The Gendered Dynamics of Sexting as Boundary Work

‘Sexting’ as a form of sexual expression and experimentation has grown increasingly ubiquitous among teens. In addition to fostering intimacy and closeness among selected partners, sexting between minors incurs considerable risks to youth mental, physical, social and emotional wellbeing. Current debates on sexting focus on the conflicting role of pressure and pleasure among youth. We highlight findings from 35 focus groups with Canadian teens examining attitudes and experiences with cyber-risk and sexting. Our results show that youth demarcate boundaries between public and private and an ordered/disordered sense of self as they seek intimacy with others through the exchange of explicit digital material. Drawing on recent conceptual and theoretical work on image-based sexual abuse, we suggest that teens express situated agency when reflecting on sexting, indicating only partial awareness of wider patriarchal contexts mediating and patterning gendered behaviours involved in sexting and its outcomes.


Introduction
The social life of teens has moved considerably into online digital spaces, such that teen relationships are increasingly starting and staying online (Lenhart et al., 2015b). Perhaps not surprisingly, communication between teens online can take on the sexually suggestive or explicit quality that is known as 'sexting'. Wolak and Finkelhor (2011, p. 2) defined sexting as 'sexual communication with content that includes both pictures and text messages, sent using cell phones and other electronic media'. 1 Practices within the broad category of sexting include words and prompting that range in sexual suggestiveness, as well as the sharing of nude or semi-nude digital images or video. The present study focuses on the sharing of intimate images between dyads in both consensual and non-consensual contexts.
Sexting among teens is not uncommon, and as it becomes more commonplace, teens may view it as a rite of passage or norm of growing up. In the United States (US) and the United Kingdom (UK), scholars studying the prevalence of teen sexting have found vast variations in the number of youth who have sexted, with estimates ranging between 4% and 40% (Bailey & Hanna, 2011;Ringrose et al., 2012). Other reports indicate upwards of 30% of youth have sent a nude picture and 45% have received a nude (Englander, 2012). Similar statistics are not as readily available in Canada, though a recent survey of 800 16-to 20-year-old Canadian youth found 40% have sent a sext and 60% have received one (Bresge, 2018). Current research conducted by Adorjan and Ricciardelli (2019a) argued that sexting may be even more common than the frequencies purported in the US.
While it is important to avoid disproportionate reactions to teen sexting (many are not engaging in it, and many who are do so consensually and privately; Wolak and Finkelhor [2011]), the sharing of nude digital images among teens does come with a set of risks, including potential social, personal and legal consequences (Bailey & Hanna, 2011;Englander, 2012), which are further delineated by gender Ringrose et al., 2013). In Canada and much of the US, although persons under 18 years of age can lawfully engage in consensual sex and sexual acts, it is a criminal offense to record such behaviours or to share or distribute the recordings (see Poltash, 2013). Section 163.1 of the Canadian Criminal Code stipulates that the possession, distribution, production or sale of 'child pornography' is an offence. Child pornography includes any visual representations of someone under or presented as under the age of 18 that shows sexual activity or is of a sexual nature. 2 The language itself in the law removes any agency, choice or legitimacy held by the purported victim (e.g., Karaian, 2012). Nevertheless, for youth, creating, distributing or possessing sexual images of themselves or a peer under the age of 18 (e.g., sexting nudes) is a criminal offense.
The plethora of risks associated with sexting is made well known to teens (both boys and girls) who are confronted with them in school programmes, educational material, news media and popular culture (Karaian, 2012;McGovern & Lee, 2018). Despite a generalized knowledge of the risks associated with sexting, many teens continue to engage in such acts as they navigate their personal and intimate lives with one another. When youth do incur harm, there are gendered differences in outcomes, with a high potentiality for a tarnished social standing, and a breakdown of emotional control (girls), and-in extreme cases-legal repercussions (boys) (Bailey & Hanna, 2011). Gendered sexual norms and taboos further compound these harms for girls, who must navigate discourses rooted in 'slut shaming' narratives in the event that their images are given or shared in a non-consensual manner (Adorjan & Ricciardelli, 2019b;Naezer & van Oosterhout, 2020;Ricciardelli & Adorjan, 2019). Young girls especially may engage in 'safety work' as a form of boundary negotiation when navigating the potentiality of sexual violence in public (or potentially public, as is the case with networked communication) spaces (Kelly, 2012;Vera-Gray & Kelly, 2020). Here, strategies to avoid or otherwise diffuse situations that put women at risk are included. For teen girls, the need for safety work online is compounded by hegemonic gendered norms that place girls' bodies under immense surveillance (Milford, 2015;cf. Kanai, 2015).
