Turning Dog Whistles into Sirens: A qualitative study into cyber bullying on twitter

The following is a report I submitted last year during my degree for a Bachelor of Social Science. It was graded with a High Distinction and I was invited to present the findings at a conference for my University.

Introduction

Since its inception in 2006 twitter has been notorious as a social media platform used for cyber bullying. Unlike Facebook, there are no requirements to use your real name and it is easy to create multiple accounts, allowing for anonymous trolls to harass other users without consequences. Attempts made to crack down on bullying on the platform have been met with disappointing results. With a statement from cofounder Evan Williams saying he “was still trying to figure out what it was by watching the way it is being used” one can’t help but feel that Twitter doesn’t even understand its own platform, let alone how to begin controlling itself (Maney. 2017 p. 48).

Much like the real world, twitter consists of millions of individuals who interact within different social spheres, each group with a different set of norms. There’s black twitter (black users primarily focussed on socio-political issues), local twitter (people who use the website as a personal account, that seem to be the focus of a majority of studies despite being only one sub-set of users), celebrity accounts, professional organisations managed by someone employed to update their social media, and Stan Twitter to name a few.

According to the top definition of Urban Dictionary Stan Twitter is defined as:

A section of Twitter that is comprised of cult-like- fandoms that worship popular artists or artist groups. This section consists of fans engaging socially to online individuals, discussing topics relating to their faves, fan girling/boying over updates on their faves, and a competitive battle ground amongst the different fandoms. — (urbandictionary.com)

This report focusses on Stan Twitter. From years of personal experience on the platform I have witnessed bullying in this community to the point where being abusive to other users is regarded as a desirable personality trait and rewarded with copious amounts of retweets and a large follower base. Many times I have been left wondering, not only how this is allowed to happen, but why have we as a community allowed it to thrive? This ultimately shaped my research question:

How has Stan Twitter fostered a culture of cyber bullying?

Literature Review

Twitter is a micro-blogging social media platform where users post “tweets” with up to 280 characters (a recent expansion upon their original 140 limit). Launched in 2006 the platform has grown to be one of the most well-known social media websites along with other household names such as Facebook, Instagram or Snapchat. However, throughout its history twitter has been plagued with a reputation for toxicity that shows no signs of stopping.

Twitter’s ability for users to easily create anonymous accounts with no accountability has led to the platform being convenient for abusive trolls to thrive (Calvin et al. 2015). A recent study found that out of 69.4% of participating college students who used twitter, 45.5% had experiences of cyberbullying. The same study found a much higher rate of bullying on twitter compared to other platforms such as Facebook (38,6%), Instagram (13.7%) and Youtube (11.4%) (Whittaker & Kowalski, 2015). These results are further validated from another study finding close to 50% of young social media users reporting bullying across various platforms. (Chatzakou et al, 2017)

Whilst cyberbullying is a problem for any social media website, twitter’s much higher rates and lack of ability to control it has made itself a standout. Unlike Instagram being successfully bought by Facebook in 2012; Twitter has struggled to find willing investors due to its negative reputation. Potential buyers have included Google. Salesforce and Disney however after researching into the company they determined it was too dangerous and flawed to own. (Maney. 2017 p.48)

Response from generations who grew up without computers is typically to scoff at the concept of any “danger” from bullying on the internet, with their solution being to “just turn off the screen” (Chatzakou et al 2017 p.1) but with the way technology has become such an integral part of millennial and Gen Z lives that isn’t so simple. The consequences of bullying extend beyond hurt feelings and can manifest in the form of internalized problems including; depression, anxiety, anger, low self-esteem and even suicidal ideation (Fahy, et al 2016).

Abuse on twitter isn’t easy to uncover and one way in which the platform has had problems controlling such content, is that bullies aren’t always overt and using detectable abusive language, but rather constant sarcasm and trolling (Chatzakou et al 2017 p.1)

When searching for journals and conducting my literature review I found minimal information that specifically dealt with the fandom side of twitter, most simply focussing on the problem of cyber-bullying on a macro scale. This gap in information on this specific subsection of twitter motivated me to further research the topic.

