Self-produced images, sexting, coercion and children’s rights

Converging sources of data would suggest that the volume of images and videos which involve minors is increasing and places a burden on law enforcement in terms of victim identification as well as prosecution. The terminology used in relation to these activities is problematic, as reference to self-produced sexual images would seem to result in victim-blaming. While there is an acknowledgement that there are diverse motives in the creating and sharing of this content, in many jurisdictions these images are seen as meeting the criteria for illegality and may result in prosecutions under child pornography legislation. This potentially criminalises (or at least problematises) the consensual creation and sharing of sexually explicit content between minors. Concerns regarding this sit alongside growing disquiet that this disregards children’s rights to sexual autonomy and decision making.


Introduction
In 2002 Nokia produced its first camera phone which was equipped with picturemessaging capabilities, known as multimedia messaging services (MMS), and which was made widely available to the general public 1 Within five years over one billion camera phones had been bought and increasingly mobile devices included a camera function. In the UK, by 2021 smartphones (the successor to camera phones) appear to be the 'base layer' of connectivity, with more users in lower socio-economic groups relying on this device for internet access without a computer. 2 OfCom also reported in 2021 that 91% of 12-15-year-olds owned their own smartphone. The EU Kids Online 2020 3 report indicated that across 19 countries there was a continuation from 2013-14 in the use of the smartphone as the way that most children accessed the Internet. Importantly, their portability and the ability to personalise them meant that smartphones were "now integrated into different social contexts and activities" (p. 18). The easy availability of camera phones and the affordances of particular technologies have made it possible for people of all ages to create and share digital artefacts that relate to all aspects of everyday life. 4 Increasingly, this includes sexual lives where digital media have transformed how we initiate, maintain, and terminate our intimate relationships. 5 A recent review found that in spite of its risks, the sharing of self-made sexually explicit images is a modern form of sexual communication and is increasingly used within dating and romantic relationships by adults and adolescents. 6 Earlier work by the same research group discussed how sexually explicit photographs can be sent through several online and mobile platforms and that these inevitably change as technology develops. 7 Earlier research 8 indicated that digital sexually explicit photographs and videos sharing by young adults largely took place through text messaging or, in a study of adolescents, through Blackberry Messenger. 9 The availability of smartphones and new communication applications has meant a move to smartphone applications such as Snapchat and WhatsApp to share sexually explicit content. Participants in Van Ouytsel et al.'s study also talked of similar applications such as Facebook Messenger, Kik, Instagram and iMessage being used to share sexual content. WhatsApp was seen to be a fast way to upload content and, of importance, the application deploys end-to-end encryption (E2EE) which means the only people who can access the data are the sender and the intended recipient.

Sexting
The sharing of sexually explicit media is often described as sexting (although this may not be a term used in everyday language 10 and is an activity most frequently seen in young adults. 11 Sexting has been seen as one of the ways in which adoles-cents use digital technologies to manage developmental tasks relating to their emerging sexuality, which include experimenting with sexual expression, constructing sexual identities, and establishing intimate relationships. 12 Taken together, the research  evidence on young people's motivations for sexting suggests that despite some notable pressures and anxieties around sexting, in the vast majority of cases the process takes place within either a romantic relationship or as a means of adolescent explorations of sexuality and identity creation. 13 The sharing of sexually explicit media by adolescents is increasingly seen as 'the norm' (which is not to say that the majority of children share sexually explicit content). 14 Sexting can be defined as the sending or posting of sexually suggestive text messages and images, including nude or semi-nude photographs, via mobiles or over the Internet, 15 although definitions vary in whether they include texts as well as images. 16 However, sexting is increasingly defined as the sending, receiving, and forwarding of nude, semi-nude, or sexually explicit images within digital forms of communication 17 although there is no consensus regarding its definition and measurements. 18 In fact, defining sexting, particularly in the context of minors, is increasingly seen as problematic. The Terminology Guidelines for the Protection of Children from Sexual Exploitation and Sexual Abuse 19 argued that there was a potential risk in using terms such as 'self-generated' or 'selfdirected' (as opposed to coercive) when talking about the sharing of compromising sexual pictures or videos as this could imply that the child was to blame for the abuse that may follow from the generation of these images and their exploitative distribution. Sexts are a form of self-generated sexually explicit content, but the practice of sexting is remarkably varied in terms of context, meaning and intention. In most instances it is unclear how 'sexually explicit' is operationalised or measured. In the US, the term 'self-produced child pornography' is still used to describe these images and videos 20 and the existing legislation clearly struggles with the fact that young people may occupy dual roles of both victim and perpetrator of this form of sexual exploitation.
