Hate Speech Research Articles

Exploring the prevalence of homophily among classes of hate speech

Exploring the prevalence of homophily among classes of hate speech

Disinformation against Crimean Tatars in Russian Social Media

This article examines the persistence of Tatarophobia against Crimean Tatars in the Russian segment of social media during the first year of the full-scale invasion (24 February 2022–24 February 2023). A total of 4,435 posts from various social media platforms, including Facebook, Odnoklassniki, Telegram, Twitter, and Vkontakte, were analysed. The discourse predominantly echoes Russia’s imperial and Soviet-era narratives, along with xenophobic and oppressive rhetoric exacerbated by the full-fledged invasion. The propaganda seeks to cultivate a negative perception of Crimean Tatars as an ethnic community, manipulating cultural and historical aspects related to them. Additionally, it endeavours to construct an adversarial image of relations between Ukraine and the Crimean Tatars, as well as of Ukraine’s current policies. Russian special services perceive the capacity of Crimean Tatars to self-organize, establish volunteer units, or organize blockades as a threat. The propaganda effort not only fosters harassment and hate speech but also encourages peninsula residents to view them as “unreliable elements” deserving of suppression.

How Far-Right Extremism Changed American Body Politic

Abstract Aggressive rhetoric was part and parcel of the rising Tea Party and the related anti-Obama conspiracy theories. The argument here is that Donald J. Trump, as the leading public promoter of Birther rumors, first-time presidential candidate, and U.S. president, played a starring role in bringing hate speech, violent threats, and political violence into the political mainstream. As a result, political violence increased significantly and several violent far-right extremist groups, domestic terrorist entities, were newly founded after Trump announced his presidential campaign in mid-2015. As ex-president and once again presidential candidate, Trump glorified the criminally convicted participants in the violent breach of the U.S. Capitol on 6 January 2021 and honored violent far-right fringe groups that were active in the late 20th century while threatening revenge against his political enemies. To assess, whether Trump, as 47th U.S. president, would be able to have his political foes investigated and jailed upon his return to the highest office, we looked for answers in the Heritage Foundation's “Project 2025,” the most detailed blueprint for the dismantling of the “deep state” and the best-laid plan for a takeover by Trump and vetted conservative “Make America Great Again” loyalists in January 2025.1 Although the sole focus here is on the rhetoric of far-right extremism and violence in the past 15 years, there were no similarly extreme far-left hate speech patterns and violent extremism related to the Democratic Party during the same time.

TurkishBERTweet: Fast and reliable large language model for social media analysis

Turkish is one of the most spoken languages in the world; however, it is still among the low-resource languages. Wide us of this language on social media platforms such as Twitter, Instagram, or Tiktok and strategic position of the country in the world politics makes it appealing for the social network researchers and industry. To address this need, we introduce TurkishBERTweet, the first large scale pre-trained language model for Turkish social media built using over 894 million Turkish tweets. The model shares the same architecture as RoBERTa-base model with smaller input length, making TurkishBERTweet lighter than the most used model, called BERTurk, and can have significantly lower inference time. We trained our model using the same approach for RoBERTa model and evaluated on two tasks: Sentiment Classification and Hate Speech Detection. We demonstrate that TurkishBERTweet outperforms the other available alternatives on generalizability and its lower inference time gives significant advantage to process large-scale datasets. We also show custom preprocessors for social media can acquire information from platform specific entities. We also conduct comparison with the commercial solutions like OpenAI and Gemini, and other available Turkish LLMs in terms of cost and performance to demonstrate TurkishBERTweet is scalable and cost-effective.

On and off line hate speech and academic performance in secondary education in Cameroon

Despite the increasing scientific attention given to hate speech, there is a lack of empirical studies examining the effect of hate speech among secondary school students and its association with academic achievement. The objective of this research is to assess the associations between online and offline hate speech based on gender, religion, ethnic affiliation, and academic achievement. The results suggest that hate speech is not spread evenly across the distribution of students, but rather is concentrated primarily among students with lower academic performance, both online and offline. More specifically, hate speech based on gender and ethnicity was found to have significant and negative associations with educational outcomes when it occurred online. In contrast, hate speech based on religion also had a significant and negative associations with educational outcomes when it occurred offline. It is worth noting that a student's ability to connect to the internet from his or her own phone is positively associated with better school results, while presence on social networks is negatively associated with these results. These results may be helpful considering policies to improve healthy school environments in the future.

