News Media and Hate Speech Promotion in Mediterranean Countries 1668484277, 9781668484272

Table of contents :
Cover
Title Page
Copyright Page
Book Series
Table of Contents
Detailed Table of Contents
Preface
Acknowledgment
Section 1: Detection
Chapter 1: European Initiatives for the Support and Counselling of Victims of Hate Crimes
Chapter 2: Creating an Online Network, Monitoring Team and Apps to Counter Hate Speech, and Hate Crime Tactics in Europe
Chapter 3: Using HurtLex and Best-Worst Scaling to Develop ERIS
Chapter 4: Spreading Organised Hate Content
Chapter 5: Approximation of Hate Detection Processes in Spanish and Other Non-Anglo-Saxon Languages
Chapter 6: The Interaction Between Offensive and Hate Speech on Twitter and Relevant Social Events in Spain
Section 2: Mediterranean Countries' Approaches
Chapter 7: The Semiotics of Xenophobia and Misogyny on Digital Media
Chapter 8: Mapping Stigmatizing Hoaxes Towards Immigrants on Twitter and Digital Media
Chapter 9: The Southernification of the Pandemic in Italy
Chapter 10: Are There Hate Speeches on Spanish Television?
Chapter 11: Hate Speech or Hate Shot?
Chapter 12: The Expression of Hate in Portuguese Digital Media
Chapter 13: Online Hate Speech and the Representations of Refugees in #VatanimdaMülteci (#RefugeeinMyCountry)
Chapter 14: Analysis of Radicalisation Prevention Policies From the Perspective of Educommunication in Mediterranean Countries
Chapter 15: New Narratives to Defuse Hate Speech
Compilation of References
About the Contributors
Index

Citation preview

News Media and Hate Speech Promotion in Mediterranean Countries Elias Said Hung Universidad Internacional de la Rioja, Spain Julio Montero Diaz Universidad Internacional de la Rioja, Spain

A volume in the Advances in Media, Entertainment, and the Arts (AMEA) Book Series

Published in the United States of America by IGI Global Information Science Reference (an imprint of IGI Global) 701 E. Chocolate Avenue Hershey PA, USA 17033 Tel: 717-533-8845 Fax: 717-533-8661 E-mail: [email protected] Web site: http://www.igi-global.com Copyright © 2023 by IGI Global. All rights reserved. No part of this publication may be reproduced, stored or distributed in any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher. Product or company names used in this set are for identification purposes only. Inclusion of the names of the products or companies does not indicate a claim of ownership by IGI Global of the trademark or registered trademark. Library of Congress Cataloging-in-Publication Data Names: Said-Hung, Elías, 1979- editor. | Montero Díaz, Julio, editor. Title: News media and hate speech promotion in Mediterranean countries / edited by: Elias Said-Hung, Julio Montero-Diaz. Description: Hershey PA : Information Science Reference, [2024] | Includes bibliographical references. | Summary: “This book will provide relevant theoretical frameworks and the latest empirical research findings about hate speech’ studies. It will be written for professionals and researchers who want to contribute to a debate about hate speech from an approach focused on digital media in a specific geographic and linguistic area (Mediterranean countries). This project can generate a relevant impact in the institutional, professional and academic fields associated with the study of the proposed topic, giving visibility to projects that are currently being advanced around it but also offering the possibility of establishing a new approach, taking into account the particularities of linguistic, media and journalistic characteristics of the Mediterranean countries and of other socioculturally related ones (Latin America)”-- Provided by publisher. Identifiers: LCCN 2023012755 (print) | LCCN 2023012756 (ebook) | ISBN 9781668484272 (hardcover) | ISBN 9781668484319 (paperback) | ISBN 9781668484289 (ebook) Subjects: LCSH: Hate speech--Mediterranean Region. | Discrimination in language--Mediterranean Region. | Discrimination in mass media--Mediterranean Region. | Hate--Mediterranean Region. Classification: LCC P95.54 .N49 2024 (print) | LCC P95.54 (ebook) | DDC 302.2309182/2--dc23/eng/20230512 LC record available at https://lccn.loc.gov/2023012755 LC ebook record available at https://lccn.loc.gov/2023012756 This book is published in the IGI Global book series Advances in Media, Entertainment, and the Arts (AMEA) (ISSN: 2475-6814; eISSN: 2475-6830) British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. All work contributed to this book is new, previously-unpublished material. The views expressed in this book are those of the authors, but not necessarily of the publisher. For electronic access to this publication, please contact: [email protected].

Advances in Media, Entertainment, and the Arts (AMEA) Book Series Giuseppe Amoruso Politecnico di Milano, Italy

ISSN:2475-6814 EISSN:2475-6830 Mission Throughout time, technical and artistic cultures have integrated creative expression and innovation into industrial and craft processes. Art, entertainment and the media have provided means for societal selfexpression and for economic and technical growth through creative processes. The Advances in Media, Entertainment, and the Arts (AMEA) book series aims to explore current academic research in the field of artistic and design methodologies, applied arts, music, film, television, and news industries, as well as popular culture. Encompassing titles which focus on the latest research surrounding different design areas, services and strategies for communication and social innovation, cultural heritage, digital and print media, journalism, data visualization, gaming, design representation, television and film, as well as both the fine applied and performing arts, the AMEA book series is ideally suited for researchers, students, cultural theorists, and media professionals.

Coverage • Visual Computing • Design Tools • Design of Interiors • Gaming • Fabrication and prototyping • Sports & Entertainment • Drawing • Computer aided design and 3D Modelling • Applied Arts • Film & Television

IGI Global is currently accepting manuscripts for publication within this series. To submit a proposal for a volume in this series, please contact our Acquisition Editors at [email protected] or visit: http://www.igi-global.com/publish/.

The Advances in Media, Entertainment, and the Arts (AMEA) Book Series (ISSN 2475-6814) is published by IGI Global, 701 E. Chocolate Avenue, Hershey, PA 17033-1240, USA, www.igi-global.com. This series is composed of titles available for purchase individually; each title is edited to be contextually exclusive from any other title within the series. For pricing and ordering information please visit http://www. igi-global.com/book-series/advances-media-entertainment-arts/102257. Postmaster: Send all address changes to above address. Copyright © 2023 IGI Global. All rights, including translation in other languages reserved by the publisher. No part of this series may be reproduced or used in any form or by any means – graphics, electronic, or mechanical, including photocopying, recording, taping, or information and retrieval systems – without written permission from the publisher, except for non commercial, educational use, including classroom teaching purposes. The views expressed in this series are those of the authors, but not necessarily of IGI Global.

Titles in this Series

For a list of additional titles in this series, please visit: http://www.igi-global.com/book-series/advances-media-entertainment-arts/102257

Examinations and Analysis of Sequels and Serials in the Film Industry Emre Ahmet Seçmen (Beykoz University, Turkey) Information Science Reference • © 2023 • 389pp • H/C (ISBN: 9781668478646) • US $215.00 Using Innovative Literacies to Develop Leadership and Agency Inspiring Transformation and Hope Limor Pinhasi-Vittorio (Lehman College, CUNY, USA) and Elite Ben-Yosef (The BYEZ Foundation, USA) Information Science Reference • © 2023 • 287pp • H/C (ISBN: 9781668456149) • US $215.00 Music and Engagement in the Asian Political Space Uche Titus Onyebadi (Texas Christian University, USA) and Delaware Arif (University of South Alabama, USA) Information Science Reference • © 2023 • 254pp • H/C (ISBN: 9781799858171) • US $215.00 Handbook of Research on the Relationship Between Autobiographical Memory and Photography Mark Bruce Nigel Ingham (London College of Communication, University of the Arts London, UK) Nela Milic (London College of Communication, University of the Arts London, UK) Vasileios Kantas (University of West Attica, Greece) Sara Andersdotter (University for the Creative Arts, Sweden) and Paul Lowe (London College of Communication, University of the Arts London, UK) Information Science Reference • © 2023 • 636pp • H/C (ISBN: 9781668453377) • US $295.00 Contemporary Manifests on Design Thinking and Practice Gözde Zengin (Karabük University, Turkey) and Bengi Yurtsever (Mugla Sıtkı Kocman University, Turkey) Information Science Reference • © 2023 • 277pp • H/C (ISBN: 9781668463765) • US $215.00 Sustaining Creativity and the Arts in the Digital Age Gilberto Marzano (Rezekne Academy of Technologies, Latvia) Information Science Reference • © 2022 • 329pp • H/C (ISBN: 9781799878407) • US $215.00 Handbook of Research on Issues, Challenges, and Opportunities in Sustainable Architecture Veronica Foong Peng Ng (Taylor’s University, Malaysia) Sucharita Srirangam (Taylor’s University, Malaysia) and Siti Norzaini Zainal Abidin (Taylor’s University, Malaysia) Information Science Reference • © 2022 • 426pp • H/C (ISBN: 9781668451199) • US $295.00

701 East Chocolate Avenue, Hershey, PA 17033, USA Tel: 717-533-8845 x100 • Fax: 717-533-8661 E-Mail: [email protected] • www.igi-global.com

Table of Contents

Preface.................................................................................................................................................. xvi Acknowledgment................................................................................................................................ xxii Section 1 Detection Chapter 1 European Initiatives for the Support and Counselling of Victims of Hate Crimes: Key Actors............. 1 César Arroyo López, Universidad de Castilla-La Mancha, Spain Beatriz Esteban Ramiro, Universidad de Castilla-La Mancha, Spain Marcell Lörincz, Foundation of Subjective Values, Hungary Chapter 2 Creating an Online Network, Monitoring Team and Apps to Counter Hate Speech, and Hate Crime Tactics in Europe........................................................................................................................ 14 Roberto Moreno López, Universidad de Castilla-La Mancha, Spain Fabienne Baider, University of Cyprus, Cyprus Chapter 3 Using HurtLex and Best-Worst Scaling to Develop ERIS: A Lexicon for Offensive Language Detection................................................................................................................................................ 32 Valerio Basile, University of Turin, Italy Stella Markantonatou, Institute for Language and Speech Processing, Athena Research Center, Greece Vivian Stamou, Institute for Language and Speech Processing, Athena Research Center, Greece Iakovi Alexiou, Institute for Language and Speech Processing, Athena Research Center, Greece Georgia Apostolopoulou, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Vana Archonti, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Antonis Balas, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Eleni Koutli, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Maria Panagiotopoulou, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece 



Chapter 4 Spreading Organised Hate Content........................................................................................................ 50 Sergio Arce-Garcia, Universidad Internacional de La Rioja, Spain Jarnishs Beltran, Universidad de Valparaiso, Chile Chapter 5 Approximation of Hate Detection Processes in Spanish and Other Non-Anglo-Saxon Languages...... 65 Juan José Cubillas Mercado, Universidad Internacional de La Rioja, Spain Óscar De Gregorio Vicente, Universidad Internacional de La Rioja, Spain C. Vladimir Rodríguez Caballero, ITAM, Mexico Chapter 6 The Interaction Between Offensive and Hate Speech on Twitter and Relevant Social Events in Spain...................................................................................................................................................... 81 Jesús Gómez, Spanish National Office Against Hate Crimes, Spain Alberto Matilla-Molina, Spanish National Office Against Hate Crimes, Spain Ma. Pilar Amado, Spanish National Office Against Hate Crimes, Spain Dimosthenis Antypas, Cardiff University, Great Britain Jose Camacho-Collados, Cardiff University, Great Britain Carlos J. Máñez, Spanish National Office Against Hate Crimes, Spain Tomás Fernández-Villazala, Spanish National Office Against Hate Crimes, Spain Alicia Méndez-Sanchís, Autonomous University of Madrid, Spain Javier López, State Secretariat for Security, Ministry of Interior, Spain Section 2 Mediterranean Countries’ Approaches Chapter 7 The Semiotics of Xenophobia and Misogyny on Digital Media: A Case Study in Spain................... 111 Max Römer-Pieretti, Universidad Camilo José Cela, Spain Julio Montero-Díaz, Universidad Internacional de La Rioja, Spain Elias Said-Hung, Universidad Internacional de La Rioja, Spain Chapter 8 Mapping Stigmatizing Hoaxes Towards Immigrants on Twitter and Digital Media: Case Study in Spain, Greece, and Italy....................................................................................................................... 136 Marta Sánchez Esparza, Rey Juan Carlos University, Spain Ignacio Vázquez Diéguez, Universidad de Beira Interior, Portugal Adoración Merino Arribas, Universidad Internacional de La Rioja, Spain



Chapter 9 The Southernification of the Pandemic in Italy: Images of the South, Fears of Contamination, and the First Wave of COVID-19 in Italy................................................................................................... 162 Marcello Messina, Southern Federal University, Russia Chapter 10 Are There Hate Speeches on Spanish Television? Methodological Proposal and Content Analysis Over the 2020 Year.............................................................................................................................. 186 Sandra Martínez Costa, University of A Coruña, Spain Teresa Nozal Cantarero, University of A Coruña, Spain Antonio Sanjuán Pérez, University of A Coruña, Spain José Juan Videla Rodríguez, University of A Coruña, Spain Chapter 11 Hate Speech or Hate Shot? Finding Patterns of the Anti-Muslim Narratives in Italy......................... 203 Alessandra Vitullo, Sapienza University of Rome, Italy Chapter 12 The Expression of Hate in Portuguese Digital Media: Ethnic and Racial Discrimination.................. 220 Inês Casquilho-Martins, Centro de Investigação e Estudos de Sociologia, Iscte-Instituto Universitário de Lisboa, Portugal David Ramalho Alves, Iscte-Instituto Universitário de Lisboa, Portugal Helena Belchior-Rocha, Centro de Investigação e Estudos de Sociologia, Iscte-Instituto Universitário de Lisboa, Portugal Chapter 13 Online Hate Speech and the Representations of Refugees in #VatanimdaMülteci (#RefugeeinMyCountry)...................................................................................................................... 237 Semra Demirdis, Cankiri Karatekin University, Turkey Chapter 14 Analysis of Radicalisation Prevention Policies From the Perspective of Educommunication in Mediterranean Countries..................................................................................................................... 260 Arantxa Azqueta, International University of La Rioja, Spain Ángela Martín-Gutiérrez, Universidad de Sevilla, Spain Angel Freddy Rodríguez-Torres, Universidad Central, Ecuador



Chapter 15 New Narratives to Defuse Hate Speech............................................................................................... 283 Maximiliano Bron, Universidad Nacional de La Rioja, Argentina Óscar Javier Arango Arboleda, Universidad de Barcelona, Spain Angelica María Rodríguez Ortiz, Univdersidad Autónoma de Manizales, Colombia Héctor Claudio Farina Ojeda, Universidad de Guadalajara, Mexico Compilation of References................................................................................................................ 306 About the Contributors..................................................................................................................... 356 Index.................................................................................................................................................... 362

Detailed Table of Contents

Preface.................................................................................................................................................. xvi Acknowledgment................................................................................................................................ xxii Section 1 Detection Chapter 1 European Initiatives for the Support and Counselling of Victims of Hate Crimes: Key Actors............. 1 César Arroyo López, Universidad de Castilla-La Mancha, Spain Beatriz Esteban Ramiro, Universidad de Castilla-La Mancha, Spain Marcell Lörincz, Foundation of Subjective Values, Hungary In this chapter, the authors introduce the readers to SHELTER, support and advice through health system for hate crimes victims, an European project implemented between 2018 and 2021, by a diversity of partners conformed by universities and NGOs. The main aim of the project was to improve the protection of victims of hate crimes and their access to resources and networks that facilitate the report, assistance and specialised support, by four specific objective: Tackling the underreporting data whithin the health system in relation to aggression and violence committed under the approach of hate crime; strengthening the medical and psyco-social care to victims of hate crime at the heath system; facilitating the access of victims to protection, assistance, and specialised support resources; and incorporating the health system institutions to a international network for supporting victims of hate crime.The results showed, among others, an improvement on the knowledge of the state of art and enhancement of the competences of the medical staff trained. Chapter 2 Creating an Online Network, Monitoring Team and Apps to Counter Hate Speech, and Hate Crime Tactics in Europe........................................................................................................................ 14 Roberto Moreno López, Universidad de Castilla-La Mancha, Spain Fabienne Baider, University of Cyprus, Cyprus In this chapter, the authors present a European initiative for the creation of an online network, monitoring team, and apps to counter hate crime tactics. The project, funded by the Directorate-General of Justice of the European Commission, aims for the promotion of cooperation on reporting and monitoring online hate speech within European countries with a synergy between strong NGOs and universities. The project was based in four axes of correlative actions, that went from the research to the design and 



implementation of trainings with stakeholders (enforcement bodies and youngsters), the creation of multiplatform tools (web and app), and specific efforts on dissemination and raising awareness with the most vulnerable communities and the general public. The results showed an improvement of the state of art on the situation in the participant countries, as well as real changes in the area of combating hate crime and on-line hate speech in some countries belonging to the consortium. Chapter 3 Using HurtLex and Best-Worst Scaling to Develop ERIS: A Lexicon for Offensive Language Detection................................................................................................................................................ 32 Valerio Basile, University of Turin, Italy Stella Markantonatou, Institute for Language and Speech Processing, Athena Research Center, Greece Vivian Stamou, Institute for Language and Speech Processing, Athena Research Center, Greece Iakovi Alexiou, Institute for Language and Speech Processing, Athena Research Center, Greece Georgia Apostolopoulou, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Vana Archonti, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Antonis Balas, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Eleni Koutli, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Maria Panagiotopoulou, Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece ERIS, a lexical resource of Modern Greek for offensive language detection, is the result of cleansing, enriching and assigning graded offensiveness values to the EL branch of HurtLex. ERIS contains 1148 entries and is openly available. Graded values were obtained with the Best-Worst Scaling methodthat was applied with the Litescale tool. Nouns and adjectives that have humans as a target were found to attract bigger offensiveness values. The classification of the terms in ERIS with the BWS method and a previous classification of a substantial subset of these terms into “offensive (context in/dependent)” with the inter-annotator agreement method are found to stand in a broad correlation, thus validating the methodology that was adopted to produce a more fine-grained and informative affective lexical resource. ERIS contains 1148 terms and their inflectional paradigms. It is openly available under the CC-BY-NC 4.0 license. Chapter 4 Spreading Organised Hate Content........................................................................................................ 50 Sergio Arce-Garcia, Universidad Internacional de La Rioja, Spain Jarnishs Beltran, Universidad de Valparaiso, Chile Hate and disinformation often go hand in hand, and there are organisations dedicated to spreading them and influencing public opinion. The aim of this chapter is therefore to expose the forms of hate-mongering and one of its means of penetration, disinformation, as well as the techniques used to spread it and the measures taken to counter it. It should be remembered that behind these hate-creation techniques, there



are companies and even government systems that have perfected and have the means to achieve their goals. These techniques and forms are used internationally, but are particularly widespread in European and Mediterranean countries, especially around issues such as immigration. The aim is to polarise societies by exploiting fissures and divisive issues. This is a dangerous game in which they end up being capitalised on by extreme and ultra-ideological groups in an attempt to change the culture. Even democracy itself is at risk. Chapter 5 Approximation of Hate Detection Processes in Spanish and Other Non-Anglo-Saxon Languages...... 65 Juan José Cubillas Mercado, Universidad Internacional de La Rioja, Spain Óscar De Gregorio Vicente, Universidad Internacional de La Rioja, Spain C. Vladimir Rodríguez Caballero, ITAM, Mexico In this chapter, the authors present how the use of artificial intelligence (AI) can help to identify and reduce the new digital crimes according to hate messages. The appearance of the internet in our lives, at the end of the last century, has meant a great technological advance, providing easier access to a huge volume of information and communication between people. The rise of communication-oriented networks has been such that true digital environments have been created, the so-called social networks, with millions of users all over the planet. This has meant, to a large extent, the modification of our personal relationships, and, unfortunately, the appearance of new ways of sending hate messages. The work presented is aimed at a digital tool built for this purpose for the automatic detection of hate (and non-hate) messages, in Spanish and other non-Anglo-Saxon languages, with AI algorithms, using training data from the Spanish language. Chapter 6 The Interaction Between Offensive and Hate Speech on Twitter and Relevant Social Events in Spain...................................................................................................................................................... 81 Jesús Gómez, Spanish National Office Against Hate Crimes, Spain Alberto Matilla-Molina, Spanish National Office Against Hate Crimes, Spain Ma. Pilar Amado, Spanish National Office Against Hate Crimes, Spain Dimosthenis Antypas, Cardiff University, Great Britain Jose Camacho-Collados, Cardiff University, Great Britain Carlos J. Máñez, Spanish National Office Against Hate Crimes, Spain Tomás Fernández-Villazala, Spanish National Office Against Hate Crimes, Spain Alicia Méndez-Sanchís, Autonomous University of Madrid, Spain Javier López, State Secretariat for Security, Ministry of Interior, Spain Hate speech is one of the major concerns of Europe. Different studies, mainly in English language, have been carried out to analyze hate speech, many of them from a theoretical perspective. Here, it is presented an observational study about hate speech poured on Twitter in Spanish regarding to five social important events: Women´s Day, International LGTBQ+ Pride Day, Spain National Day, national elections, and regional elections. Three different experiments were carried out; two used deep learning algorithms to automatically classify tweets, meanwhile, the latest tweets were classified by a human. Results showed that these events significantly triggered hate speech, yet results differed between experiments, and also depending on the nature of the events. A better understanding of the mechanisms of hate speech propagation can help improve policies in Spain or in countries with similar characteristics, and thus help law enforcement and other institutions to address the scourge of hate crimes.



Section 2 Mediterranean Countries’ Approaches Chapter 7 The Semiotics of Xenophobia and Misogyny on Digital Media: A Case Study in Spain................... 111 Max Römer-Pieretti, Universidad Camilo José Cela, Spain Julio Montero-Díaz, Universidad Internacional de La Rioja, Spain Elias Said-Hung, Universidad Internacional de La Rioja, Spain This chapter raises three questions: a) concerns a synthesis of the classic contributions of the reference semiotic authors that are considered when analysing hate speech in social media; b) entails presenting a case study that is analysed precisely with that analysis synthesis; c) shows the usefulness and interest of this type of analysis in investigations of hate speech. It offers a semiotic model for analysing misogynistic and xenophobic hate speech from digital news media on Twitter. The case study comprises the news published by El Mundo (Spain) from its users on social media, and the 33 comments generated, as a reason for this publication, during January 2021. This serves as the basis of semiotic analysis for understanding the phenomenon. The results visualise the semiotic analyses for understanding the dissemination of expressions. This approach thus helps reveal the levels of intensity, the denotative and connotative differences, the destructive-constructive and intertextual nature of messages, and sheds light on the different symbolic structures associated with hate speech. Chapter 8 Mapping Stigmatizing Hoaxes Towards Immigrants on Twitter and Digital Media: Case Study in Spain, Greece, and Italy....................................................................................................................... 136 Marta Sánchez Esparza, Rey Juan Carlos University, Spain Ignacio Vázquez Diéguez, Universidad de Beira Interior, Portugal Adoración Merino Arribas, Universidad Internacional de La Rioja, Spain This text presents a mapping of the hoaxes published on Twitter and in digital media during 2020 in Spain, Greece, and Italy after having been classified as disinformation with the intention of causing harm in the fact-checking portals of the three cited countries. Verification services that are members of the International Fact-Checking Network (IFCN) have been chosen for this analysis: Spain (Newtral.es/ Maldita.es), Greece (Ellinika Hoaxes), and Italy (Facta News/Lavoce.info). The chosen online portals belong to FactCheckEU, a European project launched by the international verification network. The validated sample presents 150 pieces of information identified as hoaxes by the verification platforms and disseminated in the current communication scenarios by the network in the three direct recipient countries of the migratory phenomenon through the Mediterranean. A qualitative methodology applied to the case study is used, which is complemented by critical discourse analysis. Chapter 9 The Southernification of the Pandemic in Italy: Images of the South, Fears of Contamination, and the First Wave of COVID-19 in Italy................................................................................................... 162 Marcello Messina, Southern Federal University, Russia Starting from February 2020, Italy was the first among the European countries, to experience dramatic rises in daily COVID-19 deaths and contagions. An important aspect that distinguished the first COVID-19



wave (Feb-Jun 2020) from the following waves of infection in Italy was the sheer imbalance, in terms of deaths and contagion, between Northern and Southern regions of the country. Despite the fact that the South was far less hit by the disease, a series of narratives that associated the spread of the epidemic with some sort of Southern infector started to appear, conveyed by social media posts, news pieces, talk shows, and even football banners. In this chapter, there is an attempt to identify and critically analyse the discourses that inscribe a characteristic “Southernification” of the pandemic in Italy, that is a partial and symbolic attempt to (1) discursively transfer the infection to the South; and/or (2) hand over the responsibilities that are behind the particularly violent first wave of infections in the country to Southern communities, polities, and cultural practices. Chapter 10 Are There Hate Speeches on Spanish Television? Methodological Proposal and Content Analysis Over the 2020 Year.............................................................................................................................. 186 Sandra Martínez Costa, University of A Coruña, Spain Teresa Nozal Cantarero, University of A Coruña, Spain Antonio Sanjuán Pérez, University of A Coruña, Spain José Juan Videla Rodríguez, University of A Coruña, Spain Few academic studies focus on hate speech on television. That is partly due to the difficulty of obtaining and analyzing broadcast content. However, the spread of those hate messages implies a high reach. For this study, the researchers propose an experimental methodology to analyze the content broadcast on the 24 hours of the five Spanish free-to-air television channels over one year (2020). The authors examined the presence of abusive or hurtful vocabulary and quantified the insults aired. They extracted and studied a sample through content analysis to detect if those insults were accompanied, in any way, by expressions of hatred. Although the messages the researchers studied for this article cannot be considered speeches or hate crimes, there are some offensive comments related to gender, race, or religion, mainly on fictional products. Chapter 11 Hate Speech or Hate Shot? Finding Patterns of the Anti-Muslim Narratives in Italy......................... 203 Alessandra Vitullo, Sapienza University of Rome, Italy Nowadays, the Muslim community is one of the most discriminated groups in Europe. Anti-Islam hate speeches circulate online and offline especially through the intense use social media, fake news, bots, and click-baiting practices. However even if Muslim discriminatory practices have been gaining more media relevance in the recent years, anti-Muslim stereotypes date back far beyond our times. Using the theoretical frameworks developed by Said and Moscovici this research aims to analyze 31 semistructured interviews conducted with the volunteers of the Amnesty International’s Hate Speech Task Force, to investigate which are the most persistent anti-Muslim representations in Europe and Italy today. In bringing the theory into practice this work will explore the dynamics occurring on Facebook among users which show polarized and intolerant positions while engaging an Islam-related conversation. This specific case study will allow to show how and why old anti-Islam stereotypes persist almost unchanged from an offline to an online world.



Chapter 12 The Expression of Hate in Portuguese Digital Media: Ethnic and Racial Discrimination.................. 220 Inês Casquilho-Martins, Centro de Investigação e Estudos de Sociologia, Iscte-Instituto Universitário de Lisboa, Portugal David Ramalho Alves, Iscte-Instituto Universitário de Lisboa, Portugal Helena Belchior-Rocha, Centro de Investigação e Estudos de Sociologia, Iscte-Instituto Universitário de Lisboa, Portugal This chapter aims to analyze the hate speech discourse in the Portuguese context, be it spread or fed through news media platforms and by their users in Portugal. Considering the need to share evidence and produce theoretical and empirical knowledge about this field of research, this study aims to contribute to this reflection and provide information to the international audience. By analyzing the Portuguese legal framework, statistical data and narratives about hate speech against immigrants and other minority groups in digital media (e.g. online news, Facebook) and traditional media (e.g. television, radio), the reader will become better acquainted with the policies associated with hate speech present through digital media and its detection. Chapter 13 Online Hate Speech and the Representations of Refugees in #VatanimdaMülteci (#RefugeeinMyCountry)...................................................................................................................... 237 Semra Demirdis, Cankiri Karatekin University, Turkey In Turkey, a dramatic event involving Syrian refugees happened because of allegations that Syrian men had harassed Turkish women. Following the case, Turkish citizens generated a popular hashtag of #VatanimdaMülteci (#RefugeeinMyCountry) to share negative opinions, feelings, and ideologies towards Syrian refugees. This study is an examination of how Twitter was used to produce and spread hate speech discourse directed at refugees and focus on the representations of refugees through the online environment to provide information about anti-refugee rhetoric for specific nations. A quantitative and qualitative content analysis was carried out of the tweets under the hashtag #VatanimdaMülteci. The results demonstrate that a significant number of tweets contained hate speech comments designed to criticise Turkish government policies regarding refugees, such as the Turkish citizenship provided to refugees and their ability to open businesses in Turkey. The study shows that the hospitality of Turkish citizens turned into hostility over time. Chapter 14 Analysis of Radicalisation Prevention Policies From the Perspective of Educommunication in Mediterranean Countries..................................................................................................................... 260 Arantxa Azqueta, International University of La Rioja, Spain Ángela Martín-Gutiérrez, Universidad de Sevilla, Spain Angel Freddy Rodríguez-Torres, Universidad Central, Ecuador The fight against radicalisation is gaining prominence on international agendas. Europe proposes multilevel actions, where “educommunication” helps to prevent hate speech, as it is a tool that contributes to the formation of a critical public opinion in the 4.0 era. The aim of this chapter is to analyse the attitudes that define interculturally competent citizenship and their presence in the radicalisation prevention policies of three Mediterranean countries: France, Portugal, and Spain. Elements related to openness, respect, civic-mindedness, self-efficacy, and tolerance are analysed. The results show that the plans analysed



show differences in 1) the presence or absence of victims in the attacks committed in the territory and 2) the presence of the Muslim population in the territory over a period of time. Furthermore, the analysis has led to the conclusion that it is necessary to promote cross-cutting policies for the prevention of radicalisation that address identity aspects. Chapter 15 New Narratives to Defuse Hate Speech............................................................................................... 283 Maximiliano Bron, Universidad Nacional de La Rioja, Argentina Óscar Javier Arango Arboleda, Universidad de Barcelona, Spain Angelica María Rodríguez Ortiz, Univdersidad Autónoma de Manizales, Colombia Héctor Claudio Farina Ojeda, Universidad de Guadalajara, Mexico According to a 2021 report by the Spanish government, hate speech has increased by 60%, with 90% of survey respondents experiencing humiliation that amounts to “hate crimes.” UNICEF has also reported a 13% increase in hate speech among young people in Latin America. Both institutions have responded with regulations and campaigns to combat hate in educational systems and society at large. This chapter presents new narratives that have been used in the Musik Thinking (Barcelona) and CoCritic.Ar (Latin America) projects, in which, through education, mechanisms are provided to strengthen critical thinking and initiate processes to deactivate hate speech, especially those directed towards immigrants. The results of the process show how music in music thinking and critical literacy, collaborative work, argumentation, and transmedia narratives in CoCritic.Ar generate spaces for citizen discussion to respond to the increase in hate speech occurring in the Mediterranean and Latin America. Compilation of References................................................................................................................ 306 About the Contributors..................................................................................................................... 356 Index.................................................................................................................................................... 362

xvi

Preface

ANALYSING HATE SPEECH: GENERAL FRAMEWORKS AND RESEARCH PROBLEMS The unleashed feelings circulating in the streets are probably the worst nightmare for any authority, even for people. As long as they limit their public nature to the setting in which they occur, these outrages can be more or less conspicuous, even constitute a crime, but little else. A surprise, a complaint, a show or an intervention by the police can follow them. They may not get a place until a fifth or sixth click in a digital newspaper. In overcoming the barriers of order, the definitive thing is the volume that it acquires. This dimension is defined by the number of people involved in that mother’s outing. When the masses are the protagonists of a movement (cultural, sports, political, social, among others), the only solution is to make way for them to reduce disasters because cancelling them is impossible. Outbursts of joy are more bearable. However, even these generate problems for lovers of order and the rest (even if they do not say so for fear of hurting sensibilities). “Popular” celebrations of sporting and political success show it quite often, but nothing compares to hate explosions’ destructive effects. Indeed, in all cultures, there have always been authentic “hate professionals.” They are the ones who generate, encourage, lead or take advantage of these outbursts. Whether they are psychopaths to lock up or tolerable elements in necessarily unstable societies are assessments that fit this highly unequal scale. In addition, getting the diagnosis right is only possible at the end and by then, things have already gone wrong, and it is only possible to rebuild with what is left. Our next history has witnessed the construction and consolidation of the announced global village. In this relatively new urban context, the public expression of emotions has also been reorganised, especially the manifestly negative ones. These have had to adapt to the dimensions and characteristics of a macro city that intertwines the real with the virtual, the analogical with the digital, the enormity of the dimensions with the neighbourhood’s proximity, and the multiplication of possible poles of identity with depersonalisation of relationships. In this complexity, merely face-to-face has lost the sense of the only channel whose rupture can drag down the rest. On the contrary, the triggering role of the virtual is increasing and it is the potential to involve elements (human or not) of a social and material nature in the new overflows that hate feeds (Nortio et al., 2020). Without ruling out maximum limits: the liquidation of the stick figure that represents the execrable and that the masses inevitably confuse with real institutions and people. A first approach to this issue, also framed in the particularly sensitive territory, is “The Theme of Hated in Modern Communication: The Case of Palestine,” one of the chapters of this book. This dangerous journey threatens to turn into a dark Matrix, presented as Mac Luhan ‘s happy Global Village. It is an increasingly widespread fear. This is shown by an example that analyses one of  

Preface

the present contributions (“The Southernification of the Pandemic in Italy: Images of the South, Fears of Contamination, and the First Wave of COVID-19 in Italy”). For this reason, it is intended to redirect the feared process to avoid destroying civilised life in the style of an enviable “European garden.” It is witnessed little by little that a colossal effort is being made to avoid the disorder that any form of hate implies, which always translates into violence: from the one generated by clean (and above all sustainable) technologies in the first world to the that appear full of filth, dirt and misery and armed to the teeth, from failed states. Academics (from all areas of knowledge that label the most diverse databases of the highest quality journals) have answered the call in this fight against hate in our societies, especially in the most advanced ones, perhaps because they live in them. Some have joined institutional organisations that try to get ahead of these crimes. Others have decided to analyse the indicators that offer clues to promote policies that prevent or counteract both crimes and incitement to them. Both of us carry out our activity in particular contexts. Of course, hatred is not absent in them. His presence is clear and probably inevitable. However, if it is measured in actions (fatalities, attacks, physical violence, brutal discrimination) and compared with the rest of the world, the levels of intensity and amplitude that they acquire in failed or unrecognised states are not even remotely reached. and legal respect for human rights. The contribution “Hate Speech or Hate Shot? finding patterns of the anti-Muslim Narratives in Italy” is an example of the importance of this first distinction. This asymmetry must be highlighted whenever the analysis of the social presence of hate is addressed: whether or not it is institutionalised. Another asymmetry that should be highlighted is that most academic production on hate speech has its origin and setting in the United States (Tontodimamma et al., 2021). In addition, the countries that follow are Great Britain, Australia and Canada. The production of others is located at a great distance (Paz et al., 2020). In the face of this overwhelming data, one can legitimately wonder if hate speech is something typical (or almost) of the Anglo-Saxon world. One might almost ask as a research question whether the rest of the Western world is less sensitive to this problem or has simply learned to live with it. Another possibility is that it has not worried those academic communities far from the privileged Anglo-Saxon campuses of the very first world. Another possible question would be whether hate speech could be described as simple dysfunctions of affluent societies. Whether academic emulation (a cynical perspective) or awareness of the problem, the academic production of hate speech has become more and more general (Di Fátima, 2023). Even the disproportion pointed out here should be tempered by the general predominance of “high-end” academic journals from the Anglo-Saxon “windows.” However, hate speech shows that something is not working well in Society. That is why they are of interest to social science analysts. These ways of saying, writing, conversing, and representing are relatively easy to detect individually. Nevertheless, it is more complicated to take charge of their leading role in specific discriminatory acts and establish cause-effect relationships, or more in accordance with the possibilities of academic studies on Society, to establish significant correlations of interest. Some of these advances and difficulties can be seen in two of the contributions in this volume led by two institutional entities that have been working for years on the detection and relationship of hate speech and crimes of the same type: “European Initiatives for the Support and Counseling of Victims of Hate Crimes” and “European Initiatives for the Support and Counseling of Victims of Hate Crimes.” There is also a prior difficulty for researchers: measuring, quantifying and counting these “discourses”, although it might initially seem elementary. They are not as far as hatred is concerned: from the outset, one hates according to a plus and a minus. The biggest problem is not establishing a scale of the intensity of hate or the expressions containing it (which is not an elementary task either). The xvii

Preface

difficulty lies in the enormity of the mass of texts that researchers face. These can indeed be reduced to specific cases, but the applicability focuses mainly on what they can contribute as a methodology, or as has been said in a metaphorical and understandable tone: what can be said about an ocean after analysing one of its drops of water. The work “The Semiotics of Xenophobia and Misogyny on Digital Media: A Case Study in Spain.” Going back to that necessary relativity in hate and its expressions, which mark different intensities, the Anti-Defamation League already proposed five general levels presented as a pyramid, distributing actions and speeches from lesser to more intense. At the base are negative stereotypes, followed by insults and expressions of discrimination. At the top are threats and genocide. Other research groups have adopted other solutions for the complete analysis of discourses: six classificatory concepts that can be reduced to three to facilitate the work of the designers of the corresponding algorithm (De Lucas Vicente, et al., 2022). Some attempts to organise and generalise these efforts are mentioned in “Creating an Online Network, Monitoring Team and Apps to Counter Hate Speech, and Hate Crime Tactics in Europe.” The presence (and detection) of hate in the expressions disseminated by the media does not necessarily imply that it is “hate speech.” The prominent role of hate could be said not to ensure that there is “hate speech” in the strict sense, in which academics who study these expressions usually handle it. Hate must refer to a vulnerable group. Vulnerability is a broad and elastic concept. More so in satisfied societies such as Western ones, in which the condition, the label, of victim constitutes a distinction with a positive tone in the discourse (not necessarily, nor usually in social reality). The powerful, at least those who are far from contempt for who they are, can be envied, insulted, or wish evil. However, these expressions directed against those who move in a habitual and relatively safe world do not properly constitute “hate speech.” Perhaps they are out of envy, fury, or a reaction to defeat... they will be in bad taste, insulting, uncivil and may even constitute crimes defined in the penal code. However, they are not sociologically, psychologically, or communicatively speaking “hate speech.” The digital age confronts social science researchers with the treatment of massive data if they want reliable studies for their colleagues. The first barrier to be overcome is establishing a fluent and clear conversation between social science academics, mathematicians, and computer scientists. It is not always overcome. A practical obstacle is the execution times of the processes. When the technicians affirm that they are achievable, the social analysts imagine them to be little less than instantaneous. In the meantime, months go by, and the data lose relevance, and the conclusions are of no interest for their application or their publication in journals with guaranteed diffusion among specialists in the area. This difficulty of understanding between specialists from one area or another appears overcome in “Using HurtLex and Best-Worst Scaling to Develop ERIS: A Lexicon for Offensive Language Detection,” where linguists and computer scientists have managed to work together with good results. It is more frequent to find mathematicians, statisticians and computer scientists in teams that are integrated, in turn, into more extensive and diversified groups, with experts in social sciences (“Approximation of Hate Detection Processes in Spanish and Other Non-Anglo-Saxon Languages.”) There are usually fewer problems in defining precisely what data has indicator value. This aspect is the responsibility of social scholars. It is fundamental. For example, establishing categories in the intensities of hate is key in the matter that concerns us. Each must allow a news reader to classify each in the corresponding “drawer.” Those who define these intensities must establish distinguishable categories and avoid classifications based on gradations of the same dimension. The limits between the intensities of hate cannot be indicated by the passage from one figure to another on a scale, and they must refer to facts or, better, to different acts: it is not the same to call for action as to insult or suggest. They are xviii

Preface

not degrees on the same scale but very clear and differentiated actions. In each case, there may also be mitigating or intensifying factors, easy to mark grammatically. It will also be possible to assess whether or not there is irony, humour, or some other quality considered significant (Paz-Rebollo et al., 2021). In addition, the perception of hate speech has a very prominent subjective component and studies are needed to understand it and the aspects that condition it (Salminen et al., 2018; Udanor & Anyanwu, 2019). The intensities, their definition and treatment constitute a starting point that adds value to the most frequent analyses limited to indicating the existence or not of hate in certain expressions in the media and social networks. But knowing what that number responds to is at least as important. In other words: does the abundance of expressions tending to hate indicate in each singularity a human being who expresses his opinion, or are we facing “mechanisms” (boots) that sow judgments with a single author (personal or institutional) on the networks? Even when it comes to people, won’t they be hired to do that job? It does not just matter for giving value to the raw digital data. The existence or not of these procedures offers clues about other issues. The text “Spreading Organised Hate Content” addresses this interesting issue. The intensity indicators are not exhausted in the studies on expressions that facilitate the diffusion of hate feelings. It also matters a lot how to define what is hated: women, immigrants, those of another religion, other races or ethnic groups, and those of another sexual orientation. The mere establishment of this “labelling,” which looks like a simple catalogue, apparently so simple, has its difficulties. For example, those who have to program to differentiate themselves must know what to do when faced with an expression that manifests contempt for multiple reasons: a black immigrant woman and Muslim religion from Africa. This offers no difficulty in reading, but it is not easy to specify a search carried out by an algorithm, let alone prepare it for that location. If each possible thematic combination of hate constitutes a new category, the possibilities and the volume of information that must be carried out “manually” to feed the future algorithm skyrocket. Mathematicians and social analysts have to speak. This book does not lack some contributions of interest on “specific types” of hate that are translated into expressions that both digital media and specific social networks disseminate, on what their characteristics are. Those linked to ethnic issues (“The Expression of Hate in Portuguese Digital Media: Ethnic and Racial Discrimination”) or immigration (“Mapping Stigmatising Hoaxes Towards Immigrants on Twitter and Digital Media: Case Study in Spain, Greece, and Italy”) and refugees (“Online Hate Speech and The Representations of Refugees in #VatanimdaMülteci (#RefugeeinMyCountry)”). Sometimes, they are only distinguishable by the focus of the study and by the methodological principles that inspire them. Mention has already been made of some of the frontiers that must be addressed when addressing studies of “hate speech.” First, what is hate speech (or not), and the relative weight that hate and speech have in them. The old principle that “a nail painted on the wall can only hang an object also painted on the same wall” helps to understand the limits of our studies on hate speech. They are, by definition, focused on the analysis of expressions, not on the analysis of facts, although they frequently denote or prepare them for even violent actions. Although they do not mark a border but an approach, studies that propose solutions to the evil of disseminating these expressions of hate are beginning to increase. Sometimes the contributions come from the field of narratives (“New Narratives to Defuse Hate Speech”). Others propose procedures that resort to training in the understanding and assimilation of communication itself in societies that could be described as digital (“Analysis of Radicalisation Prevention Policies From the Perspective of Educommunication in Mediterranean Countries”). Not only is it a legitimate perspective, but it is also an ethical responsibility that involves all of us academics who make it our profession to analyse the environment and context in which we live. If the formula already enunciated (that of the painted nail) and the principle xix

Preface

of proportionality between evils and their remedies are applied, it would be necessary to say, without a doubt, that one discourse is fought with another. Moreover, add, and not with anything else. The observatories governments and institutions have set up to analyse speeches as part of preventive policies against future (and probable) hate crimes are understandable and probably necessary. There are doubts about its effective preventive efficacy, but only doubts. Sometimes these observatories seem more interested in discovering new “pockets of social vulnerability” into which hate is unleashed in the media than in establishing effectively preventive proposals. Digital media also establish control systems, although they are not always fully effective (Paz-Rebollo et al., 2021). However, the development and dissemination of new integrating narratives seem to be a proposed solution that is already being worked on: the use of counter-discourses, not so much confrontation (Hangartner et al., 2021; Woodzicka et al., 2015). Other limits refer to the most appropriate methodologies to detect, measure, catalogue, and classify these discourses. In this sense, we also know that the metadata (daily, date, section, place, among others) offered by the media associated with the news does not always accurately define variables that must be considered. Let us think about the scenarios in which certain expressions occur. The newspaper sections where they are published offer first cataloguing, but they may not always be helpful. It is clear that the sports sections, for example, tend to mark “settings” in which these expressions occur. Political ones also tend to serve this purpose. However, those of Society already offer principles of ambiguity. This is not to mention that each medium labels its sections differently, which makes solving this problem even more difficult. Another option, which is becoming increasingly influential, is discourse analysis since it can interpret the ironies, metaphors and double meanings most present in these expressions. This set of observations and, above all, the contributions that follow constitute a first referential framework that has sought collaborations linked to the Mediterranean area and the need to advance in the automatic analysis of these discourses: although, unfortunately, hatred is not limited solely or mainly to narratives that we analyse. Julio Montero Diaz Proeduca, Spain Elias Said Hung Universidad Internacional de la Rioja, Spain

REFERENCES De Lucas VicenteA.Römer PierettiM.IzquierdoD.Montero-DiazJ.Said-HungE. (2022). Manual para el Etiquetado de mensajes de odio. doi:10.6084/m9.figshare.18316313 Di Fátima, B. (Ed.). (2023). Hate speech on social media. University of Beira Interior. Hangartner, D., Gennaro, G., Alasiri, S., & Donnay, K. (2021). Empathy-based counterspeech can reduce racist hate speech in a social media field experiment. Proceedings of the National Academy of Sciences of the United States of America, 50(118), 1–3. doi:10.1073/pnas.2116310118 PMID:34873046

xx

Preface

Nortio, E., Niska, M., Renvik, T., & Jasinskaja-Lahti, I. (2020). The nightmare of multiculturalism: Interpreting and deploying anti-immigration rhetoric in social media. New Media & Society. Advance online publication. doi:10.1177/1461444819899624 Paz, M. A., Montero, J., & Moreno, A. (2020). Hate Speech: A Systematised Review. SAGE Open, 10, 1–21. doi:10.1177/2158244020973022 Paz-Rebollo, M. A., Cáceres-Zapatero, M. D., & Martín-Sánchez, I. (2021). Suscripción a la prensa digital como contención a los discursos de odio. El Profesional de la Información, 30(6), e300613. Advance online publication. doi:10.3145/epi.2021.nov.13 Paz-Rebollo, M. A., Mayagoitia-Soria, A., & González-Aguilar, J. M. (2021). From Polarisation to Hate: Portrait of the Spanish Political Meme. Social Media + Society, 7(4). Advance online publication. doi:10.1177/20563051211062920 Salminen, J., Veronesi, F., & Almerekhi, H. (2018). Online Hate Interpretation Varies by Country, But More by Individual: A Statistical Analysis Using Crowdsourced Ratings. Fifth International Conference on Social Networks Analysis. Management and Security, 88-94. 10.1109/SNAMS.2018.8554954 Tontodimamma, A., Nissi, E., Sarra, A., & Fontanella, L. (2021). Thirty years of research into hate speech: Topics of interest and their evolution. Scientometrics, 126(1), 157–179. doi:10.100711192-020-03737-6 Woodzicka, J. A., Mallett, R. K., Hendricks, S., & Pruitt, A. V. (2015). It’s just a (sexist) joke: Comparing reactions to sexist versus racist communications. Humor: International Journal of Humor Research, 28(2), 289–309. doi:10.1515/humor-2015-0025

xxi

xxii

Acknowledgment

This book is the result of the collective effort of each of the authors participating in it and those who submitted it during the opening period of the call for proposals. The book results from the project ‘Taxonomy, presence and intensity of hate speech in digital environments linked to the Spanish professional news media - Hatemedia (PID2020-114584GB-I00)’, financed by the State Research Agency - Ministry of Science and Innovation of Spain.

 

Section 1

Detection

1

Chapter 1

European Initiatives for the Support and Counselling of Victims of Hate Crimes: Key Actors

César Arroyo López https://orcid.org/0000-0002-8249-5697 Universidad de Castilla-La Mancha, Spain Beatriz Esteban Ramiro https://orcid.org/0000-0002-4736-1693 Universidad de Castilla-La Mancha, Spain Marcell Lörincz Foundation of Subjective Values, Hungary

ABSTRACT In this chapter, the authors introduce the readers to SHELTER, support and advice through health system for hate crimes victims, an European project implemented between 2018 and 2021, by a diversity of partners conformed by universities and NGOs. The main aim of the project was to improve the protection of victims of hate crimes and their access to resources and networks that facilitate the report, assistance and specialised support, by four specific objective: Tackling the underreporting data whithin the health system in relation to aggression and violence committed under the approach of hate crime; strengthening the medical and psyco-social care to victims of hate crime at the heath system; facilitating the access of victims to protection, assistance, and specialised support resources; and incorporating the health system institutions to a international network for supporting victims of hate crime.The results showed, among others, an improvement on the knowledge of the state of art and enhancement of the competences of the medical staff trained.

DOI: 10.4018/978-1-6684-8427-2.ch001

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 European Initiatives for Support and Counselling of Victims of Hate Crimes

BACKGROUND This article refers to the SHELTER project, a European initiative implemented between 2018 and 2021. The Directorate-General of Justice of the European Commission funded this project within the Rights, Equality and Citizenship program (2014/2020). The project “Support and advice through the health system for hate crime victims”, or SHELTER by its acronym, lasted 25 months and was implemented in 4 European countries: Cyprus, Hungary, Malta and Spain, by a consortium conformed of different universities and non-governmental organisations (NGO’s). It drew on previous initiatives already developed by the consortium of participating organisations, expanding the scope and impact of their previous actions, generating new synergies between the partners and giving rise to new lessons learned. SHELTER project delved into the important European problem of hate crimes, incorporating new social agents to prevent them and protect their victims. The legal and theoretical framework in which the project was conceived is described below (Díaz et. al., 2020). The Organization for Security and Cooperation in Europe - Office for Democratic Institutions and Human Rights (OSCE – ODHIR) defined hate crimes as crimes’ motivated by bias or prejudice towards a group of people’ (OSCE-ODHIR, 2009). The OSCE recognised such intolerance manifestations as one of the most prevalent manifestations in Europe today (OSCE-ODHIR, 2009). The European Network Against Racism (ENAR) 2013-2014 Shadow Report on Hate Crime found that this phenomenon was on the rise and still remains dangerously present in today’s Europe. Such is the case that the European Parliament seeks to expand the list of European crimes, including hate crimes and hate speech in said list, incorporating them into those already existing in Article 83 of the Treaty on the Functioning of the European Union (TFEU) (Díaz, 2018). As mentioned above, the phenomenon of hate, either in its discursive aspect (hate speech) or in its broader facet (that which includes physical attacks, threats, graffiti, and vandalism, among others), originates from two key aspects such as stereotypes and prejudices, which end up giving rise to discriminatory acts and in its most extreme radicalisation to the commission of the hate above crimes. Stereotypes make up simplified and rigid images, beliefs, and/or ideas regarding a person, group of people, or a set of things. The creation of these simple images about reality is part of the natural cognitive processes of human beings and fulfils important functions insofar as they have a functional and adaptive value by facilitating the understanding of the world in a simplified, orderly, coherent way, as well as, they make possible the prediction of events, the adjustment of the individual to social norms and contribute to the categorisation of the reality in which they live through the economy of thought. It is worth noting the identity value stereotypes have for the individual by pointing out the characteristics of otherness compared to those that the person perceives that describe and designate him. In this sense, stereotypes vary from positive, negative or neutral, depending on the qualities assigned to those static images configured in consciousness. Moreover, they are usually learned, transmitted and shared socially and culturally (Fernández-Poncela, 2011). For its part, prejudice is an axiomatic predisposition to accept or reject people because of their social characteristics, whether real or imagined” (Light et al., 1991). It can also be defined as “a hostile and distrustful attitude towards someone who belongs to a group, simply because they belong to it” (Gordon & Allport, 1954). In this sense, prejudice is a learned attitude based on the experiences the person has had throughout his life, especially during his childhood. Prejudices are configured based on stereotypes. In the end, prejudice is a distorted way of interpreting reality since, although it may have a real basis, it contains erroneous, exaggerated information or accidental generalisations caused by a previous experi2

 European Initiatives for Support and Counselling of Victims of Hate Crimes

ence or someone else’s. Prejudice is not just a statement of opinion or belief but an attitude that includes feelings such as contempt, disgust, or outright rejection. Finally, due to its connection with the two cognitive aspects indicated, it is worth speaking of the behavioural component of prejudice, such as discrimination, which, therefore, is described as its external manifestation (Simpson & Yinger, 2013). Social scientists have traditionally conceptualised discrimination as unfavourable unequal treatment of a subject or group due to prejudice. Discrimination occurs in different ways, fields, and levels (direct, indirect, among others). We place discrimination among those intolerant behaviours and hate crimes, promoted by rejecting the other’s identity. In any case, the central obstacle which lies at the heart of a coherent approach to tackling the phenomenon is, still nowadays, the under-reporting and under-recording of such crimes within EU member States (FRA 2013; OSCE-ODHIR, 2015). Under-reporting results from the victims’ hesitancy to report such cases due to the lack of trust in law enforcement agencies (FRA, 2013). Further, ENAR found that if a case is reported, the bias motivation is not adequately investigated, so the hate affiliated with this crime is essentially unreported. Under-recording refers to the lack of data collection in most Member States on such crimes. So, a three-fold problem can be distinguished. Firstly, hate crime is prevalent and rising in Europe, as the European Parliament itself states in its Study to support the preparation of the European Commission’s initiative to extend the list of EU crimes in Article 83 of the Treaty on the Functioning of the EU to hate speech and hate crime or the data provided by the Organization for Security and Cooperation in Europe - Office for Democratic Institutions and Human Rights (Bayer, 2021; OSCE-ODHIR, 2009; 2014). However, although these reports collect official data from public administrations and non-governmental sources such as NGOs, in many cases, they do not include all incidents and hate crimes in the participating countries because, as indicated above, many are not reported (Arroyo-López et al., 2021). Secondly, victims are often hesitant to report hate crimes, and if victims report the crime, the majority of Member States do not adequately examine the bias element, as stated in the report hate crime recording and data collection practice across the EU (European Union Agency for Fundamental Rights, 2019). Thirdly, Member States, despite differences between them, are not adequately, and neither are recording hate crimes standardised, as stated by the European Union Agency for Fundamental Rights (FRA) in its report Encouraging hate crime reporting ― the role of law enforcement and other authorities (European Union Agency for Fundamental Rights, 2021). As a result, victims of hate crimes are in a dire position when seeking justice, as they remain invisible and have no adequate protection, support or treatment. All this, also taking into account that since 2012, there has been a directive that specifically takes into account hate crimes when dealing with crime victims and their rights. This is the case of directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012 establishing minimum standards on the rights, support and protection of victims of crime and replacing Council Framework Decision 2001/220/JHA (European Union Agency for Fundamental Rights, 2012). This directive notes that individual assessments of victims should consider whether he or she was a victim of a hate crime (Aguilar et al., 2015; Blázquez, 2013). Thus, there is a clear need to facilitate the reporting of hate crimes which can be achieved by improving the reporting channels to enforcement bodies. Since victims demonstrate a lack of trust in enforcement bodies, partners assured that other social agents are needed to increase the protection and support victims’ framework. Considering this, healthcare services constitute a practical framework for increasing hate crime reporting since medical attention is often sought immediately after the event. Also, a medical environment is effective when seeking to ensure the victim’s best interest. As such, doctors and nurses, 3

 European Initiatives for Support and Counselling of Victims of Hate Crimes

considering that these two profiles are present of the whole sanitary system: emergencies located at the health care centers and primary care, were thought to be active agents in hate crimes victims’ protection and support (Gil-Borrelli et al., 2020). Therefore, the consortium considers that the training of those agents was entitled to identify victims of hate crimes and subsequently inform the victims of their rights about the justice system and psychosocial support. (When the project was designed, inadequate training existed for such staff to identify hate crime victims in partner countries. Besides, to ensure enhanced support to victims, other key social agents, such as psychologists and social workers, who work within health services, can also identify and support victims from the phase of identification and throughout the reporting procedure (Gil-Borrelli et al., 2018). It was considered that this approach would enable adequate implementation of Directive 2012/29EU, which establishes minimum standards on the rights, support and protection of victims of crime, ensuring that victims receive appropriate support to facilitate their recovery and are provided with sufficient access to justice (Ypma et al., 2021). Additionally, taking into account that through coordination and improvement of the integration of relationships, as well as a cross-sectoral approach to the issue of hate crime reporting and the support of hate crime victims, it is possible to contribute to the well-being of the victims, it is considered necessary to establish strong networks had to be established between health care services, law enforcement bodies and relevant NGOs and community groups to facilitate the reporting of hate crime and support of victims (Iganski, 2008). Besides, it was essential to involve Health institutions in adapting their assistance and attention to victims to the characteristics of hate crime effects (by protocols, new reporting tools). The partners recognised that some victims might hesitate to report a crime for several reasons, such as their undocumented status. To rectify loopholes that may arise from this, the partners emphasised incorporating NGOs which work directly with communities which were potential and/or actual victims of hate crime to be able to have frameworks of support and information for victims who may not want to go to official authorities (Pulido-Fuentes et al.,2021; Rodríguez-Arenas et al., 2018). Concerning under-recording, when the project was planned, there was an issue of insufficient recording of hate crimes in many EU countries. This contributed to rendering victims invisible without justice and support. The consortium wanted to contribute to eradicating this gap by taking into advance other outputs and tools created under other European initiatives, like informing victims about App for reporting hate crimes developed during the implementation of the project CONTACT (Creating an Online Network, monitoring Team and phone App to Counter hate crime Tactics), a two-year initiative (2015 – 2017) in which some of the partners took part. The project idea was led by Universidad de Castilla La Mancha and embellished with the collaboration of several organisations, which are members of the European Network Against Racism (ENAR), which, in turn, supported this process. The idea emanated from the experience and the contacts of the already finished project CONTACT that was granted under the previous Justice Call (JUST/2014/ACTION GRANTS) and the networking at the ENAR platform itself. As such, some organisations and their staff were already involved in CONTACT. That proposal addressed the main aim to prevent and combat racism, Xenophobia and other forms of intolerance of the call and, complementary to the CONTACT initiative that was focused on an analysis of online hate speech, reporting of hate speech through an App, training of enforcement bodies and raising awareness of other collectives (media, youth). In this respect, previous achievements and experiences in this emerging field of Study were taken as a starting point for the SHELTER project. It was a commitment to expand knowledge and implement strategies to

4

 European Initiatives for Support and Counselling of Victims of Hate Crimes

deal with hate crimes and their consequences on victims. Through its development, the priorities of the call were achieved, carrying out activities related to: 1. 2. 3. 4. 5.