In June 2021 the Technology Coalition, at their first multi-stakeholder forum, suggested the use of the term 'Perceived First-Person Child Sexual Abuse Material' (PFP CSAM) which would avoid inadvertent victim blaming and covered both production and distribution regardless of intention. Their proposed definition of PFP CSAM was, "Sexualised visual depictions of a child that are generated without the full knowledge, consent and participation (for example coercion, blackmail or grooming) of the child and without the physical presence of an instigator AND/OR that may have been originally voluntarily produced by the minor child, but then is distributed or shared with others without the child's full knowledge or consent". They also noted that consent should not only relate to a legal capacity to consent to sexual activity in any jurisdiction but should reflect the "age-appropriate state of mind" and whether creating the images was coerced. 21 In the UK, the All Party Parliamentary Group Report (APPG) in September 2021 recommended the use of the term 'first person produced imagery' using the definition created by the Technology Coalition to replace the existing term used in the UK 'self-generated indecent images of children'. 22 The motive for this was to avoid any blame being placed on the child and to ensure that children could have this content removed from the Internet as easily as possible. It remains to be seen whether this term will be more widely adopted.

The role of coercion
It is not always easy to know who is behind the production of such content and whether the child has been pressured, manipulated or coerced into making it. Inconsistencies among definitions and the way that this behaviour is measured (including the use of self-reports) makes it difficult to assess sexting prevalence amongst adolescents, although a meta-analysis of 39 studies published in 2018 indicated that mean prevalence for sending and receiving sexts were 14.8% and 27.4% respectively, with higher prevalence on mobile devices rather than computers. The prevalence of forwarding a sext without consent was 12.0% and the prevalence of having a sext forwarded without consent was 8.4%. The study findings indicated that the prevalence of sexting has increased in recent years and increases with age. 23 A further meta-analysis showed similar results and concluded that there was an already high and increasing prevalence of sending and receiving sexts among young people. 24 A meta-analysis focusing on non-consensual sexting (where content is shared/forwarded without permission) indicated that across all ages, 8.8% of people have had their image or video-based sexts shared without consent, 7.2% have been threatened with sext distribution, and 17.6% have had their image taken without permission. Prevalence rates did not vary extensively by study characteristics such as age. 25 The EU Kids Online 2020 report indicated that across 19 European countries, less than 10% of children aged 12-16 had sent sexual messages or images over the previous year and there was variation across countries. 26 Some of the earliest work in this area presented a typology of sexting episodes which was derived from an analysis of over 550 cases obtained from a national survey of law enforcement agencies. 27 The cases all involved "images of minors created by minors" which met the criteria under US law for "child pornography". These episodes 21 Technology Coalition [53]. 22 APPG [2]. 23 Madigan et al. [34]. 24 Molla-Esparza et al. [37]. 25 Patel & Roesch [42]. 26 Smahel et al. [49]. 27 Wolak & Finkelhor [63]. could broadly be divided into two categories labelled as 'Aggravated' and 'Experimental'. The Aggravated incidents involved criminal or abusive elements which included adult involvement or criminal or abusive behaviour by minors such as sexual abuse, extortion, threats; malicious conduct arising from interpersonal conflicts; or creation or sending or showing of images without the knowledge of or against the will of a minor who was pictured. In Experimental incidents, young people took pictures of themselves to send to established boyfriends or girlfriends, to create romantic interest in other youths, or for reasons such as attention-seeking. There was no criminal behaviour beyond the creation or sending of images, no apparent malice, and no lack of willing participation by those youths who were pictured. This typology has been central in understanding how we can differentiate between content that is created and shared by young people in a consensual way and content that is shared without consent, or is created through an exchange with an adult, or where coercion was involved. This typology was used in 2016 in the UK Policing Guidelines on investigating sexting among minors to assist law enforcement to respond in a proportionate way to children who had been identified as producing or sharing content that met the criteria for indecent images of children (also called child pornography and child sexual abuse material across other jurisdictions). 28 This introduced a new crime outcome code for law enforcement (Outcome 21) which stated, "Further investigation, resulting from the crime report, which could provide evidence sufficient to support formal action against the suspect is not in the public interest -police decision". As will be further discussed, Outcome 21 was welcomed as an alternative to criminalising the sharing of explicit sexual images among minors, but it still raised challenges in relation to the implications of recording the incident.

