Professor: Jonathon Hutchinson
Student Yuncheng Wang
StuID:541017071
Overview
Regulation on Online platform are vital to its functioning. Platform, sometimes defined as the community that allows users to share their ideas, experiences based on Internet. The social media platforms that we use every day, such like TikTok, Instagram and Facebook, appear to be public spaces for free speech. In reality, how these platforms manage and operate hate speech impacts in everyone’s online experience. However, if platforms failed in monitoring its discourses, it could result bad damage upon users’ experience, in some case, extreme stances or actions breed from unregulated platforms, like threat, might undermine users both mentally and physically. From a report from ABC News in 2023, an organisation from Victoria’s Aboriginal treaty launched a call, which required Facebook and the operator Meta to enhance the regulation of racial and hate speech online. Because racial stances are regarded as hate speech, which obey the law of Australia.
While the platform responded that “Those behaviours did not conflict with community regulations on Facebook”. Hate speech comment might be acceptable to the platform, but the person being insulted may suffer from insomnia, fear or even be afraid to go online again, this is not an isolated case. This case addressed the necessity and severity of platform regulations, and this topic would also be talked in this journalism report below.
- “Why important?”
Racism Online and in the Media (2024) has revealed that, until the report was conducted, there are 1 out of 7 people are still encountering with hate speeches across Australia, while the amount of hate speech that Aboriginal people encountered doubled. Also, the trend started to online.
To start with, oversight on the Meta platform contributed to consequence of ongoing hate speeches. Although Meta’s (2022) page of “Hate Speech” interface prohibited racially discriminatory speech explicitly, it did not specifically define what constitutes insults to Indigenous people in Australia, which might probably indicate the vague prohibition among regulations regarding Meta’s platform, but also a loophole in the functioning of Meta that enabled very minorities of accounts with extreme users disseminate their ideas while walking on the margin.
Secondly, hate speech impacted the negative effect upon users’ experience on media platforms, that is, some extreme threats and provocations can seriously endanger the mental and physical safety of affected users. For minors or underage who are still developing their awareness and values and lack sound judgment, hate speech might poison their perceived idea fundamentally. Research by Racism Online and in the Media (2024) indicated, nearly 60% of Indigenous Australian users had ever suffered from racial hate speeches, which have caused their emotional damages and blamed reputations. Therefore, demonstrates that Australian digital media platforms still need to strengthen their management of negative information on their platforms.
- Real Stories…
In 2023, the administrator, Monica Morgan reported she was vilified by a pro-Bumby association which support wild horse issues. “She has been intimidated and bullied for several years.” Said Ms Trescowthick, Lawyer of Ms. Monica. This is also the first case, that the state’s Racial and Religious Tolerance Act was vilified upon First Nation people since the law’s foundation (Charmaine Manuel et al., 2024). As the Administrator of being responsible for aboriginal people and also an Aboriginal people herself, the circumstance which hate speech that undermined Ms Monica appealed the fact that, accounts that holding extreme stances have constructed a trend to undermine people while neglecting their gender, job or responsibilities that they are taking, as long as targeted victims have the relationship with Aboriginal people.
Moreover, in Harvey et al.’s (2022) report, Richard Swain, an outspoken environmental campaigner, a horse activist, and the guide of the Snowy Mountains River, was badly undermined by Hate speeches. “Shut up or you will be shut up permanently” Some attacks like above are purely death threats. “Growing up around here, the colonial crowd feel that they are the mountain men and women, and they have a monopoly over what this place is,” Swain says. “I guess I’m a threat. I’ve got a leg to stand on. I am a river guide in the parks. I am born and bred here. I am of Aboriginal heritage. They see that as a real threat to their argument.” Said Swain.
Monica and Richard’s cases are not isolated. Their cases reveal the systematic problems. While Meta has regulative policies towards hate speech, these policies have two fatal flaws in its enforcement.
First, the policies’ definition is vague. Meta prohibits racially hate speech, but these policies do not specify what constitutes the insults to Australian Aboriginal people. This gives those extremist users chances to express their hate speeches. As long as they don’t use racial words directly, the platform is unlikely to take action.