Research Method

In order to obtain data, a qualitative semi structured interview process was conducted. The participants were selected from accounts on twitter I follow, who follow me in return, and that I have an established trust with. Through building on an already existing rapport I was able to gain a large bank of data to analyse. (O’Leary, 2014, p.217). There were a number of ethical concerns in the selection process. The first was already knowing the participants (O’Leary, 2014 p. 219), however I deemed this necessary due to the sensitive nature of the topic. By using trust built from years of friendship with most participants I was able to successfully navigate the interviews without accidentally crossing any boundaries or bringing up triggering content. Interviews were conducted from my end with a criticalist viewpoint. From my perspective Stan Twitter exists as a product built by people in fandoms and my role was to analyse these structures. This matches the criticalist view of perceiving social structures as a product of human consciousness (Waller et al. p. 29).

I conducted 5 interviews in total, deliberately choosing participants of different age brackets, ethnicities, employment and religion in order to try and gain as diverse an opinion as possible. Regarding age, all participants were required to be adults; the youngest participant was 21 and the eldest in their mid-40s. Given these participants were “handpicked” and selected to match a pre-existing criteria, this serves as an example of non-random sampling (O’Leary, 2014 p.190).

Initially the plan was to interview everyone on skype (either by video or voice message, whatever the participant felt more comfortable with), however slight adjustments were made for 2 participants. Participant B did not have a Skype account so we conducted the interview by phone and Participant D lives in the same state so we were able to arrange a face to face meeting.

Ethically, participants needed to be aware of the fact they were actively involved in a research project, the purpose of the project, and how their interview was going to contribute towards the overall result. By informing my subjects about the nature of the report and obtaining explicit consent prior to the interviews, I was able to obtain informed consent (Waller et al. 2016 p47). As part of the consent process all interviewees were told they were not obliged to answer any questions they felt uncomfortable with and that they could stop the interview at any time. To make sure no one felt pressured to finish the interview, or else they were risking my university project grade, I had back up options for more people to speak to and the participants were also aware of this.

By interviewing people I already have pre-existing relationships with I needed to by hyper-aware of my own bias and narrative, in order to make sure I wasn’t telling my own story through other people’s eyes but rather advocating for the experiences of others (Waller et al. 2016 p.16). Ultimately my motivation in creating this report was based upon my own negative experience on Stan Twitter where my account was targeted for bullying by a fandom and my profile was viewed more than one million times in a single day. The harassment I experienced was severe and at the time very detrimental to my mental health. However I wished to turn such a horrible experience into a learning curve and found this research topic the ultimate opportunity to do so. Though my motivations were deeply personal I stressed to myself how important it was not to be “self-centric” (O’Leary, 2014. p.51). This research topic isn’t about other people’s views on how I was bullied, but rather the focus should be on the participants to tell their own experiences. Through the differences between accounts, and multiple truths that exist, all stories are equally important in order to analyse the creation of how these power structures came to exist.

A major concern in regards to ethical integrity is the confidentiality of participants. Some participants like C were outspokenly enthusiastic about their participation and had no problem with being named, but others, whose jobs can be dependent on their social media presence, were rightfully concerned (Waller et al. 2016 p.47). At first glance k-pop twitter can seem like a large place with some popular Stan accounts having over a hundred thousand followers, but when you begin narrowing people down into categories based upon defining characteristics, the number of people who fill those requirements becomes a lot smaller and easier to identify. As a result there were some quotes from the interviews that would have made great contributions to this report that were omitted based upon how easy it could have been for a potential reader to identify them.

Results

Immediately upon conducting my interviews with each participant it became clear that it wasn’t a question of if bullying was a problem but how badly? Out of 5 participants every single person had an experience of bullying on the platform, many seeing it in their timelines on a routinely basis

“It’s pretty much constant… you become aware of constant random attacks on people… I feel like every day there’s something that’s happening in terms of some

“Random account that said something they don’t like so we’re supposed to attack” — (B)

“Literally every day” — (D)

“I do see bullying all the time” — (E)

The individual experiences between each person were all different, but there were several common threads. A notable severe example is how normalised death threats were.

“I was already used to being told to go and kill myself and other shit I was just ready for it to happen” — (A)

“A lot of these people found my real life friends and sent them death threats” — (A)

“People still send me death threats” — (B)

“Oh I’ve had death threats!” — (D)

“People started sending me death threats” (E)

Whilst C was the exemption of the participants to explicitly talk about death threats, they instead had to deal with other false accusations:

“I’ve had people call me a paedophile” — (C.)