Why is this important?
As already noted, young people use technology to establish and maintain intimate relationships and explore their sexual identity, and the exchange of sexually explicit images and videos is one of the ways that this is achieved. 29 However, these images which may capture private, sexual moments shared between adolescents can then be forwarded non-consensually to others or can be created, and further used, in the context of an exploitative relationship with an adult or another minor. 30 Both are problematic as the images can be used strategically to manipulate a young person into creating further explicit sexual content, can be used to harass and bully a child, and once uploaded can be further distributed and function in the same way as other child sexual abuse material (CSAM). 31,32 This has become an increasing problem for law enforcement as the amount of identified CSAM increases year-on-year with a proportion of it associated with sexting. In 2010, the UK Child Exploitation and 28 College of Policing [12]. 29 Chatzinikolaou & Lievens [8]. 30 Bianchi et al. [5]. 31 Moritz & Christensen [38]. 32 Muncaster & Ohlsson [39].
Online Protection command's (CEOP) strategic overview mentioned the increasing number of self-produced images, largely of females, submitted to the Image Analysis and Victim Identification Unit. 33 Similarly, a threat assessment from the US Department of Justice 34 commented on offenders enticing children to display themselves in sexual activity through images and webcams. A US study of arrests for CSAI production at two time-points (2000-1 and 2006) indicated that there was an increase in adolescent victims, but no increase in the proportion involving very young victims. Many of the cases involved self-produced sexual images by adolescents and involving adult sex offenders. 35 A third sample indicated that arrests for CSAM production quadrupled between 2000-2009 which was attributed to cases of self-produced sexual images solicited from minors by adult offenders. 36 However, of note is that in 2009 most youth-produced images were distributed, usually by youth who took pictures of themselves and sent them to others. In 2015 the Global Alliance (whose aim was to sustain political momentum at a national level to improve the fight against child sexual abuse online) produced a threat assessment, which sought to evidence the nature, magnitude and global scope of online child sexual abuse and exploitation through a survey of its 54 members. 37 One trend identified was cases involving self-produced sexual images by children (27 countries) involving coercion (28 countries). By 2020 the UK Internet Watch Foundation (IWF) reported 68 000 cases of confirmed self-produced sexually explicit images of children, which was 44% of actionable cases on CSAM in that year, and a 77% increase from 2019. In 80% of these cases, the victims were 11-13 year-old girls. 38 However, while the IWF data can be used to examine the severity of the abuse within the images, it is not possible to code the images according to whether they are 'Aggravated' or 'Experimental'.
An earlier publication examined 687 UK-claimed victim-identified cases from the Interpol International Child Sexual Exploitation (ICSE) database with the aim to describe differences between cases involving self-taken images and those where images had been taken by others. The analysis showed an increase in identified victims during the study years 2006-2015. Almost two-thirds were female, the majority were white, and 44.3 per cent involved cases where images were self-produced. Of these, 34.4% would be seen as 'Aggravated' and were produced in the context of a coercive relationship. Almost 10% were Experimental where there was no evidence of coercion (although coercion may still have taken place). Between 2010 and 2015 the number of children involved in self-produced images each year had exceeded more than 40 per cent of the total number in the database. Although self-taken images may be perceived as less worrisome than the cases where children had been photographed by family members or family friends, two-thirds were classified as coercive. The authors felt that this was an important argument in favour of continuing to investigate 33 CEOP [7]. 34 USDOJ [56]. 35 Wolak et al. [65]. 36 Wolak et al. [66]. 37 Global Alliance [22]. 38 Internet Watch Foundation [27]. these cases under victim identification programmes. 39 A recent exploratory study of data related to 152 cyber-extortion offenders (where there were demands for victims to produce more sexually explicit images and threats to distribute self-produced sexual images of minors if they refused or attempted to end contact with the offender) found that the majority of cases focused on extorting minors (52.6 per cent). 40 From a victim perspective, the consequences of aggravated sexting or sextortion are considerable and have been linked for some young people to depression, self-harm, and other forms of psychopathology. 