Cybercrime in virtual spaces: An overview of the law in Indonesia

The Internet has the potential to overcome national barriers and facilitate global dissemination of information. Although its benefits are numerous, it also has detrimental consequences that could disrupt society. Engaging in cybercrime is typically required only to create a substantial number of blogs, accounts, applications, programs, and websites on various social and mass media platforms for illicit purposes. Examples of unlawful activities include, but are not limited to, fraud, data theft, unauthorized use of credit cards, distribution of inappropriate content, online gambling, and prostitution, as well as hate speech and extremist or terrorist activities. To address the issue of virtual space crime, the government enacted Law No. 19 of 2016, which amended Law No. 11 of 2008, with respect to information and electronic transactions. However, it is important to recognize that not all people who use the internet have a comprehensive understanding of the legal framework. Consequently, government, employers, and institutions involved in information and communication technology are responsible for educating the public about the law to reduce the incidence of criminal activity in virtual spaces.

Faking the war: fake posts on Turkish social media during the Russia–Ukraine war

This study aims to analyze the fake posts circulated on Turkish social media during the Russia–Ukraine war. With advancing technology, social media platforms have a profound impact on the way we perceive and interpret events and make us question the accuracy of information generated about international events such as wars. While the Russia–Ukraine war constitutes an important turning point in international relations, the reflection of these events on social media is also seen in fake posts. In this context, the main purpose of this study is to identify the common themes of fake social media posts and to reveal the general context of these posts on social media. In addition, the study aims to analyze the fake content circulating on Turkish social media and to reveal the emerging polarized discourses through the identified themes. The research revolves around five main themes that feed polarization: war reporting, ideological misrepresentation, humor, hate speech, and conspiracy theories. The findings show that fake content is particularly concentrated around ideological polarization and antagonisms. It was also found that misinformation and decontextualized humor blurred the true context of the war and that fake content combined with hate speech and conspiracy theories distorted the context of the war.

Countering online ‘regional blackening': interventions for province-targeted hate speech in the Chinese digital sphere

ABSTRACT While political communication scholars usually debate whether the internet and social media are democratizing forces for China, scarce attention has been given to the role of digital media in mediating and amplifying old and new forms of prejudice, discrimination, abuse and hate in China’s cyberspace. We focus on a context-specific stereotype and insult that attacks people based on their place of domicile – ‘regional blackening’. Based on social identity theory, this study examines the effectiveness of three audience-focused countermeasures (inducing empathy, constructing collective national identity, and increasing intergroup contact) in mitigating the negative influences of hate narratives and in promoting online civic intervention, and investigates the moderating role of multiculturalism.

Prevalence, Viewpoints, and Encounters with Cyberterrorism among College Students

This research aims to describe prevalence, perspective on cyberextremism and encounters of cyberextremism among college students. A sample of 1000 Tafila Technical University (TTU) students ) representing 12 Jordanian governates, found that 45.4% were males and 54.6% were females from 12 Jordanian governorates. Science colleges and arts and social colleges split the sample almost equally (49.6% vs. 50.4%), respectively. All students are using the internet, and most of them use it intensively (73%), compared to regular use (27%). The study found that less than half of the sample described online extremism as hate speech, violence, cyberbullying, sexual pornography, indirect hate speech, assaults, and post-support extremism. Students perceived several procedures to encounter cyberextremism, including closing websites, fines, criminalizing content, holding websites responsible for compensation, establishing a minimum age for viewing extremist content, removing hardening material, and compensation. Males had a higher mean of cyberextremism than females. ANOVA analysis showed significant differences between external attribution and cyberextremism, internal attributions, and no significant differences between males and females.

Cascading dynamics of hate speech propagation: Unveiling network structures and probability of retweeting on Twitter

Cascading dynamics of hate speech propagation: Unveiling network structures and probability of retweeting on Twitter

We are not a virus: repercussions of anti-Asian online hate during the COVID-19 pandemic on identity and coping strategies of Asian-heritage individuals

ABSTRACT Current scholarship has identified online hate speech as one of the main threats in digital spaces, leading to an ever-heightened marginalization of ethnic minorities. The COVID-19 pandemic has given racist expressions of hate a new momentum, directed particularly at Asian-heritage individuals who were scapegoated as the originators of the virus. This study explores the repercussions of this for Asian-heritage individuals living in Europe, specifically their (1) identity and (2) coping strategies. Using semi-structured interviews with N = 14 Asian-heritage individuals, the study shows how this group was hatefully othered online, with severe consequences for its members’ sense of hybrid belonging and ethnic identity. Initial defense attempts were muted, and individuals surrendered to passive coping strategies, withdrawing from mainstream discourse. The study shows how identity and coping were inextricably intertwined. Taking this case as an example for minority victimization during crises, we discuss potential empowerment perspectives for ethnic minority individuals.