C was not alone. A provided an in-depth account of how their friend (Z) was falsely accused of sexual harassment by a 14 year old account who was being manipulated by other accounts that did not like Z. Eventually the allegations were cleared but by the time that happened Z’s reputation had already been ruined.

Constant attacks, death threats and pedophilia accusations are the most severe examples that were uncovered during the interview process, but in terms of overall bullying on the website they make up just the tip of the Stan Twitter iceberg. But to go into deeper analysis of these examples or delve into other examples would quickly turn this report into a full blown thesis.

The negative impacts of bullying on twitter verify the 2016 findings of Fahy where manifestation can occur through internalized problems such as depression or anxiety.

Participants either experienced those side effects personally or were able to empathise with those who are suffering:

“I’m not the same person that I was before that whole experience. I was nowhere near as happy, as bright, as positive, after that experience. I ended up being very cynical. I still am and it sucks” — (A)

“I could see how it could end badly if you don’t have those coping mechanisms” — (B)

“I suffer from terrible anxiety and depression and every time this happens, even if it’s not really real, I get really bummed out about it… I Just sit there obsessing over it” — (E)

When asked if twitter was doing enough to combat the problem of abuse on the platform the universal response was laughter. The idea of Twitter being able to control bullying was so unrealistic that to every participant it was comical. But such a reaction isn’t hard to understand when participants explain how the tools that supposedly were set up to protect users are instead being used by their bullies for harassment

“They try to suspend your account through out of context tweets from the past” — (B)

B is referring to the system on tweets that allows any user to anonymously report something. In an attempt to save face over criticisms for lack of action on bullying, Twitter has been automatically suspending reported accounts that contain offensive language, regardless of the context. Many people on twitter have taken advantage of this to search a users’ history, find sarcastic comments between friends (such as “I’ll kill you!” said a joke to someone else who is also in on the joke and making similar comments in return) and report accounts they don’t like to get them banned.

D echoed this idea with the sarcastic paraphrase:

“Everyone block report this account for having an opinion” — (D)

“They turn dog whistles into sirens” — (E)

Chatzakou’s example of “just turn off the screen” sounds like the ideal salutation but a lot of people use twitter out of necessity. A and D have no other ways of engaging with the positive content they signed up to twitter for in the first place. B, C and E run websites that rely on social media to boost their traffic and engage with their readers. So how does one survive this “hellsite” (B)?

“What people can actually do is have conversation like what we’re actually having, and acknowledge that these things go on and call them out and have a bit of a support network going on” — (C.)

Building a ‘Hellsite’; the Bourdieusien Blueprint

Bourdieu’s overall perspective on social reality is that it is comprised of; field, capital and habitus, and that power is maintained through the usage of symbolic violence (Bourdieu, 1989). Whilst Twitter is not a physical reality in which we exist in our human bodied forms, the online platform is comprised of those foundations and self-sufficient to the point that, though there may be real life consequences, Stan Twitter exists as its’ own society.

Field.

“A field is simultaneously a space of conflict and competition, the analogy here being with a battlefield in which participants vie to establish monopoly over the species of capital effects in it” (Wacquant, 1992, p.17)

The field is the sub-section of twitter the person choses to engage with. In the case of this report; K-pop Stan Twitter

Capital.

If the Field is the battlefield, then capital are the tools an individual uses to navigate the field and use to play for power (Bourdieu, 1992, p.97). Capital is often thought of as wealth (economic) but also is present in the form of Social Capital and Cultural Capital

Social Capital.

As the name would suggest social capital is all about the networks of influence or support or social connections, which are one of the main reasons anyone engages with social media (not just twitter). On Twitter social capital exists in the follower count, active engagement of your tweets from followers, how many followers your mutuals have, and your reputation on the website. Social capital is regularly weaponised and used aggressively to try and gain power on the field;

“Having a lot of followers meant having a lot of people to use in your fights, a lot of people to use to attack people” — (A)

“To have a defence on twitter isn’t if you’re right or if you have facts, it’s about how many people believe you” – (A)

“You gain street cred, you gain street cred within the fandom and you feel more valid as a fan and you feel like a better fan and you feel that because you’re down talking this person that is a bad fan and you’re a better fan” – (E)

“They really use it, the fandom side, as like a way to enforce a certain immunity for their groups from being criticised” — (B)

Cultural Capital.