41,42 Concerns about the creation of content by exploited young people was also discussed in two earlier publications by the UK Charity Barnardos. They examined five of their children's services which indicated that 42% of the 702 children surveyed had been groomed online, which included the exchange of films and photographs. 43 A more recent US study concluded that sextortion (threats to expose sexual images to coerce victims to provide additional pictures, sex, or other favours) was an emerging online threat to young people. They examined 572 incidents involving minors where perpetrators were likely to pressure victims into producing initial sexual images, demand additional images, threaten victims over a period of time, and urge victims to harm themselves. The authors concluded that sextortion incidents were serious victimisations, and often co-occurred with teen dating violence. 44 Similar finds have been identified in other studies. 45,46,47,48

Self-produced sexually explicit images and the law
The relative newness of sexting, non-consensual sharing and sextortion means that legislation is often, "a messy patchwork of legal responses that often yield disproportionately punitive responses" (p. 563). 49 This is particularly the case where adolescents are seen as both victims and perpetrators. These authors suggest that appropriately situating youth sexting within the legal system is challenging due to its varying forms: purely consensual sexting; consensual but coerced sexting; non-consensual youth sexting; sextortion; and sexts that have been accessed or requested by adults. 50 Another challenge facing the legal system is that sexting by young people is positioned as intersecting among a variety of legal rights, such as privacy, free speech 39 Quayle et al. [46]. 40 [40]. 41 Wachs et al. [59]. 42 Gassó et al. [20]. 43 Fox & Kalkan [19]. 44 Wolak et al. [64]. 45 Kernsmith et al. [28]. 46 Titchen et al. [54]. 47 Bianchi et al. [5]. 48 Mandau [35]. 49 Lee & Darcy [32]. 50 Strasburger et al. [50]. and sexual expression. This then sits alongside associated harms (which are often confused with risk as opposed to known harm). 51 This has then resulted in the use of novel terminology in relation to these activities, as well as the creation of additional laws to address these behaviours. For example, in 2015 Sect. 33 of the Criminal Justice and Courts Act 2015 came into force in the UK. This created a new criminal offence of disclosing private sexual photographs and films with intent to cause distress and is more commonly known as 'revenge porn'. However, in relation to sexting, none of these laws sufficiently take into account the ways in which the consensual exchange of sexual images increasingly plays a part in the lives of young people or the harms that may follow from these images being forwarded or shared without permission. It has been argued that sexting laws, especially where images and videos have been consensually shared, call into question who they are meant to protect and from what. 52 An early publication by Gillespie argued that consensual adolescent sexts in which the creator, sender and receiver are willing participants in the exchange should not be categorised as child pornography and most likely should not be prosecuted under sexting statutes. While theoretically these behaviours may breach child pornography laws (as children who take images or videos of themselves are unintentionally possessing CSAM) it was argued that this behaviour and the content produced by it are an expression of the adolescent's identity and thus protected by Articles 8 and 10 of the European Convention on Human Rights. 53 It has also been argued in that in the absence of a finding malicious intent in a defendant, it is unclear what purpose sexting laws serve for young people. Many young people understandably do not realise the potential illegality of creating and sharing explicitly sexual images of themselves and criminal prosecution seems to be a disproportionate way of discouraging young people from expressing their sexuality online. In both the US and Australia, criminal law does not consistently address sexting, which means that in some jurisdictions, children who participate in sexting can be liable for offences related to child pornography or child sexual abuse material (CSAM). 54,55 In addition, in the US, numerous state and federal child pornography laws mandate that "the creation, reception, distribution, sale, solicitation, pandering, and ownership of sexually explicit images of minors are illegal, and these laws are often interpreted by courts such that they constitute youth-generated sexting images as child pornography". 56 While it may be argued that few cases involving explicit self-produced images result in charges against a minor, these cases do exist and have been a source of concern over a number of years. Data from the Third National Juvenile Online Victimization Study (NJOV-3) examining how often prosecutors charge minors indicated that 16% had cases that resulted in the defendant being sentenced to sex offender registration. 57 51 Aikenhead [1]. 52 Holoyda et al. [25]. 53 Gillespie [21]. 54 Moritz & Christensen [38]. 55 Westlake [62]. 56 Primack [44].