Ideological attitudes predicting online hate speech: the differential effects of right-wing authoritarianism and social dominance orientation

IntroductionWhile research on online hate speech (OHS) has expanded in recent years, only few studies adopt a theoretical framework to understand how ideological attitudes differentially motivate individuals to engage with OHS. Drawing on the dual-process motivational model of ideology and on previous political psychological research on OHS, this study examines how individual levels of social dominance orientation (SDO) and right-wing authoritarianism (RWA) predict the likelihood of producing OHS for online platforms users.MethodsWe used logistic regressions to analyze the survey data from a representative German sample of social media platform users (N = 7,349).ResultsThe analyses indicate that SDO is related with higher odds of producing OHS, while RWA is related with lower odds. After adjusting for socio-economic factors and controlling for alternative predictors, the odd ratios remain significant, indicating that these two ideological attitudes predict online hate speech in different directions.DiscussionThe results show that high-RWA individuals are less likely to engage with OHS, which is explained through their conservative motivation to conform to social norms and maintain social stability. High-SDO individuals are more likely to produce OHS and may use it following their competitive motivation to increase hierarchical relations and improve their social status within society. The findings are discussed taking into consideration the specificities of the German social context, and corroborate and expand previous research. From these subtle but crucial differential effects, relevant implications are drawn for the platform as well as for social and political levels.

Countering the Hate Design - Facilitating Peace and Harmony

Communal riots generally happen due to the previous existence of a communal ideology in the society. The hard-core communalists are hate mongers who, by their fiery speeches and provocative statements create a situation which is very conducive for riots o break. The role of the police during communal riots has not been found to be very satisfactory. But, in spite of all odds, there have been people, bodies and organisations which have exhibited compassion and courage during riots and have saved the lives of people. Such people and organisations are role-models who need to be emulated. In riots, the leadership of the police is very crucial. The Indian Police Service officers need to exhibit more of courage, boldness and grit in helping the poor and the victims of riots. A new riot control culture has to develop. The responsibility of maintaining communal harmony lies with all sections of people and organisations. Each of us need to contribute our might to combat the design of Hate and facilitate Peace and Harmony.

Credibility at stake. A comparative analysis of different hate speech comments on journalistic credibility and support on climate protection measures

Credibility at stake. A comparative analysis of different hate speech comments on journalistic credibility and support on climate protection measures

Semantic Safeguards: Harnessing BERT and Advanced Deep Learning Models Outperforming in the Detection of Hate Speech on Social Networks

Semantic Safeguards: Harnessing BERT and Advanced Deep Learning Models Outperforming in the Detection of Hate Speech on Social Networks

The power of connection: Resource and responsibility in the virtual community experience of Italian trans and gender‐diverse activists

AbstractThe community concept underwent a series of adaptations and integrations throughout the decades. One of these transitions was supported by the rise of virtual communities, especially social network sites (SNSs). These platforms are recognized to serve as spaces for marginalized groups like transgender and gender diverse (TGD) individuals to create safe and inclusive communities for self‐expression and activism. To investigate how TGD activists engage with SNS, 21 activists were interviewed, exploring challenges, potentials and perceptions. Reflexive thematic analysis revealed that participants perceive SNS in line with the community experience framework, viewing them as resources for accessing information, providing support and fostering belonging and investment, and as acts of responsibility aimed at nurturing reciprocity, sense of duty and social justice advocacy. However, SNS also proved to expose activists to vulnerability within the TGD community itself, where a hierarchy that marginalizes TGD identities with transnormative experiences is often reproduced. Additionally, SNS anonymity offers a safer environment for TGD individuals to come out but protects hate speech perpetrators. Thus, while pivotal in TGD activism and community building, the dual nature of SNS underscores the complexities of virtual spaces. Finally, the study calls for redefining the concept of community beyond the virtual–physical dichotomy.

*Corresponding author: suwandi.sumartias@unpad.ac.id E-ISSN: 2289-1528 https://doi.org/10.17576/JKMJC-2024-4002-01 Portrait of Hate Speech Propagators' Behaviour in Indonesia Twittersphere: A Comparative Virtual Ethnography Study

