Pro-anorexia/bulimia censorship and public service announcements: the price of controlling women

Abstract

Individuals, particularly women, are fixated on weight loss, driven by the goal of achieving a ‘skinny’ female physique that is desirable in western/ized cultures. There are online forums where individuals refer to themselves or their eating disorders as ‘pro-ana’ and ‘pro-mia’; their posts on these sites both align with, and challenge, what medical and mental health professionals define as serious mental health problems. In February 2012, the social media website Tumblr announced a policy to censor these online communities and use public service announcements (PSAs) to address ‘the problem’. Embracing contemporary ethnographic sensibilities, we present analyses that are attentive to nuanced meanings, and provide a critical feminist, sociological analysis of online comments from those who responded to the censorship and PSA policy. We argue that censorship extends the patriarchal control of women and that PSAs further the vested interests of corporate entities who profit from the marketing of services.

Keywords

censorship eating disorders feminist postmodernism pro-ana public service announcements social media virtual methods

Introduction

With the late modern normalization of dieting, the bombardment of weight-loss advertisements, and the media’s portrayal of the ‘ideal’ body type in Western culture, it is perhaps not surprising that extreme practices centered on weight control are prevalent in Western/ized patriarchal cultures (Malson and Burns, 2009). The routine depictions, and celebrations, of thinness are shaped by Western ideologies of gender that position women as inferior to men. Western philosophy has produced common images of ‘the body’ as ‘animal’ and ‘primitive’ (Bordo, 2003).¹ Furthermore, the body/mind duality has shaped Western ideologies of gender which socially construct woman as body. The social construction of woman as body in Western culture positions women as further from modernity and therefore unable to achieve the status that men maintain (Bordo, 2003). Within medical discourses, eating disorders have primarily been explained as a psychiatric problem that stems from the individual, and not as a cultural problem stemming from patriarchal cultures that have constructed women’s bodies – particularly excess ‘fat’ – as inferior and worthy of less status.

The emphasis on women’s thinness is part of a wider and longer history of feminine beautifying trends that could potentially be labeled as ‘self-harming’ or as many feminist scholars suggest, strategies for controlling women. For example, ancient Chinese society was a patriarchal culture, hierarchically structured and influenced by Confucian philosophy, that overtly proclaimed that men were superior to women (Hesse-Biber, 1996). Foot binding was a feminine beautifying practice of breaking the toes of Chinese girls so that their feet could not grow, often causing severe pain and a lack of mobility (Hesse-Biber, 1996). Hesse-Biber (1996) argues that the history of foot binding can help us understand how and why the thin ideal operates; both beauty ideals are mechanisms for controlling women in order to maintain the gender hierarchy. Hesse-Biber (1996: 24) provides a similar analysis of the practice of corseting in the 19th century in England and America, and poses the question: ‘Why did women corset themselves so willingly?’ The answer provided was: ‘like the women of ancient China, their body image largely determined their identity and the rewards they received’ (Hesse-Biber, 1996: 24). This answer provides an explanation for how the external controlling of women’s bodies became transformed into internalized social control. Self-imposed extreme dieting and exercising practices that result in a thin feminine body largely determine the identity and the rewards women receive in western/ized cultures. Communication platforms, like social media, were not available during the times of foot binding in Ancient China or corseting in the Victorian Era. Digitization now offers new possibilities for community and connection for those who have been constructed as engaging in contemporary self-harming strategies, such as anorexic and bulimic practices.

In tandem with technological developments, there has been growing evidence of online communities that provide spaces for the discussion and support of eating disordered practices. It is estimated that websites that cater to discussions about eating disorders number anywhere from 200 to 400 (Brotsky and Giles, 2007) to over 500 (Chelsey et al., 2003). Interactive messaging on these sites reveals the attitudinal and behavioral investments of those who identify as ‘pro-ana’ (supporting anorexia), ‘pro-mia’ (supporting bulimia), and ‘pro-edno’ (supporting eating disorders otherwise not specified). In this article, we use the term ‘pro-ana/mia’ to collectively refer to the individuals in these three categories.

‘Tumblr’ is a social media website that allows pro-ana/mia users, among others, to share text, photos, links, music, and videos in a creative blogging format. In February 2012, Tumblr announced a new policy that bans the posting of content that encourages its users to embrace anorexia, bulimia, or other eating disorders. In order to enforce the policy, Tumblr is monitoring users’ blogs and deleting content that is considered to be against the policy. The policy also dictates that Tumblr staff will attach a ‘public service announcement’ (PSA) to any user that searches for terms such as ‘pro-ana’ or ‘thinspiration’. These PSAs contain cautionary messages about the harms of pro-ana/mia practices, and they refer individuals to treatment services. According to Tumblr’s policy, censorship is aimed at removing content that actively promotes or glorifies ‘self-harm’. Tumblr’s rationale to tag pro-ana/mia content with PSAs is not explicitly stated, but PSAs are generally used to model desirable behaviors with the goal of changing (Alcalay, 1983), or stopping (Jung and Villegas, 2011), pro-ana/mia behaviors, and the ideologies that encourage them. Tumblr’s decision to censor these sites and promote PSAs is in keeping with the medical and social science discourses on eating disorders that are predominant in the scholarly literature and that cast pro-ana/mia individuals as mentally ill.

There are many examples of countries banning these websites. For example, France has banned the encouragement and glorification of extreme thinness on websites, advertisements, and magazines in response to the public concern about pro-eating disorder websites (Rodgers et al., 2012). In 2001, after receiving complaints, Yahoo began deleting pro-eating disorder websites (Ferreday, 2003), framing them as ‘self-harm’, and arguing that they are contrary to the terms of their service agreement. Social media websites such as Facebook have employees that monitor and delete pro-ana/mia groups on a regular basis (Peng, 2008). Some of these websites have produced guidelines and policies, such as ‘Instagram’s (2012) New Guidelines Against Self-Harm Images & Accounts’ after a similar social media website, Pinterest, did so in March 2012 (Olsen, 2012). Tumblr (2012) was the first social media website to instigate a similar policy in February 2012 after consulting with the National Eating Disorder Association (NEDA) located in the United States. Tumblr’s revised policy states,

Promotion and Glorification of Self-Harm. Don’t post content that actively promotes or glorifies self-harm. This includes content that urges or encourages readers to cut or injure themselves; embrace anorexia, bulimia, or other eating disorders; or commit suicide rather than, e.g., seeking counseling or treatment, or joining together in supportive conversation with those suffering or recovering from depression or other conditions. Dialogue about these behaviors is incredibly important and online communities can be extraordinarily helpful to people struggling with these difficult conditions. We aim to sustain Tumblr as a place that facilitates awareness, support and recovery, and to remove only those blogs that cross the line into active promotion or glorification of self-harm.

