Online violence against women on Twitter

Figure 1:  Cyber violence received by Laura Bates

“Online abuse began for me when I started the Everyday Sexism Project” (Amnesty International, 2018). With the start of the project by Laura Bates, a British feminist activist and writer, online violence ensued. In the early days when the project wasn’t getting attention, the website also received around 200 abusive messages a day (Amnesty International, 2018, Chapter 3). And as the project progressed, the amount of hate speech against her gradually spread to other social media platforms such as Twitter, and the number of hate speech spiked (Amnesty International, 2018, Chapter 3). Threats of violence, abuse, sexist and misogynistic comments are just some of the features of hate speech from social media that Laura Bates faces on a daily basis (Amnesty International, 2018, Chapter 3).

The case of Laura Bates is by no means isolated or insignificant; countless women who exist in the public sphere are facing different types of cyberviolence. Online violence against women is a significant impediment to women’s full participation in public life and ownership of their own voices.Twitter is one of the world’s most influential social platforms and a place for access to information, advocacy, debate, discussion, and dialogue, where influential women-whether in the political, entertainment, business, or activism fields – bring critical perspectives and insights that drive dialogue on important issues (Amnesty International, 2018, Chapter 3). Paradoxically, however, this platform for expanding their influence also exposes them to serious online violence: a range of harassment, threats and abuse designed to silence and intimidate (Amnesty International, 2018, Chapter 3).

The impact of such online violence against women is profound, raising urgent questions about digital policy, governance and the nature of freedom of expression online. Spaces that are supposed to encourage participation, dialogue and diversity of thought have become an extension of real-life violence and abuse against women (Economist Intelligence Unit, 2021). This blog seeks to explore the complexities of online violence against women on Twitter and find solutions to online violence.

The scale of online violence against women

Online violence against women is a widespread and profound problem. According to The Economist Intelligence Unit (2021), cyber violence against women tends to take nine forms, namely astroturfing, doxing and violent threats, hacking and stalking, cyber harassment, misinformation and defamation, hate speech, impersonation, video and image-based abuse, and these nine types of cyber violence tend to be combined. The Economist Intelligence Unit analysed the top 51 countries in the world in terms of the number of people online, and found that the global prevalence of cyberviolence against women was 85 %. The researchers categorised the 51 countries into North America, Latin America and the Caribbean, Africa, Europe and Asia-Pacific. In the Middle East, the rate was a staggering 98 %, while Europe, with the lowest rate, had a 74 % prevalence of cyber violence against women (Economist Intelligence Unit, 2021).

Figure 2 : Overall prevalence rates by region, in %  https://onlineviolencewomen.eiu.com/

While cyberbullying affects women in all fields, influential women often face more severe cyberviolence due to their visibility and prominence (Jankowicz et al., 2024). A study conducted by Amnesty International (2018) highlighted that on social media platforms including Twitter, influential women who are public figures are often targeted for cyberviolence, especially in traditionally male-dominated fields such as gaming, politics, and economics.

In addition to this, female journalists, politicians and women’s rights activists are particularly targeted by cyberviolence.Twitter is very important for journalists, who use it to keep in touch with readers and advertisers and thus secure employment. Although male journalists also face cyberviolence, the content and volume of hate speech is far less than that of female journalists (Amnesty International, 2018, Chapter 2). in 2018, British news presenter Cathy Newman had to hire an expert to validate the threats she was subjected to, including those from Twitter (Amnesty International, 2018, Chapter 2). In 2017, Amnesty International analysed the accounts of 177 female MPs active on Twitter in the run-up to the general election, who received a total of 25,688 years of abusive messages over a six-month period (Amnesty International UK, 2017). In 2021, a report from the Wilson Center collected and analysed the online conversations of 13 female politicians across six social media platforms, with over 190,000 users posting more than 336,000 insults against these female politicians over a two-month period (Jankowicz et al., 2024). A 2017 study conducted by Northumbria University in the UK found that 80 % of respondents used Twitter when engaging in feminist debate, the highest probability of use of any social platform. At the same time, the study also found that 88 % of respondents who regularly used Twitter for feminist debate had experienced cyberviolence on the platform (Amnesty International, 2018, Chapter 2).

These shocking statistics highlight the prevalence of online violence against women and the urgent need for effective programmes to address it.

Why women? Why influential women?

Online violence against women is rooted in the deeply entrenched patriarchal gender order, in which women are subordinate (Richardson-Self, 2018). Traditional gender roles and stereotypes are important reasons why women face cyber violence, and women are expected to conform to certain standards of behaviour (Richardson-Self, 2018). And when women, especially influential women, challenge these norms, they face huge obstacles (Richardson-Self, 2018). Women who engage in public discourse on platforms such as Twitter, especially in political, social or cultural discourse, are subject to cyberviolence as they are perceived as not conforming to the norms of femininity in the patriarchy (Richardson-Self, 2018). And because of their visibility, the cyber-violence they face is exacerbated, a phenomenon that can be seen as an extension of offline gender violence (Richardson-Self, 2018).

Social platforms such as Twitter further promote cyberviolence due to their design features and lack of relevant policies (Kowalski et al., 2014). Twitter and similar social platforms allow users to create accounts with minimal personal information, which favours privacy but also allows for anonymity and pseudonyms (Kowalski et al., 2014). This feature allows users to engage in cyberviolence without fear of the consequences that typically occur in face-to-face interactions (Kowalski et al., 2014). At the same time, because Twitter is structured to maximise the distribution of content (Marwick & boyd, 2014). Tweets can be retweeted easily and quickly, thus spreading rapidly across the network (including harmful messages) (Marwick & boyd, 2014). This viral spread can further amplify online violence, transforming isolated incidents into widespread harassment campaigns (Marwick & boyd, 2014). However, despite the large amount of cyberviolence on Twitter, policies remain inadequate and there are no sound remedies for those who have suffered from cyberviolence (Amnesty International, 2018, Chapter 1). Twitter executives have also publicly stated that there is still a lot of work to be done to address the issue of cyber violence against women on the platform.

The impact of cyberviolence is as serious as offline violence and can even fuel other forms of violence. Violence, from online to offline, seriously damages women’s physical and mental health and can even push them towards suicide (O’Brien, 2024). Economist Intelligence Unit (2021) shows that more than 92% of women say that cyber-violence has harmed their sense of well-being; 7 % of women surveyed face job loss or change of job as a result of cyber-violence. 35 % of women surveyed have suffered psychological problems as a result of cyberviolence, and one in ten women have suffered physical injuries as a result of cyberviolence.

Cyberviolence also restricts women’s freedom of expression and full enjoyment of their human rights (O’Brien, 2024). Research shows that more than half of the women interviewed believe that the Internet is no longer a safe place to share ideas, leading to mandatory self-censorship before speaking. This means that women’s voices on the Internet will be significantly reduced, affecting women’s freedom of expression and reducing the diversity of social discourse (Economist Intelligence Unit, 2021). For women in positions of influence, online violence has specific implications. Not only does it affect their personal and professional lives, but it also acts as a deterrent to other women who may be considering public roles.Nicola Sturgeon has stated that when she reads about hate speech on the internet, her greatest fear is that this online violence will push the next generation of women away from politics (Amnesty International, 2018).

Case Study

Figure 3: Diane Abbott    Photograph: Dan Kitwood/Getty Images

Diane Abbott is a British politician and the first black woman ever to be elected to the British Parliament. She has been a key figure in British politics since 1987. Over the years, Diane Abbott has been an advocate for a variety of social issues, including health, education and minority rights, and as a result, she has been the target of online violence and harassment.