Political life in Indonesia today cannot be separated from the social media platform Twitter (X) because it is often used to express opinions freely. Unfortunately, a handful of parties misused the freedom of communication and expression offered by Twitter (X) to spread hate speech. One of the biggest phenomena is “Cebong” and “Kadrun”. This research examines the characteristics of hate speech spreaders in the political context of "Cebong" and "Kadrun". This study employed virtual ethnography to interpret the phenomenon of political hate speech on Twitter (X) in May 2022. The hate speech made by "Cebong" and "Kadrun" can be concluded as two hostile political camps. They argued with each other to show their identity as supporters of one of the camps and aimed to bring down the opposing political camp. The tweets made by “Cebong” were dominated by arguments classified as arguments by analogy; meanwhile, "Kadrun" was dominated by discovery arguments: arguments from cause to effect. The manipulation strategy carried out by "Cebong" is dominated by connotations, and "Kadrun" is dominated by specific warrants and connotations. The fallacy expressed by "Cebong" is dominated by persuasive definitions; meanwhile, "Kadrun" is dominated by hasty generalizations. These findings could be suggestions for creating a healthier ecosystem in the digital public sphere in Indonesia. Twitter (X) users should be more aware of reporting content that indicates hate speech. By knowing the characteristics of hate speech and the characteristics of hate speech spreaders, Twitter users can stop the spread of hate speech. Keywords: Cebong, Kadrun, hate speech, propagators, Twitter.

SOCIOPRAGMATIC ANALYSIS OF HATE SPEECH IN HILARION HEAGY’S CONVERSION NEWS ON TWITTER

This research discusses the phenomenon of hate speech in Hilarion Heagy's conversion news on Twitter. Hilarion Heagy, a former Orthodox priest who later embraced Eastern Catholicism before ultimately converting to Islam, has sparked significant controversy within religious circles, eliciting varied reactions, especially on social media. This research addresses a significant gap in the literature by examining hate speech in the context of religious conversion. Therefore, this research aims to identify types of hate speech and find out the intent behind hate speech. This research utilizes the National Police Chief's circular number SE/6/X/2015 to identify types of hate expressions and uses Searle's illocutionary speech act theory to explain the meaning behind hate expressions. In this research, data was taken from comments on Twitter related to the news of Hilarion Heagy's conversion. The results of the research found that the expressions of hatred in the news about Hilarion Heagy's conversion were related to religious issues. The outcomes found five types of hate speech, with the most performed being Blasphemy (17 times), followed by Provoking (11 times), Defamation (10 times), Insult (7 times), then the least one being Spreading fake news (5 times). As for the types of illocutionary acts, the most dominant was Assertive (40 times), followed by Directive (7 times) and Expressive (3 times). There were no Commissive and Declaration types found at all.

Enhancing Transformative Competencies through the Internalization of Religious Moderation: A Vision for Islamic Boarding School's Educational Framework in 2030

Extremism, radicalism, hate speech, and even the rupture of inter-religious ties, are problems facing the Indonesian nation today. Al-Hikam as one of the boarding schools able to survive in those problems, by encouraging religious moderation. This study aims to explore the process of internalizing the value of religious moderation in Al-Hikam Islamic boarding schools and analyze the realization of transformative competencies in the process. An instrumental case study design was chosen for this research. Observation and in-depth interviews were conducted to obtain data. The results showed that there were 3 models of value transformation, verbal, symbolic, and written. The methods of imparting values are carried out in two ways, namely providing standardized sources of accurate information and structured practices through pesantren programs and activities. The internalization process is formed through joint service activities involving various elements of the moderation agency. This study arrived at the theoretical finding that maximizing the function of co-agency in pesantren consisting of the Kyai board, teachers, peer colleagues, and santri guardians at the transaction stage, is effectively facilitates the internalization of values by the agents (santri) of religious moderation obtained at the transformation stage into an inherent identity, namely moderate humans. This achieved identity also brings students to achieve transformative competencies: creating new values, reconciling tensions and dilemmas, and taking responsibility.

Mengintegrasikan Nilai-Nilai Buddhis dalam Penggunaan Media Sosial

This research aims to explore communication ethics in the use of social media, with a focus on Buddhist values ​​that can provide guidance for individuals in using social media. Ethics, as part of the moral values ​​implemented by religion, should be a guide for wise behavior, including when using social media. However, in reality, there is negligence in the use of social media in society, which involves practices such as smear campaigns, hate speech, online bullying (cyberbullying), and the spread of fake news (hoaxes). Therefore, this research is interesting to dig deeper through analysis of Buddhist values, so that Buddhist teachings can become a filter and guide for people to communicate wisely on social media. The research method used in this research is library research. Data analysis in this study highlights that social media use should reflect kindness, teachings of truth, joy, and consistency with facts. These four values ​​must be supported by other aspects, such as subtlety of language and expression, the basis of love in speaking, as well as consideration of the purpose and appropriate time in conveying the message. Manifestations can be found in content, writing or posts that have an educational, inspiring and useful character. In addition, when receiving information on social media, it is important to be selective and critical, but remain open to constructive criticism, while upholding healthy discussions on social media.