Development of tools and practices to improve responses to these phenomena. Supporting victims of hate crime and hate speech and addressing the issue of underreporting. Achieving strengthened cooperation between key actors. Capacity building and training activities. Dissemination and awareness-raising activities.

The priorities were met by placing healthcare services at the epicentre of reporting hate crimes and supporting hate crime victims to fight racism, Xenophobia and any other form of intolerance. It must be noted that the proposal was based on the premise that reporting hate crimes and ensuring victims’ access to justice empowers victims of such crimes. Thus, this project sought to increase the health services’ capacity, particularly their staff such as doctors and nurses, to identify and inform victims of their rights. It also increased the capacity of other relevant actors, such as psychologists and social workers working within the field of health care services, to partake in the identification of such victims to provide psychosocial assistance throughout the process of identification and reporting, thereby facilitating the reporting process but also ensuring high-quality support services for victims. Since the capacity above was intended to be increased through training activities and the creation of relevant resources, the project would increase awareness about the victims’ rights among healthcare staff, victims and the general public. Moreover, the project would also establish a network of health institutions that incorporate these new practices regarding victims of hate crimes that connect them with victims and other agents law enforcement authorities, relevant NGOs and grass root groups to ensure a cross-sectorial approach to fulfil the project objectives. The innovative aspect of this project was incorporating health care services as a central social agent in the ambit of facilitating the reporting of hate crime (thereby, inter alia, empowering victims) and visà-vis supporting victims. In this sense, the partnership in each country had worked (CONTACT project) or was linked (Faculty of Law Enforcement and Police Science) to enforcement bodies to ensure that interrelation and intersectional training between health services and enforcement bodies could be created. As noted above, the immediacy of health care services (following a hate crime) and the particular environment of a health care service render this framework particularly suitable for the objectives pursued. In this sense, the workstreams are designed in such a manner to improve the victim’s access to information, resources and networks that facilitate the reporting of hate crimes as well as specialised support assistance and protection throughout the process, from the moment of identification. This methodology emanated from the premise that the suitability of health services in conjunction with receiving necessary information as well as psychosocial support immediately after the occurrence of the crime and throughout the process could encourage victims to report the crime, thereby empowering them through accessing justice and, at the same time, improving the psychosocial support available for victims. The innovation placed healthcare services at the front line of hate crime reporting and victim support. In light of the above, this project deepens into the area of hate crime by empowering victims by informing them of their rights and facilitating their access to justice, all within the ambit of effective psychosocial support, with a system that places the victim’s needs first.

5

 European Initiatives for Support and Counselling of Victims of Hate Crimes

PROJECT OBJECTIVES SHELTER’s main objective was improving the protection of victims of hate crimes and their access to resources and networks that facilitate the report, assistance, and specialised support. Attention at health centres was intended to be improved based on the development of specific protocols, the training of the staff, the information available for victims at their healthcare facilities and the promotion of the communication and networking of the health institutions with other social agents that surround the victims and the vulnerable communities. To achieve that general objective, the consortium planned four specific objectives that would contribute to reaching that main aim: • • • •

Tackling the underreporting data within the health system concerning aggression and violence committed under the approach of hate crime. Strengthening the medical and psychosocial support to victims of hate crime in the health system. Facilitating victims’ access to protection, assistance, and specialised support resources. Incorporating the health system institutions into an international network for supporting victims of hate crime.

METHODOLOGY The methodology of the project was selected based on the objectives mentioned above. In this sense, the project should strengthen the professional skills and the awareness of the staff at hospitals and healthcare centres to improve attention and report cases of hate crimes. To this end, creating new tools and protocols that contribute to the better assistance of hate crime victims and implementing training would allow professionals to be better prepared to support and advise possible victims of hate aggression. It was also considered that these trainings should be part of the career training. That was the reason for the relevant role of the universities in the project, as they are the main education provider for future doctors, nurses, social workers and psychologists. As part of the project, training kits were designed to disseminate and openly offered to any institution/professional. The partnership considered that the availability of information and dissemination tools was necessary to achieve the objectives. These tools should include information about hate crime characteristics, how to proceed for and with victims, and information about protection and support institutions and organisations in the country. These tools were expected to encourage victims to report and talk about their experiences effortlessly. This information was planned to be available at the hospitals and other bodies or through their staff performance. Besides, some raising awareness events and events were developed mainly targeting youngsters. Finally, it was planned that the information must be available for actual victims and potential or threatened communities that face discrimination, as well as the general public. As part of the approach, it was considered essential to establish ways of collaboration and synergies between the different agents surrounding current and potential victims of hate incidents (hospitals, victims of hate supporting networks, communities, public institutions, and enforcement bodies) at the national level, but also the European context. From this collaboration was expected that standard operating procedures (SOP) would be lobbied to allow better reporting and communication between 6

 European Initiatives for Support and Counselling of Victims of Hate Crimes

enforcement bodies, medical practitioners, supporting networks and victims. Furthermore, to maximise the impact, an International Network around a digital platform and a certificate were planned to be created to allow institutions to adhere to this standard work (incorporating the protocols, among others) The consortium conformed by a convergence of actors increased the impact of implementing the action at grass root level but with national repercussions as an innovative approach. Starting from this methodological approach, the project was articulated in 5 complementary Workstreams. These were consecutive and correlative, so it was necessary to finish one of them to proceed with the next one (except work package 1): 1. Workstream of management and coordination of the project. It referred to every communication and monitoring activity that allows the proper development of the other workstreams in the project. 2. Workstream related to the research, collection and analysis of data from the Health System. Regarding the other workstreams, the complementarities were related to the connections we can make between them. Before addressing staff training at health care centres, it was necessary to detect how the state of the art was. That was the reason for the importance of designing and developing proper research because the team could obtain relevant information to help design the following training in each country. There were some tasks and activities under this workstream: a. Establishing the analysis levels: approach of institutions, internal procedures, perceptions and knowledge of staff and victims of hate crime regarding these bias crimes. b. Targeting audience dimension in each country c. Designing and sharing among the consortium the procedure for collecting data d. Collecting data (interviews, questionnaires) and desktop research on the state of the art. e. Analysing and evaluation of data. f. Determining conclusions. g. Elaborating reports with the main conclusions and recommendations 3. Workstream related to the training, empowering, and exchanging of best practices for staff. Based on the data, conclusions and recommendations obtained in the previous wokstream (research), each national training for health institutions staff was designed. The common contents, conceptualisations and methodological training approaches were designed based on the European Union legal framework, recommendations, and other recognised educational and training materials made by European institutions (Council of Europe and FRA). There were developed two different pieces of training based on the target beneficiaries’ profiles and, therefore, their own educational needs: a. Training for that actual professionals that were working at healthcare institutions b. Training for students at the end of their medical and nursery career would be an intro future for future professionals. The most important point was the inclusion of the interdisciplinary approach. In this sense, based on the precious experience in other international projects on hate crimes and hate speech, the consortium decided to include, as part of the training contents, the enforcement bodies’ experiences. Police representatives took part in the training, not only to show how their experience was with dealing with hate crimes victims and the difficulties in detecting and reporting, but also making possible to generate ideas and links for possible connections that can help both bodies to improve the protection of victims better. In this sense, links with enforcement bodies were already consolidated inside the partnership, as some of the partners had built links already in previous experiences (such as the CONTACT project) 7

 European Initiatives for Support and Counselling of Victims of Hate Crimes

or were part of the enforcement bodies’ training (Faculty of Law Enforcement and Police Science, was one of the main partners). Besides, this training had an interactive and participative approach to enable medical practitioners, social workers and psychologists to participate in their training actively. The training also included an experiential approach and victims’ representation during the working sessions to sensitise trainees and connect theoretical contents with real situations. There were also multimedia tools (video, audio, and interactive power points). Individual evaluations accompanied all the training to receive feedback from the trainees. 4. Workstream related to the dissemination, communication and awareness raising. It was planned to design several activities and conferences as well as different tools to reach victims of hate crimes first and then the general public and raise awareness regarding hate crimes, hate speech and the impact on the victims and vulnerable communities. They were focused on learning what hate crimes are and critic attitudes against discrimination and co-responsibility in promoting diversity and victim support and protection. This work was done through face-to-face meetings and public discussion, with an intersectoral and interdisciplinary approach with relevant stakeholders. In this sense, the dissemination activities were designed to find new ways for the cooperation between social agents, to put in the public agenda the inter-institutional and NGO collaboration to improve rights and protection of victims of hate crime and joint impulse protocols for assisting victims. 5. Workstream related to networking for supporting victims. This work package included the development of a certificate of adherence to “Stop Hate Damages”. It was based on creating international social awareness of health institutions and real support to them by developing an International Network, implemented by an online platform and public certificate of adherence “Stop Hate Damages” developed by the partners. This certificate should be given to those health bodies committed to the objectives of the project and, therefore, that were interested and implemented the training with their health staff in the detection and treatment of victims of hate crimes, applying the training kit for detecting hate crimes and assist victims, both in informing the victim and the procedures of medical and social care by health professionals, and notifying the Platform International Network of Health Institutions That Victims of Hate Crime support for public complaint. Furthermore, the platform commits: to developing and delivering the training kit for the training of health and social staff of the adhere institution; Professional advice by the network of processes for both the institutions and the victims; transfer and dissemination of the complaint by the platform International Network of Health Institutions That support Victims of Hate Crime; and delivery of the Membership certificate “Stop Hate Damages”. The workstreams were designed to improve the victim’s access to information, resources and networks that facilitate the reporting of hate crimes and specialised support assistance and protection throughout the process, from the moment of identification. This methodology emanated from the premise that the suitability of health services in conjunction with receiving necessary information as well as psychosocial support immediately after the occurrence of the crime and throughout the process could encourage victims to report the crime, thereby empowering them through accessing justice and, at the same time, improving the psychosocial support available for victims. The innovation placed healthcare services at 8

 European Initiatives for Support and Counselling of Victims of Hate Crimes

the front line of hate crime reporting and victim support. In light of the above, this project contributed to the area of hate crime by empowering victims by informing them of their rights and facilitating their access to justice, all within the ambit of effective psychosocial support, with a system that places the victim’s needs first.

RESULTS The project results were a consequence and were linked to the objectives set as a starting point for the project. In this sense, as a result of these, eight results were proposed: determined state-of-the-art in relation to hate crimes within the framework of the health system of the participating countries; the training of medical and social staff and students in regard to hate crimes with intersectional approach, the creation of different training materials and tools related to hate crimes for health personnel and students in the health field; the availability of information for victims of hate crimes on how to recognise and report hate crimes in the space of health institutions; raising awareness of the general public in regard to hate crimes and the Rights of the victims of such crimes; the construction of collaboration networks between different social and institutional agents (hospitals, health centers, enforcement bodies, NGOs, communities...); the creation and provision of tools to detect hate crimes when the victims reach the health system and the dissemination of the project among stakeholders and the general public. It was expected as a general result that the protection of victims of hate crimes and their access to resources and networks that facilitate the report, assistance and specialised support would be improved. The consortium achieved by this initiative that the attention for victims of hate crimes at health centres improved, considering that the project contributed to the development of specific protocols for attending and caring for this specific kind of victims of crime, through the training of the active staff at the own hospitals or primary care centres, by providing with information at the own health centres available for victims and by improving the communication and networking of the health institutions with other agents. In this sense, four protocols for preventing hate crime and intervention and detection by healthcare professionals were developed at the end of the project. Likewise, the project would contribute to enlarging the vision of hate crimes, their repercussions and reporting. Data were collected about unreported hate crimes that reached the health system, as well the research team deepened the perception of victims and staff at healthcare centres about attention, reporting and referral of hate crimes to enforcement bodies and other supporting agents (NGOs, lawyers, among others). The research allowed the incorporation of new aspects and complemented and updated state of the art regarding hate crimes and their victims. By the end of the project, the consortium mapped the Health systems of the participant countries, and after the research, the analysis and conclusions were published in national reports, one per country, and a common Conclusions report was written and published. For its part, one of the achievements of the project was the strengthening of the staff’s professional skills at hospitals and healthcare centers by the raising awareness actions implemented but overall, based on the activities and new tools created for such staff. In this sense, a specific training kit was designed to improve the detection, attention, and caring of victims of hate crimes and reporting them to enforcement bodies. The consortium printed more than 2,500 copies of it. Additionally, the team developed different pieces of training that allow professionals to be better prepared to support and advise possible victims of hate aggression. This training was focused on doctors and nurses, not only at the emergency sections at hospitals but at the primary care centres and the emergency sections that assist victims in the very place 9

 European Initiatives for Support and Counselling of Victims of Hate Crimes

where the hate crime is committed. There were implemented 27 training actions with health care professionals and students of medicine and nursery, reaching 384 people in total in the participant countries. The consideration that training should be a fundamental pillar to achieve the project’s achievements and that it should be focused so that it could serve to consolidate advances in future generations of professionals, both tools and protocols, as well as basic training, were developed collaboratively with teaching staff from universities, specifically teachers of careers related to community nursing. The training kits were disseminated and offered to all related institutions/professionals in this field. The consortium made all the information and dissemination tools available about hate crime characteristics, how to proceed with victims, protection and support institutions and organisations, among others. The idea was to make it easy for victims to report and talk about their experiences with healthcare professionals. This information was oriented to actual victims and potential or threatened communities facing discrimination. This information was available at the hospitals, especially through their staff performance, as they were trained in such a sense. Besides, 23 raising awareness events and events were implemented with 583 participants. There were established collaboration ways and synergies between the agents surrounding actual and potential victims of hate incidents (hospitals, Victim of Hate supporting networks, communities, public institutions and enforcement bodies) at the national level and in the European context as a result of this collaboration were designed standard operating procedures (SOP) to allow better report and communication between enforcement bodies, medical practitioners, supporting networks and victims. By the end of the project, four protocols were published. Furthermore, to maximise impact, an International Network was created around a digital platform, as well as a certificate, to offer the possibility to adhere to those healthcare institutions committed to the actions offered in the project (incorporating the protocols, training their staff). The certificate of adherence was called “Stop Hate Damages” and was only awarded after the consortium team did an evaluation process. One of the key elements of the project was the cooperation and synergies established with public institutions in public health matters. Especially in Spain and Hungary, the public authorities were very interested in the project, which led to the signing of collaborations with them. It was a paradigmatic case in Spain since the Health Service of Castilla-La Mancha (SESCAM) and the Community of Madrid Primary Care Management were highly committed to the project. The training was included in the official study plan of both institutions for the staff already working in the primary care centres and the emergencies. By making the final products and results of the SHELTER project freely available and downloadable on the website of the project (https://stophatedamages.eu/en/), especially the proposed protocol for the care of victims of hate crimes, “Health Care Guidelines against Hate Violence”, and the “Stop Hate Damages” membership certificate, the consortium made them available to any social and health institutions, and especially to NGOs and civil society organisations working on human rights and the protection of victims of hate crimes. The project was also the subject of media attention, including some interviews and in-depth articles. The team was aware that a greater impact of the project could be achieved through circulation and reception by the maximum number of people of audiovisual productions (videos). Three videos were produced, being available on the SHELTER project website. That way, the team sought to facilitate wider dissemination through the Internet, especially social networks and mobile phone applications.

10

 European Initiatives for Support and Counselling of Victims of Hate Crimes

DISCUSSION AND CONCLUSION SHELTER was an innovative project that made it possible to broaden the focus of the intervention with victims of hate crimes. On the one hand, through the incorporation of key healthcare agents in detecting and identifying victims, such as health professionals. This group is a crucial agent when it comes to detecting the identification of people who have been victims of hate aggression, as has been claimed by the social organisations that serve these groups, but to date, the actions in this sense at the European level have been very shallow. The SHELTER project deepened the knowledge and perceptions of health professionals, inquired into their training needs regarding the subject at hand and offered training, protocols and tools better to address the care of this especially vulnerable group. The investigation made it possible to consider not only those health professionals who are part of the emergency services located in hospitals but also detected the need also to incorporate primary care doctors and nurses. These toilets receive people who, in some instances, were unaware of having received an attack motivated by hate until these professionals investigated the causes of their physical injuries, emotional damage or psychological situations. Likewise, emergency services professionals were also incorporated by identifying them as those who assisted in the first place, even before the police services, potential victims. This last group comprised those who were at the emergency telephone number and the health professionals who came by ambulance to provide primary care at the very place where the violence had occurred. These multi-agent initiatives, in which the profiles of the partners are diverse and complementary, that seek synergies with key social and institutional agents, contribute with a holistic approach to protecting victims of hate crimes, to improve the keys to their detection and accompaniment and their assistance. From an approach to the issue from a broad prism with interconnected and solid actions, facing the issue starting from research, training and awareness to dissemination, all possible facets of the issue are covered to achieve objectives such as those raised SHELTER.

REFERENCES Aguilar, M. A., Gómez, V., Marquina, M., de Rosa, M., & Tamarit, J. M. (2015). Manual práctico para la investigación y enjuiciamiento de delitos de odio. Generalitat de Catalunya. https://tinyurl.com/2hbyd9w6 Arroyo-López, C., Moreno-López, R., & Flores-Martos, J. A. (2021). Dignity, equality, and human rights. En Moreno-López., R & Arroyo-López., C (Eds.). Support and advice through health system for hate crimes victims: a socio-sanitary approach (9-20). Tirant Lo Blanch. Bayer, J. (2021). High-impact hate speech by persons of authority: A lower threshold needed? Hungarian Journal of Legal Studies, 61(3), 269–284. doi:10.1556/2052.2020.00003 Blázquez, M. D. (2013). La Directiva 2012/29/UE ¿Un paso adelante en materia de protección a las víctimas en la Unión Europea? Revista de Derecho Comunitario Europeo, 17(46), 897–934. Díaz, E., Flores, J. A., Árias, E., Arroyo, C., Fernández, P., Hipólito, N., & Pulido, M. (2020). Support and advice through health system for hate crimes victims. Proyecto SHELTER, Spanish Nacional Report, Llere. https://stophatedamages.eu/wp-content/uploads/2021/01/Spanis h-National-Report-DEF-11.01.21.pdf

11

 European Initiatives for Support and Counselling of Victims of Hate Crimes

Díaz, J. A. (2018). Informe de delimitación conceptual en materia de delitos de odio. Observatorio Español del Racismo y la Xenofobia. https://tinyurl.com/spny398s European Union Agency for Fundamental Rights. (2012). Making hate crime visible in the European Union: Acknowledging victims’ rights, Luxembourg, Publications Office; FRA European Union Agency for Fundamental Rights. (2019). Informe sobre los Derechos Fundamentales 2019. Luxembourg, Publications Office, FRA. FRA European Union Agency for Fundamental Rights. (2021). Encouraging hate crime reporting — The role of law enforcement and other authorities. Luxembourg, Publications Office; FRA. Fernández-Poncela, A. M. (2011). Prejuicios y estereotipos. Refranes, chistes y acertijos, reproductores y transgresores. Antropología experimental, (11) 17-328. http://revista.ujaen.es/huesped/rae/ articulos2011/22fernandez11.pdf FRA. (2013). Opinion of the european union agency for fundamental rights on the framework decision on racism and Xenophobia—with special attention to the rights of victims of crime. European Union Agency for Fundamental Rights document. http://fra.europa.eu/sites/default/files/fra-opinion-2-2013framework-decision-racism-xenophobia_en.pdf Gil-Borrelli, C.C, Martín-Ríos, M.D., López-Corcuera, L., Reche-Martínez, B., Torres-Santos Olmo, R., Muriel-Pati, E., & Rodríguez-Arenas, M.A. (2020). Elaboración de un cuestionario de detección de casos de violencia de odio en urgencias hospitalarias. Gac Sanit, 34(2). doi:10.1016/j.gaceta.2019.01.006 Gil-Borrelli, C. C., Martín-Ríos, M. D., & Rodríguez-Arenas, M. Á. (2018). Propuesta de actuación para la detección y la atención a víctimas de violencia de odio para profesionales de la salud. Medicina Clínica, 150(4), 155–159. doi:10.1016/j.medcli.2017.06.017 Gordon, W., & Allport, G. (1954). The Nature of Prejudice. AddisonWesley Publishing Company. https:// books.google.com/books? id=u94XUyRuDl4C Iganski, P. (2008). Hate crime and the city. Policy Press. Light, D., Keller, S., & Calhoun, C. (1991). Sociología. McGraw-Hill. Oficina de Instituciones Democráticas y Derechos Humanos de la OSCE. (2009). Hate Crime Laws. A Practical Guide. OSCE. https://www.osce.org/files/f/documents/3/e/36426.pdf Oficina de Instituciones Democráticas y Derechos Humanos de la OSCE. (2014). Prosecuting Hate Crimes: A Practical Guide. OSCE. https://tinyurl.com/nckaayk2 Oficina de Instituciones Democráticas y Derechos Humanos de la OSCE. (2015). Understanding Hate Crimes A Handbook for Ukraine. OSCE. https://tinyurl.com/ye2pama3 Pulido, M. P., Arroyo, C., Flores, J. A., & Ytarte, R. M. (2021). Health Care Guidelines against Hate Violence in Support and advice through health system for hate crimes victims: a socio-sanitary approach. Tirant Humanidades.

12

 European Initiatives for Support and Counselling of Victims of Hate Crimes

Rodríguez-Arenas, M. A., Martín-Ríos, M. D., & Gil-Borrelli, C. (2018). Intervenciones en salud pública contra la violencia de odio. Gaceta Sanitaria, 32(2), 114–116. doi:10.1016/j.gaceta.2017.10.013 PMID:29370937 Simpson, G. E., & Yinger, J. M. (2013). Racial and cultural minorities: An analysis of prejudice and discrimination. Springer Science & Business Media. Ypma, P., Drevon, C., Fulcher, C., Gascon, O., & Brown, K. (2021). Study to support the preparation of the European Commission’s initiative to extend the list of EU crimes in Article 83 of the Treaty on the Functioning of the EU to hate speech and hate crime. European commission. https://op.europa.eu/en/ publication-detail/-/publication/f866de4e-57de-11ec-91ac-01aa75ed71a1/language-en

KEY TERMS AND DEFINITIONS Bias: “The action of unfairly supporting or opposing a particular person or thing because of allowing personal opinions to influence your judgment.” (Cambridge Dictionary) Hate Crimes: “Crimes motivated by bias or a prejudice towards a group of people.” (COE) Hate Speech: “All types of expression that incite, promote, spread or justify violence, hatred or discrimination against a person or group of persons, or that denigrates them, because of their real or attributed personal characteristics or status such as “race”, colour, language, religion, nationality, national or ethnic origin, age, disability, sex, gender identity and sexual orientation.” (COE) NGOs (Non-Governmental Organizations): A non-profit organisation that operates independently of any government, typically one whose purpose is to address a social or political issue. Protocol: “A protocol is the rules to be followed when doing a scientific study or an exact method for giving medical treatment.” (Cambridge Dictionary)

13

14

Chapter 2

Creating an Online Network, Monitoring Team and Apps to Counter Hate Speech, and Hate Crime Tactics in Europe Roberto Moreno López https://orcid.org/0000-0002-6238-9440 Universidad de Castilla-La Mancha, Spain Fabienne Baider https://orcid.org/0000-0002-7548-7680 University of Cyprus, Cyprus

ABSTRACT In this chapter, the authors present a European initiative for the creation of an online network, monitoring team, and apps to counter hate crime tactics. The project, funded by the Directorate-General of Justice of the European Commission, aims for the promotion of cooperation on reporting and monitoring online hate speech within European countries with a synergy between strong NGOs and universities. The project was based in four axes of correlative actions, that went from the research to the design and implementation of trainings with stakeholders (enforcement bodies and youngsters), the creation of multiplatform tools (web and app), and specific efforts on dissemination and raising awareness with the most vulnerable communities and the general public. The results showed an improvement of the state of art on the situation in the participant countries, as well as real changes in the area of combating hate crime and on-line hate speech in some countries belonging to the consortium.

DOI: 10.4018/978-1-6684-8427-2.ch002

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Online Network to Counter Hate Speech, Hate Crimes in Europe

BACKGROUND In a global, immediate, and participatory context, with its own communicative idiosyncrasy, in the last years, it has been argued to become a space for the expression and dissemination of intolerance (Cabo Isasi & Juanatey, 2016). A kind of intolerance in which determined individuals with different profiles are rejected, and their dignity is stripped away by denying and attacking their identity. Moreover, it is well known that such intolerance on the Internet has a negative impact not only on the group or individual targeted by the intolerant but also on those activists who defend freedom, tolerance, and non-discrimination (European Commission, 2016). While social media has begun to dominate the socio-political landscape in almost every corner of the world, an increment of racist acts, with old and new characteristics, is taking place on these social platforms (Matamoros-Fernandez & Farkas, 2021). Racist discourse is prevalent on social media, including covert tactics such as weaponising memes and using fake identities to incite racist hatred (Farkas et al., 2018; Lamerichs et al., 2018). It should be noted that, in this online framework, communication on the Internet and social networks acquires its specificities, among which we emphasise the experience of anonymity and the feeling of uninhibited virtuality (Barcelona City Council, 2017). Authors such as López-Berlanga & SánchezRomero (2019) highlight the importance of digital coexistence by expressing attitudes towards the virtual and digital worlds. Other authors, such as De Haro (2019), point out that social networks are forming more general independent networks, isolating them from other Internet users and creating safe spaces that catalyse different human behaviours and provide new exchanges and learning formats (Holcomb & Beal, 2010). As social media begins to dominate the socio-political landscape in almost every corner of the world, more and more racist acts, old and new, are taking place on these platforms (MatamorosFernandez & Farkas, 2021). Racist discourse is prevalent on social media, including covert tactics such as weaponising memes and using fake identities to incite racist hatred (Farkas et al., 2018; Lamerichs et al., 2018). While hate speech is prohibited by international law and regulatory policies based on respect for human beings, it is widespread and endangers values necessary for social cohesion. In some cases, hate speech can increase tensions and incite violence and can be targeted at a particular community or group of people or indiscriminate, making it a multidimensional problem that is difficult to define (Davidson et al., 2017). In Europe, as part of the global North, hate speech is seeping into public discourse, especially after the massive refugee crisis in 2015 (Ekman, 2019). In this sense, its real-life impact is also growing, as it can be a precursor and a trigger for hate crimes (Burnap & Williams, 2014). Many people have quickly realised that hate speech is a severe problem, especially with its presence on social networks, which generates more online conflicts between different people (Al Serhan & Elareshi, 2020). The distinction between hate speech and offensive speech must be clear, and definitions of legal terms help in this process. Mechanisms for monitoring and analysing abusive language have been established to identify offensive speech circulating in online media, as far as their technical possibilities allow. Following Olteanu et al. (2018), the spread of hate speech has attracted the interest of many researchers who initially investigated online content to help police and, after analysing the results, made it easier for statesmen, politicians, and policymakers to understand and find a solution. Social stereotypes fuel hate speech in offline and online lives and recent debates have begun to revolve around the provision of unruly free speech and, in some cases, rampant hate speech through digital technologies. This issue, and even social networks, have developed their services to detect and ban such rhetorical expressions (whether covert or overt (Ben-David & Fernandez, 2016; Pohjonen & Udupa, 2017). 15

 Online Network to Counter Hate Speech, Hate Crimes in Europe

This article refers to the C.O.N.T.A.C.T. project (D.G. Social Justice JUST/2014RRAC/AG/ HATE/6706), which was implemented between 2015 and 2017, and which is framed in a European context concerning hate speech that although hate crimes continue to be one of the main concerns in the member states of the European Union, has changed slightly in certain countries that make up the Union. For example, in Spain, there have been very relevant advances in legislation regulating aggravating circumstances related to hate crimes and speech and applying police protocols. In the Republic of Cyprus, it is noted that the concept of femicide has been recognised legally as a hate crime. The project framework was developed when the most recent reports on hate crimes within the European Union were deeply worrying. For instance, the E.C.R.I. report acknowledged an increase in hate crimes all over Europe (Council of Europe, 2014). Moreover, researchers and Non-Governmental Organizations noted how Web 2.0 contributed to the spread of hate globally. Indeed, the European Network Against Racism [E.N.A.R.] Shadow Report (2012) highlighted a rise in racist violence and discourse in social media and the Internet. The Europe Union had then acknowledged the issue by encouraging relevant initiatives in the previous years, such as Voxpol, Facing Facts, and the Light On’ project. Legal provisions (such as Law No. 927 in Greece) foreseen penalties for those inciting racist hate through the media, while the European Agency of Fundamental Rights defined within the Framework Decision on Racism and Xenophobia the following priorities: 1. 2. 3. 4.

The identification of hate speech and hate crime, The increasing use of the Internet as a tool of hate and propaganda, The under-reporting of hate speech and hate crimes, The rise of extremist groups and political parties in the E.U. (European Union Agency for the Fundamental Rights, 2013).

At that time, the ​E.U. Code of Conduct was still being developed to prevent and counter the spread of illegal hate speech online. The users and society had to wait until May 2016 for the Commission to agree with Facebook, Microsoft, Twitter and YouTube in signing a “Code of conduct on countering illegal hate speech online”. The C.O.N.T.A.C.T. project, led by the University of Cyprus, answered the above priorities by combining the European Court of Human Rights definition for hate speech (any speech which spreads, incites, promotes or justifies hatred founded on intolerance, including religious intolerance) and the L.H.R.M.I. hate speech definition: “the public dissemination of information (ideas, opinions, knowingly misrepresented facts) expressing contempt, inciting to hatred, discrimination, abuse, physical violence against a group of people or a member of that group because of his/her gender, sexual orientation, race, ethnicity, language, descent, social status, religion, beliefs or opinions.”. Each European country tackles the phenomenon differently because of a different context: indeed, common actions over borders are made possible and implementable when contextualising nationally. To make possible the application of proven mechanisms and to think of new solutions, C.O.N.T.A.C.T. combined complementary expertise: 1. Academics from various disciplines, especially in sociolinguistics, are working to understand the national context of hate speech and analyse the online data collected.

16

 Online Network to Counter Hate Speech, Hate Crimes in Europe

2. Experienced NGOs and stakeholders have a real social impact on creating a synergy with countries that have actively implemented mechanisms to monitor and report hate speech and those who need it the most. Moreover, C.O.N.T.A.C.T. provided powerful tools to get invaluable data for the NGOs to present their case to the authorities through its web platform and phone App. In brief, the project looked for the following: 1. Developing monitoring and reporting mechanisms of hate speech/ crime on new media for the general public (online report), 2. Training young new media users, media professionals, and police enforcement to gain awareness and/or develop ethical principles when dealing with hate speech/hate crime. 3. Encouraging the general public to report hate speech and hate crimes. 4. Disseminating scientific articles (based on the results of the project and for hate speech/hate crimes and development of standard solutions)

PROJECT OBJECTIVES As mentioned, the was granted by the Directorate-General of Justice of the European Commission within the Rights, Equality and Citizenship program (2014/2020). The project was deployed in 10 European countries: Cyprus, Greece, Poland, Italy, Denmark, Romania, Spain, Lithuania, Great Britain and Malta, with the participation of universities and social organisations. The project searched for creating an online network of social organisations and universities that could set up a team to monitor hate speech and hate crimes in various countries in the European context. Additionally, this initiative included the creation of an application for smartphones to facilitate the monitoring of hate speech and hate crimes that occur once the App was created for citizens and the institutions within the consortium of the project. This information collected with the App was to complete and contrast the official data collected by each country’s public institutions and contribute to better reporting by victims of hate crimes. The project’s general objective was to promote cooperation in reporting and monitoring online hate speech and hate crimes within European countries. To this end, four specific objectives were established: 1. 2. 3. 4.

Deepen the perceptions of young people. Classification of online speech. Training of young people and professionals. Dissemination of reporting tools to record hate speech and hate crimes.

The project results have been described in detail in the open-access book titled Online hate speech in the European Space, published by Springer (2017).

17

 Online Network to Counter Hate Speech, Hate Crimes in Europe

METHODOLOGY The methodology chosen to achieve the project’s objectives consisted of synergistic actions implemented collaboratively by consolidated NGOs and Universities known for their work against racism, xenophobia and intolerance, as well as with key stakeholders such as the Ombudsperson in some countries. Several consecutive work packages were established, so the actions implemented in each one gave way and served as the basis for the following work package and its activities. In this sense, the project included four packages: 1. Work package 1 included a focus on data analysis through desktop research, an analysis of hate speech on national news platforms, and a perceptual experiment, whose findings will be used to design training modules that make up package number 3. 2. Work package 2 focused on developing computational tools for data collection, especially the project’s website and the application development for reporting hate crimes. 3. Work package 3 consisted of training activities carried out within the framework of the project and included training actions with young people and other stakeholders (enforcement bodies, lawyers, among others). 4. Finally, work package 4 included all the dissemination actions carried out under the framework of the project, both at a national and a European level. The C.O.N.T.A.C.T. adopted the following complementary methodological approaches for each work package:

Methodology to Access Data The first step to be able to unify the data that would be collected on the web platform and in the phone App itself and thus allow the monitoring of hate crimes reported by citizens, a series of previous steps had to be taken (Assimakopoulos, et al., 2017). a) First, it was agreed to establish a standard definition of hate crimes for monitoring and datarecording purposes so that the definition could be unified among the partners and overcome the potential conceptual and legal barriers between the various frameworks. Laws of each country that is part of the consortium. The Organization for Security and Cooperation in Europe (Organisation for Security and Cooperation in Europe [OSCE], 2021) was taken as a reference, determining that a hate crime is a criminal offence committed with a biased motivation. This description incorporates two key elements. Firstly, the one that considers hate crimes to be an act that constitutes a criminal offence under ordinary criminal law. Moreover, it indicates that the offender intentionally chose a victim or target on which one or more protected characteristics converge. A protected characteristic is a characteristic shared by a group, such as race, religion, ethnicity, nationality or any other similar common factor, and this perceived characteristic is stereotyped, and the author’s discriminatory motivation converges on it (Fiscalía General del Estado, 2019). b) Secondly, a list of bias motivations on which to collect data was agreed upon and established. The list included those already reflected in most of the national legislation of the member countries and which takes as a reference the main bias motivations of the European Recommendations and Directives. In this case, the list included: race or ethnicity, religion or belief, sexual orientation, transgender identity, 18

 Online Network to Counter Hate Speech, Hate Crimes in Europe

and disability. By bias, motivations were understood as objective facts, circumstances or patterns related to a criminal act that, alone or in conjunction with other indicators, suggest that the offender’s actions were motivated in whole or in part by bias, prejudice and/or hostility. c) Thirdly, a final list was agreed upon that included a broad set of categories of criminal offences, which would be the ones on which the project monitoring team would collect data. As there was no consensus among the different countries of the European Union that were part of the project, it was decided that this list would help unify the compilation and comparison of the results. In this case, the list included: homicide, physical assault (with and without injury), harassment, public fear/alarm/distress, damage to property, serious desecrations, vandalism, threats and attacks on places of worship. The project specifically included the possibility of incorporating data referring to online hate on websites or social media. Therefore, both the website and the App incorporated an online form that would include at least the following fields: 1. 2. 3. 4. 5. 6.

Details on the incident (date, location, time). Type of criminal offence. Demographic information about the victim; about the perpetrator(s). Elements that could help prosecution (witnesses, vehicles, CCTV footage etc.). Bias motivation. Brief narrative highlighting the bias indicators.

Lastly, the research team agreed on the sources that would allow access to the data that would give rise to the project information. In the first place, data would be obtained from various sources: 1. 2. 3. 4. 5. 6.

Directly from victims of abuse. The police. Witnesses. Relatives or friends of victims. Mainstream media such as newspaper articles, T.V. or radio news items. Civil society organisations.

Based on the E.U. legislations, especially those related to privacy, and the previous experience of the universities and organisations, the consortium establishes a set of mechanisms to facilitate and enhance monitoring and reporting. In this sense, online reporting forms were used to ensure anonymity and free of charge for both victims and witnesses of possible hate incidents.

Methodology for the Organisation and Classification of Online Hate Speech This activity, which was carried out taking into account the national news websites, was deployed using various criteria that would help this organisation and classification of discourses: 1. On the one hand, the country where the online news media were published focused on the participating countries (Cyprus, Greece, Poland, Italy, Denmark, Romania, Spain, Lithuania, Great Britain and Malta). 19

 Online Network to Counter Hate Speech, Hate Crimes in Europe

2. On the other, classified speeches concerning the content and/or the target audience of the discriminatory speech: a. Hate speech against foreigners and immigrants (words searched: xenophobia, racism, immigrant, immigration, refugees, Roma, among others. b. Extremist Political Hate Speech (words searched: hate speech, hate crime). c. Sexual identity-based hate speech (homophobia, L.G.T.B., homosexual, gay, lesbian, transgender, misogyny). d. Religious belief (Islamism, Judaism). e. The typology of hate speech online. There were used classification of the speeches to identify if they were positive or negative comments using the criteria deployed by previous experiences (LightOn - http://www.lighton-project.eu/site/main/page/home or the P.R.O.X.I project - https://www.observatorioproxi.org/), which can be classified as generalisations, positive/ negative metaphors, positive/negative suggestions, insulting words, stereotypes, indirect positive /negative statements. f. Another remark used was identifying the media where the comments are identified. g. Finally, if the message incorporates derogatory, insulting terms as keywords, it could bias the qualitative analysis.

Methodology for Analysing Speech in the Online Media and by Young People During Surveys and Interviews Related to Their Perception of Hate Speech and Hate Crimes It was based on previous experiences of Discourse Mapping and Monitoring Projects using quantitative analysis as the methodology implemented by The Global Media Monitoring Project (G.M.M.P.) (https://whomakesthenews.org/gmmp). The discursive analysis was achieved by applying the theoretical framework described by Van Dijk (1990) or Wodak (2015). Finally, these approaches were used and adapted so that surveys and interviews related to the perception of hate speech and hate crimes among young people from different countries could be helpful.

The Methodology Used in the Design and Teaching of Training This methodology was based on the training actions that the F.R.A. and the ODIHR have been deploying to work on the contents of Human Rights and hate crimes they have already been deploying. Therefore, they included multi-agent, participatory approaches based on real cases, which allowed for building knowledge and working on the participants’ awareness. In this sense, three different complementary approaches were used: 1. Direct teaching approach through seminars. 2. Interactive approach, through the use of multimedia media. 3. Experiential approach, incorporating participatory activities and based on real national cases.

20

 Online Network to Counter Hate Speech, Hate Crimes in Europe

PROJECT ACTIONS Numerous actions were implemented under C.O.N.T.A.C.T. These were distributed in work packages with internal and external logic for better organisation. Each package integrated those activities directly related to each other, and, in turn, each work package was deployed consecutively. Therefore the products and outputs of one were interdependent on the previous.

Activities of Work Package 1: Data Collection and Analysis 1. Desktop analysis, which allowed the consortium of organisations that made up the project to, on the one hand, collect the definitions of hate crime and hate speech that was used in each country, according to their legislation; on the other, the relevant International Legal Provisions for each country context, and determine its link with national legislation, as well as a deepening of the national penal code for each country, also referring to hate crimes and speech. 2. A quantitative and qualitative analysis was carried out, where through the use of automated search tools (https://emm.newsbrief.eu/), an important consultation was carried out in the leading online news media whose headlines and text included keywords (see above) and the analysis of your comments to identify hate speech. The quantitative analysis collected the number of hits for the keywords in the online media included in the European automated tool for each country between April and June 2015 and December and February 2016. 3. For the quantitative analysis, a compilation of 10% of the articles with comments was made, which included three key themes of the project (hate crime, xenophobia and homophobia), also in 3 months. The comments were classified as positive, negative or neutral, following the methodology used in the LightOn project, identifying the discourse, and the context, among others. 4. These comments were subsequently used in designing the perceptual study with young people from all the consortium countries, and the training and awareness actions carried out within the training work package. 5. A Perceptual experiment was deployed, for which a questionnaire on the perception of communication and relationships on the Internet was prepared to focus specifically on the identification of hate speech, its presence in the communication environments of young people on the Internet and finally, the knowledge of these about it and the existing alternatives to it. The questionnaire had 23 questions, reaching 360 young people between 16 and 30. Likewise, 180 in-depth interviews were carried out with young people of the same age group and in which, based on the categories previously exposed for the questionnaire, the visions and perceptions of the young people were deepened based on their speeches. 6. Finally, taking advantage of this perceptual experiment’s ideas, recommendations and conclusions, training modules were prepared for the State Security Forces and Enforcement Bodies that would later be worked on in work package 3.

Activities of Work Package 2: Tools and Mechanisms Under this work package, online tools for monitoring hate crimes were deployed within a data collection framework that, as mentioned above, would allow official data collected by states to be compared and facilitate access to information on hate crimes and alternative reporting channels for victims. Without 21

 Online Network to Counter Hate Speech, Hate Crimes in Europe

forgetting that this website and the App, through the collection of data, social organisations could support state governments and the European Union in designing and implementing new policies and legal actions to prevent and fight against hate crimes. On the one hand, the website of the C.O.N.T.A.C.T. Project was designed and articulated. This web platform, which was developed in all the languages of the countries that made up the consortium (12 languages in total), included the following sections: 1. 2. 3. 4. 5.

About C.O.N.T.A.C.T, with information about the project. Report hate with a form to make a non-legal complaint online. Live data, where the data extracted from the web and the App, was collected in real-time. Legal framework, with information on European legislation on hate crimes and hate speech. The conference, where information was offered about the final conference and the different events of the project. 6. Outputs, with the possibility of downloading the documents and products arising from the project. 7. Other standard sections are common to other websites. The data protection policy section is related to Partner organisations and a contact section. On the other hand, another of the products that emerged from the project is the multilanguage application for mobile phones, which was developed for Android and iPhone. This App’s main objectives were to collect first-hand information (through victims, witnesses, or police officers) on crimes and hate speech. With this data collection, the project offers information that can facilitate the design of actions and legal plans that allow state governments and the European Union to be more effective in preventing and fighting hate crimes. In addition, this App, as was the case with the web, offered direct information to the victims so they could legally report before the state security forces and bodies. Among the functionalities included in the App, we can find (Figure 1, Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, Figure 8, Figure 9 and Figure 10): 1. 2. 3. 4. 5. 6. 7.

The possibility to record a video or take a picture(s). The possibility to send the file(s) above as a report. A GPS to locate where the event takes place. Add report title. Add report description. Add abuse type (categorise). Set the location of the report (the application will auto-centre the map on the current user location, but the user will have the possibility to choose the report location). 8. The application will have a draft report management (if you only take a picture through the application, for example, you should be able to complete the report later) and a sent reports view. 9. The application will have an account detail if the user wants to give his credentials. 10. The application will have an information section: about terms and abuse types of information, among others.

22

 Online Network to Counter Hate Speech, Hate Crimes in Europe

Figure 1. C.O.N.T.A.C.T. app opening screenshots (C.O.N.T.A.C.T. App)

Figure 2. C.O.N.T.A.C.T. app functionalities screenshots (C.O.N.T.A.C.T. App)

23

 Online Network to Counter Hate Speech, Hate Crimes in Europe

Figure 3. C.O.N.T.A.C.T. app functionalities (draft and profile) (C.O.N.T.A.C.T. App)

Figure 4. C.O.N.T.A.C.T. app functionalities (partner organisations information and data protection policy) (C.O.N.T.A.C.T. App)

Figure 5. C.O.N.T.A.C.T. app functionalities (reporting form screenshots 1) (C.O.N.T.A.C.T. App)

24

 Online Network to Counter Hate Speech, Hate Crimes in Europe

Figure 6. C.O.N.T.A.C.T. app functionalities (reporting form screenshots 2 (C.O.N.T.A.C.T. App)

Figure 7. C.O.N.T.A.C.T. app functionalities (reporting form screenshots 3) (C.O.N.T.A.C.T. App)

25

 Online Network to Counter Hate Speech, Hate Crimes in Europe

Figure 8. C.O.N.T.A.C.T. app functionalities (reporting form screenshots 4) (C.O.N.T.A.C.T. App)

Figure 9. C.O.N.T.A.C.T. app functionalities (reporting form screenshots 5) (C.O.N.T.A.C.T. App)

26

 Online Network to Counter Hate Speech, Hate Crimes in Europe

Figure 10. C.O.N.T.A.C.T. app functionalities (reporting form screenshots 6) (C.O.N.T.A.C.T. App)

Activities of Work Package 3: Training, Empowering, Exchanging Best Practices The objectives of the project, specified in this work package, are translated into three complementary axes that aim to make it easier for both police and legal agents and young people to: 1. To improve the recording and the reporting, for instance, a workshop on media to improve the reporting online complaint form to accelerate the recording. 2. To improve the handling of reporting and monitoring hate crime (workshop focused on law enforcement, information on rights and procedures to follow to report found on the Web platform, among others). 3. How to brighten the future (workshops on youth to sensitise them to the issue, workshops to police to encourage tolerance, perceptual experiment to understand motivations, strategy paper for recommendations to the authorities, among others). 4. The proposal combines developing and implementing mechanisms for monitoring/reporting (online) hate crimes while sharing best practices in educational seminars and other awareness-raising activities (W3/W4). 5. Among the contents that were worked on with the different participants were the following: 6. Conceptual clarifications and definitions of the following ideas: Immigration, asylum beneficiaries of international protection, examination systems asylum mixed migratory flows, residence status of foreign populations, Roma population, sexual orientation and gender identity, islamophobia, anti-Semitism, hate crimes, bias indicators etc. 7. Legal / Institutional framework best practices: Recognition, recording and addressing hate crimes. Redress mechanisms. National legal framework and legal challenges. European legislation, direc-

27

 Online Network to Counter Hate Speech, Hate Crimes in Europe

tives, case law, good practices. Material from ODIHR and FRA was used, along with the production of specialised educational material for the project’s needs. 8. Communication skills and prevention: Reaching out and establishing community trust, relationship and cooperation with vulnerable groups, development of communication skills, crisis management and intercultural mediation, stereotypes and discrimination, empowerment of professional experience. Desk and field research and the perceptual experiment were conducted before the training to fully assess the target groups’ educational needs. Defining training needs was a crucial stage in implementing the C.O.N.T.A.C.T project since successfully identifying such needs affected achieving its objectives and fulfilling the conditions and criteria mentioned above. The identification process resulted in defining clearly, a strictly delineated productive educational framework, which took into account the following: 1. 2. 3. 4. 5. 6.

The subject and the respective work culture. The socio-economic environment. The specific administrative structure of each service. The special educational and sociological characteristics of trainees. The particular cultural and social characteristics of trainees. As already mentioned, the production of the educational material was based on the FRA and ODIHR module and will also be adapted to the specific needs of each country (legal provisions for investigating racist motives, particularities as regards the trends of hate crimes, among others), in order to have the best possible result.

These work sessions were held in all the countries of the consortium, and as mentioned above, they were held with the police forces, youth trainers and educators, media professionals, and young people, and training seminars were held in some specific cases. Transversal training that incorporated different stakeholders.

RESULTS The results of the CONTACT project were as follows: 1. The public awareness and knowledge about existing monitoring and reporting mechanisms (including C.O.N.T.A.C.T tools) among the general public and vulnerable communities had empowered. They encouraged potential victims and witnesses to report hate-motivated crimes, incidents, and/ or complaints. The activities and outputs that allowed the project to achieve this result were distributing informative leaflets, using the Web platform, spreading the message methods, and raising awareness events. This resulted in enhanced knowledge and sensitisation of the general public about online hate speech and hate crime through awareness-raising activities such as events and public discussions/ workshops. 2. More precisely, the project improved the information of rights for vulnerable communities and for the general public; Increased awareness of potential issues affecting these rights with the forum on the Web platform; Improved knowledge about the existing mechanisms of monitoring and reporting 28

 Online Network to Counter Hate Speech, Hate Crimes in Europe

via tailored training educational sessions addressed to law enforcement authorities (police officers, judges, legal experts). 3. Sensitisation of young people and educators (schools, university students, trainers, academics through training sessions and workshops) about online hate speech and its consequent offline manifestations (i.e., hate-motivated incidents or hate crimes against an individual or a particular group). The project contributed to better preparation of the youngster to face similar incidents in virtual and real life and encouraged them to talk about such incidents and report them accordingly without being intimidated. 4. The creation of tools for monitoring and reporting encouraged victims of hate crime or online hate speech to easily and directly report their experiences; the multilingual tools accessible to non-English speakers made public sensitisation and reporting easier regarding these hate incidents. As far as research is concerned, C.O.N.T.A.C.T. allowed: 1. A common university elective online course focused on online hate speech had been developed in partnership with the universities within the consortium to actualise the synergies described above and sustain student awareness. 2. Improved state of the art on the situation in the participating countries. The identification of different forms of racism and hate speech as they unfold in different countries of Europe, from a cross-cultural perspective, gave new insights into the phenomena and will contribute to developing new, innovative methods to combat hate speech and crime as well as the publication of scientific research articles on the phenomenon. 3. It provided an innovative interdisciplinary approach to hate speech analysis and applied qualitative and quantitative research methods. 4. It reintroduced the academic debates on social justice in post-soviet studies such as Lithuania; and launched discussions and political debates on the effects of hate speech on society. 5. It classified different modes of hate speech production in social media and public discourses and finalised its typology. 6. It communicated research insights during major scientific events, discussion platforms, and peerreviewed publications.