As part of a wider study examining teen perceptions and experiences with cyberrisk and their strategies in managing these risks (see Ricciardelli & Adorjan, 2019), in this article, we explicate findings from a series of focus groups with Canadian teens centred on experiences of sexting, and particularly the gendered differences in the pressures and risks experienced when sexting. Our results show that youth demarcate boundaries between public and private and an ordered/disordered sense of self as they seek intimacy with others through the exchange of explicit digital material. Our focus group interviews emphasize the agency teens possess in their decision to engage in consensual sexting. However, we also show how much this agency is situated. Agency resides not in individuals but in the situations and wider environments in which they behave (McGann, 2014). Situated agency here refers to wider patriarchal and neoliberal patterns of iniquitous awareness and response that place superfluous emphasis on female responsibility in managing the consequences of 'consensual' sexting (Choi et al., 2019; see also Hillier et al., 2020). Thus, the boundary work our teen participants engage in is inherently gendered, as behaviours which are situated within broader and longstanding patriarchal norms governing female teen sexuality; a governance which simultaneously takes up gendered associations of private and public space and troubles this dichotomy through the presence of new information communications technologies.

Youth and Sexting: Social Connection and Purposeful Risktaking in the Context of Image-based Sexual Abuse
Today's teens are often described as 'digital natives' (Prensky, 2009), born into a world permeated with online spaces and the technologies for accessing them such as smartphones and other handheld devices (Roberts & Foehr, 2008). In contrast to the 'digital immigrants' of older generations, youth today frequently spend their lives building and maintaining relationships and exploring their sexuality in online spaces (Subrahmanyam & Smahel, 2011). One Pew Research Center study (Lenhart et al., 2015) indicates social media is a 'top venue for flirting', allowing youth to feel intimacy and closeness with romantic partners by connecting them to the comings and goings of their partner's life while facilitating emotional closeness. The potentiality for experiencing intimacy and closeness through sexting and the risks inherent in the sharing of sexual content with peers online place pressures on youth to engage in purposeful risk-taking, through which we argue they tread the boundary of control and chaos. Hart (2017), for example, studied the physical and emotional risks experienced by young adult Internet users who share naked self-shot imagery on social media. Participants in Hart's (2017) study demarcate the boundary between order and disorder by centring the physical and emotional 'self' within risk-taking. To lose one's sense of an ordered self-be it through posting distasteful 'pornographic' imagery or by audience mismanagement and context collapse (see below)-is akin to crossing into chaos and disorder. Hart (2017) acknowledged the need for research into the intersection of risk and pleasure in young people's experiences, given that research of youth 'at risk' largely focuses on negative impacts of sexual exploration and technology use (Follesø, 2015).
Research on teens' digital lives has long targeted identity development, particularly as shaped through the formation of relationships with peers both online and off-line (Crosnoe & Johnson, 2011;Spies Shapiro & Margolin, 2014). Given the importance of identity development and acceptance from peers during adolescence, it is not surprising that social media is particularly attractive for teens in their pursuit of self-discovery, including their sexual self-discovery (Oberst et al., 2016). In the case of teens, whose identities and self-values may be in flux, social media provides a space to explore and construct their sexual selves through interactions with peers (Subrahmanyam & Smahel, 2011). The space is pertinent given the heightened importance of sexual exploration and engagement in intimate relationships during adolescence (Buzwell & Rosenthal, 1996). It is within such processes of self-discovery and identity construction that youth may engage in voluntary risk-taking behaviours. Indeed, the sharing of nude digital images is one way that teenagers seek sensations or act in ways thought to encourage peer group acceptance and intimacy with select persons.
As new types of social media emerge and become ubiquitous, youth must navigate these new platforms, along with the risks and opportunities they present for self-presentation and authentication (Boyd, 2011(Boyd, , 2014Livingstone, 2008). Online spaces facilitated by social media are described as networked publics, or 'publics that are restructured by networked technologies' (Boyd, 2014, p. 8). Inherent in networked publics is the construction of social spaces as mediated by communication technologies, as well as the audiences and communities that emerge to observe and participate in these spaces (for a discussion, see Boyd, 2014, pp. 8-14;Vitak, 2012). Networked publics differ from traditional publics in that the information shared through them can be easily searched and disseminated, persists over time and is often visible to multiple (intended and unintended) audiences (Boyd, 2014). Because teens have increasingly moved their social worlds into these networked publics, they must develop a set of skills and the knowledge for navigating them and managing the information they share (Marwick & Boyd, 2010;Vitak, 2012). Of particular concern with networked publics is their propensity for context collapse, or 'the flattening out of multiple distinct audiences in one's social network, such that people from different contexts become part of a singular group of message recipients' (Vitak, 2012, p. 451). Thus, information intended for a specific, exclusive audience can be quickly diffused (intentionally or otherwise) among unintended recipients. In the context of sexually expressive pictures, video or text, context collapse presents an alarming and-especially among youth-harmful phenomenon should such messages fall into unintended hands. As such, navigating the risk of context collapse is enmeshed in sexting between teens, whose knowledge of privacy features on social networking sites, and skills at audience selection become crucial for risk management.