In addition, Tumblr staff state that alongside searches for tags, such as ‘pro-ana’, ’pro-mia’, ‘thinspiration’, and ‘thinspo’, the following public service announcement will appear:

Eating disorders are not lifestyle choices, they are mental disorders that when left untreated, can cause serious health problems, and at their most severe can even be life-threatening. For treatment referrals, information and support, please contact the National Eating Disorders Association’s Helpline at 1-800-931-2237 1-800-931-2237 FREE end www.nationaleatingdisorders.org. (Italics in original)

Tumblr (2012) staff reported that their policy post ‘provoked more than 25,000 likes, reblogs, and replies; and more than 2,500 … [Tumblr users] sent in comments by email’ (February, 2012).

Although the censorship of pro-ana/mia websites is not new (Ferreday, 2003), there has been little research on how users of social media respond online to this censorship or to PSAs. Our analysis addresses this gap in the literature by focusing on the online responses of individuals to Tumblr’s censorship and PSA policy. Furthermore, we build on the responses by arguing that censorship extends the patriarchal control of women and that PSAs further the vested interests of corporate entities who profit from the marketing of services. We begin by reviewing how the scholarly literature has analyzed pro-ana/mia websites, and we outline Tumblr’s censorship and PSA policy. We then explain our theoretical and methodological approaches followed by our findings, analyses, and conclusions. While the censorship of pro-ana/mia may intuitively appear to be a reasonable response to dealing with a social problem, using feminist postmodernism, and social constructionism’s concept of the ‘social problems game’, we critique many of the taken-for-granted responses to Tumblr’s policy.

Review of the literature

Analyses of pro-ana/mia

Pro-ana/mia, like anorexia and bulimia, are frequently portrayed as mental illnesses that require individual intervention and treatment from mental health professionals according to the medical model (e.g. see http://nedic.ca/know-facts/pro-eating-disorder-websites). Organizations like the National Eating Disorder Information Centre (2013) want to eliminate pro-ana/mia content and interactions online (Shade, 2003). The media are similarly oriented and promote medical model messages that highlight the problem of pro-ana/mia in a sensationalized fashion, arguably to sell news. For example, on his television show, Dr. Oz features pro-ana/mia websites as a ‘dangerous online movement’ (The Dr. Oz Show, 2012).

Like the medical and media analyses, most social science research studies are rooted in the assumption that pro-eating disorder websites are a problem that requires a solution (Bardone-Cone and Cass, 2007; Fox et al., 2005; Grunwald et al., 2008; Harshbarger et al., 2009; Rodgers et al., 2012; Sharpe et al., 2011; Strife and Rickard, 2011). Studies report the negative impacts that visiting pro-ana/mia websites can have on girls, including a higher drive for thinness, worsened perceptions of appearance, more perfectionism (Custers and Van Den Bulk, 2009), weight dissatisfaction (Bardone-Cone and Cass, 2007), and exposure to new skills for weight-loss techniques (Wilson et al., 2006). Pollack (2003) warns readers that pro-ana/mia websites may act as a ‘trigger’, inducing or worsening eating disordered behaviors in vulnerable women. Most of the academic literature maintains that pro-ana/mia websites actively support and promote ‘non-recovery’ from anorexia and bulimia (Csipke and Horne, 2007; Fox et al., 2005; Haas et al., 2011; Shade, 2003), while fewer studies make the case that pro-ana/mia websites encourage women to recover (Casilli et al., 2012).

Some of the social science literature has focused on describing the content of pro-ana/mia websites, noting that individuals who identify as pro-ana/mia engage in discussions online in many different forms. These include: bulletin boards, static websites, blog web pages, groups on social network sites, email groups, and instant messaging communities (Boero and Pascoe, 2012). Sharpe et al. (2011) provide a succinct summary of the types of content found on pro-eating disorder websites:

Tips on weight loss and how to keep it secret; tips on how to deceive others about eating habits; tips on purging (food that can be vomited easily, keeping it secret); images and texts promoting extreme thinness or weight loss (‘thinspiration’); a personification of eating disorders as supportive friend; creative writing glorifying eating disorders; and a list of commandments for the ana/mia ‘religion’. (p. 33)

Overall, social scientists negatively portray pro-ana/mia websites as graphic, dangerous, and ‘disgusting’ online environments (Ferreday, 2003). For example, Hammersley and Treseder (2007: 191–192) convey this message by quoting a pro-ana/mia website creator who writes,

My lifetime goal is to die from starvation.

The more weight I lose, the better I feel.

There’s no such thing as too thin.

Eating is a sign of weakness.

Perfection is achieved through restricting.

Anorexia will make you beautiful.

No one can do what I (we) do.

Protruding bones make you look awesome.

Anorexia shows you are superior.

Critics of anorexia are jealous or fat.

No matter where you go, you’re still thin.

No matter what you do, you’re still thin.

I am in control, ALWAYS.

No one can take ana from me, no one!