DISCUSION AND CONCLUSION As far as pan-European results, we foresaw: Implementation of Union law instruments and policies thanks to the training of law enforcement officers and access to a free multilingual iPhone App. Enhancement of cooperation between Member States, improvement of cross-border cooperation, elaboration, and dissemination of best practices. Synergies between universities, institutes, NGOs, and GOs were enhanced/creation of synergies between NGOs in existing networks of human rights NGOs and researchers will favour inter-state exchanges and transfer of knowledge and know-how. Partners in the project actively convened grassroots groups to effect real change in combating hate crime and online hate speech within their own countries. 29

 Online Network to Counter Hate Speech, Hate Crimes in Europe

The project supported the work of different social actors nationally and developed a European quality of exchange. For instance, the researchers involved are also teachers in universities and thanks to disseminating conferences and seminars, they will allow an immediate impact on the younger generation. Moreover, the academics involved have an experience in anthropological fieldwork and knowledge of the latest development in social sciences regarding discrimination.

REFERENCES Al Serhan, F., & Elareshi, M. (2020). New media and hate speech: A study of university students in Jordan. Opción: Revista de Ciencias Humanas y Sociales, (26), 166–184. Assimakopoulos, S., Baider, F. H., & Millar, S. (2017). Online hate speech in the European Union: a discourse-analytic perspective. Springer Nature. doi:10.1007/978-3-319-72604-5 Ayuntamiento de Barcelona. (2017). Informe Discurso del Odio. Resumen Ejecutivo. Ayuntamiento de Barcelona. Ben-David, A., & Fernández, A. M. (2016). Hate speech and covert discrimination on social media: Monitoring the Facebook pages of extreme-right political parties in Spain. International Journal of Communication, 10, 1167–1193. Burnap, P., & Williams, M. L. (2014). Hate speech, machine classification and statistical modelling of information flows on Twitter: Interpretation and communication for policy decision making. Cardiff. https://orca.cardiff.ac.uk/id/eprint/65227/1/IPP2014-Burnap.pdf Cabo Isasi, A., & Juanatey, A. (2016). El discurso de odio en las redes sociales: un estado de la cuestión. Ayuntamiento de Barcelona. Council of Europe. (2014). Annual report on E.C.R.I.’s activities. C.R.I., 26. https://rm.coe.int/annualreport-on-ecri-s-activities-covering-the-period-from-1-january-/16808ae6d4 Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated hate speech detection and the problem of offensive language. In Proceedings of the International AAAI Conference on Web and Social Media (pp. 512-515). AAAI. https://arxiv.org/abs/1703.04009 10.1609/icwsm.v11i1.14955 De Haro, J. (2019). Redes sociales en Educación. Eduredes. http://eduredes.antoniogarrido.es/uploads/6/3/1/1/6311693/redes_sociales_educacion.pdf Ekman, M. (2019). Anti-immigration and racist discourse in social media. European Journal of Communication, 34(6), 606–618. journals.sagepub.com/home/ejc. doi:10.1177/0267323119886151 European Commission. (2016). Code of conduct on countering illegal hate speech online. EC. https:// ec.europa.eu/justice/fundamental-rights/files/hate_speech_code_of_conduct_en.pdf European Network Against Racism - E.N.A.R. (2012). 2011/12 Shadow Reports on racism in Europe. Author.

30

 Online Network to Counter Hate Speech, Hate Crimes in Europe

European Union Agency for the Fundamental Rights. (2013). F.R.A. Opinion – 02/2013 Framework Decision on Racism and Xenophobia. Author. Farkas, J., Schou, J., & Neumayer, C. (2018). Cloaked Facebook Pages: Exploring Fake Islamist Propaganda in Social Media. New Media & Society, 20(5), 1850–1867. doi:10.1177/1461444817707759 Fiscalía General del Estado. (2019). Circular 7/2019, de 14 de mayo, de la Fiscalía General del Estado, sobre pautas para interpretar los delitos de odio tipificados en el artículo 510 del Código Penal. Fiscalía General del Estado. Holcomb, L., & Beal, C. (2010). Capitalising on web 2.0 in the social studies context. TechTrends, 54(4), 28–32. doi:10.100711528-010-0417-0 Lamerichs, N., Dennis, M. C., Puerta, R. R., & Lange-Bohmer, A. (2018). Elite Male Bodies: The Circulation of Alt-Right Memes and the Framing of Politicians on Social Media. Participations, 15(1), 180–206. López-Berlanga, C., & Sánchez-Romero, C. (2019). La interacción y convivencia digital de los estudiantes en las redes sociales. Revista nacional e internacional de educación inclusiva, 12(2), 114-130. Matamoros-Fernández, A., & Farkas, J. (2021). Racism, hate speech, and social media: A systematic review and critique. Television & New Media, 22(2), 205–224. doi:10.1177/1527476420982230 Olteanu, A., Castillo, C., Boy, J., & Varshney, K. (2018). The effect of extremist violence on hateful speech online. In Proceedings of the international AAAI conference on web and social media (Vol. 12, No. 1). PKP. 10.1609/icwsm.v12i1.15040 Organisation for Security and Cooperation in Europe - OSCE. (2021). Los delitos de odio motivados por el racismo y la xenophobia. OSCE-ODIHR. Hate Crime Reporting. https://www.osce.org/es/odihr/502275 Pohjonen, M., & Udupa, S. (2017). Extreme speech online: An anthropological critique of hate speech debates. International Journal of Communication, 11, 19. Van Dijk, T. A. (1990). Social cognition and discourse. Handbook of language and social psychology, 163, 183. APA. Wodak, R. (2015). The politics of fear: What right-wing populist discourses mean. The Politics of Fear, 1-256.

ENDNOTES 3 4 1 2

http://www.lighton-project.eu/site/main/page/home https://www.observatorioproxi.org/ https://whomakesthenews.org/gmmp https://emm.newsbrief.eu/

31

32

Chapter 3

Using HurtLex and Best-Worst Scaling to Develop ERIS: A Lexicon for Offensive Language Detection

Valerio Basile https://orcid.org/0000-0001-8110-6832 University of Turin, Italy Stella Markantonatou https://orcid.org/0000-0002-9256-8300 Institute for Language and Speech Processing, Athena Research Center, Greece Vivian Stamou Institute for Language and Speech Processing, Athena Research Center, Greece Iakovi Alexiou Institute for Language and Speech Processing, Athena Research Center, Greece Georgia Apostolopoulou Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece

Vana Archonti Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Antonis Balas Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Eleni Koutli Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece Maria Panagiotopoulou https://orcid.org/0000-0003-4016-5486 Department of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece

ABSTRACT ERIS, a lexical resource of Modern Greek for offensive language detection, is the result of cleansing, enriching and assigning graded offensiveness values to the EL branch of HurtLex. ERIS contains 1148 entries and is openly available. Graded values were obtained with the Best-Worst Scaling method that was applied with the Litescale tool. Nouns and adjectives that have humans as a target were found to attract bigger offensiveness values. The classification of the terms in ERIS with the BWS method and DOI: 10.4018/978-1-6684-8427-2.ch003

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Using HurtLex and Best-Worst Scaling to Develop ERIS

a previous classification of a substantial subset of these terms into “offensive (context in/dependent)” with the inter-annotator agreement method are found to stand in a broad correlation, thus validating the methodology that was adopted to produce a more fine-grained and informative affective lexical resource. ERIS contains 1148 terms and their inflectional paradigms. It is openly available under the CC-BY-NC 4.0 license.

BACKGROUND Offensive Language Detection in Modern Greek Offensive language is the use of expressions in natural language that aim to, or result in, insult, offend, or attack the recipient of the message. Some other terms refer to related phenomena. However, sometimes different phenomena of undesirable languages, such as toxic or abusive language, while several other phenomena are linked to these, but more are more specific, such as racism, misogyny, homophobia, and other phenomena that find manifestations in language expressions. Hate speech is a particular form of undesirable language phenomenon that involves attacks towards minorities and protected groups and is often, even if partially, regulated by law and policies. Despite these differences, in this text, the term “offensive language” (hereinafter: OL) is used for both offensive and hate language since a line between them is difficult to be drawn and, at the same time, terms in the two domains are used interchangeably (Davidson et al., 2017; Waseem et al., 2017). Hate speech is related to behaviours forbidden by the law (at least in some countries), such as violence or discrimination directed against a group of persons or a member of such a group, public defamation on the grounds of race, nation, ethnicity, religion or other beliefs/convictions, sex or gender and sexual orientation. It is also related to behaviours that could be considered prohibited, such as sending a message which can cause annoyance, harassment and/or needless anxiety to another person, which the sender knows to be false, for any reason. Our societies need to discover hate speech early, stop its dissemination, and even provide effective counter speech. This work presents (i) a lexical resource that could contribute to hate speech detection in Modern Greek and (ii) a methodology for resource development. In what follows, the term “offensive language” (hereinafter: OL) is used for both offensive and hate language despite their differences. This choice has been made because a line between these two types of language/speech is difficult to be drawn, and at the same time, terms in the two domains are used interchangeably (Davidson et al., 2017; Waseem et al., 2017). The literature on EL offensive language detection does not provide annotated corpora representing several registers, sizable OL lexica or annotation methods and guidelines. The Offensive Greek Tweet Dataset (OGTD) (Pitenis et al., 2020) contains 10,287 tweets marked as “offensive”, “not offensive” or “spam”, but it is extracted using an unpublished list of profane or obscene keywords. Work on racism draws on a published annotated dataset containing 4004 tweets (Perifanos & Goutsos, 2021); and work on terrorist argument on an unpublished list of 1265 words (Lekea & Karampelas, 2018). Frantzi et al. (2019) explore impoliteness in plenary sessions of the Hellenic parliament from 2011 to 2016, a period marked by the economic crisis. The authors concentrate on verbal attacks and analyze the data gathered

33

 Using HurtLex and Best-Worst Scaling to Develop ERIS

during this period by conducting a corpus-based study without making it publicly available for training models. Pavlopoulos et al. (2017) focus on moderating content generated by users and introduce a corpus for the Modern Greek language derived from a Greek sports news website. The corpus comprises comments that have undergone manual moderation and are labelled as either ‘accept’ or ‘reject’; no reference to particular annotation guidelines is made. Furthermore, while there is published work on lexicographic issues concerning OL, e.g., Efthymiou et al. (2014) and Christopoulou (2012), no sizable annotated lexical resources have been made openly available, and no concrete annotation methodologies have been proposed. Recently, Christopoulou et al. (2022) have argued that evaluative morphology in EL slang language, with or without gender alteration, results in derivatives with varying degrees of offensiveness. In particular, they observe that offensiveness increases significantly when the gender of the base is changed to feminine or when the feminine is just retained, with or without evaluative affixes. However, they only work with a few lemmas and do not provide a methodology for developing lexica with hundreds of entries, such as ERIS. In addition, another domain in which hate speech has been investigated is in the context of journalism. Charitidis et al. (2020) explore hate speech in social media with a focus on the use of language targeted against journalists. The authors examine Twitter posts for five languages (EN, DE, ES, FR, GR), including Greek, and further release a manually annotated dataset while additionally providing their experimental results on classification accuracy. In light of the above, work on OL lexical resources for EL seemed a reasonable choice for a series of reasons. Firstly, while the phenomena of offensive and hateful speech are related but not completely overlapping (Poletto et al., 2021), lexicons of offensive terms have been successfully employed to boost the performance of hate speech classifiers (Chen et al., 2012; Gitari et al.,2015; Koufakou et al., 2020; Njagi et al., 2015). Secondly, there are strong indications that keyword and lexicon-based approaches tend to perform better when there is a shortage of annotated corpora (Sazzed, 2021). Finally, and perhaps most importantly, lexica can leverage corpora (De Swart et al., 2008; Plaza-del Arco et al., 2022). HurtLex(EL) is presented in the next section as it offers an excellent starting point for developing EL lexical resources for OL detection. ERIS was obtained with a two-step procedure. In the step, HurtLex(EL) was manually cleansed, and then in the second step, the cleansed resource was enriched with the derivational and inflectional paradigms of the terms. A description of both the cleansing and the enrichment procedures is given in respective sections. The inter-annotator agreement method was first used to assign boolean offensiveness values to the lemma entries in ERIS, and then the BWS method was adopted to assign graded offensiveness values to them. The reasons for adopting BWS are discussed along with the application procedure. The results of the two annotation methods are shown to stand in a broad agreement relation, and shown in Section 7. The distribution of Parts of Speech (hereinafter: PoS) across offensiveness values are also discussed, as it turns out that the top values have been assigned to nouns and adjectives that have humans as their references or targets.

HurtLex: SOURCE OF INSPIRATION AND A STARTING POINT The HurtLex lexicon includes offensive, aggressive and hateful words in 53 different languages and was designed to be used across various domains and contexts (Bassignana et al., 2018). Its core consists of approximately 1000 words that were hand-selected and grouped into 17 different thematic categories. The starting point for the development of HurtLex was “Le parole per ferire” (De Mauro, 2016), 34

 Using HurtLex and Best-Worst Scaling to Develop ERIS

which contains three macro-categories targeting ” derogatory words”, “stereotypes”, and words which can, depending on the context, receive a pejorative meaning. The original material was enriched semiautomatically by retrieving the available senses from the multilingual dictionary BabelNet1. Next, three types of term senses were identified with manual annotation: always offensive senses, senses whose offensiveness depends strongly on the context, and not offensive ones. BabelNet was then used again to retrieve all the lemmas associated with the offensive senses in multiple languages. This two-step process ensured the automatic translation of the lexical resource to multiple languages and provided senses for HurtLex’s entries. However, manual filtering was applied to ensure accuracy since automatic translation could propagate wrongly identified candidates, namely non-offensive words. Finally, the resulting lemma-sense pairs were annotated manually with the labels “non-offensive”, “neutral” or “offensive”. Regarding ERIS, it has already been said that the resource has been derived from HurtLex(EL) with a two-phase procedure. The first phase was described by Stamou et al. (2022) and included the following steps: 1. HurtLex(EL) was initially cleansed from meaningless and morphologically wrong (sequences of) words by two linguists. Only words selected as valid by both annotators were retained and subsequently lemmatized. Eventually, a total of 2143 words (69% of HurtLex(EL)) were lemmatized. 2. Subsequently, the retained lemmas were assigned one of the values “offensive (context-independent)”, “offensive (context-dependent)”, and “non-offensive” by four independent annotators; the Fleiss kappa inter-annotator agreement score was 0.96. Only the lemmas that were assigned one of the first two labels and had the same “offensive” label assigned by all annotators were included in the final published resource, which is referred to as HurtLex(EL)-1 (hereinafter: HurtLex(EL)-1) that later contained a total of 737 lemmas2. It should be noted that HurtLex(EL)-1 has not been evaluated as to which registers of EL it represents. For instance, as illustrated in Figure 1 shows the lemmas ανθρωποκτονία ‘man-slaughtering’ and ανθρωποκτόνος ‘man-slaughterer’ that belong to the legal jargon, especially the second one, and the lemmas παλιανθρωπιά ‘roguery’, ανθρωπάκι ‘simpleton’ that belong to colloquial language. Figure 1. Part of HurtLex(EL)-1 showing the lemmas ανθρωποκτονία ‘man-slaughtering’, ανθρωποκτόνος ‘man-slaughterer’, παλιανθρωπιά ‘roguery’, ανθρωπάκι ‘simpleton’

Work on HurtLex(EL)-1 annotation had a side effect in identifying a set of 17 non-mutually exclusive offensive thematic categories. It is worth noting that these categories are not mutually exclusive. Moreover, it is interesting to observe that both HurtLex(EL)-1 and its predecessor, HurtLex(EL), share most of the thematic categories related to offensive words. While both HurtLex(EL) and HurtLex(EL)-1

35

 Using HurtLex and Best-Worst Scaling to Develop ERIS

share most of the thematic categories related to offensive words, slight differences exist between the two resources. Specifically, there are variations in the classification of translationally equivalent terms within the defined categories. However, a couple of HurtLex(EL) categories have been unified in HurtLex(EL)-1. Finally, HurtLex(EL)-1 contains two new categories, namely “Historical/social context” and “Nationality/Ethnicity”. The first category contains contemporary terms particular to Greek history, e.g., κλέφτες ‘armatole / militiamen’ (Greek armed groups of the Ottoman occupation era). The second category, namely “Nationality/Ethnicity”, comprises terms that reflect social and cultural differentiation and nationalities/minorities within the Greek ethnicity. Examples of such terms include Εβραίος ‘Jew’, γύφτος ‘gipsy’. The thematic categories in HurtLex(EL)-1 were retained in ERIS and outlined below, along with a couple of representative terms to better understand each category’s content. • • • • • • • • • • • • • • • • •

Social class and hierarchy, e.g., χωριάτης ‘peasant’, νεόπλουτος ‘nouveau riche’, φτωχός ‘poor’ Historical and social context, e.g., σχολαστικισμός ‘scholasticism’, μεσαιωνικός’ medieval’, Crime and immoral behaviour and respective agents, e.g., δολοφονία ‘murder’ and δολοφόνος ‘murderer’, τρομοκρατία ‘terrorism’ and τρομοκράτης ‘terrorist’, Religion, e.g., ειδωλολατρία ‘idolatry’, μασόνος ‘mason’, Nationality/ethnicity, e.g., Εβραίος ‘Jew’, γύφτος ‘gipsy’, Politics, e.g., φασισμός ‘fascism’, χούντα ‘junta’, αποστάτης ‘renegade Professions of low prestige and sexual occupations, e.g., σκαφτιάς ‘digger’, ιερόδουλη ‘prostitute’, ζιγκολό, ‘gigolo’, Animals, e.g., γουρούνι ‘pig’, γάιδαρος ‘donkey’, τσιμπούρι ‘tick’ Plants, e.g., αγγούρι ‘cucumber’, πατάτες ‘potatoes’, φυτό ‘nerd’ Characteristics of inanimates, e.g., σκουπίδι ‘trash’, βαρίδι ‘sinker’, Sentiments/ psychological states, e.g., τρελός ‘crazy’, δυστυχισμένος ‘miserable’, μανιασμένος ‘raging’, Behavior, e.g., κακότροπος ‘snappy’, λεχρίτης ‘asswipe’, εξυπνάκιας ‘smartass’, Physical and cognitive disabilities/ appearance, καμπούρης ‘hunchback’, τυφλός ‘blind’, Sexuality/ gender identity, e.g., ομοφυλόφιλος ‘homosexual’, λεσβία’ lesbian’, Taboo body parts, e.g., αρχίδια ‘balls’, κώλος ‘ass’, Scientific or medical terms, e.g., ναρκισσισμός ‘narcissism’, μικρόβιο ‘germ’, Places, e.g., μπουρδέλο’ brothel’.

ANNOTATION METHODS AND BWS The concept of “offence” is highly subjective and often influenced by the social and cultural backgrounds of the annotators. Therefore, there has been extensive debate and discourse within the research community regarding the most effective methods for annotating offensive content. The aim is to ensure the annotation process is reliable and consistent across different annotators with diverse social profiles. As described above, in the first editing phase of HurtLex(EL), an annotation procedure was adopted that relied on inter-annotator agreement scores to determine the offensiveness of the lemmas within a lexicon. This approach to annotation has been argued to correlate with the so-called “prescriptive approach” to resource development that try to represent few or even only one annotator stance, as opposed to the so-called “descriptive approaches” that try to represent various stances in the same resource (Basile 36

 Using HurtLex and Best-Worst Scaling to Develop ERIS

et al., 2021; Röttger et al., 2022). In particular, Basile et al. (2021) argue that the observed disagreement in the outcome of an annotation process is a compound signal originated by different sources. One factor contributing to the disagreement is the inherent subjectivity of the annotation task, a factor particularly relevant in the annotation of undesirable language phenomena such as offensiveness. However, disagreement may also originate from other sources, such as ambiguity in the guidelines and instructions for the annotators or factors that are external to the annotation task, such as the attention level of the annotators during the process or other environmental factors. The Best-Worst Scaling (BWS) method aims at providing some solutions to these issues. BWS is a ranking-based annotation methodology initially developed by Louviere et al. (2015) and proposed as an alternative to rating-based annotation. Applied to the annotation of natural language, BWS has been proven to be beneficial in terms of the quality of the resulting data, in particular for subjective annotation tasks such as sentiment and emotion analysis (Kiritchenko & Mohammad, 2017). Moreover, Poletto et al. (2019) showed how BWS applied to the annotation of hate speech in social media text yields more consistent annotation. At a more practical level, BWS outputs scores that are naturally on a graded scale, avoiding the need to map from measures such as inter-annotator agreement. Basile and Cagnazzo (2021) research highlights that natural language resources can be annotated using varying levels of complexity and abstraction. They categorize approaches to annotation into three broad families, namely (i) “categorial annotation”, whereby labels from a set of predetermined options are assigned to each instance in a dataset, (ii) “scalar annotation”, whereby numerical values on a predetermined scale are assigned to each instance and, (iii) “ranking annotation” whereby multiple instances are ordered, and judgments are based on groups of instances rather than individual ones. The approach adopted for developing HurtLex(EL)-1 (Stamou et al., 2022) adopted the categorial annotation approach. With all methods, annotators may be inconsistent with themselves and among them. Disagreements among annotators may arise due to differences in subjective beliefs and perspectives, task difficulty, ambiguity, or even simple annotator errors (Abercrombie et al., 2023). Also, they may be due to cognitive load and different mapping of sentiment scores (Kiritchenko & Mohammad, 2016). Although measures of inter-annotator agreement can offer some insight into the subjectivity of a given task, they do not provide information on the level of difficulty, ambiguity, or the quality and attentiveness of the annotators themselves (Röttger et al., 2022). Thus, it is crucial to consider these factors when evaluating the reliability and validity of annotated data. Basile et al. (2021) identify three potential sources of disagreement in natural language processing (NLP) annotation tasks into three main categories: individual differences, stimulus characteristics, and context. Individual differences refer to each annotator’s personal world perception, which can vary based on personal opinions, values, and sentiments. Stimulus characteristics include the complexities of language, mainly referring to different kinds of ambiguity (lexical, syntactic, semantic) and discourse genres, e.g., poetry, and politics, in which multiple labels are considered correct for certain instances. Finally, context plays a crucial role in shaping human behaviour and refers to the influence of external factors on annotation. This is evidenced by the fact that even the same individual when asked the same questions at different times, may provide different responses due to various factors such as changes in their mental state and inadvertent lapses in attention. The experience from developing HurtLex(EL)-1 highlighted the need for a scalar or ranking annotation. For instance, although the EL words πουτάνα ‘prostitute (slag)’ and βλάκας ‘idiot’ are both offensive, the first one is more severe than the second. The rating scale approach has been proposed as an alternative to assigning boolean values (a categorical approach). According to the rating scale approach, the annotator is given categorical or numerical values that account for measurable features of the 37

 Using HurtLex and Best-Worst Scaling to Develop ERIS

data and an average score of all the annotations of each item is calculated. As with boolean values, the method is not immune to inter- and intra- annotation inconsistencies, especially for phrase-based items. In addition to inconsistencies, annotators often prefer a certain part of the scale, usually the middle one. Intra-annotator inconsistencies can be reduced if the time between the annotations is limited (Kiritchenko & Mohammad, 2017). The methodology adopted in ERIS involved adopting a scalar approach to assess the level of offensiveness associated with different lexicon lemmas. This approach, known as the Best Worst Scaling method (BWS), was implemented using the web-based Litescale tool (Basile & Cagnazzo, 2021). With BWS, annotators are asked to select from an n-tuple of instances those two that bear a specific property, e.g., offensive force, to the most and least extent. By making just two selections, this method allows for ranking five out of six possible pairwise comparisons in a four-item trial, providing a significant advantage (Hollis & Westbury, 2018). Additionally, it is estimated that BWS requires a work effort similar to the rating scale for the same number of instances. However, using BWS has the added benefit of producing more reliable graded annotations, particularly when n-tuples with a value of n greater than or equal to 4 are annotated by more than 3 speakers (Kiritchenko & Mohammad, 2016). Litescale implements BWS as a stand-alone software tool with a console-based interface or alternatively, a web-based interface running locally. Furthermore, an online version was created and made available freely online, which supports the creation of annotation projects and their development as a collaborative process by a system of authentication and invitations of external users. Hollis (2017) created scoring methods for the best-worst scaling response format and found that best-worst scaling was a more efficient response format than rating scales for collecting human semantic judgments in a series of simulation experiments. He also provided preliminary empirical evidence from simulated experiments to validate his findings. However, compelling empirical evidence has yet to support best-worst scaling as a useful judgment format for collecting semantic norms. On a similar par, Kiritchenko and Mohammad (2017) concluded that BWS annotations are more reliable than those produced by rating scales, especially for complex phrased items and when a small number of annotations is produced, namely 5N for N annotated items. In particular, they obtained higher split-half reliabilities of sentiment annotations with best-worst scaling than with rating scales. The Split-Half Reliability method (hereinafter: SHR) is commonly employed to assess the degree of inter-annotator consistency (Kiritchenko & Mohammad, 2017). This method involves partitioning all the annotations of a given term into two sets. This is accomplished by randomly splitting the original annotations along the annotator dimension, resulting in two distinct sets of annotations. Subsequently, a score is computed for each set of annotations. By generating two sets of annotations for the same data set, it is possible to compute a statistical measure of the correlation between them, typically averaged over multiple random splits. The level of correlation between the two sets of scores is considered an indicator of the method’s reliability. The higher the correlation, the more reliable the SHR method is.

ENRICHING THE EDITED HurtLex(EL) To form ERIS, the derivational and inflectional paradigms of the lemmas in HurtLex(EL)-1 were extracted from “The Greek open source Morphological dictionary”3 and the “LEXIS- Computational Lexicon of Modern Greek”. Incomplete derivational and, mainly, inflectional paradigms were filled manually.

38

 Using HurtLex and Best-Worst Scaling to Develop ERIS

“The Greek open source Morphological dictionary” (GOSMD) contains 1.047.200 unique words in lowercase and uppercase form. The lexicon consists of normative forms (lemmas) and many of their inflectional paradigms. “LEXIS-Computational Lexicon of Modern [SM1] Greek” (Institute for Language and Speech Processing - Athena Research Center, 2021) is an Ontolex-modelled computational lexicon for NLP applications. It contains 68,000 entries, all of which are annotated with morphological information; 66,222 unique lemmas were extracted from LEXIS. In order to obtain the derivational paradigms, the entries of Hurtlex(EL)-1, GOSMD and LEXIS were stemmed with a Python implementation of Skroutz’s Ruby Greek stemmer4 (Ntais et al., 2016). Next, the fuzzy matching technique was applied with the tool thefuzz 0.19.05, which matched the contents of Hurtlex(EL)-1 with those of Lexis and GOSMD. The material retrieved from GOSM and LEXIS was manually cleansed to remove duplicates in HurtLex(EL)-1, abbreviations and nonsensical words. As a result, 763 new lemmas were obtained. The annotation process involved three rounds: •

•

•

Round 1: Annotators were asked whether the term is in use in Greek Language and replied with a “yes/no’’ answer. Out of the 763 lemmas in total, 673 were considered in use by all annotators and fed to the second round; inter-annotator agreement Fleiss kappa score was 0.94. We used Fleiss kappa and no disagreement resolution to ensure that only words recognizable by all Greek speakers would be included in the lexicon. Round 2: The lemmas were assigned one of the labels “yes/no offensive”. Annotation work was split between two teams, the first achieving inter-annotator agreement Fleiss kappa score 0.74 and the second 0.90. Out of the 673 words, 461 were assigned the label “offensive” by all annotators in each team and added to HurtLex(EL)-1, which eventually contained 1198 lemmas. Round 3: The 1198 lemmas were assigned “yes/no offensive” labels by the five annotators; the inter-annotator agreement Fleiss kappa score was 0.76. The annotators relied on the 17 thematic categories underlying HurtLex(EL)-1. Furthermore, they were instructed to use the label “yes” when they thought that the offensive value of a word was affected by the context. Ultimately, ERIS contains the 1148 lemmas considered offensive by all annotators.

We used the Fleiss kappa score in both offensiveness annotation rounds (Round 2 and Round 3) because the notion of context-dependent values could jeopardize the results; for the same reason, we did not use disagreement resolution. The eight annotators who carried out the reported editorial work are native speakers of EL, all of them of tertiary education. Their ages range from 23 to 65 years, and they live in different cities in Greece, where they speak different dialects of EL and cooperate online.

GRADED OFFENSIVENESS VALUES OBTAINED WITH BWS Before applying the BWS annotation method, the inter-annotator consistency was thoroughly checked. For this purpose, two annotations, each consisting of 117 lemmas, were subjected to a 4-tuple BWS experiment. The eight experienced annotators described in the previous sections participated in this experiment, with the annotators divided into two groups of equal size. No specific annotation instructions were given. SHR was computed by randomly splitting the annotators into two groups to assess the inter-annotator consistency. Since there were two independent projects, the BWS performance of each group was computed, followed by calculating the Pearson correlation between the two sets of scores. 39

 Using HurtLex and Best-Worst Scaling to Develop ERIS

The process was repeated 100 times with a different random split of the eight annotators. The average values over the 100 runs were 0.817 (P-value 4.5e −23) and 0.845 (P-value 6.3e −25), respectively. This was important news because it showed no annotation instructions were necessary to achieve excellent inter-annotator consistency in the particular group of native speakers of Modern Greek. Two BWS experiments were carried out with the Litescale tool. •

The BWS method was applied to the EL branch of HurtLex-Core (hereinafter: HurtLex-Core(EL)). HurtLex-Core is a smaller version of HurtLex comprising the top 500 most frequently used terms from the original lexicons. Each HurtLex-Core lexicon (one for each language) has been created automatically by collecting tweets in the respective language to approximate a frequency count and retaining the most frequent ones; in this way, HurtLex-Core contains only the most frequent word forms, but at the same time, fewer mistakes than its predecessor, because several classes of erroneous entries are filtered out by being low-frequency, such as misspellings and words in the wrong language. Two BWS experiments were conducted on HurtLex-Core(EL), one with 2-tuples and three annotators and one with 4-tuples and four annotators. Each annotator completed an entire annotation cycle, as they were required to annotate all the tuples.

The 2-tuple experiment returned unreliable results in the sense that there was no clear border value between offensive and non-offensive words. On the other hand, the 4-tuple experiment returned reliable results since all words assigned a score greater than 0.5 were identified as offensive, and all words assigned a score greater than 0.35 belonged to the lexicon of Modern Greek. In short, the BWS method using 4-tuples was found to be useful not only for assigning graded offensiveness values to the entries in HurtLex-Core(EL) but also for filtering out irrelevant entries, thus aiding in the cleansing process of the lexicon. The results obtained from these two experiments offer advice on using BWS with Litescale. The aim is to receive reliable annotations, i.e., annotations that yield similar results in repeated trials. Our results follow Kiritchenko & Mohammad (2016), who estimate that reliable annotations are obtained with 2N BWS annotations (where N is the number of annotated elements). Litescale uses 4-tuples by default and ensures that each tuple element appears in 5 different 4-tuples, each element appears (N/4) x5 times in each annotation cycle, and the total number of annotations is Mx(N/4)x5 where M is the number of complete annotation cycles. In order to have 2N BWS annotations, M must be equal or more significant to 8/5, and 2 is the smallest integer that meets this requirement. Our experiment has shown that two annotation cycles are not enough to yield reliable results, but four are. •

A group of eight annotators has assigned graded offensiveness values to the lemma entries of ERIS using a 4-tuple BWS. The annotation process was conducted meticulously to ensure the reliability of the results, with each annotator completing an entire annotation cycle. Additionally, it was assumed that inflectional paradigms shared the same offensiveness value as their respective lemmas.

In Figure 2, a fragment of the ERIS lexicon is illustrated. The lemmas are assigned graded offensiveness values. The gender of nouns and adjectives is denoted explicitly when the same type is used for both the feminine and the masculine genders (Article and Gender Type).

40

 Using HurtLex and Best-Worst Scaling to Develop ERIS

Figure 2. Fragment of ERIS showing the lemmas ανθρωπάκι ‘simpleton’, ανθρωπάκος ‘small fry’, ανθρωποκτονία ‘man-slaughtering’, ανθρωποκτόνος ‘manslaughterer’, παλιανθρωπιά ‘roguery’

LINGUISTIC TRAITS OF MODERN GREEK OFFENSIVE LANGUAGE ERIS has provided the grounds for interesting observations regarding EL offensive language. Arguably, ERIS offers a representative picture of Modern Greek OL given that: • •

Firstly, it has been developed through a thorough process of enriching HurtLex(EL)-1. It has been generated by skimming many multilingual texts and validated by native speakers using an interannotator agreement method (Stamou et al., 2022). Secondly, the BWS method was employed to determine offensive force ratings and a group of eight annotators demonstrated reliable annotations carried out the process without any prior instructions.

In ERIS, the derivational paradigms of the lemmas in HurtLex(EL) have been included. As a result, the lexicon contains four parts of speech (PoS) representations, namely nouns, adjectives, verbs, and adverbs. Figure 3 shows the PoS distribution in ERIS. It is worth noting that the PoS NOUN has been divided into two categories in ERIS. The first one, called ‘human’, includes nouns that can be used in a derogatory manner to refer to humans, thus assigning an offensive meaning (1). This category also contains nouns that are typically used to refer to humans by employing terms that refer to animals or plants (2). Finally, the second category has retained the NOUN label and contains all the remaining nouns in ERIS (3). The most populated category in ERIS is the sum of the ‘human’ and ‘NOUN’ categories, followed by ADJ(ectives). 1. αντιγραφέας ‘duplicator’, τσιγγανάκι ‘young/child gipsy’, ψεύτης ‘liar’, σαχλαμάρας ‘fool’ 2. a. μουλάρι (:literally ‘mule’): ‘stubborn, impolite person’, κότα, (:literally ‘chicken’): ‘coward’, γύπας, (:literally ‘vulture’): ‘someone who is greedy, opportunistic’ b. λαπάς, (:literally ‘overcooked rice’): ‘soft-boiled person, wuss’, παράσιτο, (:literally ‘parasite’): ‘someone who lives off of others’, φυτό, (:literally ‘plant’): ‘nerd’, μικρόβιο, (:literally ‘microbe’): ‘germ’. 3. ψέμμα ‘lie’, τσουτσούνι ‘dick’, απάτη ‘scam’, λαγνεία ‘lust’, σχολαστικότητα (:literally ‘scholasticism’): ‘someone being overly pedantic’ The differentiation between expressions that refer to or target humans versus other nouns is a fundamental aspect of human languages, with notable examples including pronouns such as “he/she” versus

41

 Using HurtLex and Best-Worst Scaling to Develop ERIS

“it” in English. Additionally, studies have shown that Modern Greek speakers tend to prioritize placing animate entities in early word order positions, regardless of their grammatical function (Branigan & Feleki, 1999). In the literature on functional typological linguistics, a hierarchical treatment of animates versus inanimate has been observed, with humans occupying the highest position, followed by animals (animate) and inanimate entities (Swart et al., 2008). This hierarchical structure underscores the unique grammatical status of expressions that refer to or target humans. In EL, several verb predicates select only human arguments. For instance, there are two verbs for dying, πεθαίνω and ψοφάω. The first one is normally predicated on humans, and the second is reserved for animals; predicating ψοφάω of a human is considered to be an offensive act. ERIS also suggests that the distinction between expressions that refer to or target humans versus inanimate and other animate entities is also relevant in the context of offensive language in Modern Greek. Figure 3. PoS distribution, including the category ‘human’, in ERIS

Figure 4 presents the harmonic means of BWS ratings of all PoS and for ‘human’. The harmonic mean was used because, since it does not give much weight to large items, the results for each PoS would not be significantly influenced by a small number of lemmas with high offensiveness values. A first interesting observation is that ‘human’ achieves the highest BWS harmonic mean and NOUN the second lowest.

42

 Using HurtLex and Best-Worst Scaling to Develop ERIS

Figure 4. Harmonic mean of BWS ratings of all PoS tags and the category ‘human’

In order to provide a more detailed analysis of the differentiation between ‘human’ and NOUN, we have included Figure 6. This graph illustrates the distribution of terms within each category about their corresponding Best-Worst Scaling (BWS) offensiveness ratings. It becomes evident that a more significant number of terms classified as ‘human’ fall within the BWS rating range of 0.4-1.0 when compared to those classified as NOUN (as it is also demonstrated in Figure 5). Figure 5. Distribution of terms in the category ‘human’ (left) and NOUN (right) wrt BWS offensiveness ratings

Back to Figure 4, the second observation is that the BWS harmonic mean of adjectives is pretty close to the ‘human’ one. This close similarity may not be unexpected, given that nearly all ERIS adjectives can be predicated on humans. During the BWS annotation process, the annotators reported that they were always mindful of whether a word denoted or targeted humans. They acknowledged that they tended to assign higher offensiveness values to nouns and adjectives that could be applied to humans.

43

 Using HurtLex and Best-Worst Scaling to Develop ERIS

A last observation of linguistic nature is that colloquial words tend to be assigned higher offensiveness values than their formal synonyms (4). In addition, colloquial language has a battery of expressive means, such as intensifying morphemes (4b), that result in high offensiveness values. 4. a. colloquial: πουτάνα (0.90), formal: πόρνη ‘prostitute’ (0.82), colloquial: παπαριά (0.70), formal: ανοησία (0.38) ‘foolishness’ b. colloquial: αδερφ-άρα ‘gay-INTENSIFIER’ (0.90), formal: ομοφυλόφιλος ‘homosexual’ (0.69), colloquial: παλιουποκριτής (0.70), formal: υποκριτής (0.66) ‘hypocrite’

COMPARISON BETWEEN INTERANNOTATOR AGREEMENT AND BWS-BASED ANNOTATIONS Lemmas in HurtLex(EL)-1, which constitutes a significant and proper subset of ERIS, were assigned the offensiveness values “offensive (context-dependent)” and “offensive (context-independent)” with a method based on inter-annotator agreement. On the other hand, lemmas in ERIS were assigned offensiveness ratings with the BWS method. Therefore, HurtLex(EL)-1 presents an excellent opportunity to compare the two methods of assigning offensiveness values. It should be noted that HurtLex(EL)-1 and ERIS were annotated by different annotators with similar social features; the HurtLex(EL)-1 group consisted of four annotators, and the ERIs one of eight annotators. Each member of these groups annotated all the lemmas in the respective lexica. Table 1 presents the distribution of BWS ratings in the categories “offensive (context-independent)” (YES on Table 1) and “offensive (context-dependent)” (CONTEXT on Table 1). The BWS offensiveness ratings were further divided into two groups, i.e., “low” ratings (ratings ranging between 0 and 0.40) and “high” ratings (ratings ranging between 0.40 and 1). Τhis decision to use this cutoff relied on the quantitative facts shown in Figures 4 and 5. Figure 5 shows that most of the lemmas in the ‘human’ category have been assigned offensiveness ratings greater than 0.4. In contrast, only a fraction of the lemmas in the NOUN category are above this score. In addition, Figure 4 shows that most of the adverbs and verbs score below this value. It should be recalled that ERIS includes the complete derivational paradigms of the lemmas included in HurtLex(EL)-1; this enrichment often resulted in adding adverbs and verbs morphologically related to the existing lemmas. The value 0.4 was chosen as a gross borderline between high and low-scoring lemma categories. By examining Table 1, it is evident that a significant percentage of lemmas assigned to the “offensive (context-independent)” category received ratings above 0.4, amounting to 81%. On the other hand, only 42% of the lemmas in the “offensive (context-dependent)” category were assigned ratings greater than 0.4. If it is assumed that the value “offensive (context-independent)” was likely to be assigned to words whose offensiveness is so strong that only a few contexts can mitigate or neutralize it. Although the inter-annotator agreement method used instructions to the annotators while the BWS one did not, Table 1 indicates that the two annotation methods are in broad agreement.

44

 Using HurtLex and Best-Worst Scaling to Develop ERIS

Table 1. Comparison of the distribution of annotations in HurtLex(EL)-1 and in the ERIS lexicon BWS offensiveness ratings vs yes/no ones Lemmas in 0-0.40

Lemmas in 0.40-1

Total Lemmas

% HurtLex lemmas

YES

83 (19%)

353 (81%)

436 (100%)

61

CONTEXT

161 (58%)

116 (42%)

277 (100%)

39

CONCLUSION AND THE FUTURE We have reported on developing ERIS, a lexicon for offensive language detection in Modern Greek with graded offensiveness values for 1148 lemmas. ERIS originates from HurtLex and has reached its present state through a series of cleansing, enriching and annotation exercises. ERIS is freely available on the Web. On the annotation front, two methods were compared and found to agree. The first approach involved assigning boolean values of “offensive (context-dependent)” and “offensive (context-independent)”, in which we calculated Fleiss kappa scores to assess inter-annotator agreement. The second method utilized the BWS approach with the help of the Litescale tool. One advantage of the BWS method was that it did not require any prior instructions on what should be considered “offensive speech”, making it more flexible than the boolean approach. Moreover, the BWS method generated graded ratings of offensiveness for each item with relatively little effort, which was a significant advantage for our research. Our research indicates that the BWS method is a safe and effective approach for both cleansing and annotating datasets of words using 4-tuples. In our study, we carefully checked the 4-tuples a total of four times to ensure the results’ accuracy. Our findings suggest that the BWS method yielded reliable and consistent results, making it a valuable tool for researchers interested in offensive language detection. During our analysis, we made certain intriguing observations that are worth mentioning. Firstly, we noticed that the annotators tended to rate nouns and adjectives that could be used to refer to humans or describe them with higher offensiveness scores. This observation seems to correlate well with the hierarchical treatment of animates versus inanimate that has been independently attested in many human languages. Secondly, when comparing parts of speech, we found that adverbs received the lowest offensiveness ratings, while nouns that did not refer to humans received the lowest scores. We are currently using ERIS to leverage OL corpora of Modern Greek derived from fiction texts and tweets. By analyzing the collected data, we aim to obtain genre-specific frequency evidence and shed light on using ERIS’ lemmas among native speakers. The efficacy of ERIS was evaluated by employing the BWS method, which was used to calculate the degree of offensiveness of the corpus material. This approach will provide a valuable quantitative assessment of ERIS’ effectiveness because it will show which lemmas help retrieve offensive language (from corpora of these two registers). Additionally, we hope to explore whether there is some correlation between sentence offensiveness ranking and lemmas in the lexicon.

45

 Using HurtLex and Best-Worst Scaling to Develop ERIS

ACKNOWLEDGMENT This work was partly supported by “PHILOTIS: State-of-the-art technologies for the recording, analysis and documentation of living languages” (MIS 5047429): “Action for the Support of Regional Excellence” that is funded by the Operational Programme “Competitiveness, Entrepreneurship and Innovation” (NSRF 2014-2020) and co-financed by Greece and the European Union (European Regional Development Fund).

REFERENCES Abercrombie, G., Rieser, V., & Hovy, D. (2023). Consistency is Key: Disentangling Label Variation in Natural Language Processing with Intra-Annotator Agreement. arXiv:2301.10684. https://doi.org/ doi:10.48550/arXiv.2301.10684 Basile, V., & Cagnazzo, C. (2021). Litescale: A lightweight tool for best-worst scaling annotation. In Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021). INCOMA Ltd. 10.26615/978-954-452-072-4_015 Basile, V., Fell, M., Fornaciari, T., Hovy, D., Paun, S., Plank, B., Poesio, M., & Uma, A. (2021). We need to consider disagreement in evaluation. In Proceedings of the 1st Workshop on Benchmarking: Past, Present and Future. Association for Computational Linguistics. https://doi.org/10.18653/v1/2021.bppf-1.3 Bassignana, E., Basile, V., & Patti, V. (2018). Hurtlex: A multilingual lexicon of words to hurt. In Proceedings of the Fifth Italian Conference on Computational Linguistics (CLiC-it). 10.4000/books. aaccademia.3085 Branigan, H. P., & Feleki, E. (2020). Conceptual Accessibility and Serial Order in Greek Speech Production. In Proceedings of the Twenty-first Annual Conference of the Cognitive Science Society. Psychology Press. 10.4324/9781410603494-22 Charitidis, P., Doropoulos, S., Vologiannidis, S., Papastergiou, I., & Karakeva, S. (2020). Towards countering hate speech against journalists on social media. Online Social Networks and Media, 17, 100071. doi:10.1016/j.osnem.2020.100071 Chen, Y., Zhou, Y., Zhu, S., & Xu, H. (2012). Detecting offensive language in social media to protect adolescent online safety. In 2012 International Conference on Privacy, Security, Risk and Trust and 2012 International Conference on Social Computing. IEEE. 10.1109/SocialCom-PASSAT.2012.55 Christopoulou, A. (2012). A lexicological analysis of slang vocabulary of Modern Greek. [PhD dissertation, University of Patras]. Christopoulou, K., Xydopoulos, G. J., & Tsangalidis, A. (2022). Grammatical gender and offensiveness in Modern Greek slang vocabulary. In N. Knoblock (Ed.), The Grammar of Hate: Morphosyntactic Features of Hateful, Aggressive, and Dehumanizing Discourse. Cambridge University Press. doi:10.1017/9781108991841.005

46

 Using HurtLex and Best-Worst Scaling to Develop ERIS

Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated hate speech detection and the problem of offensive language. In Proceedings of the 11th International AAAI Conference on Web and Social Media (ICWSM’ 17). 10.1609/icwsm.v11i1.14955 De Mauro, T. (2016). Le parole per ferire. Internazionale. De Swart, P., Lamers, M., & Lestrade, S. (2008). Animacy, argument structure, and argument encoding. Lingua, 118(2), 131–140. doi:10.1016/j.lingua.2007.02.009 Efthymiou, A., Gavriilidou, Z., & Papadopoulou, E. (2014). Labeling of Derogatory Words in Modern Greek Dictionaries. In N. Lavidas, T. Alexiou, & A. Sougari (Eds.), Major Trends in Theoretical and Applied Linguistics 2. De Gruyter Open Poland. doi:10.2478/9788376560885.p12 Frantzi, K., Georgalidou, M., & Giakoumakis, G. (2019). Greek Parliamentary Discourse in the Years of the Economic Crisis. In E. Jakasa (Ed.), Argumentation and Appraisal in Parliamentary Discourse. Information Science Reference. doi:10.4018/978-1-5225-8094-2.ch001 Gitari, N. D., Zuping, Z., Damien, H., & Long, J. (2015). A lexicon-based approach for hate speech detection. International Journal of Multimedia and Ubiquitous Engineering, 10(4), 215–230. doi:10.14257/ ijmue.2015.10.4.21 Hollis, G. (2017). Soring best-worst data in unbalanced, many-item de- signs, with applications to crowdsourcing semantic judgments. Behavior Research Methods, 50(2), 711–729. doi:10.375813428017-0898-2 PMID:28550657 Hollis, G., & Westbury, C. (2018). When is best-worst best? A comparison of best-worst scaling, numeric estimation, and rating scales for collection of semantic norms. Behavior Research Methods, 50(1), 115–133. doi:10.375813428-017-1009-0 PMID:29322399 Kiritchenko, S., & Mohammad, S. M. (2016). Capturing reliable fine-grained sentiment associations by crowdsourcing and best-worst scaling. In Proceedings of the 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics. 10.18653/v1/N16-1095 Kiritchenko, S., & Mohammad, S. M. (2017). Best-worst scaling more reliable than rating scales: A case study on sentiment intensity annotation. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL-2017). Association for Computational Linguistics. https://doi. org/10.18653/v1/P17-2074 Koufakou, A., Pamungkas, E. W., Basile, V., & Patti, V. (2020). HurtBERT: Incorporating lexical features with BERT for the detection of abusive language. In Proceedings of the fourth workshop on online abuse and harms. Association for Computational Linguistics. 10.18653/v1/2020.alw-1.5 Lekea, I. K., & Karampelas, P. (2018). Detecting hate speech within the terrorist argument: a greek case. In 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM). IEEE. 10.1109/ASONAM.2018.8508270

47

 Using HurtLex and Best-Worst Scaling to Develop ERIS

Clarin:el. (2021). LEXIS Computational Lexicon of modern Greek. Version 1.0.0 (automatically assigned). Institute for Language and Speech Processing - Athena Research Center. http://hdl.handle.net/11500/ CLARIN-EL-0000-0000-6105-D Louviere, J. J., Flynn, T. N., & Marley, A. A. J. (2015). Best-worst scaling: Theory, methods and applications. Cambridge University Press. doi:10.1017/CBO9781107337855 Ntais, G., Saroukos, S., Berki, E., & Dalianis, H. (2016). Development and enhancement of a stemmer for the greek language. In Proceedings of the 20th Pan-Hellenic Conference on Informatics (PCI’ 16). Association for Computing Machinery. https://doi.org/10.1145/3003733.3003775 Pavlopoulos, J., Malakasiotis, P., & Androutsopoulos, I. (2017). Deeper Attention to Abusive User Content Moderation. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics. 10.18653/v1/D17-1117 Perifanos, K., & Goutsos, D. (2021). Multimodal hate speech detection in greek social media. Multimodal Technologies and Interaction, 5(7), 34. doi:10.3390/mti5070034 Pitenis, Z., Zampieri, M., & Ranasinghe, T. (2020). Offensive language identification in Greek. In Proceedings of the 12th Language Resources and Evaluation Conference. European Language Resources Association. Plaza-del-Arco, F. M., Portillo, A. B. P., Úbeda, P. L., Gil, B., & Martín-Valdivia, M. T. (2022). Share: A lexicon of harmful expressions by Spanish speakers. In Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022). European Language Resources Association (ELRA). Poletto, F., Basile, V., Bosco, C., Patti, V., & Stranisci, M. (2019). Annotating hate speech: Three schemes at comparison. In the 6th Italian Conference on Computational Linguistics, CLiC-it 2019, 2481, 1–8. CEUR-WS. Poletto, F., Basile, V., Sanguinetti, M., Bosco, C., & Patti, V. (2021). Resources and benchmark corpora for hate speech detection: A systematic review. Language Resources and Evaluation, 55(2), 477–523. doi:10.100710579-020-09502-8 Röttger, P., Vidgen, B., Hovy, D., & Pierrehumbert, J. B. (2021). Two contrasting data annotation paradigms for subjective nlp tasks. arXiv preprint arXiv:2112.07475. https://doi.org/ doi:10.35848/13474065/aca256 Sazzed, S. (2021). A lexicon for profane and obscene text identification in Bengali. In Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021). INCOMA Ltd. 10.26615/978-954-452-072-4_145 Stamou, V., Alexiou, I., Klimi, A., Molou, E., Saivanidou, A., & Markantonatou, S. (2022). Cleansing & expanding the HURTLEX (el) with a multidimensional categorization of offensive words. In Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). Association for Computational Linguistics. 10.18653/v1/2022.woah-1.10

48

 Using HurtLex and Best-Worst Scaling to Develop ERIS

Waseem, Z., Davidson, T., Warmsley, D., & Weber, I. (2017). Understanding abuse: A typology of abusive language detection subtasks. In Proceedings of the First Workshop on Abusive Language Online. Association for Computational Linguistics. 10.18653/v1/W17-3012

ENDNOTES 1 2

3 4

5



https://babelnet.org/ The resource is available at https://osf.io/t5jey/?view_only=e910e28ea21e4895905aff2d0c0a c162 (Archived under: DOI 10.17605/OSF.IO/T5JEY) https://github.com/eellak/gsoc2019-greek-morpho https://github.com/skroutz/greek_stemmer/blob/master/LICENSE .txt https://pypi.org/project/thefuzz/

49

50

Chapter 4

Spreading Organised Hate Content Sergio Arce-Garcia https://orcid.org/0000-0003-0578-9787 Universidad Internacional de La Rioja, Spain Jarnishs Beltran https://orcid.org/0000-0001-6867-5950 Universidad de Valparaiso, Chile

ABSTRACT Hate and disinformation often go hand in hand, and there are organisations dedicated to spreading them and influencing public opinion. The aim of this chapter is therefore to expose the forms of hate-mongering and one of its means of penetration, disinformation, as well as the techniques used to spread it and the measures taken to counter it. It should be remembered that behind these hate-creation techniques, there are companies and even government systems that have perfected and have the means to achieve their goals. These techniques and forms are used internationally, but are particularly widespread in European and Mediterranean countries, especially around issues such as immigration. The aim is to polarise societies by exploiting fissures and divisive issues. This is a dangerous game in which they end up being capitalised on by extreme and ultra-ideological groups in an attempt to change the culture. Even democracy itself is at risk.