The development of addictive properties towards thrill-seeking behaviours such as sexting can arise from a desire for closeness, intimacy or to maintain or rekindle such feelings once previously experienced Livingstone, 2008). Youth who engage in sexting often generate a plethora of emotional experiences (e.g., freedom, control, excitement and anxiety; Shay, 2015) as they tread the boundaries of an ordered/ disordered self. Some scholars, like Crofts et al. (2015), found among focus group participants that sexting is an activity engaged in when bored-it creates excitement when none exists. The emotionally charged anticipations tied to sexting may also help pass time (e.g., dealing with boredom), given, as Boyd (2014, p. 80) described, when time is spent on social media it '… disappears, attention focuses, and people feel euphorically engaged'. The temporal influence of social media may serve to drive sensations of sexual exploration, risktaking and feelings of closeness to peers (Hart, 2017;Shay, 2015). In our study, we show that the presentation of a disordered self emerges whenever a youth's ability to regulate emotional responses online is insufficient (e.g., such as succumbing to unwanted pressures to send images), or when a youth's sexually expressive image is redistributed to an unintended audience (which can result in a damaged reputation or even legal consequences). Thus, as youth employ sexting to navigate sexual exploration and intimacy on social media, they also risk becoming consumed by the addictive properties of potential mutual intimacy and the excitement of consensual sexting (Albury et al., 2013;. In acknowledging the limitations of negative risk-based narratives in theorizing teen sexting, Albury (2017) argued that breaking down and problematizing the binary of safety/risk allows for a more nuanced understanding of pleasure, risk and consent. This is especially true in the case of young girls who must navigate both sets of risk (non-consensual distribution and oppressive gendered ideologies that lead to increased sexual scrutiny and put their social and emotional standing at risk; Naezer and van Oosterhout [2020]). Public space continues to remain masculine and private space feminine, and this gendered dichotomy is emboldened in many respects through new technologies. In this context, 'girls will not only be more harshly judged than boys for the content of their online profiles, but also for their degree of publicness' (Bailey et al., 2013, p. 107). Albury (2017) thus argued that teenagers (especially girls) are at risk of both context collapse in the sharing of sexual imagery and sexual shame and regulation as a deliberate pedagogical strategy rooted in gender norms and employed by concerned adults. Youth who engage in sexting are not immune to the same 'ageold double standard' that celebrates sexually expressive boys while shaming young girls (Adorjan & Ricciardelli, 2019b;Ringrose et al., 2012Ringrose et al., , p. 12, 2013. The double standard reifies a set of risks that are disproportionately experienced by girls, who are twice as likely as boys to report motivations for sexting based on coersion, pressure, blackmail and threats from peers (Englander, 2012). It follows, then, that sexts involving images of teen girls are highly regulated, scrutinized and penalized to a degree not afforded to teen boys (Ringrose & Harvey, 2015).

Current Study
In emphasizing teen agency as situated (especially gendered), we argue that sexting should be conceived along a broader continuum of potential harm and violence (cf. Kelly, 1987), which is better attuned to capturing heterogeneity in experiences of abuse and victimhood (McGlynn et al., 2017). This continuum approach divests binary notions such as 'consent' and 'non-consent' (discussed further below) and suggests that the implications of sexting may reside along a wider lens of 'imagebased sexual abuse' McGlynn et al., 2017). This term is preferred by Henry et al. (2021, p. 2; see also Bluett-Boyd et al., 2013) in preference to terms such as 'revenge porn' also due to the wider social contexts linked to it: We instead prefer to view technology not simply as a tool for motivated perpetrators to engage in egregious acts, but as part of an overarching system of inequality and discrimination, whereby societal norms, such as economic profit, excessive individualism, hypersexuality, heterosexuality and hegemonic masculinity, drive and shape behaviors, including those that are harmful, as well as those that are resistive, subversive and non-conformist.
In the current study, we conceptualize sexting, particularly the sharing of nude digital images, as an interpersonal and relational behaviour that accumulates gendered patterns of risk over time (e.g., as relations with peers change). However, we caution we did not operationalize sexting when discussing the practice with youth, providing opportunity for them to explain their sexting experiences and thus what, for each, constituted sexting. We explore the interdependent and relational boundary work undertaken by youth who engage in digital sexual expression with peers (Lamont & Molnar, 2002;Pachucki et al., 2007) and situate sexting as transcending both digital and real-world spaces. We argue that teens, but especially female teens, employ specialized skills to maintain the boundary delimiting a sense of an ordered and meaningful self. To accomplish this, we explore the role that context collapse and identity management play in sexting.

Methods
To capture the personal and in-depth experiences, motivations and meanings that youth attribute to sexting, we employ focus groups aimed at understanding youth risk-taking online. The use of focus groups to investigate youth experiences of cyber-risk is increasingly popular (e.g., Albury et al., 2013), given that group-based conversation helps facilitate accounts of both personal experiences and the interconnected social contexts in which these experiences are embedded.