O'Malley & Holt
A further study examined data from the Third Youth Internet Safety-Survey (YISS-3) and concluded that only a small percentage of young people appear in or create sexting images that could be considered illegal 'child pornography'. 58 This is a very central issue, as outside of the data routinely categorised by the IWF according to the UK Sentencing Guidelines there are, to my knowledge, no studies that have systematically analysed the content of sexting images and how many of these would be classified as illegal. It would seem that across most jurisdictions while there appears to be few convictions of minors there seems to be little interest in completely stopping children from being charged under laws related to the production and possession of CSAM.
In Europe, child pornography legislation has been adopted at the supranational level (Council of Europe and European Union) as well as across national jurisdictions. The Council of Europe has adopted two important conventions related to this: the 2001 Budapest Convention on Cybercrime 59 and the 2007 Lanzarote Convention on the Protection of Children against Sexual Exploitation and Sexual Abuse. 60 The latter defines child pornography as 'any material that visually depicts a child engaged in real or simulated sexually explicit conduct or any depiction of a child's sexual organs for primarily sexual purposes' (Article 20 (2)). It is clear that this definition is similar to the one used in the UNCRC Protocol and could be interpreted as being applicable to the exchange of sexually explicit material between children. 61 The Lanzarote Convention was intended as a "living instrument that has to be interpreted as applicable to the dangers known at the moment it has been drafted as well as to the dangers that have appeared since that time and to those that still are to appear in the future. The best interest of the child would otherwise risk being jeopardised". It is not surprising that when the Lanzarote Convention was adopted there was already a growing awareness that the production and exchange of sexually explicit images was taking place between minors and that in some instances it would not be appropriate to criminalise such activities. Paragraph three of Article 20 indicates that Member States can decide that sexting between minors that have reached the age of sexual consent, at least as far as this concerns the production and possession of images, should be excluded from child pornography legislation. It specifies that this is applicable only "where these images are produced and possessed by them with their consent and solely for their own private use". However, as noted by Croft and Lievens, "the wording of the article also implies that sexting between minors that have not reached this age or the offering, making available, distributing, transmitting, procuring or knowingly obtaining access to this type of material, could still fall within the scope of the national child pornography laws". 62 In 2019 the Lanzarote Committee's Opinion on child sexually suggestive or explicit images and/or videos generated, shared and received by children importantly concluded that, "The self-generation of sexually 58 Mitchell et al. [36]. 59 Council of Europe [14]. 60 Council of Europe [15]. 61 Chatzinikolaou & Lievens [10]. 62 Crofts & Lievens [17]. suggestive or explicit images and/or videos by children does not amount to "the production of child pornography" when it is intended solely for their own private use". 63 The Opinion also concluded that "when sexually suggestive or explicit images and/or videos of children are generated by them and shared for their own private use, Parties should take into account children's fundamental right to protection of privacy to relieve them from offences related to child pornography".
Chatzinikolaou & Lievens explore how trust, control and privacy serve as prerequisites for online sexual acts such as sexting as part of adolescent sexual development, and how a relevant legal framework should enable this. 64 This raises the question of, "whether legislators currently guarantee a framework which protects the legitimate exploration of sexuality in the context of today's mediatised childhood" (p. 40). 65 Trust remains ill-defined but it has been argued that it is important to explore trust in a digital context, since non-consensual distribution of personal data can be viewed as resulting from misplaced trust. 66 Control relates to the ability of self-determining one's sexual identity and (online) sexual representation. According to the UN Committee on the Rights of the Child (2016), 67 the exploration of sexual identity is considered part of the complex transition from childhood to adulthood and, the exploration of adolescents' emerging identities and sexualities is part of their development. Healthy development of children includes the respect for individuality, a dimension of which must relate to sexuality. It follows that the right to sexual exploration and behaviour related to this could be regarded as falling under the protection of Article 6 of the UNCRC (right to development). However, there is a growing consensus which rejects the assumption that everything that is digital is automatically public 68 and people have the right to protect their personal information (including intimate imagery or texts of sexual content) from being accessed or monitored. 69 These authors concluded that national legislators ought to adopt policies that balance the need to protect children from exploitation or abuse with children's integrity and autonomy rights, in view of their best interests, following an evidence-based child rights impact assessment.