BACKGROUND To determine how hate influences public opinion, it is first necessary to clearly define what public opinion is. Many specialists have long debated this concept, and no entirely accepted one exists. Following Seoane (2019), the concept of public opinion went through five stages from the eighteenth century to the democracies of the second half of the twentieth century, where most researchers see it as the aggregation of individual opinions collected through opinion polls. According to Vinuesa Tejero (1997), the contents

DOI: 10.4018/978-1-6684-8427-2.ch004

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Spreading Organised Hate Content

that determine public opinion are cognitive (since they represent mental aspects and judgements), evaluative (since they represent evaluations and attitudes) and action orientations. The lack of a good definition of public opinion leads to confusion, such as identifying it with people’s opinions or what the media expresses. Hence the problems in associating public opinion with the general will, state of opinion, ideology, values or popular will. To focus it better, three conditions must be present in what is public opinion: tolerance, ideological pluralism and freedom of expression within a complex and interdisciplinary framework. The functions of public opinion are the general and informal control of society, the legitimation of power, the monitoring and control of the authorities and their decisions, and the social stimulation and pressure on decision-making (Martín Segovia, 2003). It can be said that public opinion regulates and organises social behaviour, so it is an informal social control phenomenon (because it is not subject to legislation or jurisprudence and is not organised) with its characteristics but capable of influencing governments and organisations. Public opinion presents and receives influences from other sectors, such as religious, economic, moral, and commercial, and includes aspects that determine the individual and his or her behaviour. Going against so-called public opinion can even lead to problems. For example, Noelle-Neumann (1973) considered public opinion “opinions on controversial issues that can be expressed in public without fear of being isolated”. This gives rise to the concept of the “spiral of silence”, i.e. the person who has a different opinion from that described in public opinion tends to remain silent. As people can notice the rise or fall of certain opinions, reactions to the above point lead to the search for safer discourses or silence, and the fear of being isolated leads people to pay particular attention to the supposed majority opinion. Today, the media and social networks can exert significant influence as a means of social control. Polls or surveys are typical strategies to represent what is considered public opinion, “what people want”, to legitimise certain positions or opinions. When public opinion does not provide the desired response, it is common to seek enlightened opinions to support it or to introduce problems or solutions that have not yet been raised. This is why, according to Bourdieu (2012), to say that “the polls are on our side” is equivalent, in another context, to say that “God is on our side”. In this way, creating a climate of opinion is based on introducing and reinforcing concepts and opinions, generally by provoking basic emotions through simple and simplified expressions. This creates stereotypes and positive and negative associations that guide people’s perceptions (Jiménez Sánchez, 1994). This relationship creates friction between the media and public opinion. Problems include private interests, different tastes, politicisation, the belief that the media are spokespersons for public opinion and its representatives, and pluralism and openness to the opinions of privileged minorities are established. Social networks have increased what is known as cyber-democracy, in which certain people or institutions present themselves as representatives of the citizenry without ever having stood for election but simply appearing with many followers. One of the most widely accepted theories of social media opinion is that of filter bubbles by Pariser (2011). Due to the algorithms in social networks and web platforms, aspects are offered that are liked and confirm each person’s opinions and those that go in the opposite direction are not shown. As a result, people only receive information confirming their beliefs or from people with similar opinions and tastes, while suppressing information contradicting them. Thus, people only move within a sphere where they believe everyone thinks like them, redistributing and amplifying their messages, becoming “echo chambers”. The theories of filter bubbles and echo chambers have recently been challenged by several studies, such as Arguedas et al. (2022), who state that “the work reviewed here suggests that echo chambers 51

 Spreading Organised Hate Content

are much less prevalent than commonly assumed, finds no support for the filter bubble hypothesis, and presents a very mixed picture of polarisation and the role of news and media use in contributing to polarisation” and Jones-Jang and Chung (2022). Other authors suggest that these aspects would be better understood through the new social network study streams of gas-lighting (Shane et al., 2022). According to this theory, the generation of strong basic emotions in group communication leads to manipulation of the recipients, raising doubts about their perception of reality and disconnection from the rest of the population.

The Evolution of Technology and Social Media In the first decade of the 21st century, technology companies realised that the business of selling programs and applications had reached its limit and that they had to deal with various adversities, such as piracy and the emergence of free software. As a result, they realised that the real business was not software or operating systems but their customers’ data. The emergence of “free” applications and services, where the user pays with his or her data, began. This, together with the emergence of new technological capabilities, led to the development of applications that had not previously been possible, resulting in the use of techniques such as: 1. Classification of individuals according to their tastes and interests through cluster analysis using mathematical algorithms. 2. Machine learning techniques are used to identify behaviour patterns across a range of data. This gives rise to various algorithmic analyses: Bayesian analysis, random forest, neural networks, and regression trees. Even emotions, sentiments or positive-negative polarity can be detected, or even the determination of hatred and/or themes that are the subject of the analysis. 3. The analysis of human language using Natural Language Processing (NLP) techniques. All these techniques, combined to determine patterns and themes, will lead to the possibility of statistically determining a person’s tastes or interests. A decision will even be made automatically based on the percentage of success of the decision. This is artificial intelligence. One of the places where technology companies started to get customer data was from their tastes and shopping choices, but the perfect place was the new social media, where people very quickly exposed a lot of their lives, tastes and relationships. So at the end of the first decade of the 2000s, social media interaction buttons like “like” or “retweet” were created to categorise people better. All these variables of likes and the data obtained are called “lookalikes” in the Facebook network. Thanks to them, it is possible to create psychological profiles that characterise each person, plan specific campaigns, and sell to them more effectively. But after the tech companies, other companies and organisations have learned to do the same, not all with the best intentions. Particularly after the Arab Spring, 11M in Spain, the Occupy Wall Street movement in the US and the Russian elections, with protests over alleged manipulation of polling stations, all in late 2010 and early 2012, the world realised that social media were an ideal place not only to get data and classify people but even to run campaigns and try to manipulate. Disinformation techniques, created between the First and Second World Wars and perfected during the Cold War (Rid, 2020), received an incredible new boost from new tools, especially from 2010-2011. Companies such as Cambridge Analytica 2014 began to operate at the political level, most famously during Brexit and the US elections with Donald Trump. And not only at the political level but also at the 52

 Spreading Organised Hate Content

business level. There are several books, especially by different journalists, that give examples of what has happened in recent years in several campaigns in Italy, Finland, Russia or Ukraine and other countries (Aro, 2022; Da Empoli, 2020; Pomerantsev, 2019; Soldatov & Borogan, 2015). They provide examples of how social media and other internet media are used to influence public opinion or specific groups. Online participation is deeply damaged by the existence of trolls, campaigns organised for false purposes, and the influence of systems interested in favouring opinion environments, such as bot farms, cyber lobbies, or the proliferation of influencers who become opinion leaders even when their posts have no veracity (Sánchez-Carballido, 2008). All these practices distort conversations and, therefore, public opinion, which is exposed to hostile discourses of varying intensity, highly polarised discourses and conspiracy theories (Frischlich, 2022; Neubaum, 2022). According to Briant (2020), of the 17 companies and organisations worldwide dedicated to disinformation in 2017, there are already more than 600 in 2020. Many of these are dedicated to offering techniques to the highest bidder for disinformation campaigns against targets such as specific industries, organisations and even individuals. The toxicity of social media messages is very complex, affecting groups and even weakening governments and democracy itself. Both academics and journalists have reached a significant level of study, with increasing weight, where disinformation and fake news are the order of the day. Society faces a hybrid ecosystem environment where traditional and digital media coexist (López-García, 2016). The target audience of these organised disinformation and hate groups is very diverse. According to the article by Arce-Garcia, et al. (2022) on a possible troll farm from Philippines targeting the Spanish public during the worst covid period in 2020, the topics used to introduce and subsequently disseminate were varied: sex models (15.80%), extreme right (13.71%), teenagers and animals (8.13%), left-wing political parties (7.47%), influencers (6.98%), humour (4.95%), video games (4.80%), police and military (2.34%), football (2.17%), music (2.02%), science (1.69%), TV and film (1.48%) or religion (1.07%). All this is conducive to promoting a culture of disinformation, as it takes advantage of algorithms that present content that reinforces people’s opinions and hides those that contradict them in the so-called bubble filter theory (Pariser, 2011). In this way, we can arrive at the following definition of hoax: “all those false contents that are disseminated to the public deliberately fabricated for various reasons that can range from simple jokes or parodies to ideological controversies or economic fraud” (Salaverría et al., 2020). Today’s digital space has become a place where, more than ever, the aim is to capture and influence public opinion (Campos-Domínguez & Calvo, 2017). Social media are a new container for immediate access to content, where a favourable environment is created for each user to become a re-emitter of information, or “echoes of resonance”, often unverified (Rodríguez Fernández, 2021). Each social group takes advantage of every message to put it at its service, and paradoxically, when there are more media on offer, people use fewer sources of information than before. The advent of the Internet, especially with the emergence of Web 2.0, where sites adapt to the user and allow for user participation, and the widespread use of social media, have strengthened the capabilities of new actors in the flow of information. Users can now address organisations directly and engage with them without intermediaries. Not responding or responding late creates negative images. Crises 2.0 are produced and developed on the Internet and have a set of characteristics compared to other channels: immediacy, speed, anonymity and hypersexuality (Castillo Esparcia & Ponce, 2015). For this reason, the image and reputation of organisations are constantly at stake. In general, this involves the appearance of open and massive messages through platforms such as social media, which allow for rapid and

53

 Spreading Organised Hate Content

broad reach. This speed and breadth mean that the traditional media, which are also very attentive to the networks, will seek information on the subject, usually on the network itself. These characteristics have led to various techniques for creating disinformation and other malicious messages, adapted from previous eras, and new dissemination techniques adapted to the network, such as Astroturfing. This is not new but has been used throughout the 20th century and has been described as ‘active measures’ in the geopolitical environment (Rid, 2020). The advent of the 21st century and the presence of new technologies, such as machine learning or artificial intelligence, have led to an enormous evolution of previously used techniques, which have reached an enormous level of sophistication.

The Techniques of Hate Historically, the study of hatred has been marked by the “pyramid of hatred”, a work by Allport (1954) which sets out a series of steps or levels in which a situation is classified and how far it can go. In these successive stages, from bottom to top, we find “against-everything”, avoidance, discrimination, physical attack and extermination. In this way, hatred is built up in stages through jokes and small comments, separating “one’s own” from “the others” and opposing everything. This process can create the breeding ground for jumping to the second and third levels, where segregation and discrimination come in. The formation of bubble filters and echo chambers described above is the ideal place in social media to initiate and encourage the spread of hatred. This process fosters radicalisation, which, according to the basic theory, has three stages: motivation (to move the subject out of reluctance and vulnerability and into something personally meaningful), ideological radicalisation (through the promotion of violence, self-sacrifice and martyrdom), and finally interaction with other equally radicalised subjects which form a group with common goals. These three stages create, foster and maintain strong group relationships that contribute to increasing extremist behaviour with little regard for personal consequences. According to Williams (2021), hate speech is constructed in five ways: 1. By breaking the rules: e.g. showing a homosexual kiss in a country where it is criminalised. 2. Making the victim feel ashamed: for example, by making the person imagine how their family and friends would react if they saw a specific behaviour. 3. Instilling fear in victims through threats and intimidation. 4. Attempting to dehumanise the victim by comparing them to insects, vermin or primates. 5. By misinforming individuals or groups to which they belong. None of these five points is exclusive and can be complementary, and in social media, external and superior agents (influencers) make the victim feel subordinated and even humiliated, dominating them and attacking their sense of identity. All these points can now be categorised through an analysis of the target, the communication that must be designed to provoke emotions, especially fear and anger, and finally induce hatred, an emotion that eliminates rational thinking. Disinformation is, therefore, one of the bases for provoking all these reactions. There is a link between disinformation and hate crimes, with research showing that the Trump campaign in 2016 saw the highest number of hate crimes in US history, surpassing the attacks on Muslims after the 9/11 attacks in New York in 2001. According to StopHateUK, the data suggests the same could 54

 Spreading Organised Hate Content

be true in the aftermath of Brexit or against Asian communities during the UK’s covid pandemic. In the Philippines, following then President Duterte’s comments against drug addicts that they should feed the fish in Manila Bay, some 4,200 people were shot dead by police or vigilantes (up to 33 in a single day), according to official figures, and 12,000 according to an estimate by a human rights NGO. In contrast, the presence of Muslim footballer Mohamed Salah at Liverpool FC coincides with a reduction in hate crimes in the city (Williams, 2021).

Hate Linked to Emotions The Soviet biologist Serge Tchakhotine, a student of Pavlov, studied party propaganda before the Second World War and denounced the use of emotions; his book was censored in many countries, including his own (Tchakhotine, 1992). He stressed the effectiveness of repetitive and uniform discourse combined with the provocation of strong basic emotions. Later, among the various proposals on emotions in psychology, one of the most brilliant theories is that of Robert Plutchik (1980), and it is one of the most widely used in the algorithmic study of emotions associated with communication. Thus, most machine learning analyses assign emotions to communication as primary variables. Eight basic emotions are identified: happiness, trust, fear, surprise, sadness, disgust, anger and anticipation. These are not unique and can have different intensities. Combining several basic emotions gives rise to new emotions, which in turn give rise to secondary emotions: anticipation and joy give rise to optimism. Primary, secondary or tertiary dyads are formed by combining an emotion from one branch with its neighbours (fear and surprise are primary dyads, fear and sadness are secondary, and fear and disgust are tertiary). Following the order of these basic primary and secondary emotions, the disinformation industry primarily seeks to provoke these emotions in the user. For a while, it was thought that the nature of the language itself could convey emotion, but communication research, including the author’s own, shows that the message that reaches people is the one that provokes emotion, not the one that conveys emotion. In some cases, many of the messages exposed in these campaigns are designed to build trust, even admiration if it is very intense, around those they want to support. But if there is one emotion that irrationally reinforces the content, it is the presence of hate. Fear-based discourse is ephemeral because the basis of its emotion is not hated but terror, fear and submission. Hatred comes from a high-intensity aversion or disgust and is intensified by anger, leading to contempt. Hatred can also be produced by combinations such as sadness and anger. In addition, Plutchik (1980) identified certain behaviours, which he defined as “survival” behaviours, that promote emotions so that disgust or its intense version, hatred, activates rejection behaviour. The campaigns that are carried out, knowing people’s personalities through their profiles obtained from social media, make it possible to develop individualised campaigns. The messages are designed to provoke two primary emotions, depending on what you want to provoke in the person, mainly trust or hate, and an algorithm determines their intensity. As Christopher Wylie (2020) explains, the most sought-after messages in this type of campaign on social media, and the ones that persist over time, are those that achieve emotional thinking, displacing rational thinking, and hate is the emotion capable of achieving this. The intensity should not be too intense in the type of vocabulary used. Otherwise, it could be easily detected by the algorithms of the social media themselves, by fact-checkers (entities that verify the truth of news), regulatory bodies or academic researchers. In this way, it is possible to detect how the discourse is modulated in just the 55

 Spreading Organised Hate Content

right way so that the messages are not easily identifiable and persist in the network over time, as Vidgen (2019) has shown in his PhD thesis at Oxford University on Islamophobia among the ultra-right in the UK. Similarly, discourses can be established through dog-whistle politics-type messages to users: a double-meaning code language that appeals to the desired political audience without provoking anger in the opposing audience (as supporters only understand it). In recent years, researchers have begun to talk about the phenomenon of gas-lighting, a term borrowed from the 1938 film Gas Light, in which a psychological technique based on speeches that provoke strong and intense emotions in people leads them to lose their judgement, to be manipulated and to doubt their perception of reality. Several authors identify this phenomenon as a better explanation of echo chambers and disinformation (Rietdijk, 2021).

Disinformation and Hate Networking Techniques There areseveral writing techniques used to spread disinformation and hate on social media, and it is useful to know how to identify them. Their presence, which almost always involves the use of several of these techniques together, will lead to a strong suspicion of an organised campaign (Aro, 2022; Zabrisky, 2020): 1. 60/40: This technique was created by German Nazi propaganda minister Joseph Goebbels in the 1930s and 40s. It is based on presenting approximately 60% factual and 40% untrue information. 2. The Big Lie: This involves sending messages that are easily seen to be lies but which create a strong emotional charge in the people who hear them. It was Adolf Hitler’s favourite technique, to which he even devoted a chapter in his Mein Kampf. 3. Rotten herrings: This technique associates an organisation or company with scandals over time. It is one of the most commonly used techniques today. 4. Doxxing: this consists of publishing addresses and full private data of individuals or organisations on social media, together with false or emotional information. 5. Dog Whistle: this consists of sending messages whose full meaning can only be understood by those who belong to a particular group and/or have specific common knowledge. 6. Amplification of extreme voices and conspiracy theories: these are usually associated with groups that are generally conspiratorial and/or politically extremist (they are often linked) so that they become followers and amplify the result. 7. Influence by persuasion: Using data or statistical graphs altered or taken out of context to draw false conclusions. 8. Rumour-mongering: spreading rumours based on stereotypes about specific populations. 9. Gas-lighting: The emission of continuous, highly emotional messages in short periods, which cause people to doubt reality and facts, and subsequently allow any suggestion and influence. The name comes from a 1940 British film called Gas-light, which was remade in the United States in 1944 with the actress Ingrid Bergman, in which a husband ends up dominating his wife’s will through looks, gestures, mood changes or the way he treats her, always in a very emotional way. 10. Distraction or diversion: diverting the public’s attention from unfavourable news by broadcasting disasters, assassinations, terrorist attacks, and catastrophes, among others. 11. False equivalences: comparing two or more aspects that are logically similar but are not.

56

 Spreading Organised Hate Content

Attack Techniques and Countermeasures: DISARM According to the DISARM Foundation (2022), which comprises various American and European institutions that analyse social media attacks and countermeasures worldwide, the first step is to explain disinformation activities (Red Framework). 1. Planning: a. Strategic planning: target audiences and expected strategic objectives are identified. b. Target planning: this includes facilitating strategic propaganda, degrading the adversary, denying or dismissing someone, discrediting credible sources, distorting, distracting, discouraging or dividing particular groups or collectives. c. Audience analysis: audience segmentation, geographic, demographic, economic, psychological and/or political segmentation; information environment of the target audience; social media analysis and monitoring, media source assessment, trend identification; web traffic analysis; level and type of media access analysis, social and technical vulnerability identification, echo chamber identification, data vacuum (tapping before or after traditional media cover something), bias identification, loophole, suspicion or conspiracy theory identification, identification of core issues, adversarial audiences and existing media vulnerabilities. 2. Preparation: a. Narrative development: analysing existing narratives, developing competing narratives, developing and amplifying existing or creating new conspiracy theories, developing unverifiable evidence, interacting in times of crisis or live news, developing new narratives, and integrating audience vulnerabilities into the narrative. b. Content development: Creating hashtags and search points, generating information noise, creating fake research, hijacking hashtags, distorting facts, rephrasing contexts, modifying open source content (e.g. Wikipedia), reusing or plagiarising ready-made content, misleading translations or tagging, developing content supported by AI and machine learning, developing fake or altered documents, audio, images or news, developing memes or deep fakes. c. Creation of social assets: the creation of fake websites and groups on social media, cultivation of unknowing agents, preparation of petitions or fundraising campaigns through groomed characters or unknowing agents, preparation of radio and TV support, creation of fake human support accounts (trolls), screens, cyborgs and bots, creation of support networks and organisations, communities or sub-groups, infiltration of existing networks, butterfly attacks, preparation of farms and externalisation of content. d. Establishing legitimacy: creating fake experts, using academic or pseudo-scientific justifications for support, supporting legitimate accounts, creating supporting dossiers or reports, creating supporting news and sites, preparing astroturfing campaigns, creating parody accounts and sites, creating supporting accounts and influencer sites. e. Microtargeting: creating clickbait messages, buying targeted and localised ads, setting up and analysing echo chambers and bubble filters to be used to one’s advantage or creating them, exploiting empty data. f. Channel selection and affordances: the creation of polls and chats, live video or audio content, open and closed social media, use of hashtags, choice of media content to share, choice

57

 Spreading Organised Hate Content

of photos and videos, use of supporting blogs, use of diplomatic channels, use of traditional media (TV, radio or newspapers). 3. Execution: a. Priming the bomb: preparing test content, asking influencers or legitimate media to repeat the content, wrapping truth in lies or altered contexts, distorting reality, using fake experts, using the most popular words or hashtags, and using commercial systems to analyse results. b. Content spillage: broadcasting adverts, texts, images or videos in traditional media or social media, sharing memes, provoking with backlash content and/or deleting content, replying and provoking responders, and engaging traditional media. c. Maximising exposure: flooding the communication space, amplification and manipulation by trolls, hijacking existing hashtags, amplification by bots, use of spam, especially in images and with modified language and grammar, a crossover between campaigns, groups and platforms, use of marketing programmes, redirecting users to alternative platforms. d. Online harassment: censorship of political forces or traditional media, harassment of individuals or organisations, boycotting opponents, threatening or doxxing, controlling the information environment, deleting opposition accounts or blocking what they post (accusing them of violating social media rules), encouraging content sharing, inciting opposition accounts to actions that violate the content of the accounts or their content. e. Offline activity: fundraising or crowdfunding activities, organising events, paying for physical actions (protests or threats that are then used in campaigns), selling merchandising, capitalising on events, facilitating logistics, and provoking physical violence. f. Persistence of the information environment: Expanding the campaign without the need to do so artificially (requires prior entry into the network and the establishment of many contacts), hiding the identity or origin of the source of the campaign, using pseudonyms, distancing the people it is in the interest to protect from the operation, laundering accounts (by selling or swapping them), changing the name of the account, breaking the link with the content, deleting websites, shutting down or encrypting networks involved in campaigns, deleting the initial messages of the campaign, accusing others of the campaign, using servers and front organisations, payments in cryptocurrencies, broadcasting different content that changes the profile of the accounts. 4. Evaluation a. Effectiveness evaluation: analysing and measuring the effectiveness (KPIs) of the campaign’s reach by comparing performance before, during and after implementation; measuring the reach achieved with target individuals or organisations; analysing changes in the target behaviour; analysing the effectiveness and reach of the messages and media used. The following stages of countermeasures against a disinformation attack (blue framework) are specified below: 1. Define a strategic plan: define the desired end state and the conditions required to define the achievement. Tools: repairing social connections, reducing divisions, establishing the personal message, establishing positive campaigns, promoting healthy and civilised culture campaigns, supporting reliable institutions, developing monitoring, supporting civil society, building alliances, vaccinating the population or creating an anti-disinformation plan. 58

 Spreading Organised Hate Content

2. Set objectives: clearly defined, measurable and achievable. Every action should be linked to one or more objectives, should be measurable and should lead to success. Tools: Content should be defined, try to identify and block points of disinformation, provide good and reliable information, develop counter-narratives, accompany reliable and scientific characters and provide good sources to support the disinformation. 3. Set micro-targets: Identify the target population. Tools: Narrow down the target audience, to which is added advertising that supports the organisation, mentoring (elderly, young people), among others. 4. Develop content: text, images or other content is created or acquired. Tools: determining what to say and what not to say, updating fact-checks, blocking or discrediting sources of harm, normalising language, creating a competing narrative to the hoax or hostility, adding noise, honeypot, content moderation, warning banners, honeytrap, or determining the existence of an organised hostile network. 5. Channels, distribution and features: consider each social media and website that is most appropriate, considering technical capabilities, platform algorithms, terms of service, attributes, ease of use and accessibility. Tools: redirect traffic to controlled and verified sites (use of dark sites), outbidding those that create disinformation, and distraction accounts, among others. 6. Do a ‘priming of the pump’: a small general release is done as a ‘seed’ to test the response and refine the message. This is the step before the wider launch. Tools: talk to social media platforms, spread truth and transparency on official channels, warn the media against disinformation, identify accounts that retweet and attack with disinformation, hold press conferences to explain the facts, use disinformation experts, change images with new text from time to time, among others. 7. Dissemination of content: the message or information is sent to the general public or a specific population. Tools: renewing expiring campaigns (they usually do not last more than 48 hours), redirecting and detecting malware, using influencers against disinformation, using humour in the counter-narrative, not following or exchanging messages with trolls but identifying them publicly on the networks, alerting the media, etc. 8. Persistence in the information environment: establish measures to ensure the operation maintains its presence over time and avoids being taken down. Support with media, platforms and active users during the campaign. Tools: use images of lawyers and trials, videos on networks such as YouTube (it lasts longer), redirect campaigns and use media that channel, improve and continue to feed the content. 9. Evaluate effectiveness: evaluate the effectiveness of the action to review it and look for plans. Tools: measure and make statistical studies of the effectiveness of the actions over time and monitor continuously. 10. Creating social assets: creating contacts and tools in networks that allow active monitoring and the ability to send direct and indirect messages without the need for external entities, especially through specific pages or accounts in social media that will be influencers or agents of the organisation. Tools: obtaining third-party verifications, renewing images on social media and corporate websites, philanthropy and NGO sponsorship, establishing contacts with the media, administration and social entities as transparency measures, trying to divert followers from various social media, marginalising and discrediting attacking groups.

59

 Spreading Organised Hate Content

It should be noted that many disinformation campaigns are based on the so-called “cultivation of ignorant agents”, whereby certain groups join the campaign and act actively without knowing that they are being used. To this end, the “ignorant agents” are filtered and shaped beforehand by troll accounts that gradually provide false news and information or half-truths that are exposed in a self-serving manner.

Astroturfing According to Arce-García et al. (2022), one of the most common techniques used in attacks or campaigns against companies or other institutions is astroturfing technique. This technique initially came from a campaign at the end of the Second World War by British and American aviation in bombing raids on German cities, the most significant of which was Dresden. This campaign was called Thunderclap and was carried out after it was found that the German population did not lose morale after the bombing of German factories and the German army. The selective bombing was then prepared, seemingly randomly, with small, random bombing raids on streets, crossroads or parks over a short period. A few hours later, there would be a massive bombing of the whole area. This technique was transferred to the commercial sphere, where different people in different shops and places in different cities in Australia in the 1970s would buy or comment on a particular product over a short time. After a while, several people would come in and buy or order the product. All these people seemed to be ordinary people, not celebrities, but people who might resemble the local population. From then on, word of mouth was unstoppable, and sales soared. In 1975, this sales technique was banned in Australia as unethical. At the end of the 20th century, in the United States, an entrepreneur called this technique Astroturfing the name of a brand of artificial turf because, in the United States, it is associated with something that comes from people on the street as “from the grass” or “from the grassroots”. To study of Astroturfing, it is not enough to focus on finding bots since not all accounts are bots. Keller et al. (2019) recommend an identification strategy based on coordination patterns, arguing that “similar behaviour among a group of human-managed accounts is a stronger signal of a disinformation campaign than individual ‘bot-like’ behaviour” (p. 2). The current scientific challenge is to identify patterns in campaigns and attacks rather than individual actors’ behaviour. This requires longitudinal observations as well as data from multiple social media and online platforms for the analysis of the spatial dimension (Grimme et al., 2018) This technique is currently used in many disinformation campaigns and coordinated attacks against companies or individuals on social media. There are companies dedicated to what is known as black propaganda or black PR, offered by word of mouth or through the dark web (Rodríguez Fernández, 2023). This method makes artificial use of Granovetter’s (1973) theory of the strength of weak links since messages are presented by people who are known or considered similar without being close. This aspect is used to spread new messages within an already-formed group. Therefore, the introduction of messages on social media by the industry is carried out from the accounts of people who are considered known or close, who usually talk about topics and appear in common groups (geographical location, tastes, social group, work sector, leisure, opinions) to be identified as equals. It is helpful to know their structure to see how they work and to be able to detect possible patterns that reveal that they are campaigns created against someone, a company or an institution, as described by Arce García et al. (2022):

60

 Spreading Organised Hate Content

1. Distribution phase: it starts with accounts with no apparent connection between them, which appear to be housewives, sports people, city dwellers, etc. They are accounts of so-called micro or nano influencers because they do not stand out from the rest of the users. They launch a series of messages aimed at spreading an idea, a hoax, misinformation or accurate information, but from which false or erroneous consequences or conclusions are drawn. 2. Amplification phase: this begins a few minutes to an hour after the previous phase, from accounts that are also seemingly “normal” to ask the media, influencers, distribution lists or celebrities to see if they echo the campaign. They can use web portals like digital newspapers or specialised websites supporting false or offensive information if they do not get anyone. 3. Flooding phase: If the previous phases are successful, there comes the point where the networks are flooded with messages in support of these messages, usually coinciding with meal times. There may be several floods, although generally, no more than two floods are exceeded without some change to the message. A message will usually not exceed 48-72 hours without being updated. Many of these accounts are supported by automated bots. Astroturfing techniques involve influencing a group to redirect it, but there is also the butterfly attack technique, which consists of using the initial stages of astroturfing to launch messages and campaigns to destroy or discredit the group and its key influencers from within.

DISCUSSION AND CONCLUSION One of the differences between the way information moved before the advent of the Internet and the current digital age is that while there is more and more information available, in practice, paradoxically, there is less. The freedom and ability to get opinions from anywhere in the world are weighed down by the platforms’ algorithms that feed what they have identified as our tastes. The creation of so-called filter bubbles has been encouraged, where everyone only receives what feeds into their perception of the world. All this leads to the current situation where everyone has a mobile phone in their pocket with an internet connection, where it is difficult to have a different opinion or perception from others. Hate is not something new or exclusive to the Internet, as can be seen in the various chapters of this book, but there is no doubt that it has become a place for it to spread, and with an unprecedented reach, such as the lynchings of the Rohingya population in Myanmar (also known as Burma) in 2014 after false information was spread on Facebook. Disinformation is one of the ways to achieve hate, and hate leads to extreme polarisation, where it starts to provoke a spiral of silence and movement about what is allowed to do and talk about and what is not. There is a shift in what is politically acceptable, academically called an Overton window shift in public opinion. It is a new way of gaining political power. The development of technology has made it possible to automate the classification of groups or individuals or the characterisation of their personalities, all thanks to easy access to data as users share their daily lives on social media. This has allowed commercial and political marketing to spread their messages. All this has broken the problem of the first public opinion campaigns, where everyone was treated equally: today, we have reached the personalised campaign. This work has revealed various advanced techniques for planning, identifying target groups, preparing and writing techniques, monitoring and determining the results when preparing a disinformation and/ or hate campaign, and countermeasures against it. Projects like DISARM help to understand how they 61

 Spreading Organised Hate Content

work and how to counter them and will undoubtedly be the reference scheme for many future research and operations against hate and disinformation in social media and others. This phenomenon affects democracies and any system of governance. The outbreak of the Arab Spring, movements such as 15M in Spain or Occupy Wall Street in the United States, and the protests in Russia after the 2011 elections showed that the networks are a place of great power to overthrow or even threaten regimes of all kinds. Moreover, this ability to control governments from below and horizontally, initially highlighted and praised by authors such as Castells (2009) in his work Communication and Power, has become precisely the greatest threat to the control of society, coming closer to what was described in George Orwell’s novel 1984. Advances in psychology, sociology, communication techniques, artificial intelligence and computer equipment have reached the point where public and individual opinion can be known in real-time. Of course, not all population is susceptible to these statistical methods, but enough will be to sway a vote or change public opinion. The use of war communication strategies, as defined in the British Parliament, is harsh and emotionally charged, promoting hatred and displacing reason. The creation of hundreds of companies in this nascent sector worldwide today, and the use of these techniques by armies, testifies to their great power. Creating noise with lots of true, half-true and false information has proved more effective than censorship. It is to be hoped that new legislation and ethical control by the major social media and other Internet sites will prevent this plague from perverting democracy.

REFERENCES Allport, G. W. (1954). The nature of prejudice. Addison-Wesley. Arce-García, S., Said-Hung, E.; Mottareale-Calvanese, D. (2022). Astroturfing as a strategy for manipulating public opinion on Twitter during the pandemic in Spain. Profesional de la información, 31(3), e310310. doi:10.3145/epi.2022.may.10 Arce-García, S., Said-Hung, E., & Mottareale-Calvanese, D. (2022). Tipos de campaña Astroturfing de contenidos desinformativos y polarizados en tiempos de pandemia en España. Revista ICONO, 14. doi:10.7195/ri14.v21i1.1890 Arguedas, A. R., Robertson, C. T., Fletcher, R., & Nielsen, R. K. (2022). Echo Chambers, Filter Bubbles, and Polarisation: a Literature Review. Reuters Institute-University of Oxford. https://reutersinstitute. politics.ox.ac.uk/echo-chambers-filter-bubbles-and-polarisation-literature-review Aro, J. (2022). Putin’s trolls. Ig Publishing. Bourdieu, P. (2012). ¿Cómo se forma la “opinión pública”? Sociólogos-Blog de Actualidad y Sociología. https://goo.gl/lc5kuo Briant, E. (2020). Inside Cambridge Analytica and the Digital Influence Industry. Propaganda Machine. https://www.propagandamachine.tech/

62

 Spreading Organised Hate Content

Campos-Domínguez, E., & Calvo, D. (2017). La campaña electoral en Internet: Planificación, repercusión, y viralización en Twitter durante las elecciones españolas de 2015. Comunicación y Sociedad, 29. https://cutt.ly/jyHur81 Castells, M. (2009). Comunicación y poder. Alianza Editorial. Castillo Esparcia, A., & Ponce, D. G. (2015). Comunicación de crisis 2.0. Fragua. Da Empoli, G. (2019). Les ingénieurs du chaos. Jean-Claude Lattès. DISARM Foundation. (2022). Disarm blue framework. DISARM Foundation. https://disarmframework. herokuapp.com/ Frischlich, L. (2022). ‘Resistance!’: Collective action cues in conspiracy theory-endorsing Facebook groups. Impact of social media on social cohesion. Media and Communication, 10(2), 130–143. doi:10.17645/mac.v10i2.5182 Granovetter, M. S. (1973). The Strength of Weak Ties. American Journal of Sociology, 78(6), 1360–1680. https://bit.ly/3GnnRDD. doi:10.1086/225469 Grimme, C., Assenmacher, D., & Adam, L. (2018). Changing Perspectives: Is It Sufficient to Detect Social Bots? In G. Meiselwitz (Ed.), Lecture Notes in Computer Science: Social Computing and Social Media. User Experience and Behavior (pp. 445–461). Springer. doi:10.1007/978-3-319-91521-0_32 Jiménez Sánchez, F. (1994). Posibilidades y límites del escándalo político como una forma de control social. Reis. Revista Española de Investigaciones Sociológicas, 66(66), 7–36. doi:10.2307/40183715 Jones-Jang, S. M., & Chung, M. (2022). Can we blame social media for polarisation? Counterevidence against filter bubble claims during the COVID-19 pandemic. New Media & Society, 0(0). doi:10.1177/14614448221099591 Keller, F.-B., Schoch, D., Stier, S., & Yang, J.-H. (2019). Political Astroturfing on Twitter: How to coordinate a disinformation campaign. Political Communication, 37(2), 256–280. doi:10.1080/10584 609.2019.1661888 López-García, G. (2016). ‘Nuevos’ y ‘viejos’ liderazgos: La campaña de las elecciones generales españolas de 2015 en Twitter. Comunicación y Sociedad, 29(3), 149–167. doi:10.15581/003.29.35829 Martín Segovia, E. (2003). La campaña del euro (1996-2002): la Unión Monetaria Europea como objeto de comunicación política y opinión pública [Tesis doctoral, Universidad Complutense de Madrid, España]. Neubaum, G. (2022). ‘ It’s going to be out there for a long time’: The influence of message persistence on users’ political opinion expression in social media. Communication Research, 49(3), 426–450. doi:10.1177/0093650221995314 Noelle-Neumann, E. (1973). Return to the Concept of the Powerful Mass Media. Studies in Broadcasting, 9, 67–112. Pariser, E. (2011). The filter bubble. Penguin.

63

 Spreading Organised Hate Content

Plutchik, R. (1980). A general psychoevolutionary theory of emotion. In: Plutchik, Robert; Kellerman, Henry. (eds.). Emotion. Theory, research, and experience: V. 1. Theories of emotion. Academic Press, (pp. 3-33). doi:10.1016/B978-0-12-558701-3.50007-7 Pomerantsev, P. (2019). This is not propaganda. Faber. Rid, T. (2020). Active measures. The secret history of disinformation and political warfare. Profile books. Rietdijk, N. (2021). Post-truth Politics and Collective Gas-lighting. Episteme (Edinburgh), 1–17. doi:10.1017/epi.2021.24 Rodríguez Fernández, L. (2021). Propaganda digital. Comunicación en Tiempos de Desinformación. Editorial UOC. Rodríguez Fernández, L. (2023). Desinformación Y Relaciones Públicas. Aproximación a Los Términos Black PR Y Dark PR, Revista ICONO 14. Revista Científica De Comunicación Y Tecnologías Emergentes, 21 (1). . doi:10.7195/ri14.v21i1.1920 Salaverría, R., Buslón, N., López-Pan, F., León, B., López-Goñi, I., & Erviti, M.-C. (2020). Desinformación en tiempos de pandemia: Tipología de los bulos sobre la covid-19. El Profesional de la Información, 29(3). doi:10.3145/epi.2020.may.15 Sánchez-Carballido, J.-R. (2008). Perspectivas de la información en Internet. Zer, 13(25), 61-81. https:// ojs.ehu.eus/index.php/Zer/article/view/3574 Seoane, J. (2019). Opinión pública. Eunomía. Revista en Cultura de la Legalidad, 17, 235–248. doi:10.20318/eunomia.2019.5028 Shane, T., Willaert, T., & Tuters, M. (2022). The rise of ‘gas-lighting’: Debates about desinformation on Twitter and 4chan, and the possibility of a ‘good echo chamber’. Popular Communication, 20(5), 178–192. doi:10.1080/15405702.2022.2044042 Soldatov, A., & Borogan, I. (2015). The Red Web. Public Affairs Books. Tchakhotine, S. (1992). Le Viol des foules par la propagande politique. Gallimard. Vidgen, B. (2019). Tweeting islamophobia [Doctoral thesis, University of Oxford]. British Library Ethos. https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.786187 Vinuesa Tejero, M. L. (1997). Opinión pública y cultura política en la España democrática: un estudio empírico de las elecciones de 1993 (Tesis doctoral, Universidad Complutense de Madrid, España]. Williams, M. (2021). The Science of Hate. Faber&Faber. Wylie, C. (2020). Mindf*ck: Cambridge Analytica. la trama para desestabilizar el mundo. Roca Editorial. Zabriskym, Z. (2020). Big Lies & Rotten Herrings 17 Kremlin Disinformation Techniques You Need to Know Now. Byline Times. https://bylinetimes.com/2020/03/04/big-lies-and-rotten-herri ngs-17-kremlin-disinformation-techniques-you-need-to-know-no w/

64

65

Chapter 5

Approximation of Hate Detection Processes in Spanish and Other NonAnglo-Saxon Languages Juan José Cubillas Mercado Universidad Internacional de La Rioja, Spain Óscar De Gregorio Vicente Universidad Internacional de La Rioja, Spain C. Vladimir Rodríguez Caballero ITAM, Mexico

ABSTRACT In this chapter, the authors present how the use of artificial intelligence (AI) can help to identify and reduce the new digital crimes according to hate messages. The appearance of the internet in our lives, at the end of the last century, has meant a great technological advance, providing easier access to a huge volume of information and communication between people. The rise of communication-oriented networks has been such that true digital environments have been created, the so-called social networks, with millions of users all over the planet. This has meant, to a large extent, the modification of our personal relationships, and, unfortunately, the appearance of new ways of sending hate messages. The work presented is aimed at a digital tool built for this purpose for the automatic detection of hate (and non-hate) messages, in Spanish and other non-Anglo-Saxon languages, with AI algorithms, using training data from the Spanish language.

DOI: 10.4018/978-1-6684-8427-2.ch005

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

BACKGROUND The advent of the Internet has enabled the spread of social networks in all countries, particularly nonEnglish-speaking countries. According to Ethnologic study by Pereltsvaig (2020), it indicates the following population by the number of speakers (this classification is elaborated taking into account the mother tongue): Mandarin Chinese (917.8 million); Spanish (460.1 million); English (379 million); Hindi-Urdu (341.2 million); Arabic (280 million); Bengali ($228.3 million); Portuguese (220.7 million); Russian (153.7 million); Japanese ($128.2 million); Punjabi (92.7 million). However, the most spoken languages in the world, considering the second language, we have the following ranking: English (1.268 million); Mandarin Chinese (1.12 billion); Hindi (637 million); Spanish (537 million); French (280 million); Arabic (274 million); Bengali (US$ 265 million); Russian (258 million); Portuguese (252 million); Indonesian (199 million). While English is the most widely spoken, we have many other languages with very large populations. The following ranking considers only the languages we speak on the Internet: English (25,3%); Mandarin Chinese (19,8%); Spanish (8%); Arabic (4.8%); Portuguese (4.1%); Indonesian (4.1%); Japanese (3%); Russian (2.8%); French (2.8%); German (2.2%); Other (23.1%). Another significant factor to consider is that thanks to the automatic translators we now have multiple news items translated into other languages. This is the list of the most translated languages. 1. 2. 3. 4. 5.

English Spanish Chinese French German

This ranking differs slightly from the previous one. For example, French is in fourth place. This is not surprising given that French is the official language of the European Union, the United Nations and the International Court of Justice. At European level, in addition to English, languages such as Spanish, French, German, Portuguese and Italian are widely used in social networks and by the media. Another important factor to take into account is the ease with which news and comments can be translated online, with numerous tools such as: DeepL, Google translate, Wordreference, Bing Traslator, etc. Social networks are a powerful tool for connecting with people around the world and sharing information and opinions in real time. However, there have also been a number of problems associated with the overuse of social networks in non-English speaking countries. Here are some of the most common problems: •

•

66

Read of false and misleading news through social media is a common problem in many nonEnglish speaking countries. A UNESCO report notes that the spread of false information on social media has been particularly problematic during the COVID-19 pandemic in countries like Brazil, Mexico, India, and Nigeria (FreedomHouse, 2020; UNESCO, 2020). Threats to online privacy and security: social media also presents significant challenges in terms of online privacy and security. In many non-English speaking countries, regulation of social media is inadequate, and technology companies have little legal liability for misuse of users’ personal

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

•

•

data. This has led to several cases of data breaches and account hacks in countries like Brazil and Mexico (Kirchgaessner, 2022). Censorship and government restrictions: Many non-English speaking countries have restrictive government regulations regarding social media. In China, for example, authorities have blocked access to several Western social media platforms, such as Facebook and Twitter. In other countries, such as Turkey and India, restrictions have been imposed on social media in response to civil unrest and protests. Difficulties accessing content in other languages: In many non-English speaking countries, access to content in other languages is limited due to language barriers. This can limit the ability of social media users to interact with people and communities in other countries and perpetuate online fragmentation.

In summary, social networks present several problems in non-English speaking countries, from the spread of false information to censorship and unequal access to technology. It is important that individuals, organizations and governments work together to address these problems and encourage safer and more responsible use of social networks. The United Nations Organization (UN) recognizes three basic types of computer crimes: a) fraud due to computer manipulation; b) data manipulation; c) damage or modification of programs or data. And in particular, the most frequent crimes on social networks are: phishing, or fraud through corporate identity theft; identity theft; harassment and cyberbullying; defamation and slander; blackmail and threats; insults and slander to the crown; exaltation of terrorism and humiliation of the victims of terrorism. Thus, hate speech is likely to appear on social networks and digital media. Therefore, a legal framework oriented towards security in networks and information systems is necessary, for which the European Union created a Directive in 2016, known as the NIS Directive (ENISA, 2023) . This Directive is oriented towards the management of information security, notification to the authorities of particularly serious incidents, the obligation of supervision by member states and cooperation between national authorities. Based on this background, at the beginning of August 2022, the Statistical System of Criminality (CSS), which depends on the Spanish Ministry of the Interior, shared its latest figures related to so-called digital violence, that is, harassment carried out through social networks and instant messaging applications such as WhatsApp or Telegram. There was an increase of 17.5% compared to the previous year in this type of violence. In absolute terms, this means that the Ministry of the Interior recorded more than 28,000 complaints of harassment via social networks and messaging applications in 2021, compared to almost 24,000 in the previous 12 months (Ministerio del Interior. Gobierno de España, 2022). And to this official data we should add those crimes that are not quantified because they have not been reported, nor is there any data, for various reasons. In other words, there is a growing trend in terms of digital crimes and reports filed, and this trend must be halted in some way. In this sense, it is evident and necessary to have tools that support the identification and prevention of hate speech in digital environments, following the NIS Directive or any other similar type of directive. The paper by Schmitt et al., (2018) analyzes how exposure to hate speech in the media can lead to radicalization and violence. The study highlights the importance of monitoring hate speech in the media and promoting responsible and balanced coverage. Another major problem addressed by Cook et al., (2019) highlights another major problem about hate speech as being the rapid spread through social networks. Also, a report published by the Uppsala University Human Rights Institute in 2020, 67

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

highlights the negative impact of hate speech on society and emphasizes the importance of detecting and countering hate messages early (Pejchal, 2020). Finally, in 2022, a report published by the United Nations Office for the Prevention of Genocide, also highlights the importance of detecting hate speech in the media early and promoting responsible and balanced media coverage to prevent radicalization and violence (United Nations, 2022). Our work aims to detect hate messages at an early stage, which is a challenge that would allow in the first instance to control the spread of hate and discriminatory situations in society. By quickly identifying and addressing hate speech, it can be prevented from becoming more widely accepted ideas and harmful behavior towards specific groups. It also helps protect people’s human rights and prevent discrimination and harassment based on ethnicity, gender, sexual orientation, religion, disability, among others. Strengthening democracy as early detection of hate speech in the media helps to protect freedom of expression and prevent the dissemination of ideas that incite violence and discrimination. Finally, this can certainly help promote a more just and peaceful society, protecting the human rights of all people.

METHODS In this project it has been used the classical life cycle of a data mining project consisting of several main phases as mentioned in the work of Islam et al. (2018):

System Design Our system has been designed so that starting from a Dataset of messages extracted from different media, we can determine which of these messages are classified as hate and non-hate. For this we use a classic Machine Learning design. Our system starts with training data extracted from various media and where linguistic experts have correctly classified them as hate and non-hate. In a second phase, the data is analyzed to determine the need for data cleaning and balancing. Finally, various classification algorithms are tested to determine the effectiveness of each algorithm. The following is a detailed description of each of these phases.

Data Set The dataset consists of data collection using software or scrappers from the following media: La Vanguardia, ABC, 20MIN, El Mundo and El País. In the loading process, different code scripts were used to scrape hate speech from the media websites. The scrappers are developed in Python and are implemented and executed from a Docker virtual container. An improvement process was carried out by making modifications to all the scrappers. This made it possible to adapt to the changes made by the media on their websites to avoid scraping at the level of advertisements, headline structure, news body, etc. Currently, the main Spanish media do not have a public API (Application Programming Interfaces) to automatically download the information on the comments of the news made by readers, which makes it extremely difficult to analyze the information (Richardson et al., 2013). Therefore, it is necessary for the media themselves to provide these texts in order to carry out analysis of these comments, to detect, above all, those that are negative, and verify if in any case any action must be carried out because they are hate speech. Thus, in an initial phase of a project of this type, the texts are needed to make a 68

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

classification based on or labeling said texts between those that are “hateful” and “non-hateful”. That is, a dichotomous classification (Taher et al., 2021). In this way, the most efficient algorithms can be determined and trained with enough texts that come from the media (Ramezan et al., 2021). The main existing limitation, once the system is implemented and being able to determine with a high probability if a text contains a hate speech, is the fact of not being able to analyze the texts in real time due to not having the previously mentioned public API. This situation causes the following problems: • •

We cannot analyze text in real time, so we cannot make automatic decisions to delete messages or implement alerts to review said messages. Expressions and language change continuously and this means that algorithms must be continuously learning with new training data. In the absence of a real-time API, this process cannot be automated.

The main conclusion of this phase is that access to this type of public APIs would allow the automation of data collect, with the enormous advantages that this would bring we would be able to automatically detect hate speech; we would allow algorithms to have new training data to adapt their behavior to the use of the language.

Processing of the Data Data Cleaning and Expert Classification The raw data extracted from the three platforms (webpages, Facebook and Twitter) of the five media added a total of 3,043,790 data. This raw data had not been cleaned, no filters had been applied to it, so it could show empty data in the body of the news, which was an issue to be aware of. Therefore, it was necessary to carry out a large data cleaning process from which the total number of messages to be labeled by the linguistic experts was obtained, 1,045,418 data, that after the data cleaning process a database of 706,519 records was obtained as the initial dataset of Spanish digital mass media. To clean numeric data, you can use automatic truncation or condition techniques, for example. But, in the case of speeches or expressions, it is more complex and for this an initial expert classification by linguists of those expressions or terms that must be identified as “hate” is necessary. In this way, a repository is built from which the automatic system is capable of learning to identify hate speeches. In this phase, the problems encountered arise in messages that are personal reflections that can be mistaken for hate speech, when they are not. In that case, a rule-based sequential filtering system is implemented to discard these speeches. Such rules are provided by linguists. In this phase, misinformation is removed and data is standardized for a more consistent and useful view. This process is divided into the following steps: 1) Elimination of nulls and duplicates. This step is necessary to ensure that only comments containing relevant information are used. The accuracy of the model can be negatively affected by repetitive comments, in the same way, sentiment analysis does not benefit from empty comments, so these are removed; 2) Elimination of URLs, emojis and mentions to newspapers. Links and mentions to other websites and newspapers are removed as they do not provide relevant information about the sentiment 69

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

expressed in the comment. In the same way, emojis can be difficult for the model to interpret and can negatively affect the accuracy of the model, therefore these elements are removed to ensure that the text is as clear as possible; 3) Elimination of empty rows, that can appear for different reasons but do not provide relevant sentiment information, so they are removed in order to analysis can focus on relevant comments; 4) cleaning and homogenization of data as follow: convert the entire text to lower case, remove punctuation marks, eliminate numbers, remove extra whitespace, eliminate words with a length of less than two characters, removing of stop-words such as “el”, “la”, “los”, “las”, etc., are eliminated because they do not add semantic value, tokenization and lemmatization. Tokens are units of text separated by spaces or punctuation marks. Comments are divided into these tokens to count the number of words in each comment. This is useful when analyzing the frequency of words. Standardizing the text to get a more consistent and usable view of the comments as follows: In order to the expert classification, it looks for useful characteristics that can be used to train a machine learning model: a) positive word count, the number of positive words that appear in each comment is counted using a previously defined list of positive words; b) negative word count, the number of negative words that appear in each comment is counted using a previously defined list of negative words; c) count of the number of most common bigrams, which are pairs of consecutive words appearing in comments are counted. These bigrams can be useful for identifying specific patterns in the language used in comments; d) count of the number of mentions to other users, the number of times that other users are mentioned in each comment is counted. This can be useful to identify if the comment is directed to someone specifically or if it is to the public in general; e) sentiment category according to the ‘pysentimiento’ library in Spanish used to assign a sentiment category to each comment. This library uses natural language processing (NLP) techniques to analyze the text and determine if the sentiment is positive, negative, or neutral.

Data Balancing After the last explained procedure of expert classification, to create the binary model it is necessary to merge all the hate characteristics into one, leaving the distribution as follows: • •

No hate: 563,208 rows. Hate: 13,015 rows

Then, it is evident that one of the big problems we face in a project of this type is the imbalance of messages (Hasib et al., 2020). It is necessary to classify speeches as “hate” and “non-hate” and, fortunately, most speeches in media or digital platforms are “non-hate”. But this is a big problem since the target (main objective) is precisely to detect hate messages, so it is necessary to use various balancing techniques. Balancing a big data sample refers to the process of ensuring that the number of instances in each class or category of the dataset is roughly equal. This is particularly important when the dataset is imbalanced, like in this case (hate / non-hate), which means that one or more classes have a much smaller number of instances compared to others. An imbalanced dataset can negatively impact the performance of machine learning algorithms used to classified, making it difficult to train them to accurately classify instances. 70

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

There are several techniques that can be used to balance a big data sample: a) Undersampling. This involves removing instances from the majority class until the dataset is balanced. The downside of this approach is that valuable information may be lost if the majority class is underrepresented; b) Oversampling. This involves replicating instances from the minority class until the dataset is balanced. The downside of this approach is that it may result in overfitting, where the model becomes too specialized to the training data and performs poorly on new data; c) Synthetic Sampling. This approach generates new synthetic data points for the minority class to increase their representation. This technique is often used in combination with other methods to improve the accuracy of machine learning models; d) Costsensitive learning. This involves assigning different costs to misclassification of different classes, allowing the machine learning algorithm to give more weight to the minority class during training. In summary, balancing a big data sample is an important step in improving the accuracy of machine learning models, and there are several techniques available to achieve this goal. The best approach depends on the specific characteristics of the dataset and the machine learning algorithm being used. In this case, intuitively, one of the most widely used techniques is to discard “non-hate” speeches to achieve numerical parity with “hate” speeches. But this implies discarding a large volume of messages and reducing the sample size, not reaching the necessary statistical significance. Therefore, for this type of project, the generation of synthetic speeches is more efficient, which must be supervised by the team of linguists, to maintain homogeneity and avoid biases caused by these new speeches. The number of rows of the hate class is kept intact and only the number of rows of the non-hate class is modified. For this, the non-hate rows are divided into pieces of the same size as the hate sample, obtaining an exactly equal number of rows in both classes. Subsequently, a script is built that iterates over the total non-hate sample, so that models can be trained with all the available data. This last proposal is the most advantageous of the solutions because it allows maintaining a balance in the number of rows of both intensities, which can help to avoid overfitting and improve model performance.

Classification Algorithms Supervised learning is a type of machine learning where the algorithm is trained on labeled data, meaning that the input data has a corresponding output or target value. The goal of supervised learning is to learn a mapping function from input variables to output variables by optimizing a performance metric such as accuracy, precision, or recall. Then, supervised learning algorithms are used, which are those in which there is a predefined target variable based on labeled input and output data. Historically, supervised learning algorithms have a long history of use and have been instrumental in many significant advances in machine learning and artificial intelligence. The early development of supervised learning algorithms can be traced back to the 1950s and 1960s when pioneers such as Arthur Samuel and Frank Rosenblatt began exploring the use of machine learning algorithms to perform tasks such as game playing and image recognition. One of the earliest successful applications of supervised learning algorithms was in speech recognition, where hidden Markov models (HMMs) were used to model the probability distribution of speech sounds. This led to the development of commercial speech recognition systems in the 1980s and 1990s. In the 1990s and early 2000s, there was a significant growth in the use of supervised learning algorithms in fields such as finance, marketing, and healthcare. Considering this little historical summary, in this case, oriented towards classification, dividing the space of predictors (independent variables) into different and non-overlapping regions: random forest 71

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

(RF), gradient boosting (GB), support vector machine (SVM), multilayer perceptron (MLP), naive bayes (NB)). Each of the above algorithms approaches the problem in a different way: Random Forest (RF) is a supervised learning algorithm based on decision trees (Gislason et al., 2006). This algorithm creates multiple decision trees, each trained with a random sample of the original data set. It then combines the predictions from the trees to produce a final output. RF is robust to missing data and noise in the data. Gradient Boosting (GB) is a supervised learning algorithm based on decision trees that builds a model iteratively (Li et al., 2007). In each iteration, the model attempts to correct the errors of the previous model. GB is very effective for structured and unstructured data, but requires more time to train than RF. Support Vector Machine (SVM) is a supervised learning algorithm that classifies data by finding the best possible separation between different classes in the feature space (Bhavsar & Panchal, 2012). SVM focuses on maximizing the separation between different classes, which makes it very effective in linear and nonlinear classification problems. However, it can have problems when faced with large and complex datasets. Multilayer Perceptron (MLP) is an artificial neural network that uses multiple hidden layers to learn nonlinear relationships between data features (Chaudhuri & Bhattacharya, 2000). MLP is highly effective on complex, nonlinear classification problems, but can require significant training time and suffers from overfitting if not properly tuned. Naive Bayes (NB)) approaches classification based on Bayes’ theorem to classify data(Leung, 2007. This algorithm assumes that all features in the data set are independent of each other, which can be an unrealistic assumption. NB is simple and fast, and works well on high-dimensional data sets. As we have seen each algorithm approaches the problem in a different way but their efficiency depends on the training data set. Each of these algorithms has its own strengths and weaknesses. As a complement, unsupervised learning algorithms are used, which are those in which human intervention is minimal, oriented to those known from Deep Learning, such as recurrent neural networks and Long-Short Term Memory (LSTM) which are an extension of these (Zhou et al., 2016). In the training process the model parameters are fitted to the training data in such a way that the model can generalize and make accurate predictions about new data. In this case, it has been concluded that the model that provides the best performance and precision is Gradient Boosting with the following hyperparameters: n_estimators = 100; max_depth = 3; learning_rate = 0.1; max_features = ‘sqrt’.