Second, algorithm’s detection capabilities are limited. Extremist users begin using homophones and other methods to avoid regulation. Meta’s algorithm relies on keywords matching and is almost powerless against such methods. This is why despite Meta claims that they value the issues of hate speech, the actual situation of victims has not improved so much.
Furthermore, the lax management of a very small number of hate speech perpetrators by online platform operators and commercial companies is also one of the contributing factors to the continued existence of such incidents in Australia.
-Word transformation: “seemed implicit, but same insulting”
On one hand, according to Roberts, S. T.’s (2019) argument, those sending such remarks are shifting their perspective, moving from explicit insults to more subtle and incomprehensible language. Which is, on certain online platforms, negative messages or remarks are transforming to be borderline-wandering, the publishers of such remarks are appealing their explicit insulting ideas into some implicit jargons, through using near-synonyms or alternative meanings to express content prohibited by platform guidelines originally.
“Promised Regulation”-What happened on the other side?
On the other hand, there is no unified standard across different platforms regarding the effective control of such borderline hate speech. Although the audience of online platforms comes from all over the world, the platforms themselves are shared across different countries and regions. Therefore, Roberts (2019) pointed out again, among platform or operators, the existence of specific regulation internal policies was a reflection of concerning their stakeholders’ benefit only, while these regulations would be intersected with social justice, public desired equities, however, algorithmic regulations and platform surveillance itself is still centralised onto operator and stakeholder fundamentally.
Thus, in some senses, condoned less regulation among platform speeches, finally led to the overflow of hate speeches and other extreme stances.
Similar situations also occur on social media and gaming platforms in China. On these platforms, restrictions on specific prohibited words focused on the words themselves primarily. For Instance, words like “废物/Feiwu” (waste) are not allowed to be directly entered into normal chats, while users sometimes abbreviate the word to express a similar meaning, such as “fw” (the abbreviation of the Chinese Mandarin pinyin). However, as platform controls expand, even the spelling may be banned, so users resort to spellings like “fvv,” replacing the “W” with two repeated “v”s. Regardless of the differences in user speech control on platforms both domestically and internationally, continuous merging of hate speeches revealed the shortcomings among platform algorithms, both in China or foreign regions, are still imperfect, allowing accounts to ultimately circumvent the bans through alternative interpretations to express their extreme views. On the one hand, while these very few users holding extreme views cannot represent the entire user group on social media, their clearly accusatory and insulting remarks have indeed poisoned the environment of the entire user platform. The online environment itself has not been diluted simply because the source of hate speech is a small group of users; on the contrary, the experience of most users has been severely undermined by those minorities. On the other hand, this also clearly demonstrates that the algorithms of online platforms in China and other countries are still based on the “keywords” in users’ posts, and there is still much room for improvement. Furthermore, punitive mechanisms such as restricting user access and banning accounts are not particularly effective in reducing the behaviour of extremis accounts. Disseminators with extreme stances are only needed to abandon their banned accounts, spend a few minutes registering a new one, and they can continue to post extremist remarks online as before. which is an imbalance of cost between punishment mechanisms and disseminating hate speeches.
- Any Suggestions?
According to Roberts (2019), whether platforms are able to develop more smart algorithms to counter accounts’ sophisticated behaviours to escape from regulation, while continuing their hate speech. To start with, as the platform operator, the platform itself is likely the first to encounter and manage hate speech, making relevant regulations and punishment mechanisms particularly important. For instance, monitors among digital platforms could track the IP addresses and personal information used by those accounts, which might facilitate tracing registrants’ background and their requests to create accounts based on addresses or personal information, eventually regulating users’ inappropriate speech. Meanwhile, on the other hand, improvements should be made among platforms’ algorithms, which interpret that algorithms which can analyse accounts’ previous discourse contents, and punish with restrict or forbids.
