sexual abuse as a single inclusive concept encompasses forms of abuse as yet unimagined or, even, possible. And, in so doing, the concept of image-based sexual abuse provides law and policy makers with a stronger base from which to tackle its pervasive and pernicious grip. Finally, the abuse is gendered and sexualised being experienced in a myriad of different ways, victim-survivors experience it as a form of sexual abuse, and it is part of a wider picture of inequality for women where there is a high prevalence of violence against women. This is why image-based sexual abuse is also on a continuum with other forms of sexual violence. We hope that this twin approach may enable legislative and policy responses that better reflect the particular harms and experiences of victim-survivors, leading to more appropriate and effective educative and preventative strategies.