Extreme emaciation is a good start. (bold and italics in the original)

Other social science literature has focused on explaining pro-ana/mia websites. Often these articles highlight the positive functions of these websites and make the case that they serve as communities for those who feel isolated and are looking for social and emotional support from like-minded individuals (Boero and Pascoe, 2012; Brotsky and Giles, 2007; Canto-Mila and Seebach, 2011; Ferreday, 2003; Riley et al., 2009). For example, Brotsky and Giles (2007: 101) found that ‘a general sense of support is the central function of pro-ana sites, whether that support is directed at maintaining the eating disorder or choosing to seek treatment’. Sharpe et al. (2011) report that ‘many studies also find that members receive emotional support from the group (Mulveen and Hepworth, 2006; see Boero and Pascoe, 2012; Brotsky and Giles, 2007; Csipke and Horne, 2007; Gavin et al., 2008). Juarascio et al. (2010) report that women with eating disorders who access these websites lack strong social support systems offline and often suffer from depression and social anxiety. Some have argued that using the websites is a coping strategy for individuals (Lyons et al., 2006) and being part of this online community has been shown to improve self-esteem (Csipke and Horne, 2007). Others have argued that engaging in these websites helps negotiate, manage, and develop a sense of identity (Gavin et al., 2008; Haas et al., 2011; Rodgers et al., 2012) because of the support and approval of the online community.

There has been very little in-depth research that focuses specifically on the implications of the censorship of pro-ana/mia websites, although the general social response of ‘disgust’ has driven censorship policies. Ferreday (2003) argues that ‘the dominant emotion [of society] is not sympathy or concern for the young women involved, but disgust’ (p. 288), as evidenced by newspaper articles and comments on the internet which allege that the anorexic body, ‘evokes an instant reaction of disgust’ (p. 290), causing individuals to become ‘sick to the stomach’ (p. 291), ‘gag’ (p. 290), and feel ‘a desire to vomit’ (p. 292; see also Hammersley and Treseder, 2007). Ferreday argues that the disgust response combines with the sensationalism of harm to fuel the desire to censor pro-ana/mia.

Many have theorized the implications of the censorship of pro-ana/mia, concluding that it is a futile goal (Brotsky and Giles, 2007; Csipke and Horne, 2007; Custers and Van Den Bulk, 2009; Gailey, 2009; Shade, 2003). A recent study by Casilli et al. (2013) mapped networks of French pro-ana/mia websites over 2 years and found that censorship initiatives have caused pro-ana/mia networks to become more entrenched, isolated, exclusive, and harder to be reached by physicians, families, and charities (Casilli et al., 2013). Casilli et al. (2013) deduce from their findings that ‘censorship means bad news for health care providers and policy makers alike’ (p. 95) because health information and awareness campaigns are increasingly having a harder time finding pro-ana/mia communities. In a similar vein, others have argued that censorship has removed an important means of social support for pro-ana/mia individuals that helped them cope with ‘a life of isolation, fear and self-hatred’ (Gailey, 2009: 107). Csipke and Horne (2007) also suggest that it would be better for ‘clinicians to acknowledge the needs these sites fulfill and to address them in conventional treatments for eating disorders’ (p. 27). Shade (2003) posits that regardless of whether the websites are pro-ana/mia- or anti-ana/mia-oriented, they are ‘exemplary of young women creating their own discursive online community … to debate this issue in their own space and on their own terms’ (p. 2). Only a few analyses argue that censorship serves to deflect attention away from the larger and more germane issue of the mainstream cultural messages that valorize thinness (e.g. see Boero and Pascoe, 2012: 36; Gailey, 2009: 107).

Analyses of PSAs

There has been no research that provides evidence of the effectiveness of PSAs in reducing pro-ana/mia ‘self-harm’. In a related vein, Corrigan (2012) reports that there are very few studies that evaluate the effectiveness of PSAs for eliminating the stigma around mental illness. In addition, Corrigan (2012) argues that those who are running PSA campaigns should be responsible for determining the efficacy of their initiatives. Lienemann et al. (2013) make the case that while the spirit behind PSAs to help individuals with depression may be benevolent, there may be untoward effects of such PSAs. Their study’s findings suggested that individuals identified as having depression react differently to PSAs than non-depressed individuals, so that PSAs for depression ‘cannot be based on face value, good intentions and hope’ because the high mortality rate associated with depression is a serious risk (Lienemann et al., 2013: 725). To use PSAs uncritically is, they argue, ‘a behaviour akin to reckless endangerment’ (Lienemann et al., 2013: 726). In fact, one study that did examine the effectiveness of PSAs for suicide prevention found that they had negative consequences such as making targeted individuals less likely to seek help (Klimes-Dougan et al., 2013). Furthermore, the fact that PSAs recommend treatment services for pro-ana/mia is also problematic, because evidence-based research on the effectiveness of eating disorder treatment is lacking (e.g. see Agras et al., 2004: 519; Brain, 2006: 179; Rance et al., 2014: 111).

Methodology

Theoretical frameworks

This study is guided by grounded theorizing (Charmaz, 2006), an inductive approach to research that allows theoretical insights to emerge from the data throughout the entire research process. Sensitizing concepts (Blumer, 1969, as referenced by Charmaz, 2006) that were rooted in critical theorizing were at play, but were ‘only a place to start, not end’, as noted by Charmaz (2006: 17). As we gathered data, we drew upon theories that were pertinent as we coded and wrote analytic memos, and ‘discover[ed] that our work suggest[ed] pursuing more than one analytic direction’ (Charmaz, 2006: 10). The result was an approach to theorizing that included feminist postmodernism (Bordo, 2003), and the social constructionist concept of ‘the social problems game’ (Loseke, 2003).

Feminist postmodernism

From a postmodern perspective, ‘truth’ is understood to be privileged ‘discursive distinctions’ that have been deemed superior to others’ through the exclusionary power of elite ‘who claim to have special knowledge earn[ing] the right to decide the fate of those who not share this knowledge’ (Henry and Milovanovic, 1999: 5). These distinctions ‘are conceptual and … made through communication, particularly, but not exclusively written or spoken language, referred to by postmodernists as “discourse” (Manning, 1998)’ (as referenced by Henry and Milovanovic, 1999: 5 italicized in original), and they can have harmful effects. Our principle tool of analysis is to use ‘deconstruction’ (Derrida, 1970, 1981, as referenced by Henry and Milovanovic, 1999: 5 italicized in original) ‘to expose the socially constructed nature of privileged knowledge through what is called ‘critique’ …: a continuous process of challenge to those who claim to know or hold the truth’ (Henry and Milovanovic, 1999: 5 italicized in original).