System Architecture With Airflow Apache Airflow is an open-source tool and one of the main ones integrated into the main cloud environments on the market, to create, schedule and monitor workflows through code. That is, it allows you to programmatically automate tasks of the different defined workflows. Each of these flows is defined as a directed acyclic graph (DAG) of smaller tasks, in such a way that to execute them we can use predefined operators or develop them on demand in a simple way. The main components of the Airflow architecture are: •

72

Web server: The web server is the user interface of Airflow. It provides a web-based interface for users to manage workflows, monitor task execution and view logs.

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

• • • • •

Scheduler: The scheduler is responsible for scheduling and triggering workflows based on their dependencies and schedule. It communicates with the metadata database to retrieve information about workflows and tasks to be executed. Metadata database: The metadata database stores information about workflows, tasks and their dependencies. It is used by the scheduler to schedule tasks and track their execution. Executor: The executor is responsible for the execution of tasks. It communicates with the scheduler to retrieve information about the tasks to be executed and with the workers to execute those tasks. Workers: Workers are the machines where tasks are executed. They receive tasks from the executor and communicate the task status to the executor. Plugins: Plugins are used to extend the functionality of Airflow. They can be used to add new operators, sensors, hooks and other components to Airflow.

The Airflow architecture is based on a web server that provides the user interface, a relational metadata store (MySQL or PostreSQL database), a storage space for the DAG files, a task scheduling scheduler, a broker to distribute them and the workers, which execute them. It is a flexible architecture that can be implemented in different ways, both in virtual machines and through managed services. In resume, it is an architecture widely used in digital environments in the cloud, and with an accumulated development since 2015 that position it as a benchmark. Therefore, it is perfectly applicable in this project. Airflow has been used in numerous scientific works due to its great advantages such as: easy scalability of workflows by adding or removing nodes as needed (Harenslak & Ruiter, 2021; Lagos et al., 2022; Najah Ahmed et al., 2019). It is very flexible as it allows us to configure new workflows to adapt to the specific needs of each project. Code reuse. Easily monitor the progress of workflows and detect errors in real time. Integration with other cloud services and also has a large active community of developers and users who provide support and contribute to the development of new features.

CASE STUDY: SPANISH MASS MEDIA The data have been extracted from different social networks such as Twitter or Facebook from which comments have been extracted referring to several Spanish media such as: La Vanguardia, El Mundo, ABC, El País and 20 Minutos. on their websites. These messages are then rated on a scale of hate that varies from 0 to 6, with 0 being the absence of hate and a value greater than 1 signifying hate. The value from 1 to 6 measures the intensity of the hate, with the highest hate rating being 6 and 1 being the lowest hate rating.

RESULTS In our study, the algorithm that best performs the classification is SVM as it is the one that makes the least error when analyzing its confusion matrix (Mathur & Foody, 2008). The SVM (Support Vector Machine) classification algorithm has several strengths, such as greater effectiveness in high dimensionality spaces, which means it can handle large datasets with many features (Cortes & Vapnik, 1995; Shigeo Abe, 2005). It is useful for machine learning applications where the 73

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

number of features is much larger than the number of observations. Flexibility in the choice of kernel functions which allows it to adapt very well to each type of problem(Bahlmann et al., 2002). Efficient handling of large datasets as it uses a subset of support vectors to create the classification function, making it capable of handling large datasets efficiently. It is able to find the global optimal solution for the classification problem, which means that the classification function is the best possible in terms of class separation (Jian et al., 2017). have high error tolerance as it allows tolerating some errors in the training data, which means that it can handle noisy or incomplete data without significantly compromising its performance (Drid et al., 2022). Because of all these features, SVM is the most efficient algorithm in our problem. The initial dataset of Spanish digital mass media is 706,519 records (Twitter-358,399; Web-348,120) (El Mundo 217,396; El País 140,865; 20Minutos 157,149; ABC 138,271; La Vanguardia 52,838). The cleaned dataset is 574,760 records (Twitter-345,776; Web-228,984). The training dataset is 706,519 records. In short, there is a reasonably large sample size to have a meaningful sample. The dataset is balanced among the different media except for LA VANGUARDIA. In terms of a binary classification: Hate / No hate it is obtained an Accuracy of 90% for the Gradient Boosting algorithm, with the following evaluation metrics: Figure 1. Evaluation metrics

In Figure 1 we can see several key metrics to analyze the effectiveness of the algorithm such as the classification accuracy of the Hate and Non-hate model. In addition, the confusion matrix by Galar et al., (2012), Powers (2020), and Provost and Fawcett (2013)is showed in Figure 2:

74

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

Figure 2. Confusion matrix

Figure 2 shows that the confusion matrix is very useful for assessing the effectiveness of a classification model based on machine learning. We can explicitly see when one class is confused with another, which allows us to analyze the error made.

DISCUSSION AND CONCLUSION The changes brought about by the emergence of new forms of communication through social networks have had a significant impact on various aspects of society. On the one hand, they have driven progress in terms of connectivity and interaction between people around the world, leading to a wider dissemination of ideas, opinions and knowledge. However, new threats have also emerged in the digital realm, including crime and violence that did not exist before. To address these problems, it is essential to follow the guidelines established by international and European bodies and to develop automatic systems capable of detecting hate and violence in Spanish and other non-English languages. The study and application of these systems is not only important from a practical point of view, but also from an academic perspective, as it can help to better understand the complexity of digital communication and to improve education in this field.

75

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

In summary, while new types of online communication have been a breakthrough in many respects, they have also created new challenges that need to be addressed through the development of automatic hate and violence detection systems in Spanish and other non-English languages. Therefore, the first conclusion is that the accuracy of the binary classifier (hate / not hate) is the 90% and can better as the algorithm increases the number of replications and continues to learn with more data. With bigger databases as inputs, the goodness of machine learning increases and so does the accuracy. Then, it is reasonable to think that the percentage of accuracy will exceed 90% in a short time. Technology is rapidly changing the way people consume information and communicate. In the context of digital media, it is increasingly common for users to post comments and opinions online, and often these expressions can be negative, offensive or even constitute hate speech. These situations can have a significant impact on the image of brands or media and affect the reputation and prestige of companies. To address this problem, it is necessary to have effective tools that allow content managers to monitor online conversations and detect inappropriate messages in real time. In this respect, the aforementioned system can be of great help, as it provides content managers with the possibility to analyze and act accordingly in case of hate messages or inappropriate content being posted. In addition, the system also allows media content managers to analyze the impact of each news item or topic published in their media. This can be especially useful to understand how users are reacting to certain news or content, and adjust editorial strategy accordingly. In short, having systems in place to detect hate speech and analyze the impact of news and content in digital media is essential to protect the image and reputation of brands and companies, and to improve the quality of content offered to online users.

FUTURE RESEARCH DIRECTIONS Hate speech detection and classification algorithms are becoming increasingly important in today’s digital age, as hate speech and online harassment continue to be major issues on social media platforms. Therefore, the generalization of this type of algorithms, for any language, with the aim of being used by all state security forces and bodies, interior ministries, and defense ministries, to minimize radicalization movements derived from hate, racism and xenophobia messages. Likewise, the implementation of these tools in the digital media to identify those that constitute a crime, as a way of ensuring respectful communication between people. And, in social networks. Some of the key trends and developments that are likely to shape the future of hate speech detection algorithms include: 1) multimodal analysis, because hate speech is not limited to text-based content, and future hate speech detection algorithms are likely to incorporate analysis of other modalities such as images, videos, and audio; 2) contextual analysis, considering that the meaning of hate speech can vary greatly depending on the context in which it is used. Future hate speech detection algorithms will need to take into account the specific context in which the speech is used to accurately identify hate speech; 3) different domain-specific analysis, considering that politics, sports, and entertainment, have their own specific language and cultural norms. Future hate speech detection algorithms will need to be tailored to specific domains to effectively detect hate speech; 4) human-in-the-loop approaches, because while machine learning algorithms are powerful tools for hate speech detection, they are not perfect and, therefore, it is necessary for linguistic experts to label correctly, review the contents and create dictionaries of words and expressions, taking into account the characteristics of each language, 76

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

to improve the accuracy of the algorithms. So, future hate speech detection algorithms must incorporate human-in-the-loop approaches, and these algorithms must be constantly trained and updated. In addition, the future of hate speech detection algorithms is likely to involve a combination of machine learning, natural language processing, and social media analysis techniques. For example, the unification of these algorithms with Natural Language Processing (NLP) techniques and the increasingly widespread GPT Chat, that is an artificial intelligence chatbot prototype developed in 2022 by OpenAI that specializes in dialogue, and is a great language model, fine-tuned with both supervised and reinforcement learning techniques.

ACKNOWLEDGMENT Project PID2020-114584GB-I00, financed by the Spanish State Research Agency - Ministry of Science and Innovation. Spanish Government.

REFERENCES Bahlmann, C., Haasdonk, B., & Burkhardt, H. (2002). Online handwriting recognition with support vector machines—A kernel approach. Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition, (pp. 49-54).IEEE. 10.1109/IWFHR.2002.1030883 Bhavsar, H., & Panchal, M. H. (2012). A review on support vector machine for data classification. [IJARCET]. International Journal of Advanced Research in Computer Engineering and Technology, 1(10), 185–189. Chaudhuri, B. B., & Bhattacharya, U. (2000). Efficient training and improved performance of multilayer perceptron in pattern classification. Neurocomputing, 34(1), 11–27. doi:10.1016/S0925-2312(00)00305-2 Cook, C., Conijn, R., Schaafsma, J., & Antheunis, M. (2019). For Whom the Gamer Trolls: A Study of Trolling Interactions in the Online Gaming Context. Journal of Computer-Mediated Communication, 24(6), 293–318. doi:10.1093/jcmc/zmz014 Cortes, C., & Vapnik, V. (1995). Support-Vector Networks. Machine Learning, 20(3), 3. doi:10.1007/ BF00994018 Drid, A. B. S., Abdelhamid, D., & Taleb-Ahmed, A. (2022). Support vector machine based clustering: A review. 2022 International Symposium on iNnovative Informatics of Biskra (ISNIB), (pp. 1-6). IEEE. 10.1109/ISNIB57382.2022.10076027 ENISA. (2023). NIS Directive. European Union Agency for Cybersecurity [Topic]. ENISA. https://www. enisa.europa.eu/topics/cybersecurity-policy/nis-directive-new FreedomHouse. (2020). The Pandemic’s Digital Shadow. Freedom House. https://freedomhouse.org/ report/freedom-net/2020/pandemics-digital-shadow

77

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

Galar, M., Fernandez, A., Barrenechea, E., Bustince, H., & Herrera, F. (2012). A Review on Ensembles for the Class Imbalance Problem: Bagging-, Boosting-, and Hybrid-Based Approaches. IEEE Transactions on Systems, Man, and Cybernetics. Part C, Applications and Reviews, 42(4), 463–484. doi:10.1109/ TSMCC.2011.2161285 Gislason, P. O., Benediktsson, J. A., & Sveinsson, J. R. (2006). Random Forests for land cover classification. Pattern Recognition Letters, 27(4), 294–300. doi:10.1016/j.patrec.2005.08.011 Harenslak, B. P., & Ruiter, J. R. (2021). Data Pipelines with Apache Airflow. Manning Publications. Hasib, K. M., Iqbal, M. S., Shah, F. M., Mahmud, J. A., Popel, M. H., Showrov, M. I. H., Ahmed, S., & Rahman, O. (2020). A Survey of Methods for Managing the Classification and Solution of Data Imbalance Problem. Journal of Computational Science, 16(11), 1546–1557. doi:10.3844/jcssp.2020.1546.1557 Islam, M. S., Hasan, M. M., Wang, X., Germack, H. D., & Noor-E-Alam, M. (2018). A Systematic Review on Healthcare Analytics: Application and Theoretical Perspective of Data Mining. Health Care, 6(2), 2. doi:10.3390/healthcare6020054 PMID:29882866 Jian, Y., Huang, D., Yan, J., Lu, K., Huang, Y., Wen, T., Zeng, T., Zhong, S., & Xie, Q. (2017). A Novel Extreme Learning Machine Classification Model for e-Nose Application Based on the Multiple Kernel Approach. Sensors (Basel), 17(6), 6. doi:10.339017061434 PMID:28629202 Kirchgaessner, S. (2022, octubre 4). Mexico: Reporters and activists hacked with NSO spyware despite assurances. The Guardian. https://www.theguardian.com/world/2022/oct/04/mexico-nso-spy ware-journalists-human-rights-hacked-pegasus Lagos, A., Caicedo, J. E., Coria, G., Quete, A. R., Martínez, M., Suvire, G., & Riquelme, J. (2022). State-of-the-Art Using Bibliometric Analysis of Wind-Speed and -Power Forecasting Methods Applied in Power Systems. Energies, 15(18), 18. Advance online publication. doi:10.3390/en15186545 Leung, K.M. (2007). Naive Bayesian Classifier. Polytechnic University Department of Computer Science/Finance and Risk Engineering, 123-156. Li, P., Wu, Q., & Burges, C. (2007). McRank: Learning to Rank Using Multiple Classification and Gradient Boosting. Advances in Neural Information Processing Systems, 20. https://proceedings.neurips.cc/ paper_files/paper/2007/hash/b86e8d03fe992d1b0e19656875ee557c-Abstract.html Mathur, A., & Foody, G. M. (2008). Multiclass and Binary SVM Classification: Implications for Training and Classification Users. IEEE Geoscience and Remote Sensing Letters, 5(2), 241–245. doi:10.1109/ LGRS.2008.915597 Ministerio del Interior, Gobierno de España. (2022). Portal estadístico de criminalidad. https://estadisticasdecriminalidad.ses.mir.es/publico/portalestadistico/ Najah Ahmed, A., Binti Othman, F., Abdulmohsin Afan, H., Khaleel Ibrahim, R., Ming Fai, C., Shabbir Hossain, M., Ehteram, M., & Elshafie, A. (2019). Machine learning methods for better water quality prediction. Journal of Hydrology (Amsterdam), 578, 124084. doi:10.1016/j.jhydrol.2019.124084 Pejchal, V. (2020). Hate Speech and Human Rights in Eastern Europe: Legislating for Divergent Values. Routledge. doi:10.4324/9781003005742

78

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

Pereltsvaig, A. (2020). Languages of the World. Cambridge University Press. doi:10.1017/9781108783071 PowersD. M. W. (2020). Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation. CoRR, abs/2010.16061. https://arxiv.org/abs/2010.16061 Provost, F., & Fawcett, T. (2013). Data Science for Business: What You Need to Know about Data Mining and Data-Analytic Thinking. Ramezan, C. A., Warner, T. A., Maxwell, A. E., & Price, B. S. (2021). Effects of Training Set Size on Supervised Machine-Learning Land-Cover Classification of Large-Area High-Resolution Remotely Sensed Data. Remote Sensing (Basel), 13(3), 3. doi:10.3390/rs13030368 Richardson, L., Amundsen, M., & Ruby, S. (2013). RESTful Web APIs: Services for a Changing World. O’Reilly Media, Inc. Schmitt, J. B., Rieger, D., Rutkowski, O., & Ernst, J. (2018). Counter-messages as Prevention or Promotion of Extremism?! The Potential Role of YouTube: Recommendation Algorithms. Journal of Communication, 68(4), 780–808. doi:10.1093/joc/jqy029 Shigeo, A. (2005). Support Vector Machines for Pattern Classification. Springer. doi:10.1007/1-84628219-5 Taher, K. I., Abdulazeez, A. M., & Zebari, D. A. (2021). Data Mining Classification Algorithms for Analyzing Soil Data. Asian Journal of Research in Computer Science, 17-28. doi:10.9734/ajrcos/2021/ v8i230196 UNESCO. (2020). COVID-19—Fighting ‘infodemic’ and social stigma through community media in India. UNESCO. https://www.unesco.org/en/articles/covid-19-fighting-infodem ic-and-social-stigma-through-community-media-india United Nations. (2022). United Nations Office on Genocide Prevention and the Responsibility to Protect. Office on genocide prevention and the responsability to protect. https://www.un.org/en/genocideprevention/hate-speech-strategy.shtml Zhou, P., Shi, W., Tian, J., Qi, Z., Li, B., Hao, H., & Xu, B. (2016). Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (vol. 2, pp. 207-212). ACL. 10.18653/v1/P16-2034

KEY TERMS AND DEFINITIONS Accuracy: If a binary model has an accuracy of 90%, it means that 90% of the predictions are correct. If it is assumed that the model has a balanced performance between accuracy and recall, then accuracy and recall should also be close to 90%. In this case, the F1-score is also at 90%, as can be seen in the previous table. AI: Artificial Intelligence is a broad field of computer science that focuses on the development of intelligent machines that can perform tasks that typically require human intelligence, such as visual perception, speech recognition, decision-making, and language translation. The main idea is that machines

79

 Hate Detection Processes in Spanish, Other Non-Anglo-Saxon Languages

can learn from data, identify patterns, and make decisions based on those patterns. This is done using a combination of machine learning algorithms, deep learning techniques, and other statistical methods. AI is used in a wide range of applications, including natural language processing (NLP, see below), image and speech recognition, autonomous vehicles, and robotics. Confusion Matrix: This shows the predictions made by the model comparing them with the actual results, which allows knowing how many times the model is correct and how many times it is wrong in each of the classes that are being evaluated. In technical terms, a confusion matrix shows good results when high precision and sensitivity are observed in the classification of the data. That is, when most of the predictions made by the model are correct and there is a minimum number of false positives and false negatives in the data classification. F1-Score: The F1-score is a measure that combines both precision and recall and is defined as the harmonic mean between both values. FN: This means false negative. FP: This means false positive. NIS Directive: Ordered Attribute: European Union. “Directive (EU) 2016/1148 of the European Parliament and of the Council, of July 6, 2016”. Official Gazette of the European Union. NLP: Natural Language Processing is a subfield of artificial intelligence (AI) that focuses on the interaction between computers and human language. The goal of NLP is to enable computers to understand, interpret, and generate human language in a way that is useful to humans. NLP involves a wide range of techniques and methods, including machine learning, statistical analysis, and computational linguistics. Precision: Precision is defined as the ratio of true positives (TP) over all predicted positives (TP + FP). That is, it measures the accuracy of positive predictions. Recall: Recall, also known as sensitivity, is defined as the ratio of true positives (TP) over all true positives (TP + FN). That is, it measures the ability of the model to detect all the real positives. TN: This means true negative. TP: This means true positive.

80

81

Chapter 6

The Interaction Between Offensive and Hate Speech on Twitter and Relevant Social Events in Spain Jesús Gómez Spanish National Office Against Hate Crimes, Spain

Carlos J. Máñez Spanish National Office Against Hate Crimes, Spain

Alberto Matilla-Molina Spanish National Office Against Hate Crimes, Spain

Tomás Fernández-Villazala https://orcid.org/0000-0002-3726-4363 Spanish National Office Against Hate Crimes, Spain

Ma. Pilar Amado Spanish National Office Against Hate Crimes, Spain Dimosthenis Antypas Cardiff University, Great Britain Jose Camacho-Collados https://orcid.org/0000-0003-1618-7239 Cardiff University, Great Britain

Alicia Méndez-Sanchís Autonomous University of Madrid, Spain Javier López State Secretariat for Security, Ministry of Interior, Spain

ABSTRACT Hate speech is one of the major concerns of Europe. Different studies, mainly in English language, have been carried out to analyze hate speech, many of them from a theoretical perspective. Here, it is presented an observational study about hate speech poured on Twitter in Spanish regarding to five social important events: Women´s Day, International LGTBQ+ Pride Day, Spain National Day, national elections, and regional elections. Three different experiments were carried out; two used deep learning algorithms to

DOI: 10.4018/978-1-6684-8427-2.ch006

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

automatically classify tweets, meanwhile, the latest tweets were classified by a human. Results showed that these events significantly triggered hate speech, yet results differed between experiments, and also depending on the nature of the events. A better understanding of the mechanisms of hate speech propagation can help improve policies in Spain or in countries with similar characteristics, and thus help law enforcement and other institutions to address the scourge of hate crimes.

INTRODUCTION Hate crimes in the European Union are a growing concern as they violate fundamental rights protected by the Universal Declaration of Human Rights (Anderson & Meyer, 2016). Social media have facilitated the spread of offensive and hate speech, with users hiding behind the right to freedom of expression (Whitney, 2018; Arcila et al., 2022). Most research has focused on message text analysis in English, with fewer studies in Spanish or other languages (Aluru et al., 2020; Basile et al., 2019; Battistelli et al., 2020; del Valle-Cano et al., 2023; Florio et al., 2017; Pereira-Kohatsu et al., 2019; Plaza-del Arco et al., 2021; Poletto et al., 2021; Sreelakshmi et al., 2020). In Spain, over 22% of 2021’s internet hate crimes occurred on Social media, with rising percentages (López-Gutiérrez et al., 2021). Hate speech on these platforms has been identified as a key driver of hateful attitudes and hate crimes (Amores et al., 2021). Triggered by social events, research has shown a connection between online hate speech and offline hate crime (Williams and Burnap, 2016; Olteanu et al., 2018; Relia et al., 2019; Müller & Schwarz, 2020, 2021; Scharwächter & Müller, 2020; Williams et al., 2020; Arcila et al., 2022). This blurs the boundary between online and offline spaces, making hate speech an extension of the physical environment (Awan & Zempi, 2016). Negative impacts of offensive comments on victims’ mental health include increased stress, anxiety, and depression (Baumgartner et al., 2018). Online platforms have implemented measures to control hate speech, but regulation procedures are not standardized, leading to varied criteria across Social media (Berger & Morgan, 2016). These regulations often stem from platform policies rather than legislation, impacting freedom of expression, privacy, and community building (Gillespie, 2018). Consequently, systems on Social media can reflect and perpetuate social inequalities, especially regarding gender and race, disproportionately affecting vulnerable groups and amplified by platform characteristics (Seeta, 2019). Incitement to hatred on the Internet may not constitute a crime punishable by law depending on which country is based on the targeted group, but it has been studied to determine their behavior on Social media. In 2018, researchers from Stanford University analyzed more than 800,000 publications, observing that most people do not use offensive language. However, a small number of users were the ones who generated the most harmful and offensive content, suggesting that pernicious speech online is a phenomenon driven by cliques (Davidson et al., 2018). In the same vein, in 2017, it was found that offensive online slang tends to be more prevalent in online communities with strong group identities and social cohesion and is often used to reinforce group membership (Cheng et al., 2018). On the other hand, another research conducted at Yale University during the same year found that offensive comments poured on Social media can have a negative effect on how people perceive others, even if they are not the direct target of the comments, as participants generally rated the victims of unfavorable messages as less competent and intelligent (Moussaïd et al., 2017).

82

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

On the other hand, hate crime, which includes hate speech, has been contextualized in terms of temporal circumstances to elucidate the probability of the occurrence of criminal behavior about significant social events (Meijer & Wessels, 2019). This approach is raised from the perspective of environmental criminology, which argues that criminal behavior should be analyzed from the convergence between offenders, victims and a law broken in a given geographic-temporal space (Brantingham & Brantingham, 1991). Within this perspective is framed the criminological theory of routine activities, which examines how the convergence of space and time can be a key factor in the opportunity to commit a crime successfully (Cohen & Felson, 2018). Following this theoretical framework, hate crimes suffer the influence of temporality, particularly highlighting relevant social events, as pointed out by King and Sutton (2013). These researchers showed that the escalation in the number of hate crimes committed occurred almost immediately after the occurrence of a social event with significant relevance, observing how the tension decreased at the same rate as its occurrence. Moreover, these increases were of short duration. On the other hand, a study that used time series analysis with anti-immigration discourse suggested that, in the face of some events, such as terrorist attacks, there was an increase in messages promoting intolerance against immigrants (Hopkins, 2010; Legewie, 2013). This influence has been analyzed to determine the exact moment, within the time frame of events with socio-political repercussions, when this type of crime experiences a notorious increase. Thus, in the case of terrorist attacks (Borell, 2015; Deloughery et al., 2012; Disha et al., 2011; Echebarria-Echabe & Fernandez-Guede, 2006; King & Sutton, 2013; Hanes & Machin, 2014; Mills et al., 2017; Ivandic et al., 2019; Jacobs & van Spanje, 2021; Piatkowska & Lantz, 2021; Piatkowska & Stults, 2021), increases were recorded on the same day of the attack (King & Sutton, 2013); after two to three days (Jacobs & van Spanje, 2021; Piatkowska & Stults, 2021) or within a week (Deloughery et al., 2012), becoming significant after one (King & Sutton, 2013; Hanes & Machin, 2014; Piatkowska & Lantz, 2021) or two months (Piatkowska & Lantz, 2021). For their part, Brexit-triggered hate crimes mostly showed an increasing trend in the weeks following the event remaining significant until several days after and even months later (Albornoz et al., 2018; Carr et al., 2020; Devine 2018, 2021; Piatkowska & Lantz, 2021; Piatkwoska & Stults, 2021). On the other side, In Donald Trump’s election campaign, the highest record occurred in the same month as the election (Edwards & Rushin, 2018; Müller & Schwarz, 2020; Warren-Gordon & Rhineberger, 2021; Feinberg et al., 2022). Similarly, some of these studies concluded that the increases triggered by such an event were also manifested in periods prior to its celebration (Devine, 2018; Carr et al., 2020), which warns of its probable prior occurrence and of the importance of considering these circumstances when combating this phenomenon of great social relevance. In this vein, predictive policing has oriented its strategies to the application of anticipatory tools that allow them to prevent and provide an effective response to crime hotspots through the identification of criminal repetitions to outline a common pattern of the temporal distribution of crime (Bachner, 2013; González-Álvarez et al., 2020; Hardyns & Rummens, 2018; Perry et al., 2013). The predictive efficiency of the procedure dynamics is based on two central aspects: the high degree of concentration of crime in a specific area and its stability over time, which makes it a predictable phenomenon (Weisburd & Telep, 2011). Research that has employed this method to study the temporal distribution of different criminal typologies has been very diverse and with one purpose: to develop measures that ensure the control and reduction of different phenomena through a correct distribution of police resources to the most conflictive areas (Mohler, 2014). Crimes such as gender 83

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

violence, homicide or robbery are some examples that have subjected this type of quantitative analysis in the development of a geographical profile that allows predicting their occurrence(Herrman, 2015; Monárrez & Gómez, 2013; Moreno et al., 2021; Serra et al., 2022; Valente, 2019; Van Patten et al., 2009; Yar & Nasir, 2016; Xu & Wen, 2013). For many years, academic interest has been oriented towards determining the temporal course that a specific criminal behaviour adopts. It is no longer a question of defining only the places with the most remarkable criminal tendency at a specific time but instead of determining how the criminal phenomenon evolves, clarifying at what times there is a greater incidence of the type of crime under investigation and justifying the reason for its increase. For all of this, it is also important to study the spatiotemporal distribution of crimes that are committed on the Internet so that police resources could be better employed. This study analysed the relationship between significant social events, like celebrations advocating tolerance and equality or democratic elections, and hate speech on social media. Data mining, deep learning, and human monitoring techniques were used to examine Twitter messages during specific periods. The expectation was to find increased offensive and hate speech around these events, with a peak on the event day and a gradual decrease afterwards. By understanding hate speech distribution, public strategies can be better designed and implemented by institutions such as the Spanish National Office Against Hate Crimes (Oficina Nacional de Lucha Contra los Delitos de Odio, ONDOD) to efficiently address this issue and improve the safety and welfare of affected individuals and vulnerable groups.

MATERIALS AND METHODS To study offensive and hate speech (hereafter used as “intolerant speech” too), we designed three experiments on Twitter. This social network was chosen because it is one of the leading Social media in the world where personal opinions on current affairs are expressed not only by the general public but also by world leaders and companies (Currant et al., 2011; Nugyen, 2018; Duncombe, 2019). Besides, Twitter is mainly based on text, which makes the use of automatic classification tools based only on natural language processing (NLP) easier. On the other hand, in this study, tweets were published in Spanish and geolocated in Spain. These three experiments were conducted concerning five relevant events in Spain between 2014 and 2022 (Table 2): Women’s Day, International LGTBQ+ Pride Day, Spain National Day and relevant dates related to the celebration of the national (only three elections: 2015, 2016 and 2019) and regional elections (eleven elections between 2015 and 2022). These events were selected to represent different vulnerable groups. It is important to consider that these concrete events may not affect similarly to all the vulnerable groups protected by the Spanish Penal Code because of different discriminatory motivations. For example, hate speech poured out on Women’s Day is likely to be more related to the “sex/ gender” motivation than, for example, because aporophobic motivation. At the same time, International LGTBQ+ Pride Day is more likely to affect the “sexual orientation/gender identity” motivation than racism/xenophobia. On the other hand, it is possible that more offensive speech based on “ideology” is found on Spain’s National Day or election days. However, in the five events, multiple vulnerable groups may be affected since this type of crime are based on prejudices, a consequence of social categorization, so that the target group of intolerant behavior may be extended to other, as in the case of multiple discrimination.

84

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

For each event, a time range of 29 days was selected, two weeks before the specific date, the day of the exact day of the event and two weeks after. This vast time range was chosen to collect information not only on the days close to the event but also to obtain more information about the trends outside the event and thus be able to compare the “normal” traffic of offensive and hate speech with that which occurs in relevant social events. Once downloaded, tweets were classified using an automatic text classification tool based on NLP in the first two experiments. There are several classification tools for Spanish based on different machine learning or deep learning algorithms and we evaluate three different approaches to select the best hate speech classifier (Benítez-Andrades, 2022; Del Valle-Cano et al., 2023; Pereira-Kohatsu et al., 2019; Plaza-del Arco et al., 2021). Mainly we test (1) pysentimiento, which is based on Robertuito and trained on a large corpus of Spanish tweets; (2) twitter-xlm-roberta-base (i.e. t-xlm)) a multilingual model trained on tweets from over thirty languages, including Spanish; and (3) Bertin which is trained on the Spanish portion of mC4 () dataset (Barbieri et al., 2022, Pérez et al. 2021a, 2022b; Xue et al., 2020). All three classifiers utilised are transformer-based on language models based on RoBERTa (Liu et al.,2019). RoBERTa models are deep neural networks of 12 layers and utilise techniques such as attention masks and dynamic tokens masking during training to understand language relations (Vaswani et al., 2017). By training them in large text corpora, the models create accurate representations of words by mapping them into high-dimension embedding vectors. For our experiment, we fine-tuned the t-xlm and bertin models for hate-speech detection by utilizing the HateEval and HaterNet datasets (Basile et al., 2019; Quijano-Sánchez et al., 2019). The pysentimento tool we utilised was already trained on the HateEval data, and we refrained from further training. Both datasets provided a collection of manually annotated Spanish tweets where the annotators were asked to indicate the presence or not of hate speech. A significant difference between the two datasets is that HateEval includes only tweets related to immigrants and women, while HaterNet includes other types of hate speech (e.g., hate towards religion or gender). The predefined test split of HateEval and a custom split of HaterNet were used to evaluate our models. Table 1 displays a detailed distribution of tweets across the training and test sets used for the two datasets, where a slight imbalance between the two classes (66% hateful to 34% non-hateful tweets) can be observed. This data imbalance does make the task difficult, but it is essential to develop robust classifiers as in the real world, the presence of hate speech in tweets is even rarer, approximately 1% of tweets (Quijano-Sánchez et al., 2019). Table 1. Class distribution of tweets across sets in HateEval and HaterNet datasets Data HateEval HaterNet

Class

Training & Validation

Test

Hateful

2,079

660

Non-Hateful

2,921

940

Hateful

1,262

305

Non-Hateful

3,538

895

Our results (Table 2) show that overall t-xlm and pysentimento perform better than the bertin model tested and achieve similar (within 1% difference) accuracy and macro-F1 scores with. However, when examining the results for each dataset, we observe larger differences in their performance. Specifically,

85

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

when considering only the HateEval dataset, we observe a 5.83% difference in the F1 scores of the two models (t-xlm: 73.57, pysentimento: 79.40) in favour of pysentimento. In contrast, when looking at the results on the HaterNet t-xlm is the model that performs the best (t-xlm: 68.32, pysentimento: 53.53). The significant decrease in performance of the pysentimento tool when tested on HaterNet can be justified as it is only trained on the HateEval dataset. Table 2. Accuracy and macro-F1 scores of the models tested when using both HateEval and HaterNet (combined) and on each set individually. The best scores for each data split are underlined Data Combined

HateEval

HaterNet

Model

Accuracy

F1

pysentimento

76.00

73.36

txlm

77.01

72.46

bertin

74.04

69.85

pysentimento

79.56

79.40

t-xlm

75.00

73.57

bertin

72.50

70.65

pysentimento

71.25

53.53

t-xlm

77.58

68.32

bertin

76.08

66.53

For the rest of our analysis, we use the predictions made by t-xlm as it achieves a slightly better overall Accuracy score (77.01%) and equally importantly, it displays a more robust performance across the two datasets tested, and thus it is deemed better to identify different types of hate speech.

Hashtag-Generic Experiment A maximum of 1000 tweets per day were downloaded with the hashtags shown in Table 3. Subsequently, the tweets were classified with the previously mentioned automatic classifier. The hashtags were chosen by manually checking in the Spanish context, which was the most important on Twitter that day. Table 3. The five social events studied in this research with the hashtags and keywords used in each one to download tweets and analyze offensive and hate speech trends Social Event

Hashtag-exp (2014-2022)

Keywords-exp (2014-22)

Manual-exp (2019)

Women´s day

#díadelamujer #8M

feminazi, las charos, feminismo, trans

feminazi, las charos, feminismo, trans

International LGTBQ+ Pride day

#orgulloXXXX (XXXX was the year)

desviados, trans, lgtb, antinatural

desviados, trans, lgtb, antinatural

Spain National Day

#hispanidad #12deOctubre

fachas, hispanidad, viva españa, españita

fachas, hispanidad, viva españa, españita

National elections (2015, 2016 and 2019)

#elecciones

fachas, progres, izquierda, derecha

fachas, progres, izquierda, derecha

Regional elections (2015 to 2022; 11 events)

#elecciones

fachas, progres, izquierda, derecha

fachas, progres, izquierda, derecha

86

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Keywords-Specific Experiment A maximum of 250 tweets per day were downloaded for each keyword, with a total of 4 keywords that were chosen manually by checking beforehand which ones were widely used on Twitter (i.e., frequently used by users). The downloaded tweets were then classified using the automatic classifier. Concerning the keywords, there were chosen two “neutral” keywords and two “offensive” keywords on each social event (Table 3). Both in the generic experiment and in this specific experiment, a maximum number of words were chosen so as not to exceed the capacity of the Twitter API but at the same time to have a representative sample size per day to perform analysis.

Manual Classification Experiment One of the coauthors analyzed 200 daily tweets and classified them manually as hate speech vs nonoffensive speech. We only analyzed 200 tweets grouping all keywords, instead of 250 per keyword as in the specific experiment, due to human resources, as checking all these tweets manually by one person took many hours. The criteria selected to classify them in the two categories was based on the Spanish penal code, i.e., only those classified as crimes were classified. Unfortunately, in this experiment, it was only possible to analyze 2019 because of the same reason as explained before, resources (but see also the following reasons). This concrete year, 2019, was chosen as the last pre-pandemic year. That health crisis may be responsible for rare patterns that we wanted to avoid, yet it could be interesting to study comparisons between previous and Covid years in the future. On the other hand, previous years to 2019 were not chosen because it may increase the likelihood that tweets would have been deleted. Our study may be affected by Twitter’s policies on content not allowed, and tweets may have been deleted because of them (Twitter, 2023). Finally, 2019 was a good year because there were national and regional elections, so all the social events analyzed in the other experiments occurred that year.

Statistical Analysis To visualize the results, double-axis graphs were used to show, on the one hand, the mean and standard errors of the number of offensive tweets of all years (2014-2022, except for the elections), represented by a point and its interval, and, on the other side, a bar graph for the total number of tweets downloaded with hashtags (generic experiment) or with keywords (specific-experiment). On the other hand, statistical analysis was carried out to test differences between three periods: week 1, week 2, and 3. The 29 days were grouped into three-time frames, called “weeks” hereafter, yet they were not 7 days real weeks. These weeks were the first 9 days (i.e., week 1), the next 11 days (i.e. week 2), in which the specific day of the event was included, and the next and last 9 days (i.e. week 3). As already commented, all years were grouped to reduce noise and analyze only general trends. We chose a linear mixed-effect model to compare whether there were significant differences between weeks in the number of offensive and hate speech. The response variable was the number of offensive tweets, and the explanatory variables were two: the total number of tweets downloaded and the week as a 3-level qualitative variable. The total number of tweets was included in the models to control for its effect because the more tweets you download, the more probably is to find offensive/hate speech. The year was used as a random effect. The response variable was squared and transformed to normalize the model’s residuals. 87

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

All statistical analyses were carried out using R statistical software version 4.1.1 (R Core Team, 2021), and the significance level was set at p ≤ 0.05.

RESULTS Hashtags-Generic Experiment In the five events, there were significant Spearman correlations between the offensive/hate speech and the total number of tweets downloaded (see Table 4), and for that reason, the last variable had significant p-values in all the linear-mixed models run for all the events, except for the general experiment on Spanish national elections (Table 4). Table 4. Results of the linear mixed models were carried out in two of three experiments in this research Social Event Women’s Day International LGTBQ+ Pride day Spain National Day National elections Regional Elections

Model - Hashtag-experiment T. tweets F = 359.8, p < 0.001

Week F = 1.9, p = 0.15

CorS = 0.69, p < 0.001 F = 59.4, p < 0.001

F = 1.6, p = 0.20 F = 1.4, p = 0.26

CorS = 0.37, p < 0.001

F = 19.8, p < 0.001

F = 6.7, p = 0.001

F = 58.3, p < 0.001

F = 3.3, p = 0.037

CorS = 0.45, p < 0.001 F = 1.64, p = 0.20

CorS = 0.31, p = 0.004 F = 54.5, p < 0.001

F = 2.8, p = 0.064

CorS = 0.25, p < 0.001

CorS = 0.37, p < 0.001 F = 1.69, p = 0.20

F = 74.6, p < 0.001

Week

CorS = 0.45, p < 0.001

CorS = 0.25, p < 0.001 F = 38.2, p < 0.001

Model - Keywords-experiment T. tweets

F = 76.0, p < 0.001

F = 0.2, p = 0.82

CorS = 0.77, p < 0.001 F = 1.13, p = 0.32

F = 158.22, p < 0.001

F = 3.5, p = 0.030

CorS = 0.63, p < 0.001

Note: The response variable, the number of offensive/hate speech, was squared-root transformed and the explanatory variables were “Total number of tweets downloaded” (“T. tweets”) and a categorical variable, “Week” (3-level, see methods). Here are shown the F and p-values of the explanatory variables and the Spearman correlation between the response variable and “T. tweets”.

Offensive and hate speech rose drastically on Women’s Day. The increase was produced some days before, but the great peak was produced on the specific day of the celebration (Figure 1). However, there were no significant differences between “weeks” (see methods because they are not 7-days weeks), probably because the peak was mainly produced on 8th March.

88

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 1. Double-axis graphs to display the results of the Hashtag-generic experiment on Women’s Day (8th March)

Note: On the left y-axis is the number of offensive/hate tweets as the mean of all the years (2014 to 2022) plus the standard errors; meanwhile, on the right y-axis is plotted the total number of tweets downloaded each day (pooling all the years together). About the International LGTBQ+ Pride Day (Figure 2), there were no significant differences between weeks, the opposite of what happened with the specific experiment (see below).

89

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 2. Double-axis graphs to display the results of the Hashtag-generic experiment on LGTBQ+ Pride Day (28th June)

Note: On the left y-axis is the number of offensive/hate tweets as the mean of all the years (2014 to 2022) plus the standard errors; meanwhile, on the right y-axis is plotted the total number of tweets downloaded each day (pooling all the years together).

Offensive/hate speech rose to a well-defined peak in the Spain National Day case, as shown in Figure 3, yet the number of offensive/hate tweets was not proportionally higher when controlling for the number of tweets discussing that topic.

90

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 3. Double-axis graphs to display the results of the Hashtag-generic experiment on Spain’s National Day (12th October)

Note: On the left y-axis is the number of offensive/hate tweets as the mean of all the years (2014 to 2022) plus the standard errors; meanwhile, on the right y-axis is plotted the total number of tweets downloaded each day (pooling all the years together).

Finally, the national and regional elections were analyzed. First of all, it is important to note that we only studied three national elections between 2014 and 2022. Thus, the sample size is limited (Figure 4). We found a small amount of offensive/hate speech in both types of elections, and only in the regional elections (where the sample size was higher, n = 11 events) there was a positive relationship between the number of offensive/hate speech and the total number of tweets downloaded (Figure 5, and Table 4). However, there were no differences in both types of elections between weeks.

91

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 4. Double-axis graphs to display offensive and hate speech poured out in national elections (2015, 2016, and 2019)

Note: In this graph are shown the results of the Hashtag-generic experiment. On the left y-axis, the number of offensive/hate tweets as the mean of all the years (2015, 2016 and 2019) plus the standard errors, meanwhile in the right y-axis was plotted the total number of tweets downloaded each day (pooling all the years together).

92

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 5. Double-axis graphs to display the results of the hashtag-generic experiment in regional elections (11 elections between 2015 and 2022) Note: On the left y-axis is the number of offensive/hate tweets as the mean of all the years plus the standard errors; meanwhile, on the right y-axis is plotted the total number of tweets downloaded each day (pooling all elections together).

Keywords-Specific Experiment Offensive and hate speech rose smoothly on Women’s Day (Figure 6), yet significant differences were not found between weeks. However, on International LGTBQ+ Pride Day (Figure 7), we found significant differences between weeks after controlling by the total number of tweets downloaded (Table 4). Weeks 2 and 3 received significantly more offensive/hate speech than the first one.

93

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 6. Double-axis graphs to display the results of the keywords-specific experiment in Women’s day (8th March)

Figure 7. Double-axis graphs to display the results of the keywords-specific experiment on LGTBQ+ Pride Day (28th June)

94

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Contrary to other events, offensive/hate speech was significantly lower in week 2 on Spain National Day when we controlled for the total number of tweets downloaded each day (Figure 8 and Table 4). Figure 8. Double-axis graphs to display the results of the keywords-specific experiment on Spain National Day (12th October)

Lastly, there were analyzed the national and regional elections. A priori, watching Figure 9 and Figure 10 it seems that offensive/hate speech was more frequent the election day and later, yet we only found significant differences between weeks 2 and 3 in the regional elections, having more intolerant speech in the third week when controlling for the total number of tweets downloaded.

95

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 9. Double-axis graphs to display offensive and hate speech poured out in national elections (2015, 2016, and 2019) Note: In this graph are shown the results of the Keywords-specific experiment.

96

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 10. Double-axis graphs to display offensive and hate speech poured out in regional elections (11 elections between 2015 and 2022) Note: The results of the Keywords-specific experiment are shown in this graph.

Manual-Classification Experiment Figure 11 summarizes all the results together of this manual experiment. Both elections were social events with a higher proportion of hate speech. In this case, we can directly compare the five events because the same number of tweets, i.e., 200 hundred, were analyzed daily. On the other hand, it is noteworthy to observe that there was a decrease in all events on the exact day of the event celebration (day 15). This is a surprising result and may have a reasonable explanation related to social network policies regarding hate speech (see Discussion).

97

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Figure 11. The results of the manual classification experiment (see methods for more details

Note: 200 tweets were analyzed by day in 29 days, the duration of each event. The social events were five, as in the other experiments (see above). Then, one person classified them as hate speech or not based on the definition of a hate crime in the Spanish Penal Code. This experiment was carried out in 2019 when all the events were present.

DISCUSSION Hate speech is one of the major concerns of the European Union countries because it targets individuals and entire vulnerable groups. It is, therefore, important to study its behavior over time and its interaction with important social events. It is well known that this type of crime is mainly affected by social events of relevance or important news (Williams & Burnap, 2016; Olteanu et al., 2018; Scharwächter & Müller, 2020; Williams et al., 2020; Arcila et al., 2022). However, in the literature, this has generally been approached from the theoretical side but less often from the experimental perspective. Moreover, if it has been studied from the experimental side, researchers have often analyzed only a specific event, such as an election, a terrorist attack or similar (Hopkins, 2010; Disha et al., 2011; Albornoz et al., 2018; Müller & Schwarz, 2020; Warren-Gordon & Rhineberger, 2021; Feinberg et al., 2022). In this study, we have analyzed offensive and hate speech in Spain using a representative sample of Twitter posts in Spanish. We considered their temporal distribution in the context of significant social events and filtered the search by hashtags and keywords related to the topic of the event in question. The main result was that the five events studied produced a significant increase in intolerant speech detected manually by a human and automatically using a classifier algorithm based on NLP, yet differences have been found depending on the method used. The results of this study can be used to guide more effectively the policies taken by Spanish public institutions, such as the Spanish National Office Against Hate Crimes (ONDOD) and

98

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Law Enforcement Agencies, to combat hate crimes. However, the findings of this study may also apply to other European Union countries or other regions with hate speech and events with similar characteristics. An overall increase in content related to specific social events was detected, especially on the specific day of the celebration. These events may increase the participation of authors motivated by prejudice and, subsequently, hate speech because it may be more difficult for Social media to control illegal content (Whitney, 2018). These circumstances redefine the theory of routine activities (Cohen & Felson, 2018), shifting the physical scenario to the cyber one, in which the evolution of technology has marked the increase of interactions between potential victims and perpetrators (Miró Llinares, 2013). The increase in content on Twitter regarding the five events analyzed was obtained both in the generic experiment, filtering searches by thematic hashtags, and in the specific experiment, in which keywords were used. However, when controlling for the total volume of tweets posted on that social network with those hashtags and keywords, only in the case of the specific experiment were found significant differences between the 3 “weeks” (see materials and methods), yet trends were not the same for all events (see below).

Hashtag-Generic Experiment In the generic experiment, a lower volume of offensive and hateful speech was observed than in the specific experiment, which may indicate that users who disseminate and promote offensive speech avoid using hashtags instead of keywords. These words are usually used to refer to certain topics, promoting their popularity and facilitating searches by other users and the platform itself. Therefore, the authors of offensive and hateful speech may try to circumvent Twitter’s content moderation procedures, mainly the automatic ones by algorithms (Elmas et al., 2021; Torres-Lugo et al., 2022). However, many of these authors who dump discriminatory content online do not have the primary purpose of spreading illegal hate content. Thereby, they may use hashtags and do not take any specific precautions to be detected. On the other hand, those groups that are more active in spreading hate on Social media may try to make invisible the content they publish (Cheng et al., 2018; Davidson et al., 2018) with more elaborate techniques, such as the transformation of keywords with alternative letters and/or numbers, responding to publications through subtweets or screenshots, thus fulfilling their goal but avoiding their automatic deletion (Burrell et al., 2019; Tufekci, 2014; van der Nagel, 2018). In the first experiment, we could observe a peak in hateful content on 8th March and 12th October, Women’s Day and Spain National Day, respectively, in line with other previous studies (King & Sutton, 2013). This highlights the remarkable increase, similar to a positive kurtosis normal distribution, after a significant event and its rapid decline of short duration. This short duration may be caused because hate crimes are underpinned by prejudice, which typically, along with anger, is short-lived in response to a trigger (Smith, Phillips, & King, 2010). In this sense, the rapid escalation and de-escalation could find an explanation in the very characteristics of the celebrations since, unlike the other events studied, these festivities are primarily celebrated for only one day. Regarding the rest of the events analyzed in the generic experiment (i.e., International LGTBQ+ Pride day and both types of elections), the conclusions that can be drawn, in addition to those already mentioned, are scarce because the amount of intolerant speech that was found was low and therefore the temporal patterns are scarce.

99

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Keyword-Specific Experiment Regarding the results of the keyword-specific experiment, on Women’s Day, a slight increase in thematic content was observed the week before and on the day of the event’s celebration. Then, there appeared to be a decrease in the number of publications a few days later, yet it was not significant because that rise was proportional to the total volume of tweets detected. Contrary to that, in the analyses related to the celebration of Spain’s National Day, a significantly lower amount of offensive and hate speech was detected in the week of the concrete date. This was because in all the events studied and for the two experiments in which the automatic tweet classifier was used (i.e., the generic and specific experiment), there was a positive correlation between offensive or hate speech and the total volume of tweets. In other words, although the total number of intolerant tweets increased, this rise was positively correlated with the total number of tweets published. This does not mean that there was less harmful content because the total number of intolerant tweets increased. Regarding the International LGTBQ+ Pride Day analyzed in the specific experiment, a slight increase in the intolerant speech was observed days before the celebration, the effect of which was maintained for several days, including the week after. This may be because the celebrations of this event are usually extended days after (Official Tourism Portal for Spain, 2023). Usually, the marches through the streets of Madrid, for example, take place on the weekend after the celebration’s day. Finally, the specific experiment remains to be analyzed on national and regional election days. Although a similar pattern could be observed in the graphs for both events, the statistical models showed that there were only significant differences between weeks in the case of regional elections. In the latter case, in the third week, there was more intolerant discourse when controlling for the total volume of tweets. A possible explanation for these results is that while national elections, also called General elections in Spain, produced an important peak of offensive speech on the day of the event, then this number quickly dropped. In the case of regional elections, the election of regional and local government authorities may affect citizens more directly with their policies (Reyes, 2006), and this may have caused frustration and intolerant speech to be maintained for longer. In other words, this effect may be due to the territorial organization in Spain, where a large part of the powers is delegated to the regional/local level with competencies in areas such as health and education, affecting their laws more directly to citizens. Another explanation might be related to post-electoral agreements between different parties, yet the period analyzed in this study was short to study this effect; besides, years were polled together, which may mitigate the effect of a specific year. Therefore, this can lead to an increase in intolerant speech if their detractors get frustrated by the possible new political decisions that may be taken shortly. However, this type of political event might require studying longer periods than the one analyzed since electoral campaigns begin a few months before the event (Campos, 2017). Moreover, analyzing more variables that could affect the outcome may be important.

Manual-Classification Experiment The last experiment that was carried out was the manual, in which hate speech was classified by a person and analyzed from a criminal perspective. Since a human was in charge of labelling tweets, it was possible to differentiate offensive speech from hate speech, an aspect that unfortunately cannot be easily distinguished automatically with machines (i.e., algorithms). This is possibly due to the difficulty in understanding the whole context of the textual sentence, although probably in a few years, we will be 100

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

able to achieve this with the development of new technologies such as ChatGPT (Lund & Wang, 2023; Thorp, 2023). In this experiment, it was found that the national election and the regional elections that occurred that year (i.e., 2019) were the events that triggered the highest percentage of hate speech. This may be due to election campaigns that have more media coverage than other events. Moreover, elections typically happen every four years and greatly affect people’s life. This may produce more frustration, leading to greater hate speech in Social media. However, one result that was not expected in the manual experiment was that on the day of the event, day 15, there was no peak in the percentage of hateful content posted online. Indeed, the percentage of intolerant content was lower. This could be explained by the platform’s content moderation policies, which can intensify control on specific days when more illegal speech is expected to be generated, thus skewing the data obtained. Another explanation may be the use of Twitter’s API by some users who often take on the role of arbitrators to combat harassment on Twitter through collective block lists based on bots (Geiger, 2016), which undoubtedly contribute to the control of such posts.

