How Do Those Who Post Hate Speech Respond to Different Counter-speech Strategies? - Background

Sep 1, 2026

Upload by Maeve

Online hate speech continues to expand across social platforms, and counter-speech has been widely proposed as a remedy alongside automated detection and removal. Yet it remains unclear which counter-speech approaches shape subsequent behaviour of people who post hate. A content analysis of 3,907 hateful posts on X and 5,428 follow-up comments examined five strategy types and their links to apologies, reduced hostility, and continued hate expression.

LanguageEnglish

CategoryGeneral

FormatDOCX

Pages18

Reading Time6 min

AccessFree to download

Background

Hate Speech Online: Definition and Impact

Counter-speech as an Intervention

Counter-speech Effectiveness in Prior Research

Study Approach and Research Questions

Main Findings and Interpretation

How Do Those Who Post Hate Speech Respond to Different Counter-speech Strategies? Yue Jia Kristine Hansen Tiago Garrido Marques Sandy Schumann Zoe Wei Hate speech flourishes online. In addition to the (automated) detection and removal of hate, counter-speech has been proposed as a measure to address this harm. To date, however, it remains unclear what counter-speech strategies predict changes in the behaviour of those who express hate online. Advancing the literature, we conducted a content analysis of N = 3,907 hate speech posts on X, which targeted different outgroups, and N = 5,428 subsequent comments, including counter-speech and reactions of those who had posted hate. First, we sought to determine the prevalence of five types of counter-speech (e.g., attacking the perpetrator, off-topic comments, sharing facts that discount the perpetrator’s rationale). Second, we aimed to explore which counter-speech strategy was associated with a higher likelihood of apologies or less hateful expressions rather than persistent hate speech. Findings showed that almost half of the hate speech posts received no responses. When comments were added to hate speech, these were in almost one-third of all cases endorsing hate. The remaining replies were predominantly off-topic/ diversionary; the least common counter-speech strategy was to evoke empathy with the hate speech victim(s). Those who posted hate speech did typically not react to either of the responses within the same conversation thread. Eleven percent continued to post hate and eleven percent reduced their hateful expressions or even apologised, respectively. The likelihood of the latter was elevated when counter-speech aimed to evoke empathy with the victim. We discuss this result considering previous research on parochial empathy. Acknowledgement: This paper is dedicated to Yue Jia, whose passion and hard work initiated this study, but whose life was cut tragically short before it could be completed. We, Kirstine Hansen and Sandy Schumann, had the pleasure of supervising Yue and are proud to share his research. Online hate speech is on the rise globally (Chen, 2021; Garland, 2020; Getahun, 2023). Estimates suggest that on X (previously Twitter) alone, as many as 5 million tweets expressing intolerance, incivility, or threats (Meerson et al., 2025) are posted each day (Shepherd, 2024; Vidgen et al., 2019). Surveys further reveal that between 50 and 80 percent of teenagers have been exposed to online hate, with 20 percent being directly targeted (Oksanen et al., 2014; Winiewski et al., 2017). Counter-speech, promoted by international bodies, government agencies, NGOs, anti-hate groups, and social media platforms (Council of Europe, n.d.; eSafety, n.d.; Facebook, n.d.; INACH, n.d.; Media Smarts, 2012; UN, n.d.), is a potential solution to the proliferation of online hate speech; it refers to internet users’ direct responses to hate, to stop it or reduce its negative consequences (Benesch et al., 2016; Rieger et al., 2018). Unfortunately, despite its popularity, evidence that counter-speech can influence the behaviour of those who post hate is scarce and inconsistent. Some have demonstrated that counter-speech attenuates the frequency of subsequent hate speech (Garland et al., 2020), especially when it is coordinated (Garland et al., 2022) and the faction of hate speech supporters is relatively small (Schieb & Preuss, 2016). Others have shown that counter-speech has no effects (Devitini, 2022; Miškolci et al., 2020). A reason for these contradictory results might be that the majority of the research to date has not compared different types of counter-speech (see Hangartner et al. (2021) for an exception), focusing on only one strategy, or combining different strategies, such as, ‘supporting the victim(s)’ and ‘attacking the perpetrators’ ingroup’ (Devitini, 2022). In doing so, effective approaches might have been missed, and opposing effects might have masked impactful dynamics. We set

FAQs

What motivated this study on counter-speech strategies?

Despite counter-speech being promoted by many organizations, evidence about whether it changes the behaviour of people who post hate is scarce and inconsistent. The study addresses this by comparing multiple counter-speech types rather than focusing on a single strategy.

How was the study conducted on X?

The researchers performed a content analysis of 3,907 hate speech posts on X and 5,428 subsequent comments, including both counter-speech and reactions from the original posters.

Which counter-speech strategy showed the strongest association with reduced hateful behaviour?

The likelihood of apologies or less hateful expressions was elevated when counter-speech aimed to evoke empathy with the hate speech victim(s).

Sep 1, 2026

Upload by Maeve