We conducted 35 focus groups, consisting of two to five participants each, between July 2015 and November 2016. Scheduling challenges reduced group size to two or three in some cases, resulting in an average group size of 3.3 youth. Given our goal of capturing the phenomenological and interconnected aspects of youth sexting experiences, we opted to include interviews with two or three participants, given that they elicited meaningful conversation. In total, 115 youth (67 females and 48 males) participated, ranging in age from 13 to 19 and averaging 15 years of age. Most participants identified as white. Discussions ranged from 30 to 120 minutes in length and were led by Ricciardelli, Adorjan or trained research assistants. Focus group participants were recruited from participating schools, including public and private middle and high schools, as well as undergraduate university students, and the children of parents attending university. Recruitment was aided by a purposive, snowball sampling technique that allowed us to make the use of multiple initial contacts at schools and universities to gain references to additional participants. Focus group discussions were audio-recorded and transcribed verbatim, though to aid in comprehension, any quotes included are edited for speech fillers.
Focus groups took place in two major regions in Canada: urban Western Canada and rural Atlantic Canada, henceforth referred to as Cyber City and Cyberville, respectively. The choice to organize multiple locations under two major 'pseudoregions' allowed us to ensure greater confidentiality and anonymity of participants (especially those from Cyberville locations with smaller populations), while also enabling us to distinguish between thematic differences in the two regions. Of the 35 total focus groups, 15 groups took place in Cyber City, while the remaining 20 were conducted in Cyberville. When possible, participants were organized into groups of the same gender and age range, a sampling strategy aimed at reducing the intimidating effects of group conversation and ensuring that youth engaged with others with whom they are likely to have similar experiences (Morgan, 1997).
The sensitive and private nature of topics such as sexting required additional care on the part of investigators to mitigate potential ethical concerns. Prejudiced questions about the gendered prevalence and directionality of pressures faced by youth who sext are concerns in research designs (for a summary, see . To mitigate these presumptions and facilitate honest disclosure among participants, questions of gender identity and sexuality were not explicitly asked. Focus groups began with general questions about how much time participants spent online during an average day and what they associate with the idea of 'cyberrisk'. In response, 'sexting' became one of the more frequently referred risks that participants raised, which we explored by asking about whether participants had concerns about sexting and how they manage these concerns (e.g., manage risks entailed). Because the theme of sexting arose from the participants organically, we did not operationalize the term or ask youth to do so. However, the topics of sexting and sending/receiving nude images were navigated carefully to avoid presumptuous questions and allow participants to discuss such topics organically and in their own words. For example, with male participants, we framed questions to avoiding, suggesting that males pressure girls into sending pictures (despite existing evidence that this occurs). Often, questions regarding sexting were phrased as asking whether it was perceived as an issue among participants' friend groups or at school, rather than questions of direct involvement (in so doing, we avoided learning if anyone was in possession of nude or semi-nude photographs).
Data were analysed using an inductive, iterative approach assisted by NVivo qualitative analysis software. An initial round of open coding (Strauss & Corbin, 1990) allowed themes to emerge from, and be applied to, the data. Using the parent and child node functions of NVivo, all mentions of a given theme were captured as separate 'nodes' that could then be compared both across and within groups. Once initial themes emerged, a focused coding strategy allowed us to conceptualize patterns of responses into theoretical and thematic codes (Charmaz, 2014). Frequent meetings between the primary investigators ensured intercoder reliability among coding strategies, as well as validity of the data (Twinn, 1998). We focused on the sensations and boundary work that youth undergo when sexting.

Defining Boundaries
Our results show that youth construct and navigate two different 'boundaries' during identity work on social media: the boundary between public vs. private and the boundary between an ordered vs. disordered sense of self. Each boundary presents a different set of risks to youth that are embedded in emotion management, and relations with peers.

Public Versus private
Youth construct a space rooted between the boundaries of public vs. private presentations online. Information sharing online in networked publics always involves a complex level of audience negotiation (Vitak, 2012). Unlike Hart's (2017) sample of young adults who posted intimate pictures anonymously online at the risk of being found out by people they knew offline, our respondents largely chose a specific audience for their images (i.e., a partner or sexual interest), but risked having images passed along to others. In the context of youth sexting, then, the line between public and private is inherently relational given that it involves the exchange of private material from one party to another, requiring a level of trust and respect between the sender and the receiver. This is exemplified in response to the inquiry: 'if there is a picture that's sent, whose responsibility is it? Should the picture not be sent at all… or is the person that's receiving it, is it their responsibility to make sure that it's kept private?' Tamika and Anton, both 16 from Cyber City, echoed each others' sentiments in response: These youths' words reveal how teens in our study construct, however abstractly, an ideal ethical standard for how illicit images should be treated in the context of consensual sexting. Here, the relational boundary between public and private is constructed in congruence with a perceived mutual responsibility between the sender and the receiver to keep images private by maintaining the integrity of the intended audience. A similar ideal standard was reported by Englander (2012), whose study of 617 college freshmen from a state university in Massachusetts found that nearly 75% of participants believed that once digital sexual images are shared, they are never shown to anyone outside the intended recipient (indeed, in Englander's study, only 13% of respondents indicated experiences of nonconsensual redistribution).