Combined with this postmodern perspective is a feminist perspective that highlights the gendered nature of ‘truth’ discourses. Male-dominated power structures contribute to women’s oppression through discourses that devalue women’s experiences, perspectives, and knowledge, and that exercise social control over their bodies and voices (Wood, 2008: 328). In this vein, Malson and Burns (2009) argue that disordered eating ‘can only be adequately understood within the context of the oppressive gender ideologies and inequalities in gender power-relations operating in (western/ised) patriarchal cultures’ (p. 1). Exemplary in this regard is Green’s (2005) explanation of how neoliberalism and the media’s promotion of ‘feminine’ imagery is tied to an increase in anorexia and bulimia. Further, Moulding (2003) stresses the need for practitioners to critically reflect on how explanations and responses to eating disorders are ‘femininized’ (p. 57) to the detriment of their female patients. A central aim of our study has been to name (Wood, 2008) the taken-for-granted and accepted responses to pro-ana/mia that are often perceived as normal, ‘okay’, and necessary.

Social constructionism

The social constructionist concept of ‘the social problems game’ (Loseke, 2003) provides the framework for theorizing about the vested interests involved in controlling the social problem of pro-ana/mia through censorship and PSAs. The ‘social problems game’ is driven by claims-makers, those privileged to have a voice, and often political persuasion, who instill popular worry by telling us what is right and wrong, producing villains who are blamed, and then highlighting the consequences of the social problem. These morally charged constructions are strategically typified to be inclusive of a wide scope of innocent victims who are ‘suffering’, and they often produce an ‘it-could-be-you!’ fear among their audience (Loseke, 2003). The construction and control of social problems are driven by vested interests (Loseke, 2003), such as the financial motives behind the control of pro-ana/mia.

Grounded virtual liquid ethnography

Christine Hine (2005) contends that the Internet (World Wide Web) is both a cultural context and a cultural artifact – a field site for ethnographic inquiry. Hine argues that online textual interactions are just as important as face to face interactions. In this study, the online responses to Tumblr’s policy represent cultural artifacts that exist in virtual spaces that, while perhaps unconventional, are no less ‘real’ than other physical spaces. Castells (1996) makes the case that crimes that occur through the internet ‘all have very real local (and global) consequences, yet “take place” within, through or across the “space of flows”’ (as referenced by Kindynis, 2014: 229). Similarly, pro-ana/mia discussions, and responses to these, are taking place in spaces that also have real local and global consequences.

Because of the vastness of the Internet, and the dynamic and ever-evolving nature of online interactions, our research called for a method of data collection and analysis that was attuned to this particular environment. We became immersed in what we are calling an unobtrusive ‘grounded virtual liquid ethnography’, a type of ethnographic sensibility (Ferrell et al., 2008) that builds on the work of Hine (2005, 2011), Charmaz (2006), and Ferrell et al.’s (2008) conceptualizations of ethnography.

Hine (2005) explains that web surfing and following hypertextual links are part of a virtual ethnography strategy, an approach that ‘despatializ[es] notions of community’ (p. 58) and understands online spaces as ‘increasingly the site of the exercise of power by the elite’ (p. 84). Thus, Hine’s virtual ethnography is in keeping with Charmaz’s understanding of grounded theory in that it allows ethnographers to theorize based on the data as they emerge through following connections. Similarly, Ferrell et al. (2008) attest to the value of a ‘liquid ethnography’ which is ‘attuned to the dynamics of destabilized, transitory communities; ethnography immersed in the ongoing interplay of images; and ethnography comfortable with the shifting boundaries between research, research subjects, and cultural activism’ (Ferrell et al., 2008: 182). In adopting ethnography, we are able to produce critical cultural analyses that have ‘a sensitivity to subtleties of meaning, an openness to the orientations of others – even if those others are textual in nature’ (Ferrell et al., 2008: 189).

Our approach to ethnography has been unobtrusive because we have not responded to comments from the pro-ana/mia online community. The data collection and analysis has been an iterative process that involved repeated immersions into the World Wide Web from October 2013 to January 2014. Euphemistically referred to as ‘surfing’, this approach attuned us to the destabilized and transitory nature of online communities, and provided excellent opportunities to explore responses to the Tumblr policy.

After deciding to concentrate on comments to Tumblr’s policy, 52 responses were collected. Almost all of the responses to Tumblr’s posting of the policy focused on pro-ana/mia content. Data collection stopped at a sample size of 52 for two reasons. First, new responses online became harder to find in that relevant links became increasingly spread out. Second, category saturation had been met because fresh data no longer added new theoretical insights (Charmaz, 2006: 113). All 52 responses were organized into one Word document that included commenters’ user names and website links. The responses were stripped of their identifying information, and organized into 52 separate documents to maintain the confidentiality of the commenters. Female pseudonyms were randomly selected to replace user names because the literature to date states 90%–95% of those with anorexia and bulimia are women (Koski, 2008, as referenced by Boero and Pascoe, 2012). The 52 documents were inputted into NVivo, a qualitative data software program, where they were open-coded for major themes in order to organize and categorize the data (Neuman, 2007). The open-coded themes were axial coded to identify connections between categories (Neuman, 2007). After the ‘findings’ section was complete, selective coding, a process of drawing analytical connections between themes and applying them to theory (Neuman, 2007), helped generate additional theoretical insights.

Findings

Two major types of written comments to Tumblr’s censorship and PSA policy were identified: responses that denounce pro-ana/mia forums and support the policy, and responses that support pro-ana/mia forums and contest the policy.