LIMITATIONS AND CONCLUSION The results obtained in this research provide relevant information on the behavior of offensive and hate speech on Twitter in the temporal context of important social events. However, some limitations have been detected in this study. First of all, if the results between experiments are compared, it is observed that there are differences concerning the detection of intolerant speech. In the first two experiments, more offensive speech was detected than actual hate speech compared with the manual experiment. Moreover, the automatic classifier sometimes failed to label the text well, as we observed occasionally. This could explain some of the differences in the patterns and suggest the advantages of manually monitoring Social media. However, the effort in terms of resources that were involved in classifying tweets by a human was much higher. We could only analyze one year instead of the nine years analyzed automatically, but it could have been an even greater period. Moreover, unlike in the automatic evaluation, in the manual evaluation, only a fixed set of tweets were labelled per day, and therefore the conclusions can only be based on the percentage of hate among that limited sample. In contrast, in the automatic experiment, we could retrieve all tweets posted in the day. Therefore, both methods have advantages and disadvantages, yet new artificial intelligence tools that are evolving rapidly will probably overcome the disadvantages mentioned quickly, positioning themselves as truly useful tools (Moon et al., 2022). Secondly, data collection may be biased because of a lack of knowledge of how Twitter’s algorithm removes illegal content. This problem may be overcome by monitoring the social network in real-time, thus generating a complete map of the behaviorin the network. Although in this sense, it must be taken into account that this platform may not be completely free of bias, as many profiles publish tweets in a coordinated manner intending to generate artificial trending topics to try to manipulate the content popularized on this platform, and later these accounts and publications are subsequently deleted to avoid detection by Twitter (Elmas et al., 2021; Torres-Lugo et al., 2022). Additionally, the hashtags and keywords selected for this study have not considered the content transformation techniques employed by the disseminators of hate content to avoid control (Burrell et al., 2019). On the other hand, in some of the events, such as elections, it would be desirable to increase the period evaluated because they affect not only the days closer to the day of the vote. 101

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Efforts to combat hate crimes promoting intolerance and prejudice must continue. ONDOD works on prevention and investigation through action plans, guides, etc., and recently the two National Law Enforcement Agencies have created the Hate Crime Response Teams (REDO) and Violent Extremism and Hate Teams (EVO). Besides, ONDOD collaborates with various public and private entities and participates in EU initiatives like the Code of Conduct on countering illegal hate speech. Social media such as Twitter, Facebook, and TikTok have joined this initiative, which inspired Spain’s “Protocol to combat illegal hate speech online.” This research contributes to understanding hate speech behavior around significant social events. Though it is not the formal opinion of a public institution, yet may help to drive actions to combat hate speech more effectively. Future research may explore alternatives to combating hate speech, such as fake news detection. ONDOD is involved in the REAL-UP project funded by the European Commission (CERV-2021-EQUAL) to study various strategies. Predictive policing, using statistical models to anticipate verbal violence peaks, could also help prevent hate crimes. Research and projects like these are crucial for fostering respect and tolerance in diverse societies.

REFERENCES Albornoz, F., Bradley, J., & Sonderegger, S. (2018). The Brexit Referendum and the Rise in Hate Crime. mimeo. https://cep.lse.ac.uk/seminarpapers/19-09-18-ALB_v2.pdf Aluru, S., Cingel, D. P., Miller, C. H., & Stavrositu, C. (2020). Individual and environmental factors associated with cyberbullying perpetration: A meta-analysis. Journal of Communication, 70(6), 739–758. doi:10.1093/joc/jqaa039 Amores, J. J., Blanco-Herrero, D., Sánchez-Holgado, P., & Frías-Vázquez, M. (2021). Detectando el odio ideológico en Twitter. Desarrollo y evaluación de un detector de discurso de odio por ideología política en tuits en español. Cuadernos.Info, (49), 98–124. doi:10.7764/cdi.49.27817 Anderson, K. L., & Meyer, D. (2016). Violence against Queer People: Race, Class, Gender, and the Persistence of Anti-LGBT Discrimination. By Doug Meyer. New Brunswick, N.J.: Rutgers University Press, 2015. Pp. viii+194. $80.00 (cloth); $25.95 (paper). American Journal of Sociology, 122(2), 631–633. https://www.jstor.org/stable/26545851. doi:10.1086/687190 Arcila Calderón, C., Sánchez Holgado, P., Quintana Moreno, C., Amores, J. J., & Blanco Herrero, D. (2022). Discurso de odio y aceptación social hacia migrantes en Europa: Análisis de tuits con geolocalización. Comunicar: revista científica iberoamericana de comunicación y educación. . doi:10.3916/ C71-2022-02 Awan, I., & Zempi, I. (2016). Islamophobia and Twitter: A typology of online hate against Muslims on social media. Policy and Internet, 8(2), 133–150. doi:10.1002/1944-2866.POI364 Bachner, J. (2013). Predictive policing: preventing crime with data and analytics. IBM Center for the Business of Government. https://businessofgovernment.org/sites/default/files/Managem ent%20Predictive%20Policing.pdf

102

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Barbieri, F., Anke, L. E., & Camacho-Collados, J. (2022). Xlm-t: Multilingual language models in twitter for sentiment analysis and beyond. In Proceedings of the Thirteenth Language Resources and Evaluation Conference (pp. 258-266). ACL. https://aclanthology.org/2022.lrec-1.27/ Basile, V., Bosco, C., Fersini, E., Nozza, D., Patti, V., Pardo, F. M. R., Rosso, P., & Sanguinetti, M. (2019). SemEval-2019 task 5: Multilingual detection of hate speech against immigrants and women in Twitter. In NAACL HLT 2019 - International Workshop on Semantic Evaluation, SemEval 2019, Proceedings of the 13th Workshop (pp. 54–63). Association for Computational Linguistics (ACL). 10.18653/v1/S19-2007 Battistelli, A., Cipolletta, S., & Mocini, F. (2020). The role of empathy and moral disengagement in predicting cyberbullying perpetration and victimization among adolescents. Journal of Interpersonal Violence, 35(23-24), 5411–5430. doi:10.1177/0886260518765961 Baumgartner, S. E., Sumter, S. R., Peter, J., & Valkenburg, P. M. (2018). Online hate and harmful speech: Psychological consequences and empirical responses. Journal of Social and Political Psychology, 6(1), 183–213. doi:10.5964/jspp.v6i1.849 Benítez-Andrades, J. A., González-Jiménez, Á., López-Brea, Á., Aveleira-Mata, J., Alija-Pérez, J. M., & García-Ordás, M. T. (2022). Detecting racism and xenophobia using deep learning models on Twitter data: CNN, LSTM and BERT. PeerJ. Computer Science, 8, e906. doi:10.7717/peerj-cs.906 PMID:35494847 Berger, J. M., & Morgan, J. (2016). The ISIS Twitter Census: Defining and Describing the Population of ISIS Supporters on Twitter. The Brookings Project on U.S. Relations with the Islamic World. https://www.brookings.edu/wp-content/uploads/2016/07/The-ISI S-Twitter-Census-Berger-Morgan.pdf Borell, K. (2015). When is the time to hate? A research review on the impact of dramatic events on Islamophobia and Islamophobic hate crimes in Europe. Islam & Christian-Muslim Relations, 26(4), 409–421. doi:10.1080/09596410.2015.1067063 Burrell, J., Kahn, Z., Jonas, A., & Griffin, D. (2019). When users control the algorithms: values expressed in practices on twitter. Proceedings of the ACM on human-computer interaction, 3(CSCW), 1-20. ACM. 10.1145/3359240 Camacho-Collados, M., & Liberatore, F. (2015). A decision support system for predictive police patrolling. Decision Support Systems, 75, 25–37. doi:10.1016/j.dss.2015.04.012 Camacho-Collados, M., Liberatore, F., & Angulo, J. M. (2015). A multi-criteria police districting problem for the efficient and effective design of patrol sector. European Journal of Operational Research, 246(2), 674–684. doi:10.1016/j.ejor.2015.05.023 Campos Domínguez, E. M. (2017). Twitter y la comunicación política. El Profesional de la Información, 26(5), 785–794. doi:10.3145/epi.2017.sep.01 Carr, J., Clifton-Sprigg, J., James, J., & Vujic, S. (2020). Love thy neighbour? Brexit and hate crime (13902). IZA Discussion Papers. doi:10.2139srn.3738844

103

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Cheng, J., Danescu-Niculescu-Mizil, C., & Leskovec, J. (2018). Antisocial Behavior in Online Discussion Communities. In Proceedings of the ACM on Human-Computer Interaction, 2(CSCW), 1-27. ACM. 10.1145/3274437 Cohen, L. E., & Felson, M. (2018). Cambio social y tendencias en la tasa de criminalidad: un enfoque desde las actividades cotidianas. Revista de derecho penal y criminología, (20), 359-369. doi:10.5944/ rdpc.20.2018.26497 Currant, K., O’Hara, K., & O’Brien, S. (2011). The role of Twitter in the world of business. International Journal of Business Data Communications and Networking, 7(3), 1–15. doi:10.4018/jbdcn.2011070101 Davidson, T., Bhattacharya, A., & Weber, I. (2018). Identifying multiple types of social influence in online hate speech. In Proceedings of the 2018 World Wide Web Conference (pp. 1591-1600). International World Wide Web Conferences Steering Committee. Del Valle-Cano, G., Quijano-Sánchez, L., Liberatore, F., & Gómez, J. (2023). SocialHaterBERT: A dichotomous approach for automatically detecting hate speech on Twitter through textual analysis and user profiles. Expert Systems with Applications, 216, 119446. doi:10.1016/j.eswa.2022.119446 Deloughery, K., King, R. D., & Asal, V. (2012). Close cousins or distant relatives? The relationship between terrorism and hate crime. Crime and Delinquency, 58(5), 663–688. doi:10.1177/0011128712452956 DevineD. (2018). The UK referendum on membership of the European Union as a trigger event for hate crimes. SSRN 3118190. doi:10.2139/ssrn.3118190 Devine, D. (2021). Discrete events and hate crimes: The causal role of the Brexit Referendum. Social Science Quarterly, 102(1), 374–386. doi:10.1111squ.12896 Disha, I., Cavendish, J. C., & King, R. D. (2011). Historical events and spaces of hate: Hate crimes against Arabs and Muslims in post-9/11 America. Social Problems, 58(1), 21–46. doi:10.1525p.2011.58.1.21 Duncombe, C. (2019). The politics of Twitter: Emotions and the power of social media. International Political Sociology, 13(4), 409–429. doi:10.1093/ips/olz013 Echebarría-Echabe, A., & Fernández-Guede, E. (2006). Effects of attributional content and coping strategy on emotions and on behavioral intentions towards immigrants. Social Behavior and Personality, 34(1), 1–14. doi:10.2224bp.2006.34.1.1 EdwardsG. S.RushinS. (2018). The effect of President Trump’s election on hate crimes. doi:10.2139/ ssrn.3102652 Elmas, T., Overdorf, R., Özkalay, A. F., & Aberer, K. (2021). Ephemeral astroturfing attacks: The case of fake twitter trends. In 2021 IEEE European Symposium on Security and Privacy (EuroS&P) (pp. 403-422). IEEE. 10.1109/EuroSP51992.2021.00035 Feinberg, A., Branton, R., & Martinez-Ebers, V. (2022). The Trump effect: How 2016 campaign rallies explain spikes in hate. PS, Political Science & Politics, 55(2), 257–265. doi:10.1017/S1049096521001621

104

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Florio, E., Strapparava, C., & Bolioli, A. (2017). Analyzing message text with most algorithms trained for other languages. [TALLIP]. ACM Transactions on Asian and Low-Resource Language Information Processing, 16(2), 1–19. doi:10.1145/3094249 Geiger, R. S. (2016). Bot-based collective blocklists in Twitter: The counterpublic moderation of harassment in a networked public space. Information Communication and Society, 19(6), 787–803. doi:10.1 080/1369118X.2016.1153700 Gillespie, T. (2018). The Politics of Platforms: A Comparative Study of Hate Speech Regulation Online. Yale University Press. González-Álvarez, J. L., Hermoso, J. S., & Camacho-Collados, M. (2020). Policía predictiva en España. Aplicación y retos futuros. Behavior & Law Journal, 6(1), 26–41. doi:10.47442/blj.v6.i1.75 Hanes, E., & Machin, S. (2014). Hate crime in the wake of terror attacks: Evidence from 7/7 and 9/11. Journal of Contemporary Criminal Justice, 30(3), 247–267. doi:10.1177/1043986214536665 Hardyns, W., & Rummens, A. (2018). Analyzing the effects of social media on police legitimacy beliefs: An empirical study. Journal of Police and Criminal Psychology, 33(2), 117–129. doi:10.100711896017-9243-7 Herrmann, C. R. (2015). The dynamics of robbery and violence hot spots. Crime Science, 4(1), 1–14. doi:10.118640163-015-0042-5 Hopkins, D. J. (2010). Politicized places: Explaining where and when immigrants provoke local opposition. The American Political Science Review, 104(1), 40–60. doi:10.1017/S0003055409990360 Ivandic, R., Kirchmaier, T., & Machin, S. J. (2019). Jihadi attacks, media and local hate crime. https:// doi.org/ doi:10.2139/ssrn.3401120 Jacobs, L., & van Spanje, J. (2021). A Time-Series Analysis of Contextual-Level Effects on Hate Crime in The Netherlands. Social Forces, 100(1), 169–193. doi:10.1093foaa102 Legewie, J. (2013). Terrorist events and attitudes toward immigrants: A natural experiment. American Journal of Sociology, 118(5), 1199–1245. doi:10.1086/669605 Liberatore, F., Camacho-Collados, M., Vitoriano, B. (2020). Police districting problem: literature review and annotated bibliography. Optimal districting and territory design, 9-29. Springer. doi:10.1007/9783-030-34312-5_2 Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692. López-Gutiérrez, J. (2022). Informe sobre la evolución de los delitos de odio en España 2021. Oficina Nacional de Lucha contra los delitos de Odio (ONDOD). https://www.interior.gob.es/opencms/es/servicios-al-ciudadan o/delitos-de- odio/estadisticas/ Lund, B. D., & Wang, T. (2023). Chatting about ChatGPT: How may AI and GPT impact academia and libraries? Library Hi Tech News, 40(3), 26–29. doi:10.1108/LHTN-01-2023-0009

105

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Meijer, A., & Wessels, M. (2019). Predictive policing: Review of benefits and drawbacks. International Journal of Public Administration, 42(12), 1031–1039. doi:10.1080/01900692.2019.1575664 Mills, C. E., Freilich, J. D., & Chermak, S. M. (2017). Extreme hatred: Revisiting the hate crime and terrorism relationship to determine whether they are “close cousins” or “distant relatives”. Crime and Delinquency, 63(10), 1191–1223. doi:10.1177/0011128715620626 Miró Llinares, F. (2013). La victimización por cibercriminalidad social. Un estudio a partir de la teoría de las actividades cotidianas en el ciberespacio. Revista Española De Investigación Criminológica, 11, 1–35. doi:10.46381/reic.v11i0.77 Mohler, G. (2014). Marked point process hotspot maps for homicide and gun crime prediction in Chicago. International Journal of Forecasting, 30(3), 491–497. doi:10.1016/j.ijforecast.2014.01.004 Monárrez, J., & Cervera-Gomez, L. (2013). Spatial and temporal behavior of three paradigmatic cases of violence in Ciudad Juarez, Chihuahua México. ResearchGate, 1-23. http://dx.doi.org/ doi:10.13140/ RG.2.1.4007.7289 Moon, S., Lee, G., & Chi, S. (2022). Automated system for construction specification review using natural language processing. Advanced Engineering Informatics, 51, 101495. doi:10.1016/j.aei.2021.101495 Moreno, J., Quintero, S., Riascos, A., Nonato, L. G., & Sanchez, C. (2021). Homicide prediction using sequential features from graph signal processing. In Intelligent Computing (pp. 804–817). Springer. doi:10.1007/978-3-030-80129-8_54 Moussaïd, M., Barkoczi, D., & Helbing, D. (2017). Predicting and affecting opinion dynamics in realworld social networks. Proceedings of the National Academy of Sciences of the United States of America, 114(8), 201608473. doi:10.1073/pnas.1608473114 Müller, K. R., & Schwarz, C. (2020). Hate speech, social media, and the question of accountability. Media and Communication, 8(2), 443–446. doi:10.17645/mac.v8i2.3051 Müller, K. R., & Schwarz, C. (2021). Hate speech and the responsibility of the media. In H. Tumber & S. Waisbord (Eds.), The Routledge Companion to Media Misinformation and Populism (pp. 277–284). Routledge. doi:10.4324/9780367822847-25 Nguyen, J. (2018). Politics and the Twitter revolution: A brief literature review and implications for future research. Social Networking, 7(4), 243–251. doi:10.4236n.2018.74018 Official Tourism Portal for Spain. (2023). Celebrate pride in Madrid, Ibiza and Sitges. Official Tourism Portal for Spain. https://www.spain.info/es/descubrir-espana/gay-pride-madridibiza-sitges/ Olteanu, A., Castillo, C., Boy, J., & Varshney, K. (2018). The effect of extremist violence on hateful speech online. In Proceedings of the international AAAI conference on web and social media, 12(1). PKP. 10.1609/icwsm.v12i1.15040 Pereira-Kohatsu, W. M., Costa-Lopes, R., & Vala, J. (2019). The role of group norms in predicting cyberbullying among adolescents. Journal of Interpersonal Violence, 36(5-6), NP2385–NP2402. doi:10.1177/0886260517735826

106

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Pérez, J. M., Furman, D. A., Alemany, L. A., & Luque, F. (2021b). Robertuito: a pre-trained language model for social media text in spanish. arXiv preprint arXiv:2111.09453. Pérez, J. M., Giudici, J. C., & Luque, F. (2021a). pysentimiento: A python toolkit for sentiment analysis and socialnlp tasks. arXiv preprint arXiv:2106.09462. Perry, P. P., McInnis, B., Price, C., Smith, S., & Hollywood, J. (2013). Predictive Policing: The Role of Crime Forecasting in Law Enforcement Operations. The Role of Crime Forecasting in Law Enforcement Operations, RR233. doi:10.7249/RR233 Piatkowska, S. J., & Lantz, B. (2021). Temporal clustering of hate crimes in the aftermath of the Brexit vote and terrorist attacks: A comparison of Scotland and England and Wales. British Journal of Criminology, 61(3), 648–669. doi:10.1093/bjc/azaa090 Piatkowska, S. J., & Stults, B. J. (2021). Brexit, Terrorist Attacks, and Hate Crime: A Longitudinal Analysis. Social Problems. doi:10.1093ocpropab005 Plaza-del Arco, F. M., García-Madariaga, J., & Sánchez-Navarro, J. (2021). Influence of cyber-victimization on self-esteem, perceived social support, and life satisfaction in early adolescence. Frontiers in Psychology, 11, 608195. doi:10.3389/fpsyg.2020.608195 Poletto, M., Angelone, A., & Oreglia, E. (2021). The dark side of social media: Discrimination and prejudice in the age of digital communication. Computers in Human Behavior, 114, 106556. doi:10.1016/j. chb.2020.106556 Quijano-Sanchez, L., Kohatsu, J. C. P., Liberatore, F., & Camacho-Collados, M. (2019). HaterNet a system for detecting and analyzing hate speech in Twitter (Version 1.0) [Data set]. Zenodo. European Commission. (2021). REAL-UP project, CERV-2021-EQUA. EC. https://real-up.eu/ Relia, K., Jakubowicz, K., & Kovács, A. M. (2019). Hate speech and hate crime: The legal and regulatory framework for the EU and its member states. European Parliament. doi:10.2861/252148 Reyes, M. A. (2006). La construcción del Estado autonómico. Cuadernos Constitucionales de la Cátedra Fadrique Furió Ceriol, (54), 75–95. Scharwächter, E., & Müller, E. (2020). Does terrorism trigger online hate speech? on the association of events and time series. The Annals of Applied Statistics, 14(3), 1285–1303. doi:10.1214/20-AOAS1338 Seeta, J. (2019). Social media, race, and gender: A critical review. Council on Contemporar y Families. https://contemporar yfamilies.org/social-media-race-and-gende r-a-critical-review/ Serra, L., Vall-Llosera, L., Varga, D., Saurina, C., Saez, M., & Renart, G. (2022). Analysis of the geographic pattern of the police reports for domestic violence in Girona (Spain). BMC Public Health, 22(1), 1–11. doi:10.118612889-022-12916-4 PMID:35313832 Smith, P., Phillips, T. L., & King, R. D. (2010). Incivility: The rude stranger in everyday life. Cambridge University Press. doi:10.1017/CBO9780511781803

107

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Sreelakshmi, P. R., Binu, V. S., Kumar, R., & Ramakrishnan, J. (2020). The impact of cyberbullying on psychological well-being: A cross-sectional study among college students in Kerala. Indian Journal of Community Medicine, 45(2), 208–212. doi:10.4103/ijcm.IJCM_166_19 Thorp, H. H. (2023). ChatGPT is fun, but not an author. Science, 379(6630), 313–313. doi:10.1126cience. adg7879 PMID:36701446 Torres-Lugo, C., Pote, M., Nwala, A. C., & Menczer, F. (2022). Manipulating Twitter Through Deletions. In Proceedings of the International AAAI Conference on Web and Social Media (Vol. 16, pp. 1029-1039). PKP. 10.1609/icwsm.v16i1.19355 Tufekci, Z. (2014). Big Questions for social media big data: Representativeness, validity and other methodological pitfalls. In Proceedings of the 8th International Conference on Weblogs and Social Media, ICWSM 2014 (pp. 505–514). The AAAI Press. https://doi.org/10.1609/icwsm.v8i1.14517 Valente, R. (2019). Spatial and temporal patterns of violent crime in a Brazilian state capital: A quantitative analysis focusing on micro places and small units of time. Applied Geography (Sevenoaks, England), 103, 90–97. doi:10.1016/j.apgeog.2019.01.006 Van der Nagel, E. (2018). ‘Networks that work too well’: Intervening in algorithmic connections. Media International Australia, Incorporating Culture & Policy, 168(1), 81–92. doi:10.1177/1329878X18783002 Van Patten, I. T., McKeldin-Coner, J., & Cox, D. (2009). A microspatial analysis of robbery: Prospective hot spotting in a small city. Crime Mapping, 1(1), 7–32. http://ivanpatt.asp.radford.edu/Research/Microspatial%20Anal ysis%20of%20Robbery.pdf Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30. Warren-Gordon, K., & Rhineberger, G. (2021). The “Trump effect” on hate crime reporting: Media coverage before and after the 2016 presidential election. Journal of Ethnicity in Criminal Justice, 19(1), 25–45. doi:10.1080/15377938.2021.1895944 Weisburd, D., & Telep, C. W. (2011). The efficiency of place-based policing. Evidence Based Policing, 17, 247–262. Whitney, P. (2018). The Oxygen of Amplification: Better Living through Anonymity, Attention, and Scale. Data & Society. https://apo.org.au/node/172901 Williams, M. L., & Burnap, P. (2016). Cyberhate on social media in the aftermath of 44 Woolwich: A case study in computational criminology and big data. British Journal of Criminology, 56(2), 211–238. doi:10.1093/bjc/azv059 Williams, M. L., Burnap, P., Javed, A., Liu, H., & Ozalp, S. (2020). Hate in the machine: Anti-Black and anti-Muslim social media posts as predictors of offline racially and religiously aggravated crime. British Journal of Criminology, 60(1), 93–117. doi:10.1093/bjc/azz064 Witzenberger, K. (2018). The hyperdodge: How users resist algorithmic objects in everyday life. Media Theory, 2(2), 29-51. https://hal.archives-ouvertes.fr/hal-02047585

108

 Offensive, Hate Speech on Twitter and Relevant Social Events in Spain

Xu, Y., & Wen, J. (2013). Detecting robbery and violent scenarios. In 2013 Second International Conference on Robot, Vision and Signal Processing, (pp. 25-30). IEEE. https://doi.org/10.1109/RVSP.2013.14 Xue, L., Constant, N., Roberts, A., Kale, M., Al-Rfou, R., Siddhant, A., Barua, A., & Raffel, C. (2020). mT5: A massively multilingual pre-trained text-to-text transformer. arXiv preprint arXiv:2010.11934. Yar, P., & Nasir, J. (2016). GIS Based Spatial and Temporal Analysis of Crimes, a Case Study of Mardan City, Pakistan. International Journal of Geosciences, 7(03), 325–334. doi:10.4236/ijg.2016.73025

KEY TERMS AND DEFINITIONS Hate Speech: Some authors refer to hate speech from an offensive and discriminatory point of view, but if the Spanish penal code is taken into account, the definition would be related to the direct or indirect promotion of hatred, although that legislation includes more casuistry. Intolerant Speech: Without being defined as a criminal offence by the penal code, it affects certain groups in a discriminatory manner. Natural Language Processing: They are algorithms based on different technologies, such as neural networks, which allow the analytical processing of texts. Predictive Policing: Predictive policing allows the analysis of past data to predict behaviors or events that may occur in the future with an associated probability. Social Important Events: Throughout this chapter, these types of events have been named, taking into consideration those that significantly affect the whole society, whether or not all the individuals belong to a certain group, because their public diffusion is great. Spanish National Office Against Hate Crimes (ONDOD): The main goal of the National Office is to implement measures at the national level to enable law enforcement agencies to respond more effectively to victims of hate crimes and to combat hate crimes more effectively. Twitter: This is a free social network that allows users to stay connected through short text messages of up to 280 characters.

109

Section 2

Mediterranean Countries’ Approaches

111

Chapter 7

The Semiotics of Xenophobia and Misogyny on Digital Media: A Case Study in Spain Max Römer-Pieretti https://orcid.org/0000-0002-8049-2898 Universidad Camilo José Cela, Spain Julio Montero-Díaz https://orcid.org/0000-0002-4145-7424 Universidad Internacional de La Rioja, Spain Elias Said-Hung https://orcid.org/0000-0002-0594-5906 Universidad Internacional de La Rioja, Spain

ABSTRACT This chapter raises three questions: a) concerns a synthesis of the classic contributions of the reference semiotic authors that are considered when analysing hate speech in social media; b) entails presenting a case study that is analysed precisely with that analysis synthesis; c) shows the usefulness and interest of this type of analysis in investigations of hate speech. It offers a semiotic model for analysing misogynistic and xenophobic hate speech from digital news media on Twitter. The case study comprises the news published by El Mundo (Spain) from its users on social media, and the 33 comments generated, as a reason for this publication, during January 2021. This serves as the basis of semiotic analysis for understanding the phenomenon. The results visualise the semiotic analyses for understanding the dissemination of expressions. This approach thus helps reveal the levels of intensity, the denotative and connotative differences, the destructive-constructive and intertextual nature of messages, and sheds light on the different symbolic structures associated with hate speech.

DOI: 10.4018/978-1-6684-8427-2.ch007

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 The Semiotics of Xenophobia and Misogyny on Digital Media

BACKGROUND The research analysing hate speech on digital media has grown. This interest reflects the propagation capacity of these scenarios, which favour a predominantly anonymous character and high emotional charge. All of this causes wide dissemination of misinformation content among many users of a specific social media platform. In these scenarios, openly xenophobic or misogynistic discourses proliferate from the multiplicity of interactions and the interactions generated by the abundance of information available in these digital communication contexts (Piñeiro-Otero & Martínez-Rolán, 2021). Concerning xenophobic expressions from digital media, the number of studies on this topic is still low (Müller & Schwarz, 2020). A review of these publications on xenophobia, using the Web of Science or Scopus, in current digital media presents just under fifty papers as of 2022 in journals in the communication area. Most have focused on designing detection models for this type of hate through algorithms (Amores et al., 2021; Benitez-Andrades et al., 2022; Plaza-Del-Arco et al., 2022; Silva et al., 2020). Another group has provided descriptions of their operations on social media (Twitter) and their relationships with social media, focusing on some instances such as vulnerable groups associated with immigration and refugees) and specific situations (during the COVID-19 pandemic or Brexit) (Breazu & Machin, 2022; Giraldo Pérez, 2022; Park, 2017; Sánchez-Holgado, et al., 2022; Zamri et al., 2021). To these are added the reviews and applications of policies and legal frameworks in different geographical contexts (e.g., Europe and the United States), how political movements (e.g., nationalists and extreme right) favour the dissemination of these expressions through the internet, social media and official channels of the state, and ethnography, critical discourse rhetoric and cultural semiotics analysis (Alkiviadou, 2018; Arrocha, 2019; Bourou, 2022; Brindle, 2016; Damcevic & Rodik, 2018; Evenden-Kenyon, 2019; Guizardi & Mardones, 2020; Khoma & Oleksii Kokoriev, 2021; Tymińska, 2020; Sumalla, 2018; Yamaguchi, 2013). Finally, the complementary professional activity required by this type of hate speech has been analysed regarding digital newspapers (Merklejn & Wislicki, 2020; Paz-Rebollo et al., 2021; Slavíčková & Zvagulis, 2014). Misogynistic expressions involve contempt towards, the humiliation of or the disclosure of prejudice against a woman or a group of women (European Commission against Racism and Intolerance, 2016); similar to xenophobic expressions, they have found an environment favourable to their expression of social media (Ging, 2019). Their effects on society transcend the individual sphere, as they favour the promotion of stereotypes against the collective (Megías et al., 2020). Although the study of misogyny in the media has been given greater relevance than that on xenophobes, especially in recent years, as measured in publications indexed on the Web of Science and Scopus (with more than two hundred publications), most of these papers focus mainly on two axes. The first is detecting and classifying sexist expressions and hate speech against women (Martínez, 2023; Pamungkas et al., 2020; Priyadharshini et al., 2022; Ramponi, 2022;). The second is concerns understanding the issuer, the context of disclosure and the type of harassment via the increase in sexist expressions on social media (Arce-García & Menéndez- Menendez, 2023; Cuthbertson et al., 2019; Elias & Gurbanova, 2018). In any case, there is hardly any analysis of the expressions that manifest xenophobia and misogyny from perspectives, approaches and methodologies of a semiotic nature. Yao (2021) and Esposito (2021) have addressed the role of linguistic and semiotic resources in the construction of exclusive discourses (misogynist or xenophobic) on social media (Twitter) and specific platforms (Weibo). They express the need to advance works that account for the implications of semiotics in the current online environment. This issue requires special attention because the discursive and multimodal complexity of digital 112

 The Semiotics of Xenophobia and Misogyny on Digital Media

contexts, overwhelmingly dominant in the current communication scenario, must be assumed. Digital environments favour the manipulation of images, the falsification of identities, and the mistrust of groups such as women and immigrants. In this scenario, the agenda-setting of the media (news, advertising and others) favours the use of hate speech regarding certain situations or vulnerable groups and thus becomes a paradiscourse that is decisive when entering into specific communication agendas and thus influences public opinion, even ultimately giving visibility to stereotypes of citizens on specific topics or groups (Bonacho, 2021; Meseguer, 2022; Ortega, 2022). This set of considerations supports the need to promote academic actions that extend the debate on the ethics of the media or the normative treatment of such hate speech from and in current digital media (Contreras & Sánchez, 2021; Juárez & Restrepo, 2021).

SEMIOTICS, HATE EXPRESSIONS, AND THE MEDIA Semiotics has been transmuted from the science of signs to the science of significance (Vidales et al., 2021) and has intensified its relationship with communication. There is no communication without the mediation of meaning, especially as globalisation has resulted in ‘the creation of new values’ (Rubio, 2022, p. 1). In addition, semiotics and communication coincide around the theory of communication (Parra, 2020); the introduction of semiotics to university studies in communication (Francescutti, 2020); the study of emojis (Vela, 2020); and the significance of logos and their use in digital communication (Llorente-Barroso et al., 2021). Similarly, interpersonal and group communications on WhatsApp have been analysed with semiotics), and social interactions on Instagram have been analysed with sociosemiotics (Gallo & Galmarini, 2021; Moreno Barreneche, 2021). Although semiotics has roots in linguistics, philosophy, and psychology, it is no less accurate to state that universities have incorporated semiotics studies to apply them in communication in recent decades. This is due to the growing number of new digital communication scenarios on current social media platforms. Semiotic analysis of current digital media requires defining the interactions of users who participate on social media or leave comments. This requires a prior review of some basic concepts: Kristeva’s (1969) intertextuality, Barthes’ (1970) denotation and connotation, Greimas and Courtés’ (1979) generative trajectories, and Lyotard’s (2019) emancipation narrative. The first proposes that linguistic structures are isomorphic or analogous to other systems, such as literature and politics, which implies a perpetual transit, e.g., the study of language with other languages, which prompts the writer to inquire about his or her speech (Alirangues, 2018; Bohórquez, 2006; Espino, 2019; Ibri et al., 2021). Semiotics considers the various signifying practices deemed translinguistic as its object, produced by language but irreducible to linguistic categories (Kristeva, 1969). For Kristeva (1969), text and system are translinguistic in a constructive-destructive relationship. Thus, via the concept of intertextuality, the text participates in the transformation of reality. On the horizon of the semiotic analysis of digital discourses with misogynistic or xenophobic content, the perspective of Barthes (1970) on ‘denotation’ and ‘connotation’ is of great interest. The denotative scope is configured by aesthetic and narrative aspects (Bernal et al., 2013). The connotative refers to the semiotic analysis of culture and literature. Both concepts reflect the idea of the ‘discourse of perhaps’, in which the game has an acting role (García, 2019). In summary: ‘E’ (expression or meaning), in ‘R’ (relation) with a ‘C’ (content or signifier). Thus, each creation of meaning generates a new combination with the signifier, which in turn becomes the signified as per Nöth (2000), although other authors such 113

 The Semiotics of Xenophobia and Misogyny on Digital Media

as Kerbrat-Orecchioni (1983) and Fernández (2010) have stressed that this vision confuses metalanguage with connotation. Greimas and Courtés (1979) argue that semiotics must be a signification theory that distinguishes between a sign’s deep and superficial levels (Ponce et al., 2022). This concept makes it possible to describe discursive production as a process that integrates syntactic and semantic subcomponents. Rodríguez and Lázaro (2022) and Landowski (2022) have applied these approaches in current digital scenarios (e.g., Tweets). Lyotard, however, has not analysed current digital scenarios. As noted above, digital environments lend themselves to creating and disseminating false news or disinformation content (Rodríguez, 2022). Following Lyotard’s (2019) approaches to legitimisation by authority (in our case, the media), any statement could be included in discourse and considered valid at a social level. It would also be valid in the current communicative and social context. This raises the need to emancipate the narrative from institutionality. Today, any message published in current digital scenarios could distance itself from established power (e.g., the media and the gatekeeper role they have traditionally assumed). Thus, any denotative statement would adopt the position of ‘knowing’ (knowing what happens), while the addressee would be in the position of approving it or not. The referent would be summarised as something identified and expressed in the statement from which a message starts (Lyotard, 1979). Nevertheless, the performative (symbolic) character of the authority that publishes said message has an immediate character concerning the referent and the recipient (Zaya et al., 2015). This work follows Snoussi and Korbi’s approach (2021), analysing hate speech on digital media during the COVID-19 pandemic from a denotative and connotative perspective. In addition, according to the work of Barrios-Rubio and Fajardo (2022), Galofaro and Toffano (2022) and Nöth (2011) regarding the Greimasian orientation of digital media, there is a predominance of emotional and connotative messages and metalanguages. In summary, in any semiotic act, when using denotations as new connotations (Gaines, 2015), there is room for analyses that make it possible to recognise the different potential meanings each reader grants to each content. These principles allow analysing of some messages with misogynistic and xenophobic content posted by a user of Twitter (El Mundo), a digital communication medium.

THE METHODOLOGICAL APPROACH TO THE FOCAL SUBJECT Objectives and Case Study This chapter, part of an exploratory research approach, this work offers a semiotic analysis procedure for analysing messages expressing xenophobic and misogynistic hate from social networks. The news media El Mundo has been chosen to analyse hate messages to test the semiotic matrix used in this work for the approach of the proposed topic. As it was a theoretical model that did not exist at the level of semiotics, it was necessary to experiment with the possibilities and feasibility of usability of the criteria used in creating the matrix. Around this information medium, work was done on a case: the news published by El Mundo’s Twitter user (@elmundo.es) about Jennifer López’s speech at Joe Biden’s presidential inauguration (Figure 1), the debate that it generated on this social media platform and which messages with xenophobic and misogynistic content played a relevant role in Spain. The case has the following characteristics: first, applying conventional semiotic analysis to each of the 1,052 morphemes or hate 114

 The Semiotics of Xenophobia and Misogyny on Digital Media

messages during the focal period among the total number of users involved is impossible. Second, the circumstances and contexts of the case scenario implicate Latin, with multiple connotations for languages in Spain: Castilian (name of the language within Spain) or Spanish (name of the language in Spanishspeaking countries, coming from Latin America) and their connotations for the other co-official languages in Spain (e.g., Catalan, Galician or Basque). Each language, with different meanings and connotations around similar words, especially in the case of Spanish spoken in Spain and Latin America. Figure 1. Message used as a case study in the present work (https://twitter.com/elmundoes/status/1351980993968799749)

115

 The Semiotics of Xenophobia and Misogyny on Digital Media

Biden’s swearing-in ceremony had a particularly complex connotation, referring to the Spanish language instead of its alternative name within Spain, Castilian. It generated 83 retweets, 33 comments1 from followers of the El Mundo user on Twitter, and 472 likes. These comments are included in the analysed corpus, where the levels of hatred are first distinguished in terms of their different intensities, exclusively concerning the subject of xenophobia and misogyny. These intensity levels are qualified by the presence (or absence) of humour or sarcasm and attenuators or intensifiers of hatred (when they exist). This set of precautions allows the construction of the proposed semiotic analysis model for hate messages in digital environments.

Collection of Messages The choice of Twitter reflects the reasons indicated by the general bibliography. It is the social media platform where micronarratives have grown the most (exponentially) since their creation (García-DeTorres et al., 2011). In addition, its users are more inclined towards breaking news and following current affairs (García-De-Torres et al., 2011). On the other hand, Twitter has growing power in disseminating media-generated content and in the direct exercise of journalistic work through immediate contact (both the topics and the users). Twitter generates confidence among professional Spanish journalists regarding accessing reliable data in digital settings, but Facebook or Instagram do not enjoy similar credibility (Mayo-Cubero, 2019). It is a social media platform with the most significant presence of users, linked to traditional and digital media in Europe (González & Ramos, 2013). Finally, Twitter has more active user participation than other social media platforms (e.g., Facebook) (Aruguete, 2015; Mayo-Cubero, 2019). The data collection, from which the case study was derived, was carried out via web scraping and the Twitter API, based on Python. During January 2021, 221,454 messages published on Twitter were downloaded, associated with five of Spain’s leading digital news outlets (Statista, 2020a, 2020b, 2020c; Newman et al., 2020). This preprocessing followed the steps indicated by Zhang et al. (2018). To preserve the purity of the data collected, they were cleaned following the process indicated in Figure 2. Figure 2. Diagram of the cleaning process of messages collected on digital news media in Spain

116

 The Semiotics of Xenophobia and Misogyny on Digital Media

10,637 messages with hate expressions were identified (4.8% of the messages collected). 435 corresponded to misogynistic hate expressions (4.1%), and 617 corresponded to xenophobic hate expressions (5.8%). The identification of messages that could generate hate was performed manually. The following were taken into account: type of hate, its intensity and the presence (or not) of humour and the use of terms to intensify (e.g., Big) or attenuate (e.g., Small) these expressions. These included both the content generated by the professionals in the digital news media themselves and the comments made by their followers.

Theoretical Approaches and Focal Variables The analysis of the messages collected was based on the definition of hate speech used by the European Commission against Racism and Intolerance (2016) and the contributions of De Lucas et al. (2022) and Basile et al. (2018). The messages with hate speech are aimed at belittling, humiliating, or spreading prejudice against a vulnerable member or group (in this case, women, immigrants, and refugees). The next procedure was as follows: • •

Internal team training was carried out for the manual review and tagging of all messages. This work had the academic support of the authors of this chapter, who were in charge of all the conceptual clarifications required in this long process. Validation and quality control of the work was carried out by the team responsible for the review and labelling of messages, based on the following procedure: ◦◦ On a quantitative basis, the first validation identified the percentage of inconsistencies among taggers when classifying messages with hate speech. The work carried out was considered valid (by the authors in charge), with an inconsistency equal to or less than 10%. ◦◦ The second qualitative validation was a random review of the adequacy of the process of labelling the messages with hate speech, with a maximum level of inconsistency of 10%.

When identifying the messages with expressions of hate, both the level of intensity of the hate and the presence of attenuating terms (e.g., small) or intensifying terms (e.g., huge) were estimated to minimise them (or not). Hate intensities were categorised according to De Lucas et al. (2022): • • • • •

Intensity level 0: Messages that use hateful expressions without being considered as such. That is expressions with a clear negative social connotation that defines a trait associated with a person or group. Example: ‘Ultra-Catholic’. Intensity level 1: Messages that do not include express verbal violence but factually show data that stigmatise a specific person or social group. Example: ‘The Jews are behind all the anti-Covid vaccines. All the CEOs of the companies in charge are of this nationality’. Intensity level 2: Messages where attributions of bad intentions or abusive expressions towards a person or group are observed or made. Example: Immigrants come to rob us. Intensity level 3: Messages where verbal violence is manifestly exposed. Example: ‘Feminazi’. Intensity level 4: Messages with veiled or implicit threats towards a specific person or group. Example: ‘Be careful when you go hunting, and do not blow your fucking racist brains out’.

117

 The Semiotics of Xenophobia and Misogyny on Digital Media

•

Intensity level 5: Messages that want or call for actions with physical violence against someone or a specific group to which they belong. Example: ‘Those looks should be underground like Franco is. The best would be to kill them, kill them’.

These levels belong to fundamental semantics (Figure 3) and discursive syntax via thematisation or figurativisation. From there, as the diagram shows, the denotative and connotative are expressed (according to Greimas & Courtés (1979), Greimas (1976), Barthes (1970), Kristeva (1969), Lyotard (2019) and Nöth (2000, 2011). Based on what has been exposed by these authors, we first analyse the connotative framework of the message used as a case study in this work. Then, we move to the narrative of emancipation applied around it. In this way, it is possible to identify the associated speculative approaches to narratives used by the users who participated in the debate generated by the analysed publication. The latter is framed outside any grand narratives or those created by the focal medium (in our case El Mundo de España). Figure 3. Conceptual framework of semantic analysis applied in this study

In this way, the generative structures proposed in the poetics of Greimas (1976) are applied to messages of hate and how they unfold based on the dialogic promoted on social media platforms: A context where the image of the news media should be considered an incomplete map of a territory of contemporary life. This occurs because the media do not claim to represent reality but present their reality. Therefore, the analyst must understand the mediation and semiotics of the media and address semiosis; these communication processes must not be understood in the context of the exchange of messages around the published content but in the context of the production of senses and meanings constructed and received by the recipients of said messages (Peirce, 2007).

118

 The Semiotics of Xenophobia and Misogyny on Digital Media

RESULTS OF SEMIOTICS ANALYSIS Ninety-seven per cent of the comments associated with the tweet used as a case study include hate speech. Table 1 shows the different levels of hate intensity. Those with level one or level two intensity predominate. That is, those related to the text enunciated by Jennifer López (JLO) and titled ‘El Mundo’ are expressions of hate but do not include terms of aggressiveness, nor do they intend any specific contempt, although they draw the attention of both the newspaper and of the interactions posed by the authors and taken as referents (Figure 3). Therefore, there are no messages with expressions of hate with a precise exposure of verbal violence, a veiled threat or a call for physical violence. Table 1. Level of intensity of hate detected in messages associated with the tweet used as a case study      hate intensity _

     Tweets

     Percentage

     Intensity 0

     3

     9.09

     Intensity 1

     13

     39.39

     Intensity 2

     13

     39.39

     Intensity 3

     3

     9.09

     Intensity 4

     0

     0

     Intensity 5

     0

     0

     Not applicable

     1

     3

     Total

     33

     100

Via a global approach to the messages published by the news, from the Greimasian generative perspective, the following trajectory is observed: of the total of 33 comments, those corresponding to the numbers 2, 3, 4, 5, 8, 13, 15, 19, 20, 21, 22, 27, 28, 29, 30, 32 total 16. Around Biden, the American (as identity) and JLO, there are 17 (1, 6, 7, 9, 10, 11, 12, 14, 16, 17, 18, 19, 22, 23, 24, 26, 31). About JLO, Biden and the language, only 3: comments 22, 28 and 33. The narrative semantic and syntactic components account for a dialogue between actants in three senses: The first, related to the singer JLO and her Hispanic identity; the second, the relationship and difference between Spanish and Castilian; and the third, the inauguration of Biden. These three trajectories account for the different levels of hate intensity manifested in the discursive structures: spatiality, discoursivisation and actorialisation, without any reference to temporality. Spatiality refers to two places: the United States and the language debate in Spain. The discoursivisation makes intense manifestations of hate present that are not associated with veiled or implicit threats or intended to call for action with physical violence. Regarding the acting, the dialogue around the news could have provided more prominence among the actants. There are very well-determined ‘talk’ turns but no replies to them. Acting can be considered from the point of view of the diverse contributions among the authors of these comments. Finally, figurativisation has particular relevance in the discursive semantic structure of the two relevant figures: President Biden and the singer JLO. Both are the epicentre of a debate about speaking Spanish, the politics of the United States and Spanishness.

119

 The Semiotics of Xenophobia and Misogyny on Digital Media

This analysis, from the perspective of Barthes (1970), has to attend to three relevant elements in this structure: the denotative sign, the secondary sign and its connotation, and the metalanguage that derives from these processes. The denotations (as primary signs) provide the obvious: messages that express appropriate elements or respond to the news item published by the El Mundo user on Twitter. The connotative is associated with the figure of the ladder proposed by the author. In this case, it also attends to the content of any reaction to the news from nonlinear generative trajectories. These break the news content down via different approaches: comments directed towards Biden, comments directed towards JLO, and comments highlighting the speaking of Spanish or Castilian within the published news event. Metalanguage manifests in comments that can be understood in Spain, specifically as an opposition between Castilian and Spanish. It is approached with a certain contempt for everything Latin and is marked by the political distance between the models of Spain and the United States. The analysis of the comments published around the focal news item also shows a relationship over two weeks, following the semiotic square of Greimas (1976). The ‘structure is the mode of existence of meaning, characterised by the presence of the articulated relationship of two semes’ (p. 42): Biden’s inauguration and Spanish. Figure 4. Isotopic ratio, according to the semiotic square of Greimas (1976), about relationships associated with Biden’s inauguration and Spanish

It is necessary to define isotopic themes (the opposites in the discourse) to construct a semiotic square (Figure 4). This one, corresponding to the focal news item, shows that JLO speaks or does not speak Castilian or Spanish. Their relationship with this language is limited to whether or not they belong to the Latino community in the United States and their distant Hispanic origins. Semiotic squares are not generated in a cascade, and their relationships can be simplified (Figure 4). However, other relationships associated with the figure of Joe Biden and his policy towards Hispanic people can be considered (Figure 5).

120

 The Semiotics of Xenophobia and Misogyny on Digital Media

Figure 5. Semiotic square of Greimas (1976), about relationships associated with the figure of Joe Biden and his policy towards Hispanic people, can be considered

If the presence of xenophobia is considered, it manifests itself because JLO, although Latin in origin, hardly (little or very little) uses Spanish (Castilian) in her interventions. What stands out is that this singer appeared at Biden’s inauguration ceremony precisely as a representative of what is Latino without having any roots in Latino language and speech (Figure 6).

121

 The Semiotics of Xenophobia and Misogyny on Digital Media

Figure 6. Example of a comment published as a reaction to the news item used as a case study

If we analyse the model proposed in Figure 3, we see one of the examples of messages with level three hate intensity, that is, a message where verbal violence is manifestly exposed (Figure 7), a message where different types of hate are expressed: xenophobia, racism and ideology.

122

 The Semiotics of Xenophobia and Misogyny on Digital Media

Figure 7. Example of a comment published as a reaction to the news item used as a case study

This message on Twitter is an example of what is sought in denotative and connotative: it provides clues to the interlocutors (whether the author knows them of the Tweet) for creating generative structures. It is intended to establish a dialogic level with some of the hatred manifested in this message. If the chain of messages had continued (he stopped at this comment), perhaps the trajectory would have deviated from the deep level of metalanguage in which expressions relating to the American reality would have opened other avenues for debate. If we look at it as the final message from that dialogue, we find ourselves before a clear sign: the end of the discussion on this topic, not because it lacks interest, but because it is out of the general context of the news—JLO and her use of Castilian, as the newspaper ‘El Mundo’ suggests, via its Twitter user. If we continue to break the message down, connotative analysis of the discursive expressions, that is, their syntactic component, accounts for the hatred towards North Americans and their policies and demonstrates a hatred towards the citizens of the United States by calling them “sheep” (Figure 5). From this connotation, the message seeks a new denotation (semiosis), which has not occurred in the debate generated around the tweet of the news published by the user El Mundo on Twitter (@elmundo.es). It seeks to abandon the language debate (Spanish-Castilian) and propose new issues such as war and leftist ideology within this geopolitical behaviour in addition to moving the debate towards the racial confrontations of the police against Afro-American people. From Kristeva’s perspective (1969), the destructive-constructive is produced precisely in this phrase, as in Figure 5. Every time metalanguage leaves the discussion about languages and their good or bad use or name in America, leading to a level

123

 The Semiotics of Xenophobia and Misogyny on Digital Media

of intertextuality that seeks emancipation from the rest of the context, creating a new one, this coincides with Peirce’s semiosis (2007), bringing various types of hatred into the statement (Lyotard, 1979). Figure 6 shows messages with an intensity of level three or a manifest exposure of verbal violence. First, its direct relationship with JLO’s discourse should be considered. Only then can it be understood where the message comes from: the semiosis proposed in this comment, towards which the discourse wants to be directed—misogyny and xenophobia. The connotative analysis of this discursive expression has two critical linguistic components: on the one hand, ‘his’ (under quotation marks) and the insult ‘come on... and screw it!’ The relationship between what is supposed to be the Hispanic condition of the singer accompanied by an insult is a clear manifestation of hatred. It seeks to move the debate from the subject of language (Castilian-Spanish) to the use of speech of the singer with two optics: misogyny and xenophobia. The destructive intent of this discourse seeks the construction around the comment. It attempts one more twist on the problem of the newspaper headline and moves the dialogue to a new discursive construction under the protection of the metalanguage of insults. Some examples are the comments in Figures 8 and 9. Both have an intensity level of two (messages with bad intentions or abusive expressions attributed to a person or group). Figure 8. Example of a comment published as a reaction to the news item used as a case study

124

 The Semiotics of Xenophobia and Misogyny on Digital Media

The comment in Figure 8 is a nondescript image: an older gentleman with his hands folded in a reflective tone. Nevertheless, it shifts to a misogynist connotation due to the position of the hands. They make the gesture of a vagina (widely used by groups in favour of women’s rights), moving the metalanguage to an accusation or provocation of the dialogues in the chat against the singer. The semiosis is observed from the tone of the debate. Although the gentleman’s face tries to convey tranquillity and restraint, the gesture is eloquent and performed not by a woman but by a man. It seems to calm the collective conversation from the contempt for the femininity of the person who focuses on the analysed news (JLO). This devalues her femininity and, consequently, the group she represents (women, in general). Figure 9. Example of a comment published as a reaction to the news item used as a case study

Figure 9 denotes vomiting and connotes disgust for what is discussed concerning the news and was posted after the comment in Figure 7, involving multiple associated readings: First, a direct rejection of the comment immediately above. Second, a rejection of the language problems in all the debates generated around the news that was used as a case study. Third, it offers a new connotative element that seeks to describe the previous message and build a new message with a graphic hatred and, simultaneously, generate a semiosis that can be read with the connotations mentioned here. The metalanguage of emoticons is a clear example of the intertextuality around the news published by El Mundo through the associated comments. The faces that vomit are eloquent per se, but they also introduce a semiosis that could be read as typical of the boredom generated by the problem concerning the names of languages and the Catalan conflict itself. Thus, the breadth of the hate message in Figure 6 is typical of the metalanguages Lyotard (2019) has exposed: from the intertextuality of emancipation

125

 The Semiotics of Xenophobia and Misogyny on Digital Media

and its ability to offer a new meaning to the conversation between tweets. In short, something observable in communication processes, such as the chosen case, is generated around a piece of news in which new meanings are constructed and received and are very different from the initial meaning and develop around it (semiosis). A sign, in short, provides a dimension not considered within the group of signs, ultimately producing a break in the dialogue by introducing that new sign. The GIF shown in the comment of Figure 10 deserves special mention. Text accompanies a graduate who jumps and seems to say: ‘A Latina speaking Spanish’. At the level of words, the message contains expressions of hate with intensity level two (messages where attributions of bad intentions or abusive expressions towards a person or group are observed or made), with a component that is both xenophobic and misogynist, even racist. Here, the connotation within the communicative context of the news produces new generative structures (Greimas & Courtés, 1979; Greimas, 1976) and other types of connotations (Barthes, 1970). First, it refers to a woman. In addition, she is Latina (Hispanic American, Latin American, Ibero-American). Last, she speaks Spanish. These elements destroy and build the jumping graduate (GIF from the comment), a semiosis emancipated from the other texts (Kristeva, 1969; Lyotard, 1979). This new meaning, loaded with derision, focuses on the possibility of having a university degree since the (tacit) intensifier is that Latinas do not speak Spanish. Figure 10. Example of a comment published as a reaction to the news item used as a case study

Finally, the video of the penultimate comment (Figure 11) adds another twist to the debate. It is no longer about the language but a text that contains a Republican affirmation of the election of Biden: a scam. It is built with image overlays: a guardrail cap and moustache and a handwritten text stating, ‘I’m a Fraud’. This leads to intense hate speech at level three: verbal violence is exposed manifestly. This dialogue is definitively emancipated from the language’s subject regarding the act’s politics. This new connotation-denotation accurately expresses Kristeva and Lyotard’s proposal for the function of language as a constructor-destroyer and proponent of intertextuality that seeks emancipation.

126

 The Semiotics of Xenophobia and Misogyny on Digital Media

Figure 11. Example of a comment published as a reaction to the news item used as a case study

127

 The Semiotics of Xenophobia and Misogyny on Digital Media

FUTURE RESEARCH DIRECTIONS The analysis proposals derived from the semiotics offered in this work can be extended without limiting themselves to the focal referents. In this case, a choice has been made following the classic authors and referents. This theoretical exposition and synthesis later applied to a relatively small corpus may be surprising, but the interest is in bringing semiotic analysis closer to massive, highly variable, and transformable communication situations in terms of their “conversations”. Thus, this specific use is proposed as a reference to favouring the opening of new lines of research directed at topics and, above all, journalistic discourses in which information professionals are involved with readers who have become authors in conversations at the same level, in the sense that both can create and add new meanings. One of these new, more specific lines of work can be framed in the study of iconicity of Peirce (2007). This would make it easier to study memes and other graphic and iconic expressions associated with hate speech present in messages on Twitter and other contemporary digital scenarios.