Despite this expressed ideal of mutual responsibility and trust, it was largely female teens in our study who expressed concern over the potential of unwanted redistribution of nudes. Thus, these females are left feeling without agency, consent or legitimacy (Angrove, 2015;Fairbairn, 2015;Ricciardelli & Adorjan, 2019). Our female participants, in particular, often emphasized the gendered double standards associated with these risks. The potential for diverse and unwanted outcomes when sexting is evinced in the words of Serena, age 19 from Cyber City, that it is precisely the gendered relational nature of the public vs. private sphere that makes sexting risky: I didn't send this guy anything but he, but when I was breaking up with him, he was like 'don't break up with me or I'm going to exploit you' and stuff like that… So, for me personally, it's always been guys … really asking for you to send pics, and that's, and I think that's why a lot of girls do it, or like fall under that pressure of like, 'ok he's like begging for it, he's my boyfriend I guess, I can do it', but I think it's something that you should definitely be cautious of before doing. (see also Ricciardelli & Adorjan, 2019, p. 117) Serena's words enforce the fact that relationships-even those that begin amicablycan become abusive and exploitative, here with reference to the threat Serena received regarding 'revenge porn' (Bates, 2017;Fairbairn, 2015;Stroud, 2014). For Serena, being cautious of what (and to whom) sexual content is sent is a form of 'safety work', or self-protective behaviours that women engage in to avoid or minimize sexual risk (Vera-Gray & Kelly, 2020). Kelly's (1987) 'continuum of sexual violence' helps illuminate the heterogeneous experiences of abuse and victimhood; that is, in opposition to a particular understanding of what constitutes, in this case, consensual sexting 'vs.' revenge porn (see also McGlynn et al., 2017). We would add that this continuum could be extended to consider a hermeneutics of consent, where, rather than bifurcate 'consensual' vs. 'non-consensual' sexting, analysis should be placed on the processes through which what is understood to be consensual and private can become almost instantly, and publicly, non-consensual. As such, navigating the edge of public/private involves not only current risks but also cumulative risks-especially for female teens-that can accrue as social standings with recipients change (something that is often out of the control of senders despite their skills and perceptions). Male teens are not as likely as females to deal with fallout from redistribution that involves 'social rupture …existential threat, isolation, and constrained liberty' . This highlights an underlying risk of information sharing in networked publics more generally: a message may be intended as private, but the medium through which it is sent is always potentially public. However, risks linked to the use of information communications technologies need to be situated within wider, gendered and heteronormative social contexts. The risks associated with sexting for young people are mediated by, discourses [which] latently blame girls for attracting online risk, situating those who experience gender-related victimization as less or more blameworthy, depending on how well they self-protect against it. (Milford, 2015, p. 63) The potential fallout and consequences of these risks for young girls and women are grave in the context of a sext meant for one particular person that is distributed to social networks, family and even authority figures. For instance, Valerie, age 13 from Cyberville, spoke to the dire potential outcomes of context collapse: Just before Christmas…someone called and they told the police that one of my good friends is being blackmailed by a guy with pictures [who was sending the images to his friends] and then so the police here went in that guy's phone and swiped all the pictures and … went around six houses and erased all of it on their phones…. (see also Ricciardelli & Adorjan, 2019, p. 118) Thus, in the situation Valerie describes, the blackmailed youth's digital images became viewed beyond her intended audience. Young girls and women may incur greater potential stigmatization and an emboldened victimization when images are disclosed among unintended audiences, which may include peers, police, educators and others in positions of authority (Ringrose et al., 2013).
Our participants, then, demonstrated awareness of the public vs. private aspect of being online and, specifically, of the ability for the private to transcend into the public even within the boundary work around sexting. The relational boundary is constructed as youth choose to share 'risky' images with a specific recipient-a sensation-seeking behaviour-where risk is compounded on by the uncertainty of relationships, and the high potentiality for context collapse if relations degrade. The high propensity for private images to be shared puts teens, in particular young girls and women, in a risky position. They are vulnerable to the affordances of technologies that enable widespread sharing of content and the wider gendered norms in society that place an enormous burden of responsibility on female youth to avoid the stigma and shame they may incur from actions initially understood as trusting and private (Ringrose et al., 2013;Shariff & DeMartini, 2015). As found in related research, these are gendered risks, linked to wider societal double standards placing undue pressure on young girls and women to send nudes, often from male partners who subsequently engage in non-consensual redistribution (Adorjan & Ricciardelli, 2019b;Ringrose et al., 2013). Shariff and DeMartini (2015, p. 283) presciently noted that 'when women claim space that challenges the traditional and underlying assumption that public space is male space-and heterosexual at that-misogynist attitudes surface through various forms of online abuse'. Despite statements that responsibility should be equally shared, this is often not what most frequently occurs, based on direct experiences of both our participants and other teens they know.