One of the elements that such an impact assessment could address is the thorny issue of the age of sexual consent when, for example, determining exceptions to criminalisation of production and possession of sexual material involving children. As previously noted, there is evidence of children who are below the age of consent (often set at 15 or 16 in national legislation) who are creating and consensually sharing sexual images and for whom these exemptions would not apply. Finally, the best interests of the child and the limitation of any harmful outcomes associated with sexting also rest upon the importance of consent (and how this is determined) as a fundamental element for legitimate online sexual exploration, and avoiding blaming individuals who engage in sexting, rather than blaming those who breach the trust or engage in abusive behaviours. 70 There are always questions of power and agency in relation to sexual choice and our ability to distinguish between 'wanted and unwanted. . . interpersonal contact'. 71 Pearce developed the concept of 'coerced consent', closely aligned to concepts of 'coercive control'. Coerced consent refers to where the child is subtly manipulated into consenting to sexual activity (Pearce, 2013). 72 Regulatory frameworks that protect consensual sexual activities between young people that are congruent with normal sexual development but still address coercive or non-consensual acts or harmful sexual behaviour, would seem to be important in the digitally mediated world in which young people live. Chatzinikolaou & Lievens have argued that such frameworks should include provisions that enhance and safeguard trust, control, and privacy for children. 73 It has been suggested that sexting is a particularly complex issue for today's societies, as was clearly illustrated in the Opinion issued by the Committee of the Parties to the Council of Europe' s Lanzarote Convention, and affects both the recognition of children's autonomy and their privacy (Articles 13 and 16 of the Convention) as well as the child's right to protection from sexual exploitation or abuse (Article 34). 74 A qualitative study of perceptions and practices of sexting amongst young people concluded that shame and stigma which follows unauthorised distribution of sexually explicit images, along with the 'social capital' given to those who distribute them, involves a denial of rights to bodily and sexual integrity and privacy to the young person in the image as well as a denial of the right to meaningful choice. 75 In England and Wales any production or dissemination of 'indecent' images of young people potentially constitutes a criminal act of possessing, transmitting, viewing or downloading indecent images of children under the Protection of Children Act 1978 and this may apply to scenarios involving self-produced sexual images by minors. 76 As previously noted, in England there have been some attempts to make the law more proportionate in this respect. In relation to sexting, recent guidance from the Crown Prosecution Service and the College of Policing recommended that 'youth produced sexual imagery' shared between young people should not be routinely prosecuted and that most of these incidents should be dealt with informally, particularly where there is no evidence of 'exploitation, grooming, profit motive, malicious intent or persistent behaviour'. However, these recommendations still specify the need to record on the police intelligence databases that the young person has been involved in the creation and sharing of sexual images. Arthur has noted that rather than seeing sexting as an expression of adolescent sexual development it is seen by the criminal justice system in the UK as an illegal activity which may not result in a 70 Walrave et al. [60]. 71 Cowan [16]. 72 Pearce [43]. 73 Chatzinikolaou & Lievens [9]. 74 Rosani [28]. 75 Setty [48]. 76 Arthur [3]. criminal charge but (potentially) may restrict the young person's future life opportunities. Sexting is therefore framed as a dangerous activity and a social problem and does not acknowledge the agency of young people or their rights as sexual citizens. 77 It can be argued that the response of criminal justice to the sexual behaviour of young people needs to be reconceptualised in order to support and promote the welfare of young people. The legal response to consensual sexting by young people may stigmatise and discredit adolescents' understanding and experiences of sexual behaviour and disregards any claim for sexual rights or freedoms. This would allow the law to "challenge coercive, exploitative, abusive, deceptive and unauthorised producing and distributing of sexual images of children while also protecting young people's freedom to engage in self-determined sexual relationships" (p. 394). 78 National (and international) legislation often falls short of understanding and differentiating between various online sexual practices. Consequently, a gap exists in legislation to protect against images of child sexual abuse (i.e., 'child pornography') and consensual sexual practices, including the sharing of sexual images, between teens. Over the last twenty years, policies related to online child protection have remained fragmented, inconsistent in how they are implemented, and use both laws and penalties that were originally developed for adults rather than for minors. This takes little account of the complex and changing technological developments which have to be navigated and managed by young people and which warrant further consideration in our approaches to policy and practice. 79 There is also considerable ambivalence about adolescent sexuality and the rights of children to sexual expression, and a troubled discourse around agency and autonomy. As concluded by Arthur, the denial of children's agency, autonomy and access to knowledge may in fact increase their vulnerability to abusive and coercive practices as it "denies them the opportunity to develop the tools to recognise abusive and exploitative behaviour and situations" (p. 387). 80 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/ 4.0/.