Denouncing pro-ana/mia forums and supporting the policy

The responses that supported the policy were straightforward; the stance against pro-ana/mia was clear and commenters were happy about Tumblr’s new policy. The most common reason given for supporting censorship and PSAs is that pro-ana/mia content and interactions are harmful to Tumblr’s users. As Jennifer said: ‘the whole “reblog” feature of tumblr makes self-harm/pro-ana/etc blogs particularly insidious, because it provides that community/social encouragement towards harming yourself’. These commenters provided personal stories that exemplified the harm that they believed was caused by the Tumblr blogs. Chloe said,

There’s way too much celebration and acceptance of disordered eating here. Personally, I went through a period in college when I threw up all my food. So, when I see blogs with bulimia tips and tricks, I know how much they would have messed me up if they were available back then … it’s nice to see Tumblr confront the fact that they have been inadvertently fostering a community of sickness.

Ruth used an experience with her roommate to exemplify how eating disorders have harmed her:

The fact that anyone can post something promoting or glorifying eating disorders or self-injury is absolutely detestable; I … [lived] with a girl who weighed under ninety pounds, who exercised compulsively, who had panic attacks over eating small amounts of food or missing going to the gym, and who regularly (and unintentionally) helped aggravate my own self-image issues … seriously the worst three months ever … Sure, a policy against self-harm blogs … [is] the right kind of message that the internet needs.

Individuals who support censorship also explained that there is a need to protect vulnerable individuals from content that could ‘trigger’ or induce others to engage in eating disorders. Ruth argued that when individuals post content about eating disorders they are harming ‘every person who is out there looking for a trigger … [they are] doing it intentionally …’. Similarly, Heather stated that she is glad ‘someone has the brains to differentiate between self-expression and triggering’.

Supporting pro-ana/mia forums and contesting the policy

The responses that contested the policy, in general, were more numerous, complex, and lengthy than those that supported the policy, and commenters were passionate about the need for open online discussion. One argument is that Tumblr’s policy unequally targets pro-ana/mia voices while other online communities are allowed to retain their freedom of speech. For example, Terrilyn argues that ‘a blanket policy like this is going to do harm’ because it prevents the freedom of speech of ‘blogs recommending safer techniques for self-harm and starvation [that] serve an important function’ such as saving lives’. Commenters argued that Tumblr ‘is shutting down a community where people can talk openly without addressing the (actually evil) blogs that may have caused them to be where they are at’ (Carla). For example, Carla stated,

It’s still okay to have a blog where you take pictures of people eating and call them fat. In essence, it’s okay to have the kind of blog that causes someone to degenerate into the mental state where they might get an eating disorder, but once they get one we’ll sweep them under the rug unless they only talk about their disease in a manner a bunch of bros unfamiliar with eating disorders find acceptable.

Rebecca asked Tumblr staff: ‘do you plan to ban pro-fat accounts, too? What about pro-medical marijuana, or pro-gay rights? The Christian Right would LOVE to see you banning everything they see as unsavory or inappropriate’. Commenters also argue that the policy discriminates against women. Miranda writes that ‘it is strange how shit that female people are more likely to do needs this sort of action, but racism, sexism, homophobia, or pornbots’ do not. Carla states, ‘it’s not a secret that this new rule will target primarily women. Sick women that have finally found a community where they don’t feel alone’. The anger about perceived unequal treatment was palpable in Christina’s comment:

we know about your fucking hotlines and your non-profit organizations or whatever the fuck else. We know because their analysis and their philosophy and their very presence are culturally privileged. We know because they continually position themselves as authorities on us, and using their immense social capital they get to define our identities and experiences before we ever get the chance. Their shit is all over our tags already, and given the violence inherent in such approaches, this re-enforces the messages that we’re all worthless failures, especially for those of us who are otherwise marginalized. Many of us have been forcibly hospitalized and have heard it all before. (Bolded in original text)

Censorship was also seen as being harmful because it increases a commitment to the pro-ana/mia ideology and results in isolation from alternative ideologies. Emma noted that people already tell them all the time that what they are doing is wrong and she said ‘every time I’ve been part of a website that was deleted for its “pro-ana” content, the members have always reacted with outrage: it strengthens their belief that the entire world is against them and just doesn’t understand’. Another argument made is that bloggers who are censored will be less likely to seek or accept help because they will become more isolated and their pro-eating disordered ideologies will be strengthened, ‘driving [people] to more unsavory places’ and making it harder for family and friends to discover that their teenager is suffering (Rebecca). Patricia stated that ‘talking … [or writing] about these issues can be cathartic …’ since many bloggers do not have support systems offline.

Responses that contest censorship also provide explanations of the positive functions that are served by pro-ana/mia on Tumblr – they note that individuals ‘depend’ on Tumblr’s non-judgmental and non-censoring community to foster their well-being. Arej explains that ‘Tumblr provides these people with support and a sense of community through this open and honest communication’. Dariya begs Tumblr: ‘PLEASE PLEASE PLEEEEEEEEEEEEASE! Don’t delete/ban/trash/block any self-harm, ana-mia, suicide, or other similar blogs. Please’. She explains her plea by noting all of the help she has received from relating to people on Tumblr. Many of the commenters say that they use the support as a coping mechanism to deal with their personal struggles. For example, Sabrina explains,

[M]y best friend is a girl i met on my thinspo blog two years ago. she struggled with bulimia, and we spoke all the time …. she is helping me, she supports me when i eat toast in the morning, she supports me when i eat snow peas in the afternoon, she supports me when i go for a weigh in at the doctor’s office … our friendship couldn’t have been possible without tumblr. i recovered and it helped me so much. my first suicide attempt only failed since i told my friend on tumblr goodbye. she called the police though she was four states away and told them what was about to happen. she saved me … If i didn’t have my tumblr, i would be dead.

Some commenters situated pro-ana/mia as a symptom of broader social problems that require attention. For example, Arej argued that ‘the “tips” and “diet advice” offered by pro-ana are really not any different in their actual *content* to what you see in 100% legal girls’ and women’s magazines’. Nabila posits that it is ‘the racism. The homophobia. The ableism. The sexism. And all the other ridiculous shit that goes on in the world. Not a blog telling me that eating disorders are cool that causes someone to self-harm’. Zahra wishes that Tumblr was able to do something that would stop girls from wanting to make the blogs in the first place and argues that the problem will still exist even if the policy ‘works’ at censoring dialogues. These types of comments suggested that censorship is a superficial response that will not have real benefits. As Zahra stated,

It’s fighting the symptom instead of the cause. I find it irritating. This is a problem that is much bigger than explicitly proana/promia/proselfharm blogs and if anyone thinks this is going to make those blogs go away or magically fix everything or ‘cure’ everyone, they have a seriously skewed idea of what the problem is in the first place.