Ordered Versus Disordered Sense of Self
Enmeshed with the boundary between public and private, youth also demarcate a boundary between an ordered vs. disordered sense of self based on their interactions with peers online, and especially when sexting. The boundary between disclosing the right amount and type of information (i.e., ordered self) vs. oversharing (i.e., disordered self; see Agger, 2015;Veinot et al., 2011) is under constant negotiation as the desired self is constructed and presented in online spaces by young people while being mindful of the intended audience. Milford (2015, p. 67) noted: it can be impossible for girls and young women to adhere to both sets of media expectations for gender performance-to simultaneously be private and "responsible" as well as public and "mediatized"-in response they can self-censor or go offline, even at the expense of the increased social and economic opportunities associated with a greater online presence.
It should be recognized that for many youth, 'going offline' is not an option, as youth cannot monitor the 'self' being constructed or authenticated by others in online spaces without being online themselves. Thus, information sharing on social media plays an increasingly crucial role in how teens navigate identity formation (Oberst et al., 2016;Subrahmanyam & Smahel, 2011).
Sexting is one rather personal yet exciting and emotionally stimulating way youth can explore their sexual identity in digital spaces. The role of romantic and sexual relationships is integral to navigating identity development during the adolescent years (Bouchey & Furman, 2004). In our study, this rings true particularly in the context of teens in romantic relationships. For example, Riley, a male aged 17 from Cyberville, explains that sexting has its purposes, particularly 'in long distance rela-tionships… you're like half a country away'. Edward, age 18, agrees with Riley that sexting can be beneficial in 'relationships where you can't see the person everyday'. In another group, Denise, age 18 from Cyber City, explains that sexting images can be desirable in instances where 'you're like swapping pics at the time, you're like, it's like 11 at night, you're like maybe dating… long distance, maybe you're dating this person, maybe it's a fun thing for you'. Thus, sexting is viewed as a desirable medium for sexual exploration, especially in instances where opportunities to navigate these experiences in real life are limited. We can here turn again to the work by Boyd (2014), who posited that teens are attempting to-with whatever means available to them-engage socially with one another in light of cultural dynamics that limit their time spent together. Our focus group participants sought opportunities to engage with peers without adult surveillance (especially parents; see Adorjan & Ricciardelli, 2019a) and create spaces where they have the agency to prepare and employ identity work free from adult intervention. The intimate and private nature of sexual self-expression further intensifies the desire to demarcate these spaces and the images produced within them from adult supervision.
When sexting, however, the boundaries between the ordered and disordered self are under constant negotiation and require emotional management as the ideal self is constructed and then presented online. Said emotional management is enmeshed in the amount, frequency, timing and type of information shared. In consequence, by oversharing information, or going 'overboard', youth risk negative feedback from their intended recipient, such as being insulted, blocked or labelled an attentionseeker. Oversharing can involve audience mismanagement and the miscalculation of relationships, but it can also be as simple an act as sending or receiving a digital sexual image at an inappropriate time-experiences only raised by female teens in our interviews. As Janiya notes, 'if it's like 2pm, I'm at a lecture, it's a Wednesday, like what are you doing?' (Ricciardelli & Adorjan, 2019, p. 119). Janiya's words show that online sharing-particularly sexual expression-requires a specific specialized skill set that makes it possible to traverse the boundaries, including a mastery of how, when and what to share. Oversharing or risking a 'disordered self' is a risk amplified among female teens when compared to their male peers. A group of three females, all 13 years of age from Cyberville, discussed this: Irene: Sometimes, girls send photos and then the guys saves them and then when they break up, they kind of just. Amelie: Send them Irene: They float around … people get mean and start calling each other names Researcher: Oh, about the pictures? Irene: About the people. Amelie: They want you to do it [send images], and then the boys want you to do it and then after they're like; they ask you for it, and soon as you do … they block you. Greta: They block you. Amelie: And then, you can just delete them and then send it around, and then the boyfriend will be like 'gross' and all that, but he's the one who wanted it done. (see also Ricciardelli & Adorjan, 2019, p. 118) As the above excerpt shows, the line between an ordered and a disordered self is rarely clearly marked. Instead, it fluctuates at the discretion of often-fickle social relationships, and the repercussions for losing control can be especially detrimental for female teens involved in heterosexual relationships. In this context, girls especially face possible harassment, bullying and ostracization from peers, while the desired intention of fostering intimacy and vulnerability with a potential sexual partner is lost. The lack of societally projected responsibility placed upon male teens (who may declare at best a dismissive 'gross') is evidenced by our female participants' experiences.
Beyond the skills of knowing how, when and what to sext, teens (and again, female teens especially) are incentivized to foster a certain level of awareness to manage any negative feedback when sexting, especially among LGBTQ2+ identifying young girls and women. Ally, a female aged 17 from Cyberville, demonstrated this need for additional awareness in her account of an instance where her semi-nude photo in the possession of an ex-boyfriend was shared among her peers. She explains that her ex-boyfriend was angry: cuz I left him … I came out as a lesbian and all the guys were mad at me, and like, my previous boyfriends are like 'what?', 'I didn't know', 'it wasn't my fault'… I dated them and I realized that I didn't like guys at all, right, and they were mad [so they tried to get back at me].
Here, the risks associated with context collapse are compounded on by both gender and sexual orientation. Teens who participate in networked publics are constantly negotiating ever-collapsing contexts to both positive and negative ends (Boyd, 2014;Marwick & Boyd, 2010;Vitak, 2012). Rather, a disordered self involves the loss of mental and emotional control regarding negative feedback, such as being blocked. Negotiating this is compounded on for LGTBQ2+ youth who face additional challenges associated with homophobic attitudes among peers, which is important when considering the extent relational aggression is fuelled by homophobia (Bailey & Steeves, 2015;Boyd, 2014; see also Pascoe, 2011). Fairbairn (2015, p. 234) observes that: when people's safety and integrity is compromised online, they are marginalized and/or pushed out of these spaces. When this happens repeatedly based on gender identity and/or sexual orientation (among other factors, including racism), these patterns of discrimination exclude certain social groups from full participation in society.
The gendered and heteronormative dynamics, here demonstrated regarding iniquitous risk distribution, are apparent among our female participants' experiences and perceptions with sexting. A marked, though not unanticipated, absence of such reflections and experiences among our male teen participants suggests the highly gendered boundaries that permeate 'agentic' choices in consensual sexting.

Discussion
The teens in our study construct, negotiate and tread the boundaries between private and public information sharing, as well as an ordered and disordered sense of self in their quest to foster intimacy and emotionally charged relationships with peers through the sharing of digital sexual material. Given that sexting takes place in networked publics and requires both a sender and a recipient, we argue that these boundaries are relational in their construction. Similar to Hart (2017) who studied young adults who anonymously post nude selfies online, our participants experience physical and emotional risks to their overall wellbeing. These risks may be cumulative over time given that once an image is sent, there is no way to know for certain who may receive it. The technological underpinning of networked publics suggests that content produced 'through social media are far from ephemeral', instead enduring through time and potentially becoming visible to multiple unintended audiences (Boyd, 2014, p. 11). Teens who sext risk their social standing and reputation among peers, adults and-in extreme cases-authorities. Thus, the 'spreadability' and 'searchability' (Boyd, 2014) of digital information means that teens also risk future opportunities lest employers or university recruitment find said images, and also risk criminal sanctions if such material is discovered by parents and reported to authorities.
Participants discuss the nuances involved in demarcating these boundaries during sexting and the mental and emotional work that goes into maintaining them. In doing so, they develop an illusion of control over how their sexual imagery is responded to by gauging whether the recipient will disseminate the image to unintended audiences. For the 'digital natives' (Prensky, 2009) of our study, sexting has becoming increasingly ubiquitous with coming-of-age and is normalized as a means of sexual experimentation and exploration during adolescence. We recognize that sexting leads to numerous sensations among teens, including the thrill, excitement, freedom and anxiety of potential intimacy and sexual discovery during adolescence. The uncertainty but hopefulness of sending private images with the intention of fostering a relationship with someone can be emotionally stimulating. These sensations, however, are braided with the riskiness of sexting-risks that we found, in line with numerous other studies, are highly gendered, with female teens experiencing the bulk of the shame and stigma that may accrue from non-consensual redistribution of sexts.
Our sample includes youth ranging from 13-to 19-year-old who have a range of experience levels with managing content spread through networked publics. The inexperience and naivety of most teens regarding intimate relationships puts them at additional risk of harm, both physical and emotional. The sting of having a trusting relationship violated for the first time may hurt substantially more than later experiences, meaning that age and experience likely influence the level of harm incurred. Our results reinforce many studies indicating that female teens are at particular risk of incurring harm as a result of sexting (Karaian, 2014;. Indeed, there is a marked disparity among the teens we interviewed regarding sexting experiences and perceptions. Female teens speak of walking a tightrope between being 'sexy enough' and 'being too sexy' (Steeves & Bailey, 2016, p. 74), frequently experiencing nonconsensual redistribution of their nude images by male partners. The males in our sample do not speak of sexting outside the frame of a mutually beneficial practice, evidencing a marked lack of responsibility for managing private contexts and preventing their exposure and collapse (for a discussion of male teen views on non-consensually sent 'dick pics', see also Ricciardelli & Adorjan, 2019). Sexting, in other words, is not often framed as risk-taking behaviour among male teens, in stark contrast to female teens, who frequently raise the theme of risk directly linked to male peers and partners. Clearly, wider gendered norms impact the narratives guiding teen sexting dynamics; norms related to patriarchy, gendered double standards as well as heteronormativity. Studying 'bystanders' to sexting, Harder (2021) found contradictory findings among male bystanders in particular, who expressed both excitement (receiving images of female nudes as a 'success' or 'checkmark') and moral dilemma (expressing degrees of sympathy for female victims of image-based sexual abuse, but under pressure not to take a stand due to the draw of bonding with male peers), (suggesting the influence of hegemonic masculinities; (Connell, 2005).