Cultural capital is less about follower count and more to do with knowledge, skills and credentials. A notable example of a “blue check mark”, something that gives someone an official verification on the website. Based on observation a lot of accounts with high social capital are threatened by those with cultural capital. Whilst social capital relies on the ability to rally supporters through whatever means necessary, cultural capital is based a lot more on knowledge, facts and critical thinking (which often goes against the populist narrative)

Habitus.

Habitus refers to an individual’s ingrained habits, skills, how they embody the world around them and react to it (Bourdieu, 1992 p.126).

Symbolic Violence.

Symbolic violence is where the dominant group in society preserves their power in a way that the dominated group accepts with consent. Taken for granted people internalise this as just being how the world, or in this case bullying on twitter, works. (Bourdieu, 1992). Every interviewee spoke of the bullying on twitter as if they understood it as a part of the experience, regardless of whether they agreed with it or not.

Talking about Z being targeted A described them by saying; “The word oppression, especially as you get older, people don’t want to hear it” — (A)

In regards to their own bullying, whilst obviously not happy at the prospect, A spoke about being ready for death threats whenever posting a controversial opinion, or logging in to the website and seeing a lot of notifications. The eldest participant C had the hardest stance on bullying that was built on how they overcame physical abuse in high school by fighting back and punching their bully in the face.

“When you are a bully victim you don’t realise that you are presenting yourself as a soft target and that’s what bullies get drawn to, and that’s I think that’s true on the internet, I think that’s true in real life as well” — (C.)

E spoke about struggling to deal with the bullying and feels that because they’re so often targeted then there has to be some legitimacy to it;

“I’m always wondering what did I do so wrong” — (E)

Conclusion

Having been active on twitter and interacting with the people I was interviewing for years I was prepared for each participant to have experienced bullying on the platform. But as each person opened up in detail, the reality of how brutal twitter is for a social networking website became staggering.

Paedophilia accusations were laughed off by C as if we were discussing the weather, death threats and doxing were also common findings. The impact the “barrage” (B) of criticism and bullying can have is overwhelming at times, and several interviewees relied on therapy or previous counselling experience in order to cope. Considering the participants are all grown adults, it baffles me that the minimum age to create an account on twitter is only 13, and if this report were in any position to be making recommendations I would heavily suggest raising that to 18 (this isn’t even factoring in porn accounts and nudity being allowed to exist freely on the platform).

Since conducting these interviews, transcribing the audios and analysing the findings I have found myself hyper aware of how twitter functions whilst continuing to recreationally use the website in my spare time. A specific example E brought up is the treatment of female songwriters who work for BTS saying “there is definitely a gendered element to it”. After retweeting an article attempting to give credit to female cowriters, the fandom instead reacted with backlash and accused them of “taking away attention from people of colour [BTS] for not giving them credit for their creative input in the album”.

On June 7 Charli XCX and Ryn Weaver were spotlighted as co-writers on a new BTS single ‘Dream Glow’. When Weaver replied “lolol” on twitter to one of her fans saying they were pretending to like the BTS song because she had written it, the BTS fandom reacted by attacking her mentions.

“You can hear my hey, heys [smiling heart emoji] very amped to have written on this song [kisses emojis] @bts_bighit @charli_xcx”

At the time of writing this report Weaver’s tweet has 3678 replies, the majority of which are negative. Replies include vomiting emojis, clown emojis, reaction images of clowns, accusations of “using the boys for clout” and the top reply is a woman getting pepper sprayed (implying the BTS fan would want to pepper spray Weaver). Weaver attempted to clarify

“I wrote on the song I love it I love BTS I thought that as a silly tweet”

But that tweet was inundated with 1014 replies, again mostly negative. In the two weeks that passed at the time of writing this report, Weaver has not tweeted again.

(Note: As I review my writing to publish on medium, I cannot help but compare the experience of Weaver to male songwriter @SleepDeez — who has garnered himself a following on twitter from BTS loyalists and now joins the fandom in engaging in regular petty fanwars)

As a criticalist, research is conducted ideally to crate social change (Waller et al. 2016 p 21). Realistically I do not expect this report to have any long term impact upon such a large platform that Twitter is. However I would like to hope that this has demonstrated the way social media can exist beyond our day to day lives, as a society of its own, and how there is a need to for it to be studied further.