Commenters who talked about Tumblr’s PSA policy are opposed to the marketed messages because they think that the referrals will not help. Nabila stated: ‘[She] does not need a glib little message of “Go here for help!” because that won’t do shit’, a comment that provoked 25,482 likes and reblogs as of 8 April 2014. Rebecca also challenged the treatment solution that is being advertised through Tumblr’s PSAs, arguing that ‘abstinence-only addiction rehabilitation does not work for everyone’.

Discussion

Many commenters denounce pro-ana/mia forums, and support Tumblr’s censorship and PSA policy. However, there are just as many, if not more, commenters whose commitment to contesting the policy is portrayed through the depth and length of their comments. This second group argues that these forums serve as important sources of support and many individuals reflect on the larger cultural contexts that fuel eating disorders. Furthermore, they question the effectiveness of PSAs.

Drawing on feminist postmodernism, we argue that those who support the policy are discursively producing knowledge that silences the vulnerable, a defining trait of late modernity (Ferrell et al., 2008). Tumblr’s PSAs advertise primarily medical treatments for pro-ana/mia, and these discourses are endorsed institutionally by authoritative bodies such as the National Eating Disorder Association (NEDA). From a social constructionist perspective, these messages are promoted by claims-makers who have been privileged with the ‘expert’ voice in regard to eating disorders and by extension, pro-ana/mia: They have ownership over the ‘problem’ (Loseke, 2003). Denouncing alternative discourses becomes a moral act, through positioning pro-ana/mia members as simultaneously both the victims of eating disorders and the villains who are subjecting others to life-threatening behaviors. As such, censorship is justified (Ferreday, 2003).

As we discussed, medical discourses define pro-ana/mia as harmful, and Tumblr’s censorship policy builds on that idea by saying that pro-ana/mia discussions are harmful. While some of the social science literature argues that pro-ana/mia discussions are harmful, there appears to be no scholarly evidence to back up these claims – they are seemingly based on common-sense notions of what is harmful. Furthermore, there is no ‘proof’ that censorship and PSAs prevent any of the suggested harms that are alleged to come from pro-ana/mia discussions. In fact, some evidence suggests that the traditional approaches to treatment have low success rates (Agras et al., 2004; Rance et al., 2014; Starr and Kreipe, 2014; Suokas et al., 2014). Still, no one with authority questions censorship or the PSA-style advertisements that are being posted on Tumblr.

Those who contest the policy support their position by focusing on harm reduction, but their argument is that the prevention of open discussions engenders harm. These commenters make persuasive points about the negative impacts of censorship: the prevention of freedom of speech, the entrenchment and isolation of women with eating disorders, and the destruction of supportive communities. Their critiques of censorship are in keeping with feminist analyses that alert us to the many ways in which women’s bodies and voices are controlled in patriarchal society. It is primarily men who represent the powerful elite of consumer society, and censorship and PSAs reflect how male authority maintains and contributes to gendered power relations. As Wood attests, ‘the greater power generally held by men is a primary reason that women’s experiences, perspectives, and knowledge have been devalued, and that women’s voices have been suppressed’ (Wood, 2008: 328). To give legitimacy to the voices that contest censorship and PSAs would, according to these commenters, serve a positive function by providing a sense of community, catharsis, and life-support.

Tumblr’s policy also reinforces the idea that eating disorders are confined to a specific group of women, and in so doing masks the extent to which women experience disordered perceptions of their bodies and disordered eating practices that follow. As Bordo (2003) noted: ‘… eating disorders, far from being “bizarre” and anomalous, are utterly continuous with a dominant element of the experience of being female in this culture’ (p. 57). Indeed, day-to-day life is saturated with routine depictions, and celebrations, of the thin female body in Western culture, a form that is unattainable for the vast majority of women. Censoring pro-ana/mia is about silencing women, and shutting down critical analyses that might expose both the motives of those who profit from women’s dissatisfaction with their bodies and the patriarchal ideologies that keep women in an ‘endless pursuit of want’ (Hayward, 2004: 458) that cannot be satiated.

The characterization of eating disorders as needing medical services has economic implications that are enormous. The PSA advertisements for recovery organizations financially benefit the agencies that produce the PSAs, and organizations whose services are featured in the PSAs. Countries spend up to 100 million dollars annually on medical services for eating disorders (British Columbian Ministry of Health Services, 2009–2010; Simon et al., 2005); hence, there is a lot of money at stake in owning the solutions for this ‘problem’. With respect to Tumblr, its staff works in collaboration with NEDA who gets free or paid advertising for their organization’s services anytime someone uses a keyword such as ‘pro-ana’, ‘anorexia’, or ‘thinspiration’ (http://www.nationaleatingdisorders.org/press-room/press-releases/2012-press-releases/national-eating-disorders-association-partners-tumblr). Tumblr claims they are censoring and putting up PSAs to protect their users, but the closeness of the ties between Tumblr and the NEDA raise questions about the ways in which each are profiting from the policy. Even though the organizations featured in the PSAs may be non-profit, they benefit from the direct advertising because it leads to additional ‘business’ which helps to legitimize the need for outside funding and donations. Just because an organization is not-for-profit does not mean that their services are automatically humanitarian and benevolent.

Conclusion

Substantively, this study contributes to a gap in the scholarly literature by providing an analysis of the online comments that address Tumblr’s policy regarding pro-ana/mia discussions. Our study also makes theoretical contributions by providing a critical, sociological analysis of responses to Tumblr’s censorship and PSA policy through the triangulation of feminist postmodern and social constructionist perspectives. Our analyses add to the online comments by addressing: how medical discourses dominate understandings of pro-ana/mia; the role of claims-makers in owning the ‘problem’; the impact of Tumblr policy on women; and the vested interests of corporate elites who stand to benefit from the promotion of services to treat pro-ana/mia individuals.