DISCUSSION AND CONCLUSION The limitations inherent to studies such as this work (limited to a single medium and a single news item published by it on Twitter), the denotative and connotative analysis applied to the 33 comments related to the focal news item has allowed the development of a matrix analysis of each2. This allows us to understand, via semiotics, the symbols, signs and forms that users use when expressing xenophobic hatred and misogyny statements. Each comment studied separately cannot account for what Barthes (1970) has suggested in his studies. However, as a whole, they allow us to observe the connotative character associated with the news, which allows us to delve, via semiotics, into how hate speech is disseminated by this type of actor (users related to the news media in Spain) who moves in digital scenarios. The approaches of Greimas (1976) and Greimas and Courtés (1979) that have been used here in terms of their generative trajectory allow assessing the meaning of chained messages, discursive production processes developed in several stages and with syntactic and semantic components specific to each case. These messages are intertwined with each other via different meanings (both superficial and deep). As a whole, they comprise the corpus of news and comments that confront each other to transform the reality that the news shows across the exchanges of meanings in any Twitter “conversation”. The proposal offered here thus enables delving into the discursive production associated with the dissemination of hate speech, which, in this case, has also been labelled based on intensity and is thematically limited to xenophobia and misogyny. Rodriguez’s approaches (2022) remain open when he follows Lyotard (2019) in analysing the current digital scenario by focusing on one fact: It is a new space that favours the creation and dissemination of news or content at unknown levels. In addition, disinformative messages, hate speech, and news are published with equal “rights” in them. This is due to the inherent legitimacy of the content published by users who participate and assume the position of ‘knowing’ the issues, giving rise to the different debates generated in this current digital scenario. These are characterised by being, many times, based on a lack of internal consistency and experimental verification. Nöth’s contributions to self-referentiality have not been ignored: The actants actively create their profiles and seek notoriety in the dialogical processes that digital environments facilitate. This approach 128

 The Semiotics of Xenophobia and Misogyny on Digital Media

allows analysing hate speech in these specific contexts. It is thus possible to value the authors and their uses (voluntary or not) of certain expressions to enhance their position in the debates concerning the rest of the participants. Additionally, there is the democratic spirit that these scenarios theoretically favour. In the case studied here, this fosters the loss of values associated with tolerance and coexistence and other values related to respect for particular groups (and their members): women and immigrants. Other authors, including Barrios-Rubio and Fajardo (2022) and Snoussi and Korbi (2021), point out that this type of expression on social media platforms and current digital communication scenarios abounds. The interaction codes in these scenarios act as extensions of individual identity in society (Gaines, 2015). Thus, developing relationships around news in digital contexts can provide context (Galofaro & Toffano, 2022; Greimas & Courtés, 1979). The semiotic analysis thus helps identify the structures associated with the connotations and denotations related to hate speech by users and in the news published in this context in Spain. This is a framework where users who interact with tweets published on Twitter and messages on other digital communication platforms use connotations and denotations to reaffirm their positions towards, prejudices of and stereotypes about certain people or the groups they represent, especially women and immigrants.

ACKNOWLEDGEMENT The chapter results from the project ‘Taxonomy, presence and intensity of hate speech in digital environments linked to the Spanish professional news media - Hatemedia (PID2020-114584GB-I00)’, financed by the State Research Agency - Ministry of Science and Innovation.

REFERENCES Alirangues, M. (2018). La palabra precaria. Prolegómenos para una teoría negativa de la ficción. Escritura e imagen, 14, 149. Alkiviadou, N. (2018). The Legal Regulation of Hate Speech: The International and European Frameworks. Politička misao, 55(4), 203-229. doi:10.20901/pm.55.4.08 Amores, J. J., Blanco-Herrero, D., Sánchez-Holgado, P., & Frías-Vázquez, M. (2021). Detectando el odio ideológico en Twitter. Desarrollo y evaluación de un detector de discurso de odio por ideología política en tuits en español. Cuadernos.Info, 49(49), 98–124. doi:10.7764/cdi.49.27817 Arce-García, S., & Menéndez-Menéndez, M.-I. (2023). Inflaming public debate: A methodology to determine origin and characteristics of hate speech about sexual and gender diversity on Twitter. El Profesional de la Información, 32(1), e320106. doi:10.3145/epi.2023.ene.06 Arrocha, W. (2019). Combating xenophobia and hate through compassionate migration: The present struggle of irregular migrants escaping fear and extreme poverty. Crime, Law, and Social Change, 71(3), 245–260. doi:10.100710611-019-09833-w Aruguete, N. (2015). El poder de la agenda. Política, medios y público. Biblos.

129

 The Semiotics of Xenophobia and Misogyny on Digital Media

Barrios-Rubio, A. & Fajardo, G. (2022). The Media in the Construction of Reality in the Context of Colombian Social Nonconformity. Information, 13(8), 378, 1-14. https://doi.org/ doi:10.3390/info13080378 Barthes, R. (1970). L’ancienne rhétorique. Communications (Englewood), 16(1), 172–229. doi:10.3406/ comm.1970.1236 Basile, V., Bosco, C., Fersini, E., Nozza, D., Patti, V., Rangel, F., Rosso, P., & Sanguinetti, M. (2019). SemEval-2019 Task 5: Multilingual Detection of Hate Speech Against Immigrants and Women in Twitter. En Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), (pp. 54–63). Association for Computational Linguistics. 10.18653/v1/S19-2007 Bernal, P., Villadiego, M., & Urbanczyk, M. (2013). Narrativas publicitarias de la modernidad en la prensa colombiana de los años treinta. Poliantea, IX(16), 211–239. Bohórquez, D. (1997). Julia Kristeva: teoría, proceso e interpretación del sentido. Signa: revista de la Asociación Española de Semiótica, 6. Bonacho, M. (2021). La responsabilidad de los medios de comunicación al informar sobre la comisión de delitos de odio. Economist & Jurist, 29(255), 68–73. Bourou, M. (2021). Diversity is a code word. And what it means is white genocide! Journal of Language Aggression and Conflict, 10(1), 219 239. https://doi.org/ doi:10.1075/jlac.00060.bou Breazu, P., & Machin, D. (2022). Racism is not just hate speech: Ethnonationalist victimhood in YouTube comments about the Roma during Covid-19. Language in Society, 1–21. doi:10.1017/S0047404522000070 Brindle, A. (2016). Cancer has nothing on Islam: A study of discourses by group elite and supporters of the English defense league. Critical Discourse Studies, 13(4), 444–459. doi:10.1080/17405904.201 6.1169196 Contreras, J.M., & Sánchez, R. (dir.) (2021). El tratamiento normativo del discurso del odio. Thomson Reuters Aranzadi. Cuthbertson, L. (2019, Nov 25). Women, politics and Twitter: using machine learning to change the discourse. arXiv preprint. https://arxiv.org/abs/1911.11025 Damcevic, K., & Rodik, F. (2018). Ready for the Homeland. Romanian Journal of Communication and Public Relations, 20(3), 31. doi:10.21018/rjcpr.2018.3.264 De LucasA.RömerM.Izquierdo-AlegríaD.Montero-DíazJ.Said-HungE. (2022). Manual para el etiquetado de mensaje de odio en medios digitales en España. doi:10.6084/m9.figshare.18316313 Elias, S., & Gurbanova, N. (2018). Relocating Gender Stereotypes Online: Critical Analysis of Sexist Hate Speech in Selected Social Media. Proceedings of the International Conference on Language Phenomena in Multimodal Communication (KLUA 2018). https://www.atlantis-press.com/proceedings/ klua-18/25900116 Espino, T. (2019). Palabras extrañas a sí mismas una aproximación desde la hermenéutica intercultural al bilingüismo de Julia Kristeva. RILCE. Revista de Filología Hispánica, 1(35), 117–136.

130

 The Semiotics of Xenophobia and Misogyny on Digital Media

Esposito, E. (2021). The visual semiotics of digital misogyny: Female leaders in the viewfinder. Feminist Media Studies, 1–17. doi:10.1080/14680777.2022.2139279 European Commission against Racism and Intolerance. (2016). Recomendación General Nº 5 relativa a la lucha contra el discurso de odio y memorándum explicativo. Unión Europea. https://rm.coe.int/ ecri-general-policy-recommendation-n-15-on-combating-hate-speech-adopt/16808b7904 Evenden-Kenyon, E. (2020). Educating the English: The role of universities in tackling hate speech and Islamaphobia in post-EU-Referendum Britain. Papeles de Europa, 32(2), 129–144. doi:10.5209/pade.68027 Fernández, I. B. (2010). Metalenguaje interactivo. Herramientas en la red para los nativos digitales. ICONO 14, Revista de comunicación y tecnologías emergentes, 8(1), 274-288. Francescutti, L. F. (2020). La semiótica de la comunicación masiva en España: Sus avatares académicos contados por sus protagonistas. Revista Signa, 29, 391–414. doi:10.5944igna.vol29.2020.23455 Gaines, E. (2015). Undestanding the Codes and Assumptions of New Media. In P. P. Trifonas (Ed.), International Handbook of Semiotics (pp. 369–376). Springer., doi:10.1007/978-94-017-9404-6_16 Gallo, G. I. ., & Galmarini, M. R. . (2021). Reactivación del turismo rural post cuarentena 2020: TIC y adaptaciones aplicadas por emprendedores. Revista Intersección. Eventos, Turismo, gastronomía Y Moda, 2(4), 72–82. Galofaro, F. & Toffano, Z. (2022). Enunciation and topic/comment structure: the offensive replies to Pope Francis’ tweets. RIFL, 243-253. doi:10.4396/SFL2021A23 García, G. (2019). Quizças… Barthes y Derrida. Acta Poética, 40(1), 63–85. doi:10.19130/iifl. ap.2019.1.846 García-De-Torres, E., Yezers’ka, L., Rost, A., Calderín, M., Edo, C., Rojano, M., Said-Hung, E., Jerónimo, P., Arcila, C., Serrano-Tellería, A., Sánchez-Badillo, J., & Corredoira, L. (2011). Uso de Twitter y Facebook por los medios iberoamericanos. El Profesional de la Información, 20(6), 611–620. doi:10.3145/ epi.2011.nov.02 Ging, D., & Siapera, E. (2019). Gender hate online: understanding the new anti-feminism. Palgrave Macmillan. doi:10.1007/978-3-319-96226-9 Giraldo Pérez, S. (2022). Delitos de odio. Incidencia de la pandemia COVID en los delitos de odio en España. Sociología y tecnociencia, 12(1), 216- 240. doi:10.24197/st.1.2022.216-240 González, S., & Ramos, F. (2013). El uso periodístico de Facebook y Twitter: Un análisis comparativo de la experiencia europea. Historia y Comunicación Social, 18(0), 419–433. doi:10.5209/rev_HICS.2013. v18.44253 Greimas, A. J. (1976). Semántica estructural. Editorial Gredos. Greimas, A. J., & Courtés, J. (1979). Semiotics and Language. Indiana Press University. Guizardi, M. L., & Mardones, P. (2020). Las configuraciones locales de odio. Discursos antimigratorios y prácticas xenofóbicas en Foz de Iguazú, Brasil. Estudios Fronterizos (México), 21. Advance online publication. doi:10.21670/ref.2003045

131

 The Semiotics of Xenophobia and Misogyny on Digital Media

Ibri, I. A., Alves, M. A., & De Tienne, A. (2021). Semiótica e Pragmatismo: Interfaces Teóricas-Vol. II. Editora Oficina Universitária. doi:10.36311/2021.978-65-59541-28-7 Juárez, J., & Restrepo, N. (2021). El auge de los discursos de odio y el papel de la ética y los medios de comunicación. En: Almudena Barrientos-Báez (coord.). XIII Congreso Internacional Latina de Comunicación Social 2021: Libros de Actas. Comunicación y Nuevas Tendencias. Kerbrat-Orecchioni, C. (1983). La enunciación. De la subjetividad en el lenguaje, 117-135. Hachette. Khoma, N., & Oleksii Kokoriev, O. (2021). Compliance of the Baltic States with the Principle of Tolerance as Condition for the Development of the United Europe. The Age of Human Rights Journal, 16(16), 186–206. doi:10.17561/tahrj.v16.6038 Kristeva, J. (1969). Le langage, cet inconnu. Seuil. Landowski, E. (2022). As metamorfoses da verdade, entre sentido e interação. Estudos semióticos, 18(2), pp. 1-23. doi:10.11606/issn.1980-4016.esse.2022.198273 Llorente-Barroso, C., Kolotouchkina, O., & García-García, F. (2021). The meaning of the logo from its semiotic construction and its reliance on new formats of digital communication: The case of Apple: El caso de Apple. Revista Latina de Comunicación Social, 79, 333-356. https://www.doi.org/10.4185/ RLCS-2021-1529 Lyotard, J-F. ([1979] 2019). La condición postmoderna. Informe sobre el saber. Cátedra, ebook. Martínez, L. (2023). Mensajes misóginos en lo perfiles de Tik Tok de la prensa española: Estudio y percepción por parte de los jóvenes. Ámbitos, 59, 110–127. doi:10.12795/Ambitos.2023.i59.08 Mayo-Cubero, M. (2019). Uso de las redes sociales en la cobertura periodística de crisis, desastres y emergencias en España. Revista Española de Comunicación en Salud. 43-54. doi:10.20318/recs.2019.4428 Megías, I., Amezaga, A., García, M. C., Kuric, S., Morado, R., & Orgaz, C. (2020). Romper cadenas de odio, tejer redes de apoyo: los y las jóvenes ante los discursos de odio en la red Juventud. Fad. doi:10.5281/zenodo.4288486 Merklejn, I., & Wiślicki, J. (2020). Hate Speech and the Polarization of Japanese National Newspapers. Social Science Japan Journal, 23(2), 259–279. doi:10.1093sjj/jyaa015 Meseguer, S. (2022). Incitación al odio por razón de las creencias en los medios publicitarios. In F. Pérez-Madrid. Discurso de odio y creencias (pp. 91–114). Thomson Reuters Arazandi. Moreno Barreneche, S. (2022). Identidades estetizadas, interacciones mediatizadas: Un abordaje de Instagram desde la sociosemiótica. Comunicación y Medios, 31(45), 129–141. doi:10.5354/07191529.2022.64515 MüllerK.SchwarzC. (2020). Fanning the Flames of Hate: Social Media and Hate Crime. https://ssrn. com/abstract=3082972 doi:10.2139/ssrn.3082972 Newman, N., Fletcher, R., Schulz, A., Andi, S., & Kleis, R. (2020). Digital News Report 2020. University of Oxford.

132

 The Semiotics of Xenophobia and Misogyny on Digital Media

Nöth, W. (2000). Handbook of semiotics. Indiana Press University. Nöth, W. (2011). Self-referential postmodernity. Semiotica, 183(1), 199–217. 10-1515/semi.2011.010 Ortega, C. (2022). Los medios de comunicación ante el discurso de odio: ¿Plataformas de difusión o “watchdogs” de los derechos fundamentales? En: A. Daunis Rodríguez. Odio y discriminación en tiempos convulsos, 69-80. Comares. Pamungkas, E.-W., Basile, V., & Patti, V. (2020). Misogyny detection in Twitter: A multilingual and cross-domain study. Information Processing & Management, 57(6), 102360. Advance online publication. doi:10.1016/j.ipm.2020.102360 Park, S. (2017). Inventing aliens: Immigration control, ‘xenophobia’ and racism in Japan. Race & Class, 58(3), 64–80. doi:10.1177/0306396816657719 Parra, E. (2020). ¿Por qué es necesaria una historia de la semiótica?: En busca de la identidad y la memoria a reconstruir. CIC: Cuadernos de información y comunicación, 25, 47-60. Paz-Rebollo, M. A., Cáceres-Zapatero, M. D., & Martín-Sánchez, I. (2021). Suscripción a la prensa digital como contención a los discursos de odio. El Profesional de la Información, 30(6), e300613. doi:10.3145/epi.2021.nov.13 Peirce, Ch. S. (2007). La lógica considerada como semiótica: índice del pensamiento peirciano. Biblioteca Nueva. Piñeiro-Otero, T., & Martínez-Rolán, X. (2021). Say it to my face: Analysing hate speech against women on Twitter. El Profesional de la Información, 30(5), e300502. doi:10.3145/epi.2021.sep.02 Plaza-Del-Arco, F. M., Molina-González, M. D., Ureña-López, L. A., & Martín-Valdivia, M. T. (2020). Detecting Misogyny and Xenophobia in Spanish Tweets Using Language Technologies. ACM Transactions on Internet Technology, 20(2), 1–19. doi:10.1145/3369869 Ponce, F., García, C., Cárdenas, G., Gómez, M., & Peña, A. (2022). Acercamiento al cuento “La careta” de Julio Ramón Ribeyro a partir de la semiótica de Greimas y del Grupo Entrevernes. Sincronía. Año, XXVI(82). doi:10.32870incronia.axxvi.n82 Priyadharshini, R. (2022). Overview of Abusive Comment Detection in Tamil. ACL. https://aclanthology. org/2022.dravidianlangtech-1.44.pdf Ramponi, A., Testa, B., Tonelli, S., & Jezek, E. (2022). Addressing religious hate online: From taxonomy creation to automated detection. PeerJ. Computer Science, 8, e1128. doi:10.7717/peerj-cs.1128 Rodríguez, A., & Lázaro Pernias, P. (2022). Nuevas narrativas en publicidad. El storydoing como modelo de comunicación que favorece la cocreación del relato de marca entre los usuarios de Instagram. AdComunica, 24, 57–80. doi:10.6035/adcomunica.6554 Rodríguez, W. J. (2022). De la posverdad a la hiperrealidad: ¿Qué hay detrás de la difusión viral de “Fake News”? Quaestiones Disputatae: Temas En Debate, 15(30), 86-105. http://revistas.ustatunja.edu. co/index.php/qdisputatae/article/view/2599

133

 The Semiotics of Xenophobia and Misogyny on Digital Media

Rubio, D. (2022). Protocolo y semiótica como estrategia de comunicación política. SIGNUM: Revista Internacional De Investigación En Eventos, Protocolo Y Relaciones Institucionales, 1(1), 1–10. https:// publicaciones.protocoloimep.com/signum/article/view/5 Sánchez-Holgado, P., Amores, J. J., & Blanco-Herrero, D. (2022). Online Hate Speech and Immigration Acceptance: A Study of Spanish Provinces. Social Sciences (Basel, Switzerland), 11(11), 515. doi:10.3390ocsci11110515 Silva, S., Castro Ferreira, T., Silva Ramos, R. M., & Paraboni, I. (2020). Data Driven and Psycholinguistics Motivated Approaches to Hate Speech Detection. Computación y Sistemas, 24(3). doi:10.13053/ cys-24-3-3478 Slavíčková, T., & Zvagulis, P. (2014). Monitoring anti-minority rhetoric in the Czech print media. Journal of Language and Politics, 13(1), 152–170. doi:10.1075/jlp.13.1.07sla Snoussi, T., & Korbi, W. (2021). Cyber anti-Hate speech during the Covid-19 pandemic: Semiotic Analysis. The Asian ESP Journal, 17(4.2), pp. 169-189 Statista. (2020a). Ranking de los 10 medios de comunicación con más seguidores en Twitter en España en abril de 2020. Statista. https://es.statista.com/estadisticas/518802/Twitter-perfiles-de-medios-decomunicacion-con-mas-seguidos-en-espana/ Statista. (2020b). Páginas oficiales de medios de comunicación más populares en Facebook en España en abril de 2020, por número de seguidores. Statista. https://es.statista.com/estadisticas/518580/paginasdemedios-de-comunicacion-de-Facebook-con-mas-seguidores-en-espana/#statisticContainer Statista (2020c). Ranking de las principales marcas de medios de comunicación online según el porcentaje de población que las usaba de forma semanal en España en 2020. Statista. https://es.statista.com/ estadisticas/670743/principales-medios-online-porcentaje-de-usuariossemanales-espana/r Sumalla, J. M. (2018). Los delitos de odio en las redes sociales. IDP Revista de Internet Derecho y Política, 27(27). doi:10.7238/idp.v0i27.3151 Vela, C. (2020). A semiotic approach to the study of emojis. Círculo de Lingüística Aplicada a la Comunicación, (84), 153–165. Vidales, C., Parra, E., & Lanigan, R. L. (2021). Presentación: Comunicación y Semiótica. Comunicación y Sociedad (Guadalajara), 1-3(0), 1–3. Advance online publication. doi:10.32870/cys.v2021.8112 Yamaguchi, T. (2013). Xenophobia in Action. Radical History Review, 2013(117), 98–118. doi:10.1215/01636545-2210617 Yao, X. (2021). Face masks, materiality and exclusion in the COVID-19 semiotic landscape. Social Semiotics, 1–21. doi:10.1080/10350330.2021.2016032 Zamri, N. A. K. (2021). Coronavirus Exacerbates Xenophobia: Deconstructing Otherness. In The Twitter. European Proceedings of Social and Behavioural Sciences., doi:10.15405/epsbs.2021.06.02.43 Zaya, G. M., Vázquez, Á. F. J., & Morgunova, E. (2015). Metodología para la enseñanza del análisis semiótico del texto poético. Didasc@ lia. Didáctica y Educación, 6(1), 167–190.

134

 The Semiotics of Xenophobia and Misogyny on Digital Media

Zhang, Z., Robinson, D., & Tepper, J. (2018). Hate Speech Detection Using a Convolution-LSTM Based Deep Neural Network. In Proceedings of ACM The Web conference (WWW’2018). ACM. https://doi. org/10.475/123_4

ENDNOTES 1



2



Comment number 25 was not analysed, as it was written in Catalan. This message contains “intolerapla”, which translates as intolerable. Although it could generate intertextual emancipation, it does not generate any interaction on the part of the actants regarding the news item studied in this paper. Of course, the comment in itself demonstrates that the discourse related to languages places Spanish as the language spoken in part of the Spanish territory and not as the language of all of Spain. This distinction or note in the chat in terms of another of the co-official languages of the Spanish State is not trivial and should be considered a point of construction attention regarding what was destroyed by the members of the debate on the news item studied in terms of language. Access to the analysis table carried out by the authors https://acortar.link/tymjPi

135

136

Chapter 8

Mapping Stigmatizing Hoaxes Towards Immigrants on Twitter and Digital Media: Case Study in Spain, Greece, and Italy Marta Sánchez Esparza https://orcid.org/0000-0001-6525-0148 Rey Juan Carlos University, Spain Ignacio Vázquez Diéguez https://orcid.org/0000-0002-7938-5446 Universidad de Beira Interior, Portugal Adoración Merino Arribas https://orcid.org/0000-0002-3294-9996 Universidad Internacional de La Rioja, Spain

ABSTRACT This text presents a mapping of the hoaxes published on Twitter and in digital media during 2020 in Spain, Greece, and Italy after having been classified as disinformation with the intention of causing harm in the fact-checking portals of the three cited countries. Verification services that are members of the International Fact-Checking Network (IFCN) have been chosen for this analysis: Spain (Newtral.es/ Maldita.es), Greece (Ellinika Hoaxes), and Italy (Facta News/Lavoce.info). The chosen online portals belong to FactCheckEU, a European project launched by the international verification network. The validated sample presents 150 pieces of information identified as hoaxes by the verification platforms and disseminated in the current communication scenarios by the network in the three direct recipient countries of the migratory phenomenon through the Mediterranean. A qualitative methodology applied to the case study is used, which is complemented by critical discourse analysis.

DOI: 10.4018/978-1-6684-8427-2.ch008

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

INTRODUCTION In a world context marked by global migration movements and multiple information channels on the net, a study on the stigmatising hoaxes towards immigrants in Europe is necessary, specifically in three Mediterranean countries receiving the population (Spain, Greece, and Italy). Keep in mind that in the three nations previously mentioned, conflicts have led to the rejection of those who arrive. In this study, we align ourselves with Berger and Luckman (1966) to determine that rejection is a socially constructed category. From this derives the importance of verifying that the information related to immigration, disseminated in digital media and Twitter, respects deontology and does not intend to stigmatise people or migrant groups. Precisely, stigmatising hoaxes fall within the scope of disinformation, which, according to a report by the European Commission (HLEG, 2018), refers to inaccurate, misleading, or false information, always to cause harm. To better understand the case study, we align ourselves with Magallón (2019), who encompasses, in what has been called ‘unfaking news’, not only disinformation but also fake news, misinformation, false stories, news verification, and false news, to clarify the process of circulation of information that is not true and that affects current communication scenarios that are disseminated through the network. In addition, in line with UNESCO (2015), the European Union also considers disinformation as verifiably misleading information, that is, information that, when going through verification platforms, is detected as a hoax (Salaverría et al., 2020). For this reason, in this work, reliable verification platforms are taken as a reference that is accredited in each of the analysis countries and that have signed a code of principles that commit them to detecting hoaxes about immigrants published in digital media as reliably as possible. Disinformation made a massive splash in public opinion in 2016, with two events: Brexit and the election of Trump (Alonso-García et al., 2020; Bakir & McStay, 2018). These two events of a political nature have been interpreted as disinformation strategies to obtain political revenue (Molina, 2019). Since then, the audience has used the term fake news to refer to false news of all kinds. However, there is only misinformation if there is an intention to deceive (Fallis, 2015). New academic research on fake news specifies that this false or distorted information (disinformation) imitates the content of reliable media, always to deceive the final recipient, especially through social media, where they find the ideal channel as part of the “new discursive order” (Camargo, 2021; Chenzi, 2021). This is the main argument that justifies the need for verification systems to detect hoaxes, that is, false information that carries the intention of harming. For this reason, fact-checking platforms were born to explain to the audience what information they have in front of their eyes. Academic studies such as Wardle and Derakshan’s (2018), Kapantai et al.’s (2021), or Ireton and Posseti’s (2018) study how fact-checking platforms detect erroneous information, classified as misinformation, disinformation, and mal-information, and certify that it is not true, because they act as the only real firewall in the situation of disinformation (Carr et al., 2020). Other investigations progress to 11 forms of misinformation (Kapantai et al., 2021): clickbait, misleading connections, fabricated misinformation, impostor misinformation, hoaxes, biased or one-sided information, pseudoscience, conspiracy theories, trolling, fake reviews, and rumours (Kapantai et al., 2021). Precisely, this is the context of news reproduction through social media, where the focus must be placed to discern if this information is true or, on the contrary, since it is news reproduction, greater 137

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

vigilance must be observed to fight against pseudo-fast news through Facebook or Twitter, which are the hegemonic social media, but knowing that these networks are not journalistic media, subject to accountability, but simple means of transmission(Waisbord, 2018). Twitter has 400 million users worldwide and We are social/Hootsuite (2021) reveals the sentiment of public opinion through the 500 million short messages per day that are broadcast on that platform. Social media thus positively impacting the population (Sayce, 2020). In this regard, social media are already the primary setting for hoax dissemination, according to Masip et al. (2020). More specifically, regarding migration crises, for example, in 2015 up to 7.5 million tweets were collected with the hashtags #refugees and #refugeecrisis (Siapera et al., 2018). This provoked the rejection of foreigners in society while the study of the strategies followed in social media by anti-immigration movements or xenophobic groups was fed (Ekman, 2019). In these scenarios, populist political parties take the opportunity to gain electoral gain and focus their discourse on Twitter on provocation and polarisation or the dissemination of hoaxes against immigrants (Vampa, 2020). Due to the ability of social media to influence public opinion, verification initiatives have been developed, both for social media and for journalistic media. In this paper, we will focus on those three European countries that are the object of our study: the hoaxes spread about immigrants, which have been detected as such by fact-checking platforms in Spain, Greece, and Italy. It also analyses whether disseminating this information through social media generates stigmatisation and can even cause hate speech about the migratory phenomenon and the people that comprise this group. For this, an analysis sheet of the terms that entail rejection towards migrants or stigmatising stereotypes towards people or groups is prepared by exposing a corpus that helps to measure this rejection. However, we recognise that there is no universally accepted definition of hate speech since it is a term with imprecise and equivocal profiles, which fits well with journalistic grammar (Rey, 2015). For their part, Aguilera-Carnerer and Azeez (2016) point out another added difficulty in defining hate speech because it is multifaceted, above all, in a technological society where media discourse develops in a context favourable to interpretation, many times without argument, and the reader, based on the shared ideology, does the rest (Van Dijk, 2021). In this research, we share what the United Nations (2013) affirms when it explains that hate speech is understood as any communication in speech, writing, or behaviour that attacks or uses pejorative or discriminatory language concerning a person or group, based on their religion, ethnicity, nationality, race, colour, ancestry, gender, or other identity factors. The European Commission against Racism and Intolerance [ECRI] (2017) also warned about fears and the growing uncertainties monopolised by nationalist and xenophobic populist movements throughout Europe, where foreigners appear as a threat, causing racist insults in society regularly, showing xenophobic hate speech to unprecedented heights. Hence, an important part of the analysis includes anti-immigrant political discourses issued in exclusion migration policies, which disproportionately impact migrants in the European Union, turning them into victims of violence and racist discourses. For her part, Elósegui (2017) qualifies hate speech as discriminatory harassment because there is no right to insult. Among the hoaxes detected in this work are headlines that manifest this harassment and insult towards immigrant people or nationalities. Bauman (2016) also draws attention to the speeches to exacerbate the fear of the stranger, a frequent practice throughout history. Due to the immediacy and diffusion of social media and the Internet, the impact of these discourses is relevant for social cohesion and peaceful coexistence.

138

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

According to Estrada-Cuzcano et al., (2020), we move in a culture of misinformation, in which Twitter is the protagonist in the dissemination of messages, which can lead to misinformation and digital scenarios make it difficult to access accurate information (López & Rodríguez, 2020; Mazzoleni & Bracciale, 2018). In conclusion, it is necessary to show citizens the misinformation and provide tools for network users to acquire skills to fight against false messages. The denials guide citizens not to consume hoaxes and also to bring to light that the beneficiaries of disinformation are hate speech and ideological polarisation that feed radicalism or extremism (Pérez-Escolar & Herrero-Diz, 2023).

ORDEVAL APPROACH As a general objective, this work proposes to map (represent graphically and visually a set of elements of the same type) the main stigmatising hoaxes launched around immigrants who arrive in Spain, Greece, and Italy and that have been detected by the platforms of fact-checking as misinformation with the intent to harm, after being published on Twitter and digital media during 2022. This case study investigation takes as a methodological reference that of Notario and Cárdenas (2020), which typifies and quantifies the content of misinformation on migration based on a sample of Maldita Migración. For this, verification portals, members of the International Fact-Checking Network (IFCN), have been chosen and, therefore, have committed to the code of conduct that obliges them to have an open and honest verification policy, together with the commitment to be impartial, non-partisan, and transparent regarding its sources, financing, organisation and working methodology (IFCN, 2021). Specifically: Spain (Newtral.es/Maldita.es), Greece (Ellinika Hoaxes), and Italy (Facta news/Lavoce.info/Pagella Politica). In addition, these chosen online portals belong to FactCheckEU, a European project launched by the international verification network. A qualitative methodology is used, like Hernández-Sampieri and Mendoza (2018) or Gamir-Ríos et al. (2021), which is complemented by critical discourse analysis, in line with Baker et al. (2011), resorting to specific approaches or concepts of anthropology, rhetoric, conversation analysis, semantics, pragmatics and sociolinguistics, suitable for approaching complex social phenomena, such as stigmatising labels towards immigrants. Hate expressions towards immigrants are determined and quantified here, like in works like that of Said-Hung et al. (2021), whereas the datasets of Baviera et al. (2019). When analysing whether the hoaxes thrown by the fact-checkers contain explicit language or language associated with hate speech against immigrants, lemmatisation techniques are used (Figure 1). To achieve this goal, three specific objectives are defined: 1. Detect hoaxes through fact-checking sites in the three Mediterranean countries (Spain, Greece, and Italy), whose verification platforms belong to IFCN. 2. Compare the discourse used in hoaxes against immigrants in each country. 3. Prepare a scale of levels of hate in the expressions used in hoaxes of the different countries.

139

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Figure 1. Process of processing the analysed messages

The results of the case studies are obtained by applying quantitative tools on the analysis sheet. These quantitative tools are applied to each case (Spain, Greece, and Italy) where the hoax detected by the corresponding fact-checking platform is collected. The comparative study of the three countries makes it possible to evaluate the quantity and quality (content) of the hoax already denied and the format used (text, image, video, and audio). Finally, to detect the clichés associated with immigrants, we align ourselves with Wardle (2017), who exposes the scale that measures the intention to deceive: parody (it does not intend to cause harm but has the potential to deceive); false connection (when the headlines, images, or subtitles do not support the content); misleading content (misleading use of information to incriminate someone or something); false context (genuine content that is shared outside of its original context); impostor content (when genuine sources are impersonated); manipulated content (when genuine information or images are manipulated to add, a format that can be extended to video or audio contents); and fabricated content (content that is 100% false and is designed to mislead and harm).

RESULTS In the results that are extracted from the analysis sheet, the presence of expressions of rejection towards immigrants, which have been certified as such by the verification platforms, is contemplated. This is the main classification of the present study, which identifies the stigmatising terms published and clearly

140

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

intends to provoke rejection of the migration phenomenon, whether directed at migration in general or pointing to a specific group. Unverified or distorted data and even premeditated errors have also been identified, as have impersonations of journalistic media accounts, always with the aim of causing harm. Among the stereotypes detected, some indicators are repeated, such as the association of immigration with the economic burden they cause to the host country, taking away social benefits from nationals to the point of blaming them for the host country’s lack of employment or subsidies. Another repetitive aspect is the threat to security, blaming the migrant collective for insecurity in all areas, including terrorism. Similarly, the threat to the identity of European culture is recurring and is related to strengthening European borders to prevent “invasion.” One of the categories detected in the language is the hostility or pejorative expression to speaking of immigrants as “irregulars” or “refugees” when used as an insult. It is also detected that the verification date for a hoax is carried out between 24 and 48 hours after disseminating the malicious content. However, it is almost immediate regarding cases linked to current affairs such as Easter processions, physical or alleged attacks, or politician statements. The origin of the population to which the fake news alludes opts for the generic term of the country (for example, Moroccan), as opposed to derogatory terms such as “Moorish.” It is striking that information media are also supplanted in their digital editions: newspapers such as Bild, El Mundo, El País, among others, to which are added false profiles on Twitter. Attention to the axes of xenophobic discourse is important since it allows its classification into three large areas related to security: the alleged economic privileges of migrants and the alleged attack on social, cultural, or religious traditions associated with Spanish citizenship. The next axis of the most common xenophobic discourse is the one that alludes to migrants as recipients of subsidies, the minimum vital income, or other public aid. It could be affirmed that hate speech against the migrant population would aim to activate protection and defence mechanisms against three factors in which the fear of losing them would be present: security, social and economic well-being, and national identity. It can be concluded that the misinformation associated with migrants develops a xenophobic discourse related to the basic and survival needs of citizens and that it may be capable of provoking alert reactions and concern in those willing to believe it. Regarding the first point of the study, detecting hoaxes through the fact-checking sites of the three selected Mediterranean countries (Spain, Greece, and Italy), we found that the three platforms analysed in this work verify a total of 150 pieces of information. Valid on immigrants, classified as hoaxes, each one explains why in detail. It is considered that 50 works from each country can make up a number of representative pieces, especially if we look at the works that have been taken as references: Vorobyeva et al. (2020) analyse 59 hoaxes; Cheddadi (2020) selects 62 tweets; Rodríguez-Fernández (2019) increases this value to 168 “information”, although Notario and Cárdenas (2020) achieve important contributions with a sample of 20 false or fraudulent content related to migrants (Table 1).

141

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Table 1. Validated sample of the hoaxes analysed Country

Fact-checking platform

Number of hoaxes checked

Spain

Newtral.es/Maldita.es

50

Greece

Ellinika Hoaxes

50

Italy

Facta.news/Lavoce.info

50

To elaborate on the scale of levels of hate in the expressions used in the hoaxes of the different countries, a difference is made between expressions that are lies, half-truths, hyperbole, exaggerations, vulgarisms, profanity, euphemisms, associations of ideas, aggressive or pejorative terms, decontextualisation, falsification of data, decontextualised images, exaggerations, etc (Newtral.es, 2022). All of them create a public opinion contrary to the immigrant, which has been highlighted by fact-checking. Linguistic patterns and metadata are detected in all cases, leading the audience to assume a wrong message. To exacerbate the problem, migratory movements have become a topic of political debate in the three countries analysed in this study, causing division, criticism, and controversy among citizens precisely because they receive messages that are classified as fake news, intended to provoke rejection of immigrants, to the point of seeing immigration as “a problem,” as reflected in the surveys that have been carried out, for example, in Spain in CIS (2022) and corroborated by the Eurobarometer (2022), where Europeans believe that there are 9.5% more migrants and refugees in Europe than are actually. Some of the recurring topics refer to immigrants receiving much more from the host country than they contribute. This is how a negative perception of the collective is built. However, considering the news verification platforms demonstrate the opposite and dismantle the process, going so far as to classify this information as a hoax. With these discourses, one moves from the problem to the threat and xenophobia, exacerbated in times of economic crisis. A special case is made up of the image (photography and video), which is present in 90% of the sample and is used in almost all cases as support for the text, or as a tool for the supposed verification or testimony of the textual content. Fake news is also found, based exclusively on audio content and distributed through social media and WhatsApp. Video is the format used in 50% of cases, with text support. Those videos present poor-quality, unedited images, such as those spontaneously recorded with a smartphone or those collected by security cameras. This realism could give the appearance of truth in the scene, but videos can also be manipulated. Some more significant examples of each of the selected fact-checking projects will follow.

Spain: Newtral.es/Maldita.es We chose two verification portals for Spain since they coincide and complement each other, which helps us guarantee the cataloguing of hoaxes more accurately. Newtral is a semi-startup founded by journalist Ana Pastor in 2018, but it was the first Spanish verification platform to join IFCN in June 2017, when the journalist led the television program El Objetivo (LaSexta) before creating the current company. It verifies, to a greater extent, political speeches and hoaxes of journalistic interest, such as those related to immigration in Spain. This portal boasts of having artificial intelligence applied to language (Data Science Linguistic) to detect lies automatically.

142

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Newtral establishes a collaborative relationship with its audience and invites citizens to alert them of suspicious content, hoaxes, fake news, links, or images they wish to ask for confirmation about its veracity and send through their WhatsApp channel (Figure 2). Figure 2. Significant hoax recovered from Neutral.es

For its part, the verification portal Maldita.es was opened in 2016 and explicitly reflected its methodology to check hoaxes, arguing that it is based on two variables: the importance of the message’s virality and the danger it entails when appearing in special situations delicate. Regarding dissemination, the portal differentiates between the possible hoax that reaches the portal only one way to be verified and the one that is received on a recurring basis and through different channels (WhatsApp, Twitter, or Facebook) or the one that has been issued in an account with hundreds of followers (politicians, public figures, impact social media, media, among others). What is more, this portal does not publish denials of a hoax that has little impact, precisely to not amplify it, except in flagrant cases (Maldita.es, 2019, 2022). Regarding dangerousness, Maldita.es considers crisis situations, especially delicate ones, such as those related to migration, which can affect social coexistence. This fact-checking portal performs a double-check and contacts the sources to verify the information. If necessary, it also uses video or audio identification technology. Regarding the hoaxes published on Twitter, Maldita.es verifies if the account is a troll that tries to imitate or impersonate the original account of an institution or person (Figure 3).

143

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Figure 3. Visualisation of the most significant hoaxes recovered from Maldita.es

One of the bloodiest examples refers to the viralization of a map of Spain indicating the autonomous communities with a figure in euros, corresponding to the monthly payment offered to unaccompanied foreign minors. After the verification of Maldita.es, which contrasted with every one of the Spanish autonomies, the portal explained that the financial aid that appears on the map is a mixture of social benefits for people with limited resources, along with aid for the elderly over 18 years of age, after being supervised by the autonomous communities (both foreigners and nationals). Another of the viral examples, which was also replicated in Portugal, France, Italy, the United Kingdom, Germany, Denmark, the Netherlands, Croatia, Sweden, and Greece, refers to the hoax published on December 16, 2022, assuring that a Moroccan had destroyed a Christmas nativity scene in Calahorra (La Rioja). This hoax was published on Twitter by the president of the Alternative France party, Gilbert Collard, who is also a member of the European Parliament, although hours after publishing it, he deleted

144

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

it from his account without explaining that it was a hoax. Attention is drawn to these discursive strategies since some political parties take advantage of this event to obtain electoral returns. They do not fabricate the hoax but spread any of the networks to generate hatred against immigrants for the representations built by the social network without assessing the consequences, such as causing outbreaks of violence. Precisely, the followers of a specific account feel identified with the messages and with the community that shares them through what is known as “group emotions,” and everyone thinks and shares the group’s orientation. Even on Twitter, the excitement is growing with the comments and retweets of the conversation. In this case, given the magnitude of the viralization of the event, the Calahorra city council itself officially reported that a racist hoax was being spread. In the previous examples, disinformation alluding to immigration is observed through intentionally disseminating information with errors in all or part of the message. They are characterised by a lack of rigour and a clear intention to distort the facts to influence public opinion. In this way, the audience is destabilised (Romero, 2019). Academic research such as Peirano’s (2019) concludes that fake news is designed to outrage. Lastly, it is significant that this information does not comply with what is stipulated in the journalistic codes of ethics referring to news on immigration, where it is made explicit, among others, that it is not necessary to include the nationality of the people so as not to stigmatise them or indicate the sources to compare.

Greece: Ellinika Hoaxes (EH) This Greek fact-checking agency is an independent initiative in the field of news-checking founded in 2013 to verify news in the media and social media. It does not belong to any medium. Legally, it is constituted as a non-profit organisation, with its registered office in Thessaloniki. Ellinika Hoaxes is also a member of the IFCN global data verification network and aligns with its principles of political independence, source transparency, funding, methodology, and commitment to open and honest corrections, as stated on its website. EH is also a member of the European Observatory against Disinformation and the European Observatory for Digital Media and participated in the European program FactCheckEU. info, with entities dedicated to verifying false information from all over Europe. The founding team of EH is made up of verification experts, and they address disinformative content related to politics, pseudoscience, immigration, and outbreaks of xenophobia, among other topics. The two main factors that determine whether the EH team prioritises an investigation are the spread or reach of a story and the severity of the effects of misinformation, such as an article with potentially inaccurate medical claims. Once an issue is prioritised, EH identifies potentially suspicious material in an article, photo, or video. Next, the content and signs of possible falsehood are carefully reviewed. Sometimes contact is made with the original source. In the case of audio-visual material, it is checked whether that image has been used other times through the Google Image service, and if so, the page containing the original video or image is archived. Likewise, geolocation tools and other tools are used to detect if the images have been manipulated. Scientific sources and studies are also examined in the case of science-related hoaxes, and other members of international verification networks are used. Among the hoaxes denied by EH between January and October 2022 are several related to immigration, which could be considered stigmatising hoaxes, associating the immigrant collective with social phenomena such as crime and violence, marginalisation, or the illegitimate use of social resources, to 145

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

the detriment of the aid and benefits owed to the population of national origin. These are, then, messages that certainly propagate hate speech around immigrants in a country like Greece, which currently hosts more than a million immigrants from non-EU countries and which is one of the leading countries of entry to Europe for immigrants by sea (Calderón et al., 2022). In this context and during the first ten months of 2022, the following hoaxes were denied in EH: 1. video of an alleged immigrant residing in Greece who appears to be shooting: This video is circulated on social media, and it shows a man listening to music, singing, and, at one point, pulling out a gun and shooting. The video was released along with the statement that it was an event from the previous day recorded in a house in Greece and carried out by an illegal immigrant from Pakistan. The image was presented as evidence that networks of illegal immigrants with access to weapons and out of control had developed in the country. Along with this statement, it was added that among those immigrant networks that enter Greece every day, there would be Turkish espionage agents and terrorists from ISIS or Al Qaeda. From EH, the video was analysed through a reverse image search, and it was verified that this video had been playing on YouTube since 2019 and came from Pakistan or India. The verifier thus concluded that it was not recent and had not occurred in Greece (Figure 4). Figure 4. Ellinika Hoaxes verified that the video was from 2019 and from Pakistan or India

146

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

2. Video of a group of immigrants arriving in Greece to supplant Greek workers: the images show a group of foreigners with blue documents and thanking Greece for regularising them. These images have been shared on digital media and in social media posts since July 2022. Even some articles on Facebook present immigrants as “replacements” for Greek workers and as “illegal immigrants” regularised by the Greek prime minister, Kyriakos Mitsotakis. These messages incorporate a threatening element in the face of the arrival of these immigrants, spreading fear of the risk of Greek citizens losing their jobs. However, EH’s work made refuting some of this information possible. According to the journalists from the verification company, the images in the video in question show people with travel documents, which are granted based on decisions of the European Parliament and international treaties and not by a decision of the Greek prime minister. In addition, this documentation is granted to beneficiaries of refugee status or applicants for international protection and does not serve to regularise their presence in the country but allows them to leave the borders of Greek territory (Figure 5). Figure 5. EH verified that the documents only allow you to leave Greece

3. Foreigners with high-end mobile phones and privileged seats at a New Year’s concert by a music star in Syntagma Square, organised by the Athens City Hall: this hoax aims to question the role of victims associated with immigrants, stressing that the aid and support they receive grant them certain privileges compared to the rest of the national population.

147

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

This hoax was published on various websites and social media, accompanying photographs of the concert with these words collected on the EH website: Pakistanis in the Constitution: Suffering was painted on the iPhone 13s in their hands. From afar, they smiled like unhappy people from war zones. People who lost their homeland, unhappy, in ragged clothes, hungry and thirsty, danced to the sound of chiftetelia and belly dance that sounded from the speakers of the poor iPhone 13 pro and iPhone 13 pro max. Damn poverty. Their only consolation is that they saw Sákis Rouvás up close. You don’t even say it. It was a balm for their aching souls. Even less that they saw him without paying 265,000 euros with VAT. The use of irony and some incorrect information gives rise to the indignation of the Greek readers, who cannot understand that these people enjoy privileges they are not supposed to enjoy. In its denial, the EH portal analyses the reported facts and concludes that this concert did not take place in Syntagma Square, but on Lycabettus Hill and that, moreover, it was carried out without the presence of the public for reasons of public health related to the COVID-19 pandemic (Figure 6). Figure 6. EH denies that immigrants enjoy privileges

4. Behaviour of refugees that is harmful to the country’s citizens: among the hoaxes detected by EH concerning immigration, some associate refugees of various origins with strange behaviours and

148

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

events. This is the case of a video comprising three small video clips, which ensures that Ukrainian refugees set their houses on fire while trying to burn a Russian flag. This hoax was disseminated through the fraudulent use of the image of Bild, a German media outlet whose management denied having disseminated this information. In some of the published publications that contain the video, statements like this were made: Refugees from Ukraine in Europe continue to behave stupidly and destroy the property of the locals who hosted them. The first denials about this message began to appear on verification websites in May 2022. The verifiers discovered that one of the clips that make up this video had been recorded by a student at the University of Prague in the Czech Republic and that he was unaware that it had later been used to construct a negative message about refugees. The young man even removed the video from his accounts on social media. The other clips were from older videos, not related to the refugee community but to other fires (Figure 7). Figure 7. EH verified that the Bild newspaper had denied having published this information

5. Refugees full of Nazi tattoos: another of the stigmatising hoaxes detected and denied by EH was the one that linked refugees of Ukrainian origin to a photograph of people full of Nazi-oriented

149

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

and aesthetic tattoos. The rejection that such a link provokes in the minds of those who receive this message is clear. The images appeared on social media in July 2022, and, in them, you can see two men with tattoos of Nazi symbols such as the swastika and the eagle emblem. The users who shared the images pointed out that they were refugees from Ukraine. However, the verification work showed that the images were taken on a beach in Croatia and that their protagonists were Hungarians, not Ukrainians. Specifically, they are members of a neo-Nazi network called Blood and Honour, who participate in an annual music festival in Croatia in memory of a member who lost his life in a car accident. These people attend this event annually and were photographed walking along the promenade of a Croatian coastal city (Figure 8). Figure 8. EH verified that they are not Ukrainians, and the image was captured on a beach in Croatia

150

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Italy: Facta News The case of Italy is of particular interest since, in the months covered by this study, it was in the preelection and electoral period before the general elections held on September 25. The discourse on immigration is one of the most relevant issues of the electoral campaign and is part of the agenda of all the parties and the messages that their circles of sympathisers, affiliates, and voters launch on social media. In this context, between January and October 2022, numerous stigmatising hoaxes around immigrants were detected by verifiers. In this paper, we analyse some of these specific hoaxes, denied by Facta and Pagella Politica, and some of the key stigmatising messages launched against these immigrants, and that the verifier Lavoce.info calls into question. Facta.news is a fact-checking company associated with the International Data Verification Network that has users’ reports and complaints as its fundamental source (Facta News, 2023). It is part of European projects in the field of fact-checking and disinformation, such as the European Observatory for Digital Media (EDMO), the Social Observatory for Disinformation and Analysis of Social media (SOMA), and the Italian Observatory for Digital Media (IDMO). Pagella Politica, for its part, is an editorial project born in 2012 that deals with the verification and analysis of political news and that, since 2017, has been part of the International Fact-Checking Network (IFCN), the leading international network in project data verification. As for Lavoce.info is a fact-checking project owned by the La Voce Association, founded in 2002 and has 43 professionals on the payroll dedicated to this work, in addition to a large group of collaborators. One of the hoaxes disseminated and denied in the months prior to the Italian elections asserted that 83,000 immigrants would have disembarked on the island of Lampedusa “with the basic citizenship income already paid into their account.” This message, which even estimated the damage to the public treasury caused by the landing at 3.4 million euros, was widely shared on social media, citing an alleged article in the Libero Quotidiano newspaper, which was indeed false. The verification firm Facta was in charge of working on this message, discovering that the hoax contained a photomontage of an image previously used to report on the landing of immigrants in Messina and that it had never been published in the Libero newspaper. In addition, according to Facta, the figure of 83,000 immigrants is flatly false since the Italian Ministry of the Interior data estimates that 50,000 immigrants arrived in Italian territory between January and August 2022. On the other hand, the Facta portal recalls that to access basic income, citizens from countries outside the European Union need to prove a residence permit, which implies a stable presence on Italian territory (Figure 9).

151

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Figure 9. Facta denies the photomontage spread on networks

Among the disinformative content denied in the months prior to the elections also included the statements of the Fratelli di Italia candidate, Giorgia Meloni, through a tweet released on August 12 in which it was stated that the Superior Court of Justice of the European Union forced Italy to “guarantee food, accommodation, and subsidies to immigrants responsible for crimes and violent acts.” The verifier Pagella Politica produced an article clarifying that the message sent by Meloni was wrong since what the European Court had established was the state’s obligation to come to the aid of destitute asylum seekers, regardless of whether they had committed any violent acts. The Italian State can impose sanctions on individuals who have committed crimes, but it cannot suddenly withdraw all measures related to reception, leaving this person completely helpless. This conduct, recalls the verifier, would be incompatible with human dignity, so the CJEU prevents it from occurring (Figure 10).

152

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Figure 10. Denial of the Italian verifier Pagella Politica

Along with denying hoaxes such as the above, the activity of the large fact-checking agencies in Italy, during the first ten months of 2022, focused on countering prejudice against the group of immigrants who appeared recurrently on social media and in the media and that were fanned by the crusading propaganda of the months prior to the electoral contest. An example is the association between crime, delinquency, and immigration, which was mentioned in the campaign by the candidate of the Fratelli d’Italia and later prime minister, Giorga Meloni. In the middle of the electoral campaign, the politician advocated establishing a “naval blockade” to contain the arrival of immigrants, which became a national security problem. The leader of the Italian League, Matteo Salvini, behaved in the same way in the campaign. Lavoce.info responded to these ideas using Eurostat data on asylum requests in Italy, which went from 17,110 in 2012 to 45,200 in 2021. Despite this increase, crime figures not only did not increase but instead decreased by the same period. In the same way, the verification portal analysed the data of foreigners interned in prisons in Italy, which account for a third of the total, emphasising that most of these foreign convicts have been for minor crimes. According to this portal, the association of immigrants and crime made by various political leaders in the campaign was unfounded (Figure 11 and Figuren12).

