Can Women Fight hate speech only by being brave?

Images credit:

A girl called Linghua Zheng died not long ago, just six months after she received her offer letter. on 13 July 2022, Zheng received her master’s offer letter and took a photo with her grandfather to post on the Chinese social media, Xiaohongshu. Then some accounts from other platforms reposted her photos and it went virus. Some Chinese netizens seemed more concerned about her appearance than the acceptance notice. Her picture was wildly reposted with some calling her a ‘nightclub girl’ and a ‘demon’ (Ng, 2023). Rumours of her relationship with her grandfather were even more disgusting. The most she expressed was that she did not know why anyone would use such filthy language or why they did it. She has tried to defend her rights and appealed. However, she did not receive effective support and the abusers intensified. The incident festered again, exposing her to more abuse and insults. As a result, she dyed her hair back to black. On 23 January 2023, Linghua passed away. In her suicide note, she apologised to her family and friends, blaming herself for not having the courage to live. Everything she loved was ruined by those who hid behind a screen. To this day, the memory of the internet has not receded. Upon learning of her death from depression, internet users were quick to recall her iconic pink hair, the photo of her with her grandfather in his hospital bed, and the rumours and internet violence that followed. Over the past six months, the rhetoric about the “pink-haired girl” has never ceased, interweaving into a vast web of opinion that “bound” Linghua to the end of her life. The unbridled online violence of the perpetrators and the long struggle of the victim to save herself are in constant opposition.

Linghua Zheng’s death is the latest case of hate speech and online harm. According to the United Nations (2022), hate speech is speech that expresses hatred against a specific target or group of targets based on inherent characteristics; in addition to being speech that hurts people’s feelings, it is also speech that can cause immediate and sustained harm (Sinpeng et al., 2021). In recent years, hate speech has come to be focused on the Internet context, referring specifically to online speech that expresses negative emotions and directs hostility. If we use the Internet buzzword for it, hate speech may be more like a combination of “hater” and “troll”. There are even many people who make a living from it. On the face of it, just as swearing is considered rude by everyone, hate speech should definitely not be a popular thing at all. However, the internet is an “extra-legal” place where the boundaries of morality are highly blurred. With many people talking about themselves using anonymous IDs, hate speech has found a perfect place to live. Users are becoming more comfortable with the censorship mechanism (Roberts, 2019). Clearly Chinese social platforms are very good at blocking keywords, so why does a large amount of harmful speech still get seen? There are platforms on which vetting practices are guided by their externally oriented community guidelines (Roberts, 2019). That is, the internal logic of these platforms does not define some obvious hate speech as hate speech. On the platform where Linghua initially posted, Xiaohongshu, the comments underneath was still friendly. Xiaohongshu is a social media outlet for sharing shopping, beauty and fashion, and other lifestyle topics. Users can post, find inspiration and purchase goods. To ensure the quality of content, it uses a strict vetting system and KOL operations. Based on my experience of using it, the user base is mostly young, well-educated women with high spending power. In such an environment, it is difficult for users to argue. However, being illegally ported photos is one thing, being cyberbullied is another. When content is republished on other platforms, things are out of control. According to Linghua’s Xiaohongshu account, in addition to accounts selling training courses using her images for promotional purposes, there are also news accounts posting them to Douyin with strange captions. It is well known that Douyin has become a particular cultural phenomenon. Douyin’s sinking market covers a wider range of people, and the lower quality and literacy of its users makes vulgar content easier to enjoy more. These pose a great challenge to Douyin’s content review and management. In addition, Douyin users’ fragmented understanding of information directly determines what kind of comments they will post.

In addition to the education level of users, malice from the male population as a whole is even scarier. Malice against women displayed by social networks is at a high level. On the one hand, the internet is an important platform for women to express themselves and find opportunities for advancement. On the other hand, the internet has become a tool for abusers to attack women. According to a survey (Vogels, 2021), women are more likely than men to be sexually harassed online. Misinformation and defamation are common means of attack in the context of online violence experienced by women. In some cases, sex can be a laughingstock for men, but a nightmare for women. It is all too easy to discredit a woman and damage her character by spreading rumours. If she is silent, the rumour is assumed to be true; if she refutes it, she is just too sensitive. So, is hate speech against women related to morality? On the point of men’s moral demands on traditional women, there is no difference between Eastern and Western countries. Since the feminist movement brought about a breakthrough in society’s moral demands on women, women have sought sexual liberation. Women are gradually breaking away from the social status and social morality imposed on them by patriarchy. However, the social structure of male dominance and female weakness still exists. It is still a certain segment of men who hold the power of sexual choice. They want to dismantle all ideas of sexual liberation for their own benefit. They intentionally or unintentionally bring sexual liberation to “promiscuity”; they create a social atmosphere in which “promiscuity” is femininity. In fact, they achieve the goal of playing with women at low cost. In the eyes of these people women should not dye their hair or be photographed with any men. These statements may be motivated by an inherent bias towards gender roles or a sense of threat to women’s autonomy. Slut-shaming and unrestrained rumourmongering ensues. Perhaps many women are used to receiving private messages and lewd comments with sexual innuendo. Unfortunately, gender, appearance, size, and race can all enable women to be attacked. Logic or no logic, it is unlikely that these people who make hate speeches will change their minds – ration is too difficult to convince emotions.

In a context where everyone seeks freedom of expression, should hate speech be protected? Even the freest speech is toxic when expressing an opinion devolves into spreading hate. These toxic cultures are centred on specific issues or events and often rely on the harassment of others (Massanari, 2017). What makes them unique is their use of social technology platforms as a conduit for coordination and harassment, and their seemingly organised quality. Members of these communities often demonstrate technological prowess when engaging in morally dubious behaviour, such as aggregating public and private content about their behavioural goals and utilising platform policies. Individuals associated with toxic technological cultures embrace the power of community as a way to change or express their dislike of those they dislike (Massanari, 2017). In this case, the business models and algorithms of online platforms may reinforce hate speech, as platforms may prioritise the display of content that is highly controversial or has extreme views, which helps to increase click-through rates and visits. In addition, the business models and algorithms of online platforms may reinforce hate speech. When platforms prioritise the display of content that is highly controversial or has extreme views, click-through rates become higher and the impact becomes greater, as Douyin has done. When algorithms continue to recommend similar content, a vicious circle is created. When freedom of expression is abused, the antagonism between different groups can increase, creating another destabilising factor in society.

In this case, technology companies and regulators have the ability and responsibility to regulate the speech that occurs in the public domain. Technology companies have an obligation to manage hate speech posted on their platforms, including content moderation, algorithm optimisation, and social media education. Better user reporting mechanisms, and better online community management practices. Regulators can enact laws and regulations to prohibit harmful speech. However, the challenge is that governments may abuse this power to suppress the freedom of expression of opposition or dissidents. Therefore, social media needs to work with regulators. Facebook, for example, has both an automated system to quickly detect hate speech and violent content and a human review to improve the accuracy and depth of content review (Sinpeng et al., 2021). local governments, media, and civil society organisations to comply with local laws and regulations in order to adapt to local political requirements, while improving understanding and adaptation to local culture (Sinpeng et al., 2021). In addition to these, social media platforms can provide access to complaints and grievances, such as online forms, emails, hotlines, etc., so platforms need to be optimised based on genuine feedback from users (Roberts, 2019).

Hate speech and the online harm it causes is one of the greatest challenges facing modern society. The unique semantics of the women in Linghua’s case reflect the gender violence character of hate speech. The need to express negative emotions combined with the convenience of the Internet has made the Internet a low-cost outlet for negative emotions. The internet often allows ordinary people to use more extreme language to attack extreme dissent. Perhaps society does not give people enough security, or perhaps people are becoming increasingly unaware of their inadequacy. If netizens were simply expressing themselves, we might see just a few expressions of emotion. But when a single incident draws the attention of thousands of netizens to one place, and millions of people start cursing at the same person or the same incident, we can only see the accumulation of hatred. The so-called “cyber violence” is in fact a lack of rationality. Violence can either lead to verbal confrontation or turn into a real legal case. Whether it is an attack against a group of women or pure malice, it is extreme. The use of freedom of expression must be carefully considered. Freedom of expression should not be achieved in a way that undermines the protection of other human rights. We therefore need more effective regulatory policies to protect women. At the same time, everyone has an obligation to recognise and avoid hate speech and to engage in positive online social behaviour. Only the joint efforts of society can build a harmonious cyberspace where everyone can express their opinions freely and free from any form of intimidation or threat.


Massanari, A. (2017). Gamergate and The Fappening: How Reddit’s algorithm, governance, and culture support toxic technocultures. New Media & Society, 19(3), 329–346.

Ng, K. (2023, March 26). Online trolls are taking a toll in China. BBC News.

Roberts, S. T. (2019). Behind the Screen: Content Moderation in the Shadows of Social Media (pp. 33–72). Yale University Press.

Sinpeng, A., Martin, F. R., Gelber, K., & Shields, K. (2021). Facebook: Regulating Hate Speech in the Asia Pacific. Department of Media and Communications, The University of Sydney.

United Nations. (2022). What is hate speech? United Nations.

Vogels, E. A. (2021, January 13). The state of online harassment. Pew Research Center.

Jidanji. Xiaohongshu. Retrieved April 10, 2023, from

Be the first to comment

Leave a Reply