Our analysis calls into question the censoring of online discussions about pro-ana/mia, and the connections between censoring, PSAs, and broader issues related to the control of women’s bodies and voices. Those who contest censorship argue that Tumblr’s policy isolates women from community members, alternative ways of knowing, and life-saving coping mechanisms. Furthermore, the unquestioned motives that underlie Tumblr’s response to pro-ana/mia raises serious questions about whose interests are fundamentally at stake. Those involved in pro-ana/mia censorship and related PSAs have pecuniary vested interests in perpetuating and monopolizing the social problem of eating disorders – by casting individuals as sick and in need of services, corporate entities stand to profit financially. The more that individuals are convinced that they need the services that are marketed through PSAs, the more power afforded to those involved in the production and promotion of the PSAs – there is a lot to be gained economically by controlling this ‘problem’.

Our research signals the importance of supporting open, online discussions around pro-ana/mia and related topics that involve women’s self perceptions and practices. The medical model need not be the ‘only game in town’, nor should it be given its questionable efficacy. Discussions online create the potential for ‘cultural resistance and transformation’ (Bordo, 2003: 218) by allowing alternative discourses and interactions that provide opportunities for women to re-frame analyses of, and responses to, eating ‘disorders’. Pro-ana/mia is not a stand-alone problem for individual women – many, if not most women experience ‘issues’ with respect to their bodies even if they do not identify with pro-ana/mia. It is a logical response in a society that valorizes thinness. Digitization offers an ability for women who are engaging in supposed self-harming practices to collectively communicate, connect, and potentially resist in ways that were not possible during the times of foot binding in Ancient China or corseting in the Victorian Era. As such, there is a need for research that further examines whether, and in what ways, online environments develop and extend critical consciousness about eating disordered ideologies, practices, and solutions to these.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Agras

Brandt

Bulik

. (2004) Report of the National Institutes of Health workshop on overcoming barriers to treatment research in anorexia nervosa. International Journal of Eating Disorders 35(4): 509–521.

Alcalay

(1983) The impact of mass communication campaigns in the health field. Social Science & Medicine 17(2): 87–94.

Bardone-Cone

Cass

(2007) What does viewing a pro-anorexia website do? An experimental examination of website exposure and moderating effects. International Journal of Eating Disorders 40(6): 537–548.

Boero

Pascoe

(2012) Pro-anorexia communities and online interaction: bringing the pro-ana body online. Body & Society 18(2): 27–57.

Bordo

(2003) Unbearable Weight: Feminism, Western Culture, and the Body. Oakland, CA: University of California Press.

Brain

(2006) Feeding anorexia: gender and power at a treatment center. Feminist Review 83: 177–180.

British Columbian Ministry of Health Services (2009–2010) Action plan for provincial services for people with eating disorders. Available at: http://www.health.gov.bc.ca/mhd/pdf/ed-services-action-plan-master.pdf (accessed 18 October 2014)

Brotsky

Giles

(2007) Inside the ‘pro-ana’ community: a covert online participant observation. Eating Disorders 15(2): 93–109.

Canto-Mila

Seebach

(2011) Ana’s friends. Friendship in online pro-ana communities. Sociological Research Online 16(1): 152–164.

10.

Casilli

Pailler

Tubaro

(2013) Online networks of eating-disorder websites: why censoring pro-ana might be a bad idea. Perspectives in Public Health 133(2): 94–95.

11.

Casilli

Tubaro

Araya

(2012) Ten years of ana: lessons from a transdisciplinary body of literature on online pro-eating disorder websites. Social Science Information 51(1): 120–139.

12.

Charmaz

(2006) Constructing Grounded Theory: A Practical Guide through Qualitative Analysis. Thousand Oaks, CA: Sage.

13.

Chelsey

Alberts

Klein

. (2003) Pro or con? Anorexia nervosa and the internet. Journal of Adolescent Health 32(2): 123–124.

14.

Corrigan

(2012) Where is the evidence supporting public service announcements to eliminate mental illness stigma? Psychiatric Services 63(1): 79–82.

15.

Csipke

Horne

(2007) Pro-eating disorder websites: users’ opinions. European Eating Disorders Review 15: 196–207.

16.

Custers

Van Den Bulk

(2009) Viewership of pro-anorexia websites in seventh, ninth and eleventh graders. European Eating Disorders Review 17(3): 214–219.

17.

Ferreday

(2003) Unspeakable bodies: erasure, embodiment and the pro-ana community. International Journal of Cultural Studies 6(3): 277–295.

18.

Ferrell

Hayward

Young

(2008) Cultural Criminology: An Invitation. London: Sage.

19.

Fox

Ward

O’Rourke

(2005) Pro-anorexia, weight-loss drugs and the internet: an ‘anti-recovery’ explanatory model of anorexia. Sociology of Health and Illness 27(7): 944–971.

20.

Gailey

(2009) ‘Starving is the most fun a girl can have’: the pro-ana subculture as edgework. Critical Criminology 17(2): 93–108.

21.

Gavin

Rodham

Poyer

(2008) The presentation of ‘pro-anorexia’ in online group interactions. Qualitative Health Research 18(3): 325–333.

22.

Green

(2005) Dialogue with Chilean novelist, Diamela Eltit. Feminist Review 79: 164–171.

23.

Grunwald

Wesemann

Rall

(2008) Pro-anorexia websites: an underestimated and uncharted danger! Child and Adolescent Mental Health 13(2): 96–98.

24.

Haas

Irr

Jennings

. (2011) Communicating thin: a grounded model of online negative enabling support groups in the pro-anorexia movement. New Media & Society 13(1): 40–57.

25.

Hammersley

Treseder

(2007) Identity as an analytic problem: who’s who in ‘pro-ana’ websites? Qualitative Research 2: 283–300.

26.

Harshbarger

Ahlers-Schmidt

Mayans

. (2009) Pro-anorexia websites: what clinicians should know. International Journal of Eating Disorders 42(4): 367–370.

27.