153

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Figure 11. Lavoce.info detects anti-immigration messages in the electoral campaign

Figure 12. Lavoce.info collects Salvini’s tweet that associates complaints with immigrants

154

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

In the same way, among the statements made viral by supporters of one and other political sectors, including leaders of formations that were running in the elections, is the association of aid to immigrants with unjustified and abusive spending of public coffers, to the detriment of other investments and state aid to the population. As a result, other analyses conducted by the verification company Lavoce.info concentrated on demonstrating that this idea was unfounded. To this end, Lavoce.info uses data from the XII Annual Report on the Economics of Immigration from the Leone Monressa Foundation, according to which the majority of the immigrant population is of working age and contributes to public coffers, and only 1.8 per cent of those over 65 years of age are immigrants, with this age group representing the highest percentage of aid and public spending. In the same way, the presence of immigrants is analysed among the schoolchildren who study in the Italian educational system and that barely amounts to 10% of the total. It is indicated that the funds destined to aid the reception of immigrants from the Ministry of the Interior have been significantly reduced since 2017 (Figure 13). Figure 13. Lavoce.info publishes positive information on immigrants

155

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Lavoce.info also responds to the fear, fuelled by some political formations, that the immigrant vote will alter the results of the elections in Italy. In this sense, he analyses the trends in the vote of secondgeneration immigrants settled in Italy, although he separates the orientation of the vote from the fact that the supported parties have defended specific measures in favour of immigration. Of course, the verifier points out that the migratory phenomenon will have a long-term electoral effect in one sense, while in the short term, the anti-immigration messages favour, on the contrary, the vote for right-wing and extreme-right parties. In this way, it can be concluded that the verifiers at Pagella Politica, Facta and Lavoce.info have been forced to respond to hoaxes and misinformation that stigmatised the immigrant group and depicted it in public opinion as a social group that would pose a threat to the welfare, health, and safety of the national citizens of Italy. These messages have been especially disseminated on social media and digital media in the months before the general elections, held in Italy on September 25. To better understand the critical discourse analysis of the hoaxes analysed, a table that responds to the stigmatising concepts detected by the fact-checking platforms is shown. We point out that Maldita Migración was born in 2019 after noticing that a third of the hoaxes circulated through the networks were related to refugees or immigrants. In this way, followers collaborate on social media to share the denials and simultaneously be part of the community that warns of hoaxes that are spread on networks (Table 2). Table 2. Mapping of stigmatising hoaxes Facta. news/ Pagella Politica/ Lavoce. info

Newtral. es/Maldita Migración

Elinica Hoaxes

Immigrants threaten security

X

X

X

02. Abusive use of social assistance programs and threats to health and well-being

X

X

X

03. Refugees cause more economic crisis

X

X

X

04. Treatment of refugee or immigrant arrivals as “invasion”

X

X

X

05. Association between immigrants-crime-delinquency

X

X

X

06. MENAS are a problem

X

X

07. Illegal stay in the country

X

X

08. Immigration problems at times of electoral campaign

X

X X

09. The vote of second-generation immigrants can influence the results

X

10. The nationality of the immigrant appears

X

X

11. Explicit association between immigrants-crime-delinquency

X

X

X

13. Decontextualised content

X

X

X

14. Misleading content

X

X

X

15. Pejorative and contemptuous terms

X

16. Expressions of rejection

X

12. Immigration Skits

X

17. Audio-visual (photo, video, audio) without source 18. Impersonation of a media outlet or social media account

156

X X

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

DISCUSSION AND CONCLUSION This work has relied on the veracity and credibility of the selected fact-checking platforms, contrasting misleading and false content. The conclusions are presented following the objectives set out in this investigation. Thus, concerning the first general objective, the conclusions of this study confirm the presence of topics that appear in the fake news related to migrants, seasoned with hate speech towards foreigners. The intention to damage the collective imagination of migrants takes precedence over reality, for example, in terms of false information related to public aid or subsidies. The results of this research offer an accurate quantitative and qualitative picture of the case studies (Spain, Greece, and Italy). The current disinformation crisis on social media, such as Twitter, or direct messaging systems, such as WhatsApp, is also evident. Specifically, Twitter became the third source of messages submitted to verify the migratory phenomenon. On the other hand, it is concluded that there is a significant difficulty in detecting fake news because, based on true information, they produce decontextualised and confusing content for the citizen. Regarding the categorisation of the types of disinformation applied in the cases analysed, it is verified that the contents are manufactured or manipulated, which leads one to think that there is a will to deceive. The disinformation content of permanent news stands out, which, as has been pointed out, is maintained and allows its systematic reuse in matters related to the supposed privileges in terms of public aid and alleged violent actions by migrants. The perception of economic threat is the one that has the most force in determining the rejection of immigrants. It is no coincidence that it is linked to erroneous approaches, a lack of argumentation, and the indeterminacy of the data. Unaccompanied foreign minors are also the centre of false information, which focuses on emotional issues, very easy to share on social media. There is a metaphorization in this information, with constant recourse to figures and comparisons, to count the “others” in comparison with the native society. This is more worrying if it is seen that there are not enough resources for everyone and hostility is generated towards those who arrive, wanting them to leave as soon as possible. Coupled with the fact that, in this context of social media, the narratives are easily detected as false and malicious because they lack a reliable source. The supposed sources from which the hoaxes originate have a similar narrative scheme: they take some news out of context, alleged witnesses to the events appear, and events or situations from different contexts are mixed and disseminated through WhatsApp, Twitter, or other networks, where users copy or forward. It is a way of disguising hoaxes, where facts or situations are mixed to give a decontextualised interpretation. When survey data or opinion barometers on immigration are released, citizens react with unease, concern and concern at the perception that border policies and controls do not stop clandestine arrivals. If they also coincide with electoral campaigns, this issue is an ideal scenario to blame immigrants. The conclusions are more significant in the case of the videos broadcast on Twitter. This investigation has encountered limitations, specifically when accessing the original hoax, as the social media have removed it. In our case, we have trusted the sample taken by the fact-checking platform.

157

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

However, we value the work of the verification services, and it is concluded that they are valid for research projects, such as the one presented to study the hoaxes about immigrants in Spain, Greece and Italy. These platforms provide enough information for citizens to be aware of disinformation as an unfair practice and understand the hoaxes about the immigrant community.

REFERENCES Acciai, E. & del Rizzo, S. (2023). L’Italia è l’unica a garantire la protezione umanitaria? La Voce. https:// lavoce.info/archives/category/rubriche/fact-checking/ Aguilera-Carnerero, C., & Azeez, A. H. (2016). ‘Islamonausea, Not Islamophobia’: The Many Faces of Cyber Hate Speech. Journal of Arab & Muslim Media Research., 9(1), 21–40. doi:10.1386/jammr.9.1.21_1 Alonso-García, S., Gómez, G., Sanz, M., Moreno, A. J., & Rodríguez, C. (2020). The Impact of Term Fake News on the Scientific Community. Scientific Performance and Mapping in Web of Science. Social Sciences (Basel, Switzerland), 9(5), 73. doi:10.3390ocsci9050073 Naciones Unidas Alto Comisionado de Derechos Humanos de Naciones Unidas. [Naciones Unidas] Alto Comisionado de Derechos Humanos de Naciones Unidas (2013). Plan de Acción de Rabat. OHCHR. https://www.ohchr.org/Documents/Issues/ Baker, P., & Gabrielatos, C., KhosraviNik, M., Krzyanowski, M., McEnery, T., & Wodak, R. (2011). ¿Una sinergia metodológica útil? Combinar análisis crítico del discurso y lingüística de corpus para examinar los discursos de los refugiados y solicitantes de asilo en la prensa británica. Discurso & Sociedad, 5(2), 376–416. Bakir, V., & McStay, A. (2018). Fake news and the economy of emotions. Digital Journalism (Abingdon, England), 6(2), 154–175. doi:10.1080/21670811.2017.1345645 Bauman, Z. (2016). Extraños llamando a la puerta. Paidós. Baviera, T., Sampietro, A., & García-Ull, F. J. (2019). Political conversations on Twitter in a disruptive scenario: The role of “party evangelists” during the 2015 Spanish general elections. Communication Review, 22(2), 117–138. doi:10.1080/10714421.2019.1599642 Berger, P. L., & Luckman, T. (1966). The Social Construction of Reality. Anchor Books. Calderón, B. G., Córdoba-Cabús, A., & López-Martín, Á. (2022). Las fake news y su percepción por parte de los jóvenes españoles: El influjo de los factores sociodemográficos. Doxa Comunicación. Revista Interdisciplinar de Estudios de Comunicación y Ciencias Sociales, 36, 19–42. doi:10.31921/ doxacom.n36a1741 Camargo, L. (2021). El nuevo orden discursivo de la extrema derecha española: De la deshumanización a los bulos en un corpus de tuits de Vox sobre la inmigración. Cultura. Lenguaje y Representación, 26, 63–82. doi:10.6035/clr.5866

158

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Carr, P. R., Cuervo, S. L., & Daros, M. A. (2020). Citizen Engagement in the Contemporary Era of Fake News: Hegemonic Distraction or Control of the Social Media Context? Postdigital Science and Education, 2(1), 39–60. doi:10.100742438-019-00052-z Cheddadi, Z. (2020). Discurso político de Vox sobre los menores extranjeros no acompañados. Inguruak. Revista Vasca de Sociología y Ciencia Política, 69(69), 57–77. doi:10.18543/inguruak-69-2020-art04 Chenzi, V. (2021). Fake news, social media and xenophobia in South Africa. African Identities, 19(4), 502–521. doi:10.1080/14725843.2020.1804321 CIS. (2022). Barómetro de diciembre de 2022. CIS. https://www.cis.es/cis/export/sites/default/Archivos/ Marginales/3380_3399/3388/es3388mar.pdf Eurobarómetro Comisión Europea. (2022). Eurobarómetro Standard 96: Opinión pública en la Unión Europea. EC. https://cutt.ly/3GAkK5L ECRI. (2017). Recomendación de Política General nº 7 sobre la legislación nacional para combatir el racismo y la discriminación racial, aprobada el 13 de diciembre de 2002. ECRI. https://www.coe.int/t/ dghl/monitoring/ecri/activities/GPR/EN/Recommendation_N7/REC7-2003-8- ESP.pdf Ekman, M. (2019). Anti-immigration and racist discourse in social media. European Journal of Communication, 34(6), 606–618. doi:10.1177/0267323119886151 Ellinika Hoaxes. (2023). Ellinika Hoaxes. Ellinika Hoaxes. https://www.ellinikahoaxes.gr/ Elósegui, M. (2017). Las recomendaciones de la ECRI sobre discurso del odio y la adecuación del ordenamiento jurídico español a las mismas (No. ART-2017-103148), Revista General de Derecho Canónico y Derecho Eclesiástico del Estado, 44. Estrada-Cuzcano, A., Alfaro-Mendives, K., & Saavedra-Vásquez, V. (2020). Disinformation y Misinformation, Posverdad y Fake News: Precisiones conceptuales, diferencias, similitudes y yuxtaposiciones. Información, Cultura y Sociedad, (42), 93–106. doi:10.34096/ics.i42.7427 Facta News. (2023). Antibufale. Facta. https://facta.news/antibufale/. Fallis, D. (2015). What is disinformation? Library Trends, 63(3), 401–426. doi:10.1353/lib.2015.0014 Gamir-Ríos, J., Tarullo, R., & Ibáñez-Cuquerella, M. (2021). La desinformación multimodal sobre la otredad en Internet. Difusión de bulos racistas, xenófobos e islamófobos en 2020. Anàlisi, 64(64), 49–64. doi:10.5565/rev/analisi.3398 Hernández-Sampieri, R., & Mendoza-Torres, C. P. (2018). Metodología de la Investigación. Las rutas cuantitativa, cualitativa y mixta. McGraw-Hill. HLEG. (2018). A multi-dimensional approach to disinformation: report of the independent high-level group (HLEG) on fake news and online disinformation. European Commission. https://ec.europa.eu/ digital-singlemarket/en/news/final-report-high-level-expert-group-fake-news-andonline-disinformation IFCN. (2021). The commitments of the code of principles. IFCN. https://bit.ly/3dyo5uw

159

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Ireton, C., & Posetti, J. (2018). Journalism, Fake News & Disinformation. Handbook for Journalism Education and Training. Unesco. Kapantai, E., Christopoulou, A., Berberidis, C., & Peristeras, V. (2021). A systematic literature review on disinformation: Toward a unified taxonomical framework. New Media & Society, 23(5), 1301–1326. doi:10.1177/1461444820959296 López, F., & Rodríguez Rodríguez, J. M. (2020). El Fact Checking en España. Plataformas, prácticas y rasgos distintivos. Estudios sobre el Mensaje Periodístico, 26(3), 1045–1065. doi:10.5209/esmp.65246 Magallón, R. (2019). Verificado México 2018: Desinformación y fact-checking en campaña electoral. Revista de comunicación, 18(1), 234-258. Maldita.es. (2019). Por qué este proyecto. Maldita Migración. https://bit.ly/3UasWp7 Maldita.es. (2022). ¿Qué es Maldita Migración? Maldita Migración. https://maldita.es/migracion/ Masip, P., Aran-Ramspott, S., Ruiz-Caballero, C., Suau, J., Almenar, E., & Puertas-Graell, D. (2020). Consumo informativo y cobertura mediática durante el confinamiento por el Covid-19: Sobreinformación, sesgo ideológico y sensacionalismo. El Profesional de la Información, 29(3), e290312. doi:10.3145/ epi.2020.may.12 Mazzoleni, G., & Bracciale, R. (2018). Socially mediated populism: The communicative strategies of political leaders on Facebook. Palgrave Communications, 4(1), 50. doi:10.105741599-018-0104-x Molina, J. P. (2019). Unfaking News. Cómo combatir la desinformación. Estudios sobre el Mensaje Periodístico, 25(2), 1267–1269. Newtral.es. (2022). Newtral. https://www.newtral.es/tag/inmigracion/ Notario, M. L., & Cárdenas, M. L. (2020). Xenophobia and Fake News. Analysis of the confused news flow. Maldita Migración Project. IROCAMM: International Review of Communication and Marketing Mix, 3(1), 75–91. doi:10.12795/IROCAMM.2020.v01.i03.06 Peirano, M. (2019). El enemigo conoce el sistema: Manipulación de ideas, personas e influencias después de la economía de la atención. Debate. Pérez-Escolar, M., & Herrero-Diz, P. (2023). Capítulo 5. El desmentido como instrumento para mejorar la calidad del debate público en el escenario digital. Espejo de Monografías de Comunicación Social, (10), 89–108. doi:10.52495/c5.emcs.10.p96 Rey, F. (2015). Discurso de odio y racismo líquido. In Revenga Sánchez, M. (dir.), Libertad de expresión y discursos del odio (pp. 51–88). Universidad de Alcalá. Rodríguez-Fernández, L. (2019). Desinformación y comunicación organizacional: Estudio sobre el impacto de las fake news. Revista Latina de Comunicación Social, 74, 1717–1728. doi:10.4185/RLCS-2019-1406 Romero, J. A. O. (2019). Desinformación: concepto y perspectivas. Real Instituto Elcano (RIElcano), 41.

160

 Mapping Stigmatizing Hoaxes to Immigrants on Twitter, Digital Media

Said-Hung, E. M., Merino-Arribas, M. A., & Martínez-Torres, J. (2021). Evolución del debate académico en la Web of Science y Scopus sobre unfaking news (2014-2019). Estudios sobre el Mensaje Periodístico, 27(3), 961–971. doi:10.5209/esmp.71031 Salaverría, R., Buslón, N., López-Pan, F., León, B., López-Goñi, I., & Erviti, M. C. (2020). Desinformación en tiempos de pandemia: Tipología de los bulos sobre la Covid-19. El Profesional de la Información, 29(3). doi:10.3145/epi.2020.may.15 Sayce, D. (2020). The number of tweets per day in 2020. https://bit.ly/3CZbUB2 Siapera, E., Boudourides, M., Lenis, S., & Suiter, J. (2018). Refugees and network publics on Twitter: Networked framing, affect, and capture. Social Media + Society, 4(1). doi:10.1177/2056305118764437 UNESCO. (2015): Countering online Hate Speech. UNESCO Digital Lybrary. https://unesdoc.unesco. org/ark:/48223/pf0000233231 Vampa, D. (2020). Competing forms of populism and territorial politics: The cases of Vox and Podemos in Spain. Journal of Contemporary European Studies, 28(3), 304–321. doi:10.1080/14782804.2020.1 727866 Van Dijk, T. A. (2021). Discurso y poder. Gedisa. Vorobyeva, T., Mouratidis, K., Diamantopoulos, F. N., Giannopoulos, P., Tavlaridou, K., Timamopoulos, C., Peristeras, V., Magnisalis, I., & Shah, S. I. H. (2020). A Fake News Classification Framework: Application On Immigration Cases. Communication Today, 11(2), 118-131. https://tinyurl.com/y6zasn49 Waisbord, S. (2018). Truth is what happens to news: On journalism, fake news, and post-truth. Journalism Studies, 19(13), 1866–1878. doi:10.1080/1461670X.2018.1492881 Wardle, C. (2017). Fake news. It’s complicated. First Draft Foot Notes. Medium. https://medium.com/1stdraft/fake-news-its-complicatedd0f773766c79 Wardle, C., & Derakhshan, H. (2018). Thinking about ‘information disorder’: formats of misinformation, disinformation, and mal-information. In C. Ireton & J. Posetti (Eds.), Journalism, ‘fake news’ & disinformation (pp. 43–54). Unesco.

161

162

Chapter 9

The Southernification of the Pandemic in Italy: Images of the South, Fears of Contamination, and the First Wave of COVID-19 in Italy Marcello Messina https://orcid.org/0000-0002-8822-3342 Southern Federal University, Russia

ABSTRACT Starting from February 2020, Italy was the first among the European countries, to experience dramatic rises in daily COVID-19 deaths and contagions. An important aspect that distinguished the first COVID-19 wave (Feb-Jun 2020) from the following waves of infection in Italy was the sheer imbalance, in terms of deaths and contagion, between Northern and Southern regions of the country. Despite the fact that the South was far less hit by the disease, a series of narratives that associated the spread of the epidemic with some sort of Southern infector started to appear, conveyed by social media posts, news pieces, talk shows, and even football banners. In this chapter, there is an attempt to identify and critically analyse the discourses that inscribe a characteristic “Southernification” of the pandemic in Italy, that is a partial and symbolic attempt to (1) discursively transfer the infection to the South; and/or (2) hand over the responsibilities that are behind the particularly violent first wave of infections in the country to Southern communities, polities, and cultural practices.

INTRODUCTION After the first surge of Covid-19 in China and East Asia, at the end of February 2020, cases and deaths related to the infection of the SARS-CoV-2 virus started to occur in Italy, which was the first, among European countries, to experience dramatic rises in daily deaths and contagions. Characterised by a very severe lockdown, Italy’s first wave of the Covid-19 pandemic lasted until May/June 2020. According DOI: 10.4018/978-1-6684-8427-2.ch009

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 The Southernification of the Pandemic in Italy

to Balasco et al. (2021, p. 1), “a total of 240,331 cases (confirmed infections) and 34,892 deaths from pneumonia were registered as of Jun 28, 2020, identifiable as the end of the first wave of the Italian outbreak”. An important aspect that distinguished this first wave from the successive waves of infection in Italy was the sheer imbalance, in terms of deaths and contagion, between the Northern and Southern regions of the country. Southern regions experienced a minimal number of infections and deaths, while the North of Italy emerged as one of the global epicentres of the pandemic (Sebastiani et al., 2020). Incidentally, according to these authors, it was most probably thanks to “government measures” that “the Covid-19 epidemic in central and southern regions [did not] rise to the high levels that were already occurring in the North” (202, p. 343). Other explanations for the lower death toll in the South of Italy include the health benefits of higher forestation in the region (Roviello & Roviello, 2021). On the other hand, the higher atmospheric pollution in the North has been listed as “an additional co-factor of the high level of lethality recorded in that area” (Conticini et al., 2020, p. 1). Enlisted above are plausible explanations offered by some of the scientific literature to make sense of the glaring imbalance between the North and the South of Italy in terms of deaths and contagions during the country’s first wave of the COVID-19 pandemic. Crucially, in the national debate, these explanations were counterpointed by all sorts of ludicrous discussions and analyses of the situation, connected with the perception of a sort of underlying injustice behind the fact that the South had been somewhat “spared” from the sufferings undergone by the North. As will become clear as the present chapter unfolds, what is referred to as “national debate” is largely (but not exclusively) represented by news items, press releases, and informative talk shows, among others. As argued by Van Dijk, “most of our social and political knowledge and beliefs about the world derive from the dozens of news reports we read or see daily. There is probably no other discursive practice, besides everyday conversation, that is engaged in so frequently and by so many people as news in the press and on television” (Van Dijk, 1991, p. 110). Van Dijk (1991) continues to show how reproducing negative images of specific social groups can engender and eventually become part of this shared social and political knowledge. In Italy, before it became clear that the contagions and deaths would remain very limited in the South, a series of narratives that associated the spread of the infection with some sort of Southern scapegoat started to appear, not only in the news but also conveyed by social media posts, talk shows and even football banners. All these different manifestations contribute to reproducing crystallised images of Southern Italy, which derive from a North-South dialogue, constantly characterised by “the imbalance between the two parties involved” (Gribaudi, 1997, p. 83). This paper identifies four different types of discursive stratagems aimed at systematising the various South-blaming enunciations that populated the Italian national debate between February and June 2020. These stratagems are somewhat ordinated chronologically, in a way to unsystematically reflect a sort of sequence of different collective sentiments that followed each other as the events unfolded. First came a paradoxical handing over of the label of infector to Southern subjects; afterwards came a series of bizarre, if racially charged, explanations for the difference in contagions/death rates between North and South; this was, in turn, constantly associated with a sort of “waiting”, in the deliberate hope that the pandemic “finally” came to hit the South, too; finally, the delusion of this perverse yearning led to significant complaints against a Government that was formed by an allegedly high ratio of Southern-born ministers, and that was held responsible for an alleged pro-Southern bias. The author of this chapter argues that what is in operation in these different stratagems can be subsumed under the traditional representation of the South as Italy’s internal anomaly (Dickie, 1994; Gribaudi, 1997), endowed, in nationally produced stereotypes, with suspicious and impure ethno-racial 163

 The Southernification of the Pandemic in Italy

identities (Pugliese, 2008), as well as “backwards” socio-cultural values and behaviours (Banfield, 1958), predisposition to crime and illegal conduct (Lombroso, 1896; 1900), laziness, incompatibility with civic coexistence and democratic participation (cf. Putnam, 1993), among others. This assigned attribute set codifies Southern Italy’s Otherness in the national (and global) debate. This characterisation has been linked by John Dickie to a characteristic Italian national ethnocentrism (Dickie, 1994) and by Gabriella Gribaudi to the representation of the South as the negation of the positive values that allegedly characterise the Italian nation (1997). Other authors, such as Teti (1993) and Pugliese (2017), have associated this Otherness attached to the South to racism and the operativity of whiteness as a paradigm of Northern caucacentric oppression (Pugliese, 2008). Part of this Otherness, at times, has to do with a sense of dirt, repugnancy, and filth, associated precisely with images of disease, insanity and decay (Chiaro, 2017; Messina, 2020b). These images unleash fears of contamination and contagion that often also penetrate the normal politico-economic jargon, where the cyclical preoccupation of an industrious, rich and prosperous North that ends up being infected by the endemic decision-making ineptitude of the South has become a cliché (Messina, 2019). Indeed, it is possible to inscribe these dynamics within Roberto Esposito’s critical dichotomy immunitas vs. communitas (Esposito, 2006), whereby “if communitas is that relation, which in binding its members to an obligation of reciprocal gift-giving, jeopardises individual identity, immunitas is the condition of dispensation from such an obligation and therefore the defence against the expropriating features of communitas” (2006, p. 27). However, the author of the present chapter argues that there is something eminently — and perversely — civic to these Northern attempts to achieve immunitas from the Southern “disease”, something along the lines of what Suvendrini Perera describes as “civic violence” (2014, p. 3) — Northern immunisation from the South is a “powerful social practice” (Perera, 2014, p. 4), that enhances an us vs. them mentality, which in turn retraces “a geopolitical fault line that split the peninsula and its islands along a black/white axis” (Pugliese, 2008, p. 3). Incidentally, through this theoretical prism, it is possible to visualise the paradoxical scapegoating of the South as infector as much as the unconcealed desire for a more expressive outbreak of the infection in the South; the indignation for the higher contagion/death rate in the North as much as the dissent towards an allegedly “South-centric” government. In previous scholarship, the chapter author has argued that a widely observed demonisation of the South needs to be complemented by what he defined as a characteristic “Southernification of evil” (Messina, 2018a, p. 204), which entails assigning to the South a series of national flaws, problems and troubles that might even have little to do with the South itself. In this paper, the “Southernification of the pandemic” relates to both a perverse desire to see the Southern populations suffer more from the pandemic and a literal and paradoxical singling out of the South as the very source of the infection.

“NAPOLETANO CORONAVIRUS” The disease far less hit the South, and a series of manifestations that associated the spread of the infection with some Southern scapegoats started to appear, conveyed by social media posts, news pieces, talk shows and even football banners. An early, paradigmatic example of this scapegoating happened a couple of weeks before the progressive shutdown of the country. On Feb 21, 2020, during a Serie A match played in Brescia between the local team, Brescia Calcio, and the away side Società Sportiva Calcio Napoli, a football banner exposed by Brescia supporters, to offend the Neapolitan supporters, read “Napoletano 164

 The Southernification of the Pandemic in Italy

Coronavirus” [Neapolitan Coronavirus]. The banner associated the incipient Covid-19 surge in Italy with Neapolitans, while it was precisely in the Brescia province and the neighbouring province of Bergamo that the pandemic would find its most deadly epicentre a few days later (Spagnolo, 2020). The “Napoletano Coronavirus” football banner drew upon the more famous “Napoli Colera” [Naples Cholera] adage, an ever-recurring invective in the last fifty years (Barcella, 2018).1 The very same idea of “Napoli Colera” had been exploited by an even earlier football banner, shown in Milan by Internazionale [Inter Milan] supporters on Feb 1 2020, that read “Napoletani figli del colera vi mettiamo in quarantena” [”Neapolitans, children of cholera, we put you into quarantine”] (Riccio, 2020). Both football banners here conflated the racist association of Naples with cholera with the incipient COVID-19 pandemic, and while the Brescia banner, at the end of February, drew upon something that had knowingly already started in Italy,2 the Milan banner, exposed roughly three weeks earlier, entirely referred to a pandemic outburst that had so far hit East Asian countries exclusively.3 However, it is quite safe to say that both banners were exposed when one could hardly predict that Italy would become the first European country to be severely and tragically hit by the COVID-19 pandemic. In this sense, Misuraca et al. (2022) talk about an “optimism bias”, according to which “Italians might have considered it less likely for the virus to hit Italy, compared to other countries, or for themselves to catch the disease, compared to other people” (Misuraca et al., 2020, p. 4). Here Naples needs to be intended as a metonymy of the entire South of Italy, a region that, as mentioned above, is constantly narrated, and characterised as the internal negation of the values that allegedly define Italianness (Gribaudi, 1997; Pizzo, 2009). Naples, as the “aberrant city” (Dines, 2013) par excellence, is often (mis-)characterised in advance of the rest of the South, as an authentic frontline of the entire Southern anomaly, at times sharing this infamous reputation of “avant-garde of the aberrance” with the two Sicilian metropolises (Palermo and Catania), as well as Bari, Reggio Calabria, Taranto, among others. In a world that was becoming more and more defined by ideas of contagion, infection and contamination, the Italian nation, represented by the unfiltered voice of football fans, was starting to “naturally” associate these ideas with a South seen as eternally filthy, diseased, and repugnant. In line with Roberto Esposito’s already mentioned “immunisation paradigm” (Esposito, 2006), if the optimism bias made Italians think it would be unlikely that the virus would hit Italy, it probably also made many of them think that if the virus was to hit Italy, it would have hit the South preferentially. A plethora of suspected infection cases, prevalently from Southern hospitals, started populating the headlines at the end of January: first in Bari then in Reggio Calabria, then a few cases in Naples (Gazzetta del Sud, 2020; Mautone, 2020; Pierini, 2020; Positano News, 2020). Two audio messages started circulating in instant chat services, announcing that a confirmed case of COVID-19 had been hospitalised in Lecce (Puente, 2020; Bufale.net, 2020). None of these suspected cases was confirmed, and the two first confirmed cases in Italy were registered in Rome on Jan 31, 2020 (Giovanetti et al., 2020; Reuters, 2020).

“LA GRANDE FUGA” [“THE GREAT ESCAPE”] The two first confirmed cases in Rome were Chinese citizens travelling to Italy for tourism. For some days, the idea that the infection was still limited to a matter of isolated cases from people who had come from East Asia arguably reinforced the optimism above bias, linked to an “illusory superiority” according to which “Italians in the coronavirus outbreak could have erroneously believed that their ability to contain and fight the disease was higher than the ability of the Chinese or other countries to do so” (Misuraca 165

 The Southernification of the Pandemic in Italy

et al., 2022, p. 4). Articles about the virus psychosis (Mazzarella cit. Spena, 2020), the infodemic “that does more harm than the Coronavirus”4 (Fanpage, 2020), and the “talk-show virus”5 (Minuz, 2020) started to appear. On Feb 3, 2020, during his intervention on TV programme L’aria che tira, prominent virologist Roberto Burioni invited the Italian population to stop panicking: At the moment there is no virus in Italy, and so it would be better for us to worry about lightnings and flood, which instead exist right now. [...] today, if one wanted be infected (by the Coronavirus) in Italy, if they woke up in the morning and said “I want to be infected”, they cannot do that, so there is no point in talking about that” (Burioni, 2020).6 This wave of reassurance quickly gave way to increased worries, starting from the first COVID-19 death on Feb 21. In two weeks, on Mar 7, the total number of deaths had skyrocketed to 233, and, as it is well known, this was just the beginning. On that same evening, a draft of the prime ministerial decree putting in a red zone of alert the entire region of Lombardy, together with other 14 provinces, was shared by national media. According to the day’s headlines, this announcement triggered the panic of hundreds of people who rushed to the city’s main train stations to return to their home regions (De Riccardis and Pisa, 2020; Il Messaggero, 2020b). National newspaper Il Giornale declared that “hundreds of people are already heading towards the south of the country from Milan, which poses the concrete risk of helping the diffusion of the epidemic” [“centinaia di persone si sono già messe in marcia da Milano verso il sud del Paese col rischio concreto di diffondere agevolmente l’epidemia”] (Scognamiglio, 2020). The online newspaper Open compared the alleged exodus to the internal migration that has characterised Italian history: “Hundreds of people towards the stations, on night trains, a symbol of commuting throughout the country, of the emigration from the South to the North, but this time of a counter-emigration from the North to the South. Not worrying about the fact that, together with this mass of people, the Coronavirus and the contagion will also move south.” (Open, 2020). Southern regions and their governors started to panic about the spread of the infection, brought by the “infectors” coming by train from the North. The President of Apulia, Michele Emiliano, immediately ordered a 14-day quarantine for all those who had been coming from Lombardy and the other Northern provinces since 7 March (Regione Puglia, 2020). He also pleaded the people who were travelling towards Apulia to stop and go back to the North: “Vi parlo come se foste i miei figli, i miei fratelli, i miei nipoti: Fermatevi e tornate indietro. Scendete alla prima stazione ferroviaria, non prendete gli aerei per Bari e per Brindisi, tornate indietro con le auto, lasciate l’autobus alla prossima fermata. Non portate nella vostra Puglia l’epidemia lombarda, veneta ed emiliana scappando per prevenire l’entrata in vigore del decreto legge del Governo” (Gazzettino, 2020). Similar appeals were made by the President of Calabria, Jole Santelli, who said that getting back from the North was ‘madness’, and by the Mayor of Naples, Luigi de Magistris (Adnkronos, 2020; Fatto Quotidiano, 2020). The President of Campania, Vincenzo De Luca, predicted that the inflow of people to the region was going to be a major turning point in Campania’s infection curve: Yesterday we reached a turning point in terms of the contagion in our region. In other words, yesterday we recorded hundreds of arrivals of our fellow citizens from Northern Italy in Campania. So, there is a “before” and an “afterwards”. [...] A truly extraordinary prevention work has been done. Now, with the arrival of hundreds of people, who obviously have not all been controlled by the North, we will be in trouble. [...] And therefore we must expect a peak of infections in our region in the month of March.

166

 The Southernification of the Pandemic in Italy

[...] We have entered a second phase because we have had a wave of returns to Campania. It does not depend on us7 (Regione Campania, 2020). It was later reported that the great exodus of 7-8 March had been overstated and that most people wishing to return to their home regions had already done so by the end of February (D’Alessandro, 2020). Anyhow, the few hundred people who had travelled from the North to the South on 7-8 March were identified mainly as students and workers fuori sede, that is, people who live, study and work outside of their home region(De Luca, 2020). It is essential to explain that often these people remain, for the officiality of Italian authorities, residenti [“residents”] of their home regions and are only provisionally domiciliati [“domiciled”, “temporarily living”] in the (Northern) regions where they have moved. Such situations of precarious residence status often last for several years, and, as very well clarified by Lara Palombo, the official residence status is a critical component in a complex “territorial logic that regulate[s] the provision of services for “residents” within the region” (Palombo, 2010, p. 39). Palombo continues to explain that this situation de facto creates “a border that draws a regional sovereign line between southern and northern lives” and that “in this policy setting, ‘national sovereignty’ through Italian citizenship is not enough to access services, which are reserved for the residents of the region” (Palombo, 2010, p. 39). Health services are among the most important provisions reserved to the residents of a particular region, or that are more difficult to access for non-residents.8 Despite the apparent fact that the people escaping from the North were, in their vast majority, Southerners trying to reach their home regions, a situation in which Southern regions could even timidly suggest their willingness to reject these people triggered the general perception that the world was going upside down. As early as the end of February, Il Messaggero had already produced a headline about “the ‘vengeance’ of the South”9 (Messaggero, 2020a). Even the Spanish newspaper El Pais talked about a “paradigm shift”: “For the first time since the unification of the country, since the times of the growth of FIAT in Turin and the massive immigration, the exodus was produced in the opposite direction”10 (Verdú, 2020). Similar collective, not-so-remote memories of Southerners being rejected upon “immigrating” in the North were summoned in order to portray a historical moment when, according to an op-ed piece in the financial newspaper Il Sole 24 Ore, the confidence, certainties, and assurances of the Italians were “subverted” (Prisco, 2020): Once upon a time, there were the two “Italies”, the productive North and the orphaned South of the Kingdom that was, the Southern question (Billia) and “Garibaldi made trouble” (Totò), the “we do not rent to southerners” [signs] and the significant tolerance of the South, which is a resource but sometimes even a problem. Because of too much tolerance – against the illegality next door – you can die. Some events change perceptions, overturn commonplaces, and subvert all certainties, good or bad. We still do not know precisely how the coronavirus emergency will impact our lives in the medium to long term. We do not know if it is an event of historical significance, and we hope not. However, it is undoubtedly killing the confidence of many, up and down the boot [i.e., Italy]: in a handful of hours, Lombards and Venetians, who had embodied, until the day before yesterday, the platonic idea of productivity applied to the Italian territory and at the same time played the role of the “big spenders” pampered by the tourist industry, are no longer welcome (Prisco, 2020).11 The repeated references to the endless histories of Northern intolerance towards Southerners in Unified Italy are significant and will be discussed further in this chapter. For now, it is essential to clarify 167

 The Southernification of the Pandemic in Italy

that the perception of an unprecedented reversion of the North vs South power dialectics coexisted with more “traditional” episodes of overt Northern racism against the Southerners who were escaping. Again in the realm of football banners, on Mar 9, Varese Calcio supporters exposed their racist views on the events outside the local sports arena, roughly translatable as: “Thanks to those who left, without you Lombardy is now clean”12 (VareseNoi, 2020). As mentioned above, the association between Southerners and dirt is deeply rooted in Italian national consciousness and international contexts (Chiaro, 2017; Messina, 2020b; Dines, 2013). The idea of the exodus of Southerners as a way of “finally freeing” the North from a burden also operated in a meme that went viral on social media hours after the perceived exodus: the meme read “Anyway, in just one night Conte13 managed to do what Bossi14 did not manage to do for thirty years! Closing Lombardy’s borders and freeing it from the terroni,15 who even paid for their tickets!!!”.16 Similar manifestations, coming from prominent political authorities, made the headlines in national newspapers, as on the front page of the 9 March Florence edition of La Repubblica, where it was claimed that the then President of Tuscany Enrico Rossi had pleaded non-Tuscan people to leave Tuscany: “Those who are not Tuscan, please go back home”17 (La Repubblica, 2020). Rossi’s speech referred to people from the red alert zone in Lombardy and its vicinities, but the sensationalistic newspaper headline made it appear as anti-Southern rhetoric. Again, Roberto Esposito’s immunisation paradigm can be helpful here, precisely as it allows us to perceive the contiguity between immunitas, sovereignty and the urge of “‘nonbeing’ or [...] ‘not-having’ anything in common” (Esposito, 2006, p. 28). In this sense, “sovereignty” is a key term that refers both to Roberto Esposito’s formulation of immunitas, whereby sovereignty emerges as the “negative relation that exists between unrelated entities” (2006, p. 35), and to Lara Palombo’s argument about the production and reproduction of a “white northern form of Italian sovereignty” (Palombo, 2010, p. 34) within the Italian national space. Because of all the details above, it is essential to understand the positionality of these Southern people in the national space, especially in such a critical historical moment: on the one hand, as shown above, they were perceived as a threat by their home people and regional governors, who thought they should stay where they were; on the other hand, the Northern cities and regions they were escaping might have seen their exodus as a relief. What was in operation in this situation, I argue, was an entrenched resentment towards the Southern immigrant, which is a significant mark of Italian society as a whole (Dickie, 1994; Pugliese, 2008): hardly tolerated in the Northern cities they have moved to, they are never really in the condition of being fully reaccepted in their home regions.18 They experience Abdelmalek Sayad’s “double absence” (Sayad, 1999) without exiting the borders of their own country.

THE VIRUS “CONQUERS” THE SOUTH Once it became clear that the contagions and deaths were very limited in the South, all sorts of ludicrous explanations were attempted by TV news and social media, boiling down to the idea that Northern people’s alleged vocation for operosity and work responsibility, in contraposition to an implicit and selfevident “laziness” of Southern people, had led them to experience a more severe spread of the infection. On Mar 20, 2020, during the TV programme “Stasera Italia” on the channel Rete 4, anchor Barbara Palombelli wondered why the vast majority of the COVID-19 victims died in the North: “Everybody wonders why, but today it turns out that 90% of the deaths are from the regions… basically from the Northern regions. What might have occurred there? Something different? Behaviour. What could have 168

 The Southernification of the Pandemic in Italy

been more? More responsible people, who therefore all go to work? Why is there this outbreak in the North?”19 (Palombelli, 2020). Palombelli’s speculations here need to be associated with a persistent Lombrosian conception (Lombroso, 1896; 1900), whereby alleged Southern laziness and the complementary operosity of Northern people are understood as atavic characteristics, unquestionable a priori that are treated as self-evident truths and do not need further explanation — once again, this resonates with Van Dijk’s previous remarks as to the role of news and the media as loci of production and reproduction of what is held as crystallised, self-evident and undisputed knowledge (Van Dijk, 1991). During an interview on the radio programme “La Zanzara” on the channel Radio24, the then Member of the Chamber of Deputies Gianfranco Librandi tried to argue that Africans, as well as Calabrians and Southerners in general, were “genetically more resistant”20 against COVID-19. Claiming to be a Calabrian himself,21 Librandi continued to explicate that “we [the Southerners] are white Africans”22 (Librandi, 2020). Here Librandi’s undoubtedly good intentions may clash with the racist overtones and historical implications of his words. First of all, he deliberately assigns to Africans a superior body constitution, capable of escaping a disease that, at the moment of the radio interview, had already killed dozens of thousands of people worldwide: this is perhaps subsumable to Fanon’s critique of the fixation on the corporeality of Black subjects and a supposedly superior biological functioning assigned to them (Fanon, 1986, p. 164-166). Secondly, by making sense of Southerners through the prism of Africa and vice versa, Librandi implicitly draws upon a trite colonial imagery that has historically characterised Unified Italy: as Pugliese argues, “the deployment of the loaded signifier “Africa,” as the lens through which the South was rendered intelligible for Northerners, marks how the question of Italy was, from the very moment of unification, already racialised by a geopolitical fault line that split the peninsula and its islands along a black/white axis” (Pugliese, 2008, p. 3). In other words, the conflation between the South of Italy and Africa is instrumental to the legitimation a “civilising” colonial intervention over both these cartographical entities, that in turn identifies, in both contexts, the presence of “racial anomalies” to be domesticated, neutralised and, when needed, exterminated (cf. also Del Boca, 2011). In terms of extermination, it is perhaps impossible to dissociate Librandi’s use of the concept of “white Africans” from the label of “white niggers” that served to encourage and legitimise the deployment of mass violence (lynchings) against Italians (mainly Sicilians and Southerners) in the US (Jacobson, 1999; Staples, 2019; Salvetti, 2017). Paradoxically, the aforementioned preposterous explanations as to why COVID-19 had not spread in the South as much as it was expected coexisted with continuous narratives of Southern infractors of the lockdown, who were portrayed as irresponsibly enjoying physical proximity with other people, thus facilitating the spread of the infection. On Apr 12, local news outlet Palermo Today released a video of Palermitan people meeting up for an Easter barbecue on the roof of a residential building in the low-income Sperone district in Palermo. The images show the clandestine party being interrupted by a police helicopter and other law-enforcing units reaching the building by car. The following sequences show the partying people trying to escape from the roof, the police reaching the top of the building, and finally, a waste collection unit confiscating and disposing of all the chairs and barbecue grills found on the roof (Palermo Today, 2020). The powerful metanarrative of the transgressor and of the exemplary punishment that is going to re-establish the state order reproduces the powerful “scopic regimes” (cf. Feldman, 1997; Perera, 2014) that inscribe the repression, domestication and neutralisation of perceived Southern menaces such as the so-called briganti, punished, incarcerated, killed and exposed by the Italian (Piedmontese) authorities in the first years post-Unification (cf. Messina and Di Somma, 2017; Pugliese, 2017; Rossano, 2011). Myriads of similar stories appeared elsewhere in the local Southern press: “hordes 169

 The Southernification of the Pandemic in Italy

of people” (Francklin, 2020) in Bari, photos of the “crowds on the streets” (VirgilioNotizie, 2020) in Naples’s Quartieri Spagnoli, “illicit activities” (VeraLeaks, 2020) in Taranto’s fish market, among others. On Apr 15, a national TV crew went to Naples in order to film transgressors for the channel Raitre show “Agorà”; unable to find any substantial group of people disobeying the physical distancing rules, the news correspondent declared that “we were unlucky actually, at the moment they are behaving… But until a few minutes ago there was an intense movement of passers-by”23 (Francioni, 2020). There is undoubtedly something perverse in the very idea of lurking outside in order to catch transgressors, only to feel “unlucky” when there are not any around: eventually, the correspondent’s words unleashed the outrage of Neapolitan commentators on social media, as reported by Francione (2020), and by the online news platform NapoliToday (2020), which reported the resentful remarks by the representative of the Naples City Council Flavia Sorrentino: If we asked for explanations, the journalist would probably say that she has been misunderstood. There was no sensational scoop and unfortunately for those who usually observe us with a “Nordic gaze”, the Neapolitans are demonstrating that prejudice is truth’s sworn enemy. We are not talking about finding a reason to be offended at all costs, but to ascertain every time how internalised the cultural preconception towards us is. The half-pronounced sentences, the embarrassment mixed with the displeasure for not having been able to catch the Neapolitans in the act of incivility, all represent modules of a linguistic and gestural register that we know well and fight against. [...] If there is one thing that this emergency has unearthed — with the thousands of deaths and the investigations of Bergamo and Milan for a negligent management of the epidemic — it is that you [Northern Italians] have no moral stance to give us lessons in civilisation. You probably will not sleep at night because of that, but it would be good if you started to get over it (Sorrentino cit. NapoliToday, 2020).24 Sorrentino’s long rant is symptomatic of a growing frustration towards the specific modes of representation that would have had Neapolitans and Southerners as the most odds-on victims of mass infection, all the more as COVID-19 is deemed to be more likely to hit those who transgress specific social norms, a behaviour that is traditionally assigned to Southerners in internal and external narratives (cf. Banfield, 1958; Putnam, 1993). On top of the undeniable pride showcased by Sorrentino for the lower rate of infections and deaths registered in the South, in her speech, it is also possible to identify the clear consciousness of an underlying, diffused frustration with the very idea of the North suffering more than the South. From time to time, this frustration somewhat found an escape valve in the expectation that a more expressive and incisive spread of the infection would finally hit the South in a matter of days or weeks. This was a constant characteristic of media discourse on social networks and traditional news outlets: during the days of the aforementioned big escape towards the South, the grim confidence in an imminent outbreak of contagions in the South was diffused among the media. On Mar 4, the front page headline of Libero showed a sort of perverse excitement by heralding “The Virus Set Out to Conquer the South”, preceded by a kicker proudly announcing that “The Infections Creates Italian Unification”, and followed by a subheadline celebrating the “Thirty infections in Campania, 15 in Lazio, 5 in Sicily and 6 in Puglia: now we are finally all brothers, stop with the search of the Northern infector” (Libero, 2020). Here the events and circumstances that led to the Italian Unification emerge as an unresolved node that generates resentment up to the present day. On the one hand, the Libero headlines, redacted from a Northern perspective, are symptomatic of a generalised sentiment of intolerance towards the very idea of sharing 170

 The Southernification of the Pandemic in Italy

the same national space with Southerners — once this is also accompanied by a “softer” impact of the pandemic in the South and by increased levels of sufferings in the North, generalised indignation for what is perceived as an injustice kicks in. The genocidal desire that inscribes Libero’s overt celebration of increased infections in the South, in this context, may be felt as a justified compensation for such a perceived injustice. On the other hand, the above-quoted declarations by Flavia Sorrentino demonstrate that a generalised resentment towards the patterns of cultural and material domination that inscribe Northern privilege — or white Northern sovereignty, as Palombo puts it (2010, p. 34) — within Unified Italy exists in the South as well.

“I FEEL UNDERREPRESENTED” Anti-Unification resentments have characterised the political debate in Southern Italy since the very moment of unification, with successive waves of intellectual and material insurgency that characterise the South from the brigandage mentioned above to, for instance, Sicilian Separatism in the 1940s,25 from Gramsci’s political writings26 to Nicola Zitara’s (1971, 1973) radical anticolonial calls for liberation in the 1970s, among others.27 A new wave of Southern Anti-Unification sentiments has arisen in proximity to the 2011 celebrations for the 150th Anniversary of the Italian Unification (cf. Monsagrati 2014). Not without controversies and editorial vicissitudes, the chapter author has elsewhere identified this period as “post-Italian” (Messina, 2016; 2018b; 2018c). Among the key works towards the construction of “post-Italian” narratives, best-selling volumes such as Pino Aprile’s Terroni28 (2010) and Marco Esposito’s Separiamoci (2013) have been identified. Marco Esposito’s book, in particular, visualises political independence for Southern Italy as a possible means of achieving social and territorial justice for the region and its inhabitants.29 Among the numerous arguments and data offered by Marco Esposito, Separiamoci contains a witty and well-argumented remark about the fact that, at the very moment of writing the book, more than 20 years had passed since the last time someone who was born in the South of Italy had held the position of Prime Minister, namely, Ciriaco De Mita, born and raised in the province of Avellino, whose mandate had ended in July 1989 (Esposito, 2014, p. 27-37). Marco Esposito himself clarifies that looking at politicians’ details would be, in itself, ridiculous, were it not for the increased attention that the Italian political debate had dedicated to the matter in the last decades, starting with the racial slurs suffered in 1990 by Bettino Craxi on account of his father’s Sicilian origins, which were connected in turn with the political rise of the Northern League30 (Esposito, 2014, p. 30-31). After the 1989 fall of De Mita’s Government, the next Southern Prime Minister was Giuseppe Conte, from the Apulian province of Foggia, whose first mandate started in June 2018. For a strange, statistically unlikely coincidence, when such an epoch-making event as the COVID-19 pandemic kicked in, the Government in office was headed by a Southern subject. Furthermore, the first Conte Cabinet had already fallen before the pandemic even started, in August 2019, and a new Government had been formed a few days later, with Conte reconfirmed as its head. This “Conte II” Cabinet had a record figure of 12 Ministers out of 22 from the South, an unprecedented occurrence at least since the 1990s (Fausti & Forti, 2019). These unusual circumstances immediately stimulated the vitriolic sarcasm of the then editor-in-chief of the Libero newspaper, Vittorio Feltri, a well-known figure in Italian journalism. Days after the formation of the Conte II Government, in a Libero editorial published on Sept 5, 2019, Feltri had incited his readers to “leave Conte alone with his zoo full of terroni31 who are hostile to the North that supports all of them financially”32 (Feltri, 2019). 171

 The Southernification of the Pandemic in Italy

During the pandemic, the fairly efficient central management of an unexpected and unprecedented emergency was commended by many observers, such as Sebastiani, Massa and Riboli (2020): The comparison of the curves of the epidemic in different Italian regions in relation to the time when the government measures were introduced (9th to 12th of March 2020), suggests that the earlier the measures were taken in relation to the phase of the epidemic in that particular region, the lower the cumulative incidence achieved during this epidemic wave. These observations indicate that the government measures were effective to both slow down the epidemic that was rampant in the North of Italy and prevent the epidemic in the centre and South of Italy the rise to the detrimental levels that were already present in the North of Italy in mid-March 2020 (Sebastiani, Massa and Riboli, 2020, p. 345) Maslova and Savino (2020) document that Conte’s consensus among Italians rose from 48% before the pandemic kicked in in February 2020 up to 60% (2020, p. 40). Bull (2021) noted that “Conte reached higher levels of approval than any of his immediate predecessors in the course of exercising a level of authority and control over the lives of Italian citizens not seen since the Second World War” (2021, p. 149-150). Furthermore, while highlighting the several areas of controversy that characterised Conte’s management of the pandemic, Bull recognised significant merits, such as the ability to perform a “leap into the unknown” (2021, p. 156) by enforcing a national lockdown (1) by then “was an unprecedented step take in peace-time Europe” (2021, p. 156), and that (2) “did not just emulate Wuhan but surpassed it” (2021, p. 156). More cautious analyses exist, such as Rullo’s argument about the “highly personalised nature of government” (2021, p. 204) emerging as a “growing prime ministerial dominance in Italian politics” (2021, p. 204) brought about by the COVID-19 outbreak. Regardless of the good or bad assessment of the Conte II Government’s management of the pandemic, it is important to note here that criticism against the Executive took, at times, the shape of direct attacks against the regional background of Conte and his Ministers. While this may indeed be an inevitable side-effect of the personalisation of Government flagged by Rullo (2021), the fact that some of these attacks — such as the editorial above by Feltri (2019) — pre-dated the pandemic attests to the persistence of generalised anti-Southern sentiments in Italy, that was only exacerbated by the COVID-19 outbreak (Figure 1).

172

 The Southernification of the Pandemic in Italy

Figure 1. A social media post showing the hometowns of the Conte II ministers

Source: https://bit.ly/3ljJ0Zd

A viral image started to appear on social media, where a table associated the surnames of the Ministers with their respective role in the Cabinet and hometown. The last row also showed the President of the Republic, although that is not a position within the Executive. Prominence was given to thirteen Southern ministers, plus the Deputy Minister Vito Crimi from Palermo, two central Italian Ministers from Rome — Roberto Gualtieri and Lorenzo Fioramonti — and Dario Franceschini as the group’s

173

 The Southernification of the Pandemic in Italy

only Northern subject from Ferrara (Fig.1). The table conveniently omitted the surnames of seven other in-office Ministers from the North. A standard social media post displaying this image is shown in Fig. 1: the user announces in Venetian that “I had a feeling of being underrepresented” (Fig.1) and adds a further, sarcastic comment below: “Among other things, the only one who is from North of Rome is Franceschini [sad emoticon]. But let’s not get upset: the regional criterion is unimportant; they get selected for their competence! [three appalled emoticons]” (Figure 1). Posted by numerous users and groups across various social media, this table was perhaps meant to convey the same feelings the above user verbalised. What is striking here, on top of the omission of several Northern Ministers, is the omission of the fact that this was a unique situation within the history of Italian Governments where a majority of Northern Ministers — and an overwhelming majority of Northern Prime Ministers — has consistently been the most typical situation (Esposito, 2013; Fausti and Forti, 2019). The post shown in Fig. 1 seems to stage a protest against the sacrilege of (momentarily) losing a privilege to Southerners, something utterly unbearable for many Northerners. Again, this feeling of lèse-majesté resonates with the perception that the deaths and sufferings related to the COVID-19 outburst in the South had been much lower than in the North. One could perhaps imagine that the social media post shown in Fig. 1 came with the (absurd) insinuation that the Southdominated Government had somewhat favoured this tragic imbalance to the detriment of the North. The very same Vittorio Feltri who had ranted about Conte’s “zoo full of terroni” (2019), on Apr 21 2020, on the TV show “Fuori dal coro”, voiced the frustration of the overthrown North by suggesting that “envious” Southerners were happy about the COVID-19 crisis in Lombardy, and then closed his tirade by affirming that Southerners were “inferior”: The fact that Lombardy fell into disgrace due to the Coronavirus has excited the hearts of many people who are obviously nourished by a feeling of envy or anger towards us because they suffer from a sort of inferiority complex. I don’t believe in inferiority complexes, I believe that Southerners are inferior in many cases33 (Feltri, 2020) In Feltri’s case, the resentment about the North’s unexpected primacy in the COVID-19 tragedy becomes the source of envy towards the South, which he then transfers to the South itself by formulating the perverse and absolutely unproven accusation that Southerners could rejoice by watching Northerners die and suffer. What is more, the “envious” and “angry” Southerners by then had also managed to “sneak into” the Government, causing Feltri’s further indignation and perhaps nurturing the suspicion that Conte’s “zoo full of terroni” had somewhat determined the sheer imbalance between North and South in terms of infections and deaths. What the Government measures had in fact managed to do, as seen above concerning Bull (2021) and Sebastiani, Massa and Riboli (2020), was contain the epidemic by preventing a more expressive spread of infections and deaths from the North to the South. Furthermore, this was undeniably the right thing to do.

FUTURE RESEARCH DIRECTIONS This chapter is part of a larger corpus of research covering different aspects of the COVID-19 pandemics. Previous contributions by the chapter author have offered a critical grasp on the experience of physical isolation in Brazil from a first-person (almost autoethnographic) perspective (Messina, 2020a), as well 174

 The Southernification of the Pandemic in Italy

as experimental, speculative, and practice-based insights on the paradigmatic post-2020 shifts in creative and intellectual activities (Messina, Costa and Scarassatti, 2020; Keller, Costalonga and Messina, 2020). The approach adopted in this chapter is partly an attempt to complement these previous research experiences by shifting the focus on broader social and geopolitical disruptions brought to the surface by the pandemic. All these research directions will be continuatively pursued and combined in future research. Undoubtedly, some important aspects of the North vs South dialectics during Italy’s first wave of COVID-19 have not been at the forefront of the present article or have been only briefly mentioned due to space constraints. For example, the issues emerging from football chants and banners and the way they interact, on the one hand, with the historical and social particularities of the Italian national space, and, on the other hand, with the specific contingency of the pandemic may deserve a separate work. Similarly, a limited part of this chapter has been dedicated to speeches and releases by politicians occupying national and local positions: a comprehensive work on the multiple tensions, contentions, resentments, and contrasts that emerge from these speeches during the first half of 2020 and beyond, and on what they signal – once again – specific North vs South conflicