The UK’s new Online Safety Act, programmed for implementation in the second half of next year, could reshape how social media platforms deal with harmful online content. This legislation aims to protect users by requiring platforms to swiftly remove illegal material such as hate speech and incitement to violence. While this marks a significant step forward in addressing online abuse, our research offers further insight into why such measures are necessary.
Our April 2024 study, published in the Journal of Consumer Research, shows that even leisure-oriented spaces can become breeding grounds for verbal violence that can lead to harassment and disturbing patterns of abuse, including what we call “blood games” – verbal attacks for entertainment. While our research zooms in on one specific community, it reveals broader patterns that echo across today’s social media platforms.
Online hostility isn’t confined to politically charged or marginalised communities, however. For 12 years, we studied the spread of direct, cultural, and structural violence within an unexpected online group: the British electronic dance music community. This verbal brutality is at odds with EDM’s ethos, which, according to a SiriusXM report, aims to “promote happiness” and stimulate dopamine in the brain to evoke “good and happy times”
Our research explored a popular British online dance community focused on Hard House music. We found that voluntary, leisure-oriented communities can become toxic spaces rife with verbal violence, sometimes lasting for years. This behaviour – marked by sadistic entertainment, clan warfare, and mob justice – offers insights that apply to other online platforms such as Reddit, Twitch or Discord, known for harbouring toxic and abusive behaviour.
Verbal violence in a music-centred online community
To explore why online consumption communities – groups of people who connect and engage with others sharing their interests from anywhere at any time – sometimes turn hostile, we studied “John” (a pseudonym), a semiprofessional DJ and the founder of an electronic dance music community we refer to as Hard House (HaHo). This is a hybrid community, composed of 60% males, mostly in their 30s and 40s. In its early days, starting in 2001, friends used the online platform to discuss events, DJs, and music. They also advised each other on private matters and occasionally met at clubbing events – 15 of which we attended to add an ethnographic dimension to our study.
By the time we immersed ourselves in this world, however, HaHo had developed patterns of endemic verbal violence between members, which peaked in 2005-2006, before dissipating in 2018. Over 20,000 members generated more than 7 million posts, and our analysis of 170 discussion threads revealed the emergence of harmful behaviour often rooted in boredom and status anxiety.
Fifteen distinct clans emerged within the community, each defined by a shared identity rooted in common values, rituals, and a collective sense of moral obligation toward its members. These clans often clashed in a form of social warfare, competing fiercely for dominance and prestige. Their conflicts were shaped by differences in class background, musical preferences, or distinct interaction styles within the broader group.
Prestige within the community was largely earned through verbal aggression. This culture of hostility was further institutionalised by the creation of annual awards that celebrated and amplified the behaviour. The “Dark Side” award honoured the community’s most ruthless members, while the “Flounce” award mocked those who left the group in dramatic, rage-filled exits following verbal attacks. Meanwhile, the “Meltdown” award recognised the most entertaining and chaotic conflicts between members.
These awards normalised toxic interactions, reinforcing a cycle of aggression and ridicule. Moderators played a limited role in curbing the hostility, which emboldened members to engage in punitive campaigns against those who failed to conform to the prevailing norms. As a result, dissenting voices were often silenced or driven out, deepening the divisions and entrenching the culture of verbal brutality.
Our study describes the community’s most prevalent form of violence as a “verbal bullfight,” staged to entertain members by enraging newer or more vulnerable participants. In one instance, a newcomer, “Marc”, sought advice from veteran member “Tony”. Instead of receiving help, Marc was bombarded with insults like “idiot” and “quitter.” Tony’s posse quickly joined in, mocking Marc for 540 posts until he left the platform, feeling “like a complete loser.”
This kind of toxic behaviour, which we describe as “hedonic Darwinism,” fosters exploitation for amusement. Such behaviour has been linked to severe consequences, including depression, anger, and even suicide. Despite the adult demographic, this juvenile cruelty was a way for members to bond and reinforce group solidarity.
Online “clans” mirror real-world social structures
Our research identified such widespread structural violence as often being characterised by unequal power distributions and systemic forms of exploitation. Neuroscientist Bernhard Bogerts suggests this sadistic behaviour combats boredom, while neuropsychologist Thomas Elbert argues it can be “fascinating and emotionally arousing,” producing euphoria and reinforces group solidarity.
HaHo’s clans mirrored real-world social structures, formed around class, musical taste, drug use, and interaction style. While some exchanges were clever, many descended into crude tyranny, resembling structural violence seen against immigrants and stigmatised groups online, as witnessed dramatically this summer in Europe.
When asked about this online brutalisation, HaHo founder John admitted he did not approve but prioritised community growth over the protection of individuals. His attempts to engage moderators to prevent these attacks had little effect until the situation escalated, forcing him to intervene cautiously to avoid a rebellion among the members. Engagement in HaHo declined rapidly after moderators banned clan leaders and promoted respect for diversity. Further decline occurred due to competition in the electronic music scene and the aging of its members, though the platform continues to exist to this day.
Members have since redefined HaHo as a supportive family, viewing past “blood games” and violent awards as immature. They have also come to realise that such actions are not harmless, even introducing a “Redeemed Member” award for those previously violent members who have abandoned their sadistic behaviour. This musical community mitigated brutalisation, clan warfare, and a libertarian style of justice by replacing punishment narratives with a healing-oriented approach based on restorative justice.
Can this be a template for the kind of online violence the UK’s Online Safety Act is trying to address? Such a question is well beyond the scope of our research. But we do raise important questions about the role of moderation and community management in preventing such brutalisation. As we consider the broader impacts of online behaviour, it becomes clear that further ethics-based research is needed to explore whether society should continue to tolerate forms of “harmless play” that mask deeper issues of verbal violence – and whether policy interventions like the Online Safety Act are enough to protect young and vulnerable users in today’s digital world.