Hayward

(2004) City Limits: Crime, Consumer Culture and the Urban Experience. London: The Glass House Press.

28.

Henry

Milovanovic

(1999) Constitutive Criminology at Work: Applications to Crime and Justice. Albany, NY: State University of New York Press.

29.

Hesse-Biber

(1996) Am I Thin Enough Yet? Oxford: Oxford University Press.

30.

Hine

(2005) Virtual Methods: Issues in Social Research on the Internet. New York: Oxford International Publishers Ltd.

31.

Hine

(2011) Towards ethnography of television on the internet: a mobile strategy for exploring mundane interpretive activities. Media, Culture & Society 33(4): 567–582.

32.

Instagram (2012) Instagram’s new guidelines against self-harm images & accounts. Available at: http://blog.instagram.com/post/21454597658/instagrams-new-guidelines-against-self-harm-images (accessed 18 October 2014)

33.

Juarascio

Shoaib

Timko

(2010) Pro-eating disorder communities on social networking sites: a content analysis. Eating Disorders: The Journal of Treatment & Prevention 18(5): 393–407.

34.

Jung

Villegas

(2011) The effects of message framing, involvement, and nicotine dependence on anti-smoking public service announcements. Health Marketing Quarterly 28(3): 219–231.

35.

Kindynis

(2014) Ripping up the map. British Journal of Criminology 54: 222–243.

36.

Klimes-Dougan

Klingbeil

Meller

(2013) The impact of universal suicide-prevention programs on the help-seeking attitudes and behaviours of youths. The Journal of Crisis Intervention and Suicide Prevention 34(2): 82–97.

37.

Lienemann

Siegel

Crano

(2013) Persuading people with depression to seek help: respect the boomerang. Health Communications 28: 718–728.

38.

Loseke

(2003) Thinking about Social Problems, 2nd edn. Hawthorne, NY: Aldine De Gruyter.

39.

Lyons

Mehl

Pennebaker

(2006) Pro-anorexics and recovering anorexics differ in their linguistic internet self-presentation. Journal of Psychosomatic Research 60(3): 253–256.

40.

Malson

Burns

(2009) Critical Feminist Approaches to Eating Dis/Orders. East Sussex: Routledge.

41.

Moulding

(2003) Constructing the self in mental health practice: identity, individualism and the feminization of deficiency. Feminist Review 75: 57–74.

42.

Mulveen

Hepworth

(2006) An interpretive phenomenological analysis of participation in pro-anorexia internet site and its relationship with disordered eating. Journal of Health Psychology 11: 283–296.

43.

National Eating Disorder Information Center (2014) Understanding statistics on eating disorders. Available at: http://nedic.ca/know-facts/pro-eating-disorder-websites (accessed 18 October 2014)

44.

Neuman

(2007) Basics of Social Research: Qualitative and Quantitative Approaches. Boston, MA: Pearson Education, Inc.

45.

Olsen

(2012) Pinterest bans thinspiration and pro-ana content … But that doesn’t mean that it’s going away. Available at: http://www.blisstree.com/2012/03/27/beauty-shopping/pinterest-terms-of-service-bans-thinspiration-787/ (accessed 18 October 2014)

46.

Peng

(2008) Out of the shadows, Newsweek, 22 November. Available at: http://www.newsweek.com/pro-anorexia-groups-spread-facebook-85129 (accessed 18 October 2014)

47.

Pollack

(2003) Pro-eating disorder websites: what should be the feminist response? Feminism and Psychology 13: 246–251.

48.

Rance

Clarke

Moller

(2014) ‘If i see somebody. i’ll immediately scope them out’: anorexia nervosa clients’ perceptions of their therapists body. Eating Disorders: The Journal of Treatment & Prevention 22: 111–120.

49.

Riley

Rodham

Gavin

(2009) Doing weight: pro-ana and recovery identities in cyberspace. Journal of Community & Applied Social Psychology 19(5): 348–359.

50.

Rodgers

Skowron

Chabrol

(2012) Disordered eating and group membership among members of a pro-anorexic online community. European Eating Disorders Review 20(1): 9–12.

51.

Shade

(2003) Weborexics: the ethical issues surrounding pro-ana websites. ACM SIGCAS Computers and Society 33(4): 2.

52.

Sharpe

Musiat

Knapton

. (2011) Pro-eating disorder websites: facts, fictions and fixes. Journal of Public Mental Health 10(1): 34–44.

53.

Simon

Schmidt

Pilling

(2005) The health service use and cost of eating disorders. Psychological Medicine 35(11): 1543–1551.

54.

Starr

Kreipe

(2014) Anorexia nervosa and bulimia nervosa: brains, bones and breeding. Current Psychiatry Reports 16(5): 441.

55.

Strife

Rickard

(2011) The conceptualization of anorexia: the pro-ana perspective. Affilia: Journal Of Women and Social Work 26(2): 213–217.

56.

Suokas

Suvisaari

Grainger

Raevuori

Gissler

Haukka

. (2014) Suicide attempts and mortality in eating disorders: a follow-up study of eating disorder patients. General Hospital Psychiatry Psychiatry, Medicine and Primary Care 36(3): 355–357.

57.

The Dr. Oz Show (2012) Women of the pro-ana movement. Available at: http://www.doctoroz.com/videos/women-pro-ana-movement-pt-1 (accessed 18 October 2014)

58.

Tumblr (2012) Follow-up: Tumblr’s new policy against pro-self-harm blogs. Available at: http://staff.tumblr.com/post/18563255291/follow-up-tumblrs-new-policy-against-pro-self-harm (accessed 18 October 2014)

59.

Wilson

Peebles

Hardy

. (2006) Surfing for thinness: a pilot study of pro-eating disorder web site usage in adolescents with eating disorders. Pediatrics 118(6): 1635–1643.

60.

Wood

(2008) Critical feminist theories giving voice and visibility to women’s experiences in interpersonal communication. In: Leslie

Dawn

(ed.) Engaging Theories in Interpersonal Communication Multiple Perspectives. Thousand Oaks, CA: Sage, pp. 323–335.