Editorial: Political communication strategies and the instrumentalization of hate speech in electoral campaigns

Contemporary electoral communication unfolds within hybrid media systems whose attention logics reward conflict and emotional intensity. Political actors may use hostile discourse to stigmatize groups, delegitimize opponents, intimidate journalists, and gain visibility. Within that hostility, we may find different and more specific strategies: hate speech attacks people on identity grounds; toxicity covers abusive language; polarizing discourse sharpens moral and affective distance; and digital political violence includes threats, harassment, and intimidation. These strategies of hostility can have numerous effects at the political and deliberative levels. For example, Waisbord (2020) describes ‘mob censorship' as networked harassment that seeks to silence journalists, while Bilewicz and Soral (2020) show that repeated derogatory language can normalize prejudice and desensitize audiences.

Nevertheless, research in this field faces conceptual and measurement problems. Definitions of hate speech vary across legal, normative, discourse-analytic, and computational traditions, while coverage remains uneven across languages and platforms (Amores et al., 2021; Di-Fátima, 2023; Matamoros-Fernández and Farkas, 2021). However, the literature points out and warns that hate speech, particularly in political communication, has been on the rise in recent years, fueled not only by certain extremist parties, but also by the algorithms of digital platforms; which, in turn, makes confrontation particularly appealing to the parties themselves, creating a worrying toxic cycle (Amores et al., 2026; Tasenţe, 2026).

Against this background, this Research Topic examines how these forms of hostility operate in political and electoral communication without conflating them. It asks how political actors and supporters construct enemies, mobilize fear and anger, exploit visibility, and undermine rivals, journalists, minorities, or institutions. Its eight Original Research articles cover Brazil, Indonesia, Nigeria, Portugal, Spain, and the 2024 European Parliament campaigns in France, Germany, Italy, and Spain. Together, they connect messages to the actors, institutions, platforms, and publics through which hostility is produced, amplified, contested, and normalized.

A first area of contribution concerns target construction. Dalben et al. examine 579,708 posts about three women journalists during Brazil's 2022 presidential campaign. Bolsonaro's attacks and supporter activity coincided with bursts of attention and gendered hostility. The case connects personal intimidation with attacks on journalism's democratic legitimacy, consistent with “mob censorship” (Waisbord, 2020). In Lagos, Ademosu et al. show how election-related messages and replies constructed Yoruba and Igbo identities through territorial gatekeeping, blame, shaming, insults, and threats. Comparing Spanish press coverage from 2006 to 2024, Estrada Villaseñor conceptualizes “migratory crisification” as the coupling of alarmist political narration, exceptional governance, and affective mobilization; contrasting African migration with Ukrainian displacement exposes selective allocations of threat and deservingness. Across four European countries, Segado-Boj et al. identify migrants and racialized people as central targets of social hate on TikTok during the 2024 European Parliament campaign. These studies treat hate as a relational process defining belonging, threat, and legitimate voice.

A second line of contribution lies in differentiating affective mechanisms. Segado-Boj et al. find that social hate relies chiefly on fear, othering, and dehumanization, whereas political hate more often mobilizes anger, blame, and hostile metaperceptions. Villegas Lirola et al., surveying more than 1,100 adolescents in Spain, report an association between social extremism and the justification of violence and illegal action, moderated by political self-positioning and gender and especially pronounced among boys identifying with the far right. Electoral hostility thus follows different pathways: fear can securitize out-groups, whereas anger and moral accusation can turn disagreement into antagonism. Content-based and cross-sectional evidence, however, cannot establish that exposure causes radicalization or violence.

A third line of contribution locates antagonistic communication within infrastructures of visibility and authority. Analyzing more than 200,000 Spanish-language posts from ten main Spanish parties between 2015 and 2023, Amores et al. find substantial interparty and temporal differences in algorithmically scored toxicity, with the far-right party Vox recording the highest averages and 2021 the main peak. Toxicity is positively but weakly associated with engagement, cautioning against equating visibility with persuasion. Costa shows that Chega's 2024 newspaper Folha Nacional combined leader-centered personalization with attack-dominant messaging on immigration, insecurity, corruption, and elite privilege; party-owned media can therefore complement social platforms in message discipline and agenda construction. In Indonesia, Aswar et al. interpret electoral surveys, visualizations, and media headlines as Foucauldian technologies of normalization, legitimization, and subject formation. The case does not equate polling with hate speech; it shows how apparently neutral knowledge formats can define viability and confer authority on political narratives.

Across these cases, three recurrent processes emerge: boundary making, affective activation, and amplification. Political actors construct targets and hierarchies of belonging; emotionally charged rhetoric makes those distinctions salient; and media organizations, party publications, survey institutions, platforms, and users shape circulation. This ecological reading also guards against conceptual inflation. Hate speech, toxicity, harassment, ethnic rhetoric, crisification, and extremism overlap, but are not interchangeable. Comparative accumulation requires explicit definitions, transparent coding, multilingual validation, and careful separation of offensive language from attacks on protected or historically marginalized groups.

The Research Topic's methodological pluralism—computational and manual content analysis, critical discourse analysis, qualitative case study, and survey modeling—is a major strength, capturing phenomena from individual posts to institutional knowledge production. Its designs nevertheless warrant calibrated inference. Several studies rely on purposive or platform-specific samples; automated toxicity scores may reproduce linguistic and contextual biases; and descriptive, interpretive, and correlational results cannot establish effects on vote choice, trust, or polarization. The Research Topic is strongest as a comparative map of mechanisms and sites, not as a causal estimate of electoral impact.

Future research should connect production, circulation, exposure, and consequence. Multiplatform, multilingual, and longitudinal designs could track how antagonistic frames travel between speeches, party media, news, and social networks. Experiments and quasi-experiments could test effects on candidate evaluations, out-group attitudes, participation, institutional trust, and willingness to justify violence. This is particularly relevant among younger citizens, for whom political disengagement may coexist with distrust in political institutions and alternative forms of participation (Sánchez-Holgado et al., 2026). Audits should examine recommendation, moderation, monetization, and counter-speech while protecting targeted communities and researchers from renewed harm. Transparent documentation is crucial as data access and automated classifiers change. Comparisons should also extend beyond the electoral moment, because strategic hostility may have cumulative antecedents and durable afterlives. The democratic challenge is to alter the political and technological incentives that make dehumanization, exclusion, and intimidation electorally useful. The articles in this Research Topic provide a geographically diverse foundation for that task.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited and reviewed by: Gianfranco Pasquino, Johns Hopkins University, Italy