Therefore, to reduce similar circumstances from submerging on digital platforms, more effective methodologies need to be implemented on the platform, either socially managed by the government or being manipulated by individual account users themselves. For instance, formal regulations to restrict companies or operators would also facilitate reducing the amount of hate speech cases through enhancing the law, or external methods to prevent similar cases. According to Erica et al. (2024), they argued that although there are existing methods to manipulate regulations among digital platforms, the limitations, the lack of standards and the possibility of differentiated definitions between public standards and private companies’ standards made the barrier to preventing macro-viewed digital platform regulations. For instance, verdictive tools might be implemented among platforms, people are more likely to focus on new tools, including tools generated or advanced by private companies, which differentiated with tools implemented by the UN, especially in defining or verifying through standards. This stresses the necessity of legislative monitoring and being monitored by the government.
Additionally, from individuals on platforms, another methodology that applicable are breed their motivation in monitoring other’s behaviour, while enhancing the mechanism of punishment on platforms, this might also facilitate reducing hate speech behaviours’ appearance on social media. According to Couch et al.’s (2015) argument, rather than being panoptical surveillance by the “individual”, the synoptical enabled people who are the “many” but also “being surveillance” switched into the people who are able to monitor others. In other words, every individuals are not only people who monitor, but also the people who are under the monitor. This applied and effective methodologies to audiences on digital platforms, based on measurements that enabling individuals with power to report other accounts, which obeyed or possibly marginalised insulting behaviours, could help audiences’ self-regulation among platforms.
- Final ideas
The cases of Monica and Swain being victims of hate speech, and the vulnerability of the Meta platform in handling racially discriminatory content both demonstrated that although the social media platforms driven by technological advancements such as algorithmic recommendations, there is still considerable room for improvement in managing extreme speech and user behaviour within the platform. The scope of prohibited content regarding racial issues or offensive language should be broader.
Furthermore, the potential tacit approval from operators or the platform’s own loosened oversight contributes to the indiscriminative attacks of hate speech against Indigenous users or related accounts.
In addition, the imperfections in the Meta platform’s algorithm also prevent the effective filtering of hate speech that targets specific groups or uses borderline descriptions that fail to be flagged by the algorithm.
Above all, hate speech always exists on social media platforms, not only because of technological limitations, but also social media platforms lack the sufficient measures to regulate it. Social media platforms and companies should promote their responsibilities, and relevant government departments and social media users also need to monitor these platforms and companies.
The problems of hate speech among online platforms, especially towards Indigenous people in Australia have not been effectively given enough attention. When considering the future, how to improve the situation among behaviours of hate speech and platform regulations must be a topic that digital media platforms, operators, and the national level behind them should explore in-depth and discuss in the long term.
References
Couch, D., Han, G.-S., Robinson, P., & Komesaroff, P. (2015). Public health surveillance and the media: a dyad of panoptic and synoptic social control. Health Psychology and Behavioral Medicine, 3(1), 128–141. https://doi.org/10.1080/21642850.2015.1049539
Charmaine Manuel, Jonathon Kendall, & Callum Marshall. (2024, December 4). Brumby activists racially vilified former head of Yorta Yorta corporation, VCAT finds. Abc.net.au. https://www.abc.net.au/news/2024-12-04/vcat-racial-vilification-barmah-national-park-brumby-yorta-yorta/104683066
Harvey, A., Fallon, M., & Carter, L. (2022, February 21). Communities torn apart as a fight over horses descends into “madness.” ABC News. https://www.abc.net.au/news/2022-02-22/kosciuszko-brumby-battle-turns-feral-mountain-culture-war/100830536
Meta. (2022). Hateful Conduct | Transparency Center. Meta.com. https://transparency.meta.com/policies/community-standards/hateful-conduct/
Roberts, S. T. (2019). Behind the screen: Content moderation in the shadows of social media. Yale University Press.
Erica Gaston, Fiona Mangan, Cristal Downing, Raphael Bodewig, Lauren McGowan, Emma Bapt, and Adam Day, 2024 PBF Thematic Review: Synergies between Human Rights and Peacebuilding in PBF-supported Programming (New York: United Nations University, 2024).
Racism online and in the media. (2024, November 28). Vic.gov.au. https://www.vic.gov.au/victorias-anti-racism-strategy-2024-2029/case-change-why-we-need-strategy/racism-community-and-public-places/racism-online-and-media
Be the first to comment