Diane Abbott says she receives a large number of abusive messages compared to other parliamentarians. A report by Amnesty International in 2017 revealed that almost half of the abuse tweets received by 177 female parliamentarians in the six months before the UK general election were received by Diane Abbott alone, a shocking 45.14 %. (Amnesty International UK, 2017). The nature of the abuse is often racist and sexist, including derogatory remarks and threatening messages, as well as racist and gender-based insults (Amnesty International UK, 2017).

The online violence against Abbott not only affected her personal work, but also caused wider concern among the general public, particularly women and ethnic minorities, about the safety and inclusiveness of online spaces (Amnesty Watch, 2017). For many, social media platforms are an important tool for expressing personal views, but many women have had to choose to shut down social platforms due to widespread cyberviolence against women (Amnesty Insights, 2017). Cyber violence can drive women off social platforms such as Twitter, and women who persist in speaking out are likely to self-censor, which severely impacts the freedom of speech for citizens (Amnesty Insights, 2017).

The revelation that Diane Abbott was subjected to cyberviolence has raised media and public scrutiny of social media platforms’ policies regarding the protection of their users.Twitter has acknowledged and stated that cyberviolence against women on a large scale should not exist on social media.Twitter has a responsibility to respect human rights and to safeguard women’s freedom of expression, and has announced that it will be taking a hard line against cyberviolence (Amnesty Insights, 2017). However, according to a study published by the Fawcett Society and Reclaim the Internet, a week after they reviewed a range of abusive and violent content on Twitter and reported anonymous accounts to the platform, posts involving cyberviolence remained on Twitter and the anonymous accounts received no further notification (Amnesty Insights, 2017).

Critical analysis

One of the key factors contributing to serious cyberviolence against women is the persistence of social norms that marginalise certain voices. Women, especially those from minority backgrounds, can often only challenge entrenched social norms by engaging in public discourse (Jane, 2016). However, such behaviour occupies a space that has long been dominated by men, which can provoke hostility in some groups, which can lead to the use of anonymous accounts on social platforms to target women (Pariser，2011).

The anonymity of Twitter contributes significantly to violence. While anonymity is intended to protect privacy and freedom of expression, it also allows individuals to separate their online behaviour from their real-world identity (Citron, 2014). This allows users to engage in behaviours that they would not do when communicating offline, including harassment and threats (Citron, 2014).

Considering the challenges caused by online violence, especially against women, Twitter and similar social media platforms must take multiple measures to effectively address these issues. At the same time, social media platforms must also consider how to balance freedom of expression with scrutiny of online speech (Flew, 2021).

Technologically, Twitter should invest in advanced AI technologies to better detect tweets that engage in online violence. These tools are able to understand contextual and linguistic nuances and distinguish between harmful content and benign interactions (Gorwa et al.) Additionally, AI technologies can be used for automated auditing, where AI systems can analyse multiple streams of data at the same time, this feature is crucial for the timely detection and reduction of cyberviolence (Gorwa et al., 2020).

In terms of policy, Twitter must ensure that the anti-online violence policy is clear, comprehensive, and transparent in its implementation (Matias, 2016), and the policy should include specific rules against gender-based violence and harassment. The platform should also simplify the reporting process, ensure timely responses to complaints, and communicate clearly with users about the status and outcome of their reports (Matias, 2016).

Conclusion

In conclusion, there is a need for social media platforms such as Twitter to provide a strong and multifaceted response to the widespread problem of cyber violence, especially against women. As we have seen from the harrowing experiences of influential women such as Diane Abbott, online violence not only affects the well-being of individuals, but also damages the public’s freedom of expression. In this situation, Twitter must better detect and respond to cyberviolence by adopting advanced AI technology, and improve policy clarity and enforcement, maintaining a transparent communication process. All of our efforts are aimed at creating an online environment where all users can participate freely and safely, leading to a richer and more diverse space for public discussion.

References

Amnesty International. (2018, March 20). Toxic Twitter – Women’s experiences of violence and abuse on Twitter. Amnesty International. https://www.amnesty.org/en/latest/news/2018/03/online-violence-against-women-chapter-3-2/

Amnesty International. (2018, March 20). Toxic Twitter: Triggers of violence and abuse against women on Twitter. Amnesty International. https://www.amnesty.org/en/latest/news/2018/03/online-violence-against-women-chapter-2-3/

Amnesty International UK. (2017, September 6). Diane Abbott talks about “sheer levels of hatred” she receives online. Amnesty International UK. https://www.amnesty.org.uk/press-releases/diane-abbott-talks-about-sheer-levels-hatred-she-receives-online

Amnesty International. (2018, March 21). Toxic Twitter – A toxic place for women. Amnesty International. https://www.amnesty.org/en/latest/research/2018/03/online-violence-against-women-chapter-1-1/

Amnesty International. (2018, March 21). Nicola Sturgeon: A case study in violence against women online. Amnesty International. https://www.amnesty.org/en/latest/news/2018/03/nicola-sturgeon-online-violence-against-women/

Amnesty Insights. (2017, September 4). Unsocial media: Tracking Twitter abuse against women MPs. Medium. https://medium.com/@AmnestyInsights/unsocial-media-tracking-twitter-abuse-against-women-mps-fc28aeca498a

Citron, D. K. (2014). Hate crimes in cyberspace. Harvard University Press.

Economist Intelligence Unit. (2021). Prevalence of online violence against women: Methodology. Economist Intelligence Unit. https://cdn.vev.design/private/WbTNgdOVVvgyq5TIBiYpWVmMCJQ2/hyw1xhPZO6_EIU_METHODOLOGY_PREVALENCE%20OF%20ONLINE%20VIOLENCE%20AGAINST%20WOMEN_FINAL.pdf.pdf

Flew, T. (2021). Regulating platforms. Polity Press.

Gorwa, R., Binns, R., & Katzenbach, C. (2020). Algorithmic content moderation: Technical and political challenges in the automation of platform governance. Big Data & Society, 7(1), 205395171989794-. https://doi.org/10.1177/2053951719897945

Jankowicz, N., Hunchak, J., Pavliuc, A., Davies, C., Pierson, S., & Kaufmann, Z. (2024). Malign creativity: How gender, sex, and lies are weaponized against women online. Wilson Center. https://www.wilsoncenter.org/publication/malign-creativity-how-gender-sex-and-lies-are-weaponized-against-women-online

Jane, E. A. (2016). Online misogyny and feminist digilantism. Continuum (Mount Lawley, W.A.), 30(3), 284–297. https://doi.org/10.1080/10304312.2016.1166560

 Kowalski, R. M., Giumetti, G. W., Schroeder, A. N., & Lattanner, M. R. (2014). Bullying in the Digital Age: A Critical Review and Meta-Analysis of Cyberbullying Research Among Youth. Psychological Bulletin, 140(4), 1073–1137. https://doi.org/10.1037/a0035618

Marwick, A., & boyd, danah. (2014). “It’s just drama”: teen perspectives on conflict and aggression in a networked era. Journal of Youth Studies, 17(9), 1187–1204. https://doi.org/10.1080/13676261.2014.901493

Nathan Matias, J. (2016). Going dark: Social factors in collective action against platform operators in the reddit blackout. Conference on Human Factors in Computing Systems – Proceedings, 1138–1151. https://doi.org/10.1145/2858036.2858391

 O’Brien, M. (2024, January 4). Online violence: real life impacts on women and girls in humanitarian settings. International Committee of the Red Cross. https://blogs.icrc.org/law-and-policy/2024/01/04/online-violence-real-life-impacts-women-girls-humanitarian-settings/

Pariser, E. (2011). The filter bubble: what the Internet is hiding from you. Penguin Press.

Richardson‐Self, L. (2018). Woman‐Hating: On Misogyny, Sexism, and Hate Speech. Hypatia, 33(2), 256–272. https://doi.org/10.1111/hypa.12398