Strategies that female youth may engage in to resist often do little 'to change [wider] gendered double standards, where women's sexual agency is framed as shameful' (Harder, 2021, p. 663). Attention to broader societal messaging is obviated. The focus of policymakers and educators on youth sexuality and risk is evinced in programmes that purport abstinence or 'responsibilization' messages, and the criminalization of (often white, female, heterosexual) youth who are caught engaging in sexting or possessing sexts (Karaian, 2014;. One limitation of our research design is that youth are unlikely to share personal experiences of sending nudes during focus groups with other peers present. Oneto-one interviews or anonymous online surveys may better elicit such personal experiences. Furthermore, the ethical considerations of using focus groups led us to avoid asking teens in our sample about their sexual orientation or sexual experiences, meaning that demarcating the boundaries between risks experienced specifically by LGTBQ2+ youth went largely unexplored. Moreover, our sample of teens is largely white identifying, and a more inclusive sample with respect to race and ethnicity would undoubtedly reveal differing experiences with risk and reward (Fleschler Peskin et al., 2013;Houck et al., 2014). Future research is needed to understand how the risks of sexting are experienced by other groups; in Canada, this should extend to Indigenous youth specifically. Situating youth sexting in terms of intersecting identities is an avenue of research we hope will be explored in the future.
We acknowledge the agency of teens who are exploring their sexuality and relationships with others. In considering the pleasurable, voluntary aspect of sexting, we recognize that there are indeed cases where sexts are sent without incurring harm. Existing research suggests that there are cases of sexting that are motivated by pleasure and which do not lead to negative consequences (Albury et al., 2013;. Additionally, future efforts to mitigate risks associated with sexting might include rethinking approaches to youth sexuality and sexual exploration. As Lee and Crofts (2015, p. 457) noted in their review of extant research on sexting, 'pleasure and pressure are not necessarily mutually exclusive feelings-just as risk can be both exciting and dangerous'. In essence, we are engaged in a society that reproduces both the circumstances by which youth feel compelled to sext and the dynamics that make navigating boundaries difficult. It is not that sexting is unavailable as exciting and emotionally stimulating for female youth-it is that to access the excitement and emotional stimulation (the positive outcomes) females have to navigate a particular, gendered set of risks that males largely do not have to engage with.
The understandings of sexting that youth, especially female youth, hold require situating in the wider context of societal narratives which are frequently gendered and mediated by patriarchal assumptions, including gendered risk narratives received through cyber-safety messages in schools, and reinforced by messaging from parents. Educational responses that fail to attend the patriarchal and gendered context of sexting will not be effective (see Karaian, 2014) and may only act to embolden the notion that young females are solely responsible for any consequences. Young males, moreover, may not be disincentivized from redistributing intimate images given the persistent risk-linked to norms of hegemonic masculinity-of peer exclusion if they fail to brag about their possession of such images (Ringrose et al., 2012). Until these deeply rooted social norms are rendered more explicit in school curriculum, significantly more female youth than male will continue to perceive the Internet as an unsafe space (Johnson, 2015).
Research on how sexting is being experienced and responded to under the context of the current COVID-19 pandemic is also essential, as the risks we explicate here are likely further underscored through the shifts to 'new normal' social distancing and interpersonal communications from home via information communication technologies, especially social media.

Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding
The authors received no financial support for the research, authorship and/or publication of this article.

Rosemary Ricciardelli
https://orcid.org/0000-0002-0905-8968 Michael Adorjan https://orcid.org/0000-0002-1900-2087 Notes 1. We recognize this definition of sexting as diffuse and incorporating instances of both consensual-and non-consensual-based image/text sharing. The variance in the definition is in part due to limitations in our sample, specifically the nature of focus groups and the ethical considerations of asking minors to share their own experiences with sexting. We never asked youth to define for themselves what they consider sexting. Future research is needed to explore how young people themselves conceptualize sexting. 2. See https://laws-lois.justice.gc.ca/eng/acts/c-46/section-163.1.html. Despite the severity of being charged with child pornography, it may be the case that police are reluctant to formally charge youth as sex offenders, suggested by the lack of such prosecutions in the Canadian context. We also note, regarding the use of the term 'child pornography', that the language is problematic within the law, Here, image-based child sexual abuse better captures the nature of the offence and the harm experienced by victims.