The Bourdieu comparison came after every interviewee struggled to answer exactly how Twitter had become so prone to bullying, in a way that was far more severe than any other platform. Beginning the research, I was already aware that bullying existed. My priority was to explore the depths, in order to explain how this had come to thrive. The only rational explanation that seems plausible is the concept that a very large group of people have created our own society and bullying became a part of how this particular society is upheld. Whilst individuals (such as myself) may question this, so long as the dominant power players within the field continue to thrive based on this structure, the odds are it will not change. Instead the best people can do is learn coping mechanisms, form support networks within their own smaller groups, or avoid the platform all together.

Twitter needs to be viewed by anthropologists as a new society, rather than merely an extension of the ones we currently live. Previous studies into cyber bullying on twitter are clearly done by people who don’t possess an understanding of how the platform operates. In McHugh et al’s 2019 research into cyberbullying on the platform, analysed tweets were selected on the basis of including ‘#cyberbullying’ or other such variants. This is essentially the equivalent to the police doing an investigation into murder but only interacting with people who publicly announced their crimes. Instead social scientists need to take an immersive approach and study twitter, specifically the different fields within, through the use of ethnography. Ethnographic research requires much greater effort than mere observation but the reward is a far deeper understanding of the culture.

References

Bourdieu, P. (1989) Social Space and Symbolic Power Soiological Theory, 7,1,pp 14–25.

Bourdieu P, & Wacquant, L. (1992) An Invitation to Reflexive Sociology. Chicago: University of Chicago Press

Calvin, A. J., Bellmore, A., Xu, J. M., & Zhu, X. (2015). #bully: Uses of hashtags in posts about bullying on Twitter. Journal of School Violence, 14, 133–153.

Chatzakou, D., Kourtellis, N., Blackburn, J. De Cristafaro, E., Stringhini, G., Vakali, A. (2017) Mean birds: detecting aggression and bullying on twitter. arXiv preprint, arXiv:1702.06877

Corcoran, L., Guckin, C.M., & Prentice, G. (2015) Cyberbullying or cyber aggression?: A review of existing definitions of cyber-based peer-to-peer aggression. Societies, 5, 2

Fahy, A. E., Stansfeld, S. A., Smuk, M., Smith, N. R., Cummins, S., & Clark, C. (2016). Longitudinal associations between cyber- bullying involvement and adolescent mental health. Journal of Adolescent Health, 59, 502–509.

O’Leary, Z. (2014), The Essential Guide to Doing Your Social Research Project (2nd edn). London: SAGE

McHugh, M., Saperstein, S., & Gold, R. (2019) OMG U #Cyberbully! An exploration of public discourse about cyberbullying on twitter. Health Education & Behavior, 46, 97–105.

Maney, K. (2017), 140 character assassination. Newsweek Global, 26, 48–49

Pieschl, S., Porsch, T., Kahl, T., & Klockenbusch, R. (2013) Relevant dimensions of cyberbullying — Results from two experimental studies. Journal of Applied Developmental Psychology, 34, 5.

Urban Dictionary: Stan Twitter. (2019). Retrieved from https://www.urbandictionary.com/define.php? term=Stan%20Twitter

Waller, V., Farquharson, K., & Dempsey, D. (2016) Qualitative Social Research: Contemportary Methods for the Digital Age. London: SAGE

Weaver, Ryn (@RynWeaver). “You can hear my hey, heys [smiling heart emoji] very amped to have written on this song [kisses emojis] @bts_bighit @charli_xcx” 8 Jun 2019, 1:53 am. Tweet.

Weaver, Ryn (@RynWeaver). “I wrote on the song I love it I love BTS I thought that as a silly tweet” 8 Jun 2019, 2:25am. Tweet.

Whittaker, E., & Kowalski, R. M. (2015). Cyberbullying via social media. Journal of School Violence, 14, 11–29.

Sample Questions

· How long have you been on twitter for?
· What motivated you to create your account?
·Have you personally been a victim of cyber bullying before? If yes, why do you think you were made a target?
· How did you cope with the abuse?
· Are you a member of any group chats? If yes, have you ever seen any planned targeting of an account to collectively bully by the group?
· How often do you see bullying in your twitter feed?
· When observing bullying on twitter have you noticed any recurring patterns This can be in the form of the same accounts instigating the abuse, the same account being targeted, the format of the abuse itself, etc.
· What do you think is the difference between bullying and acceptable criticism over a tweet?
· What do you think twitter should be doing to better manage this?
· Do you feel there is a difference in bullying on Stan twitter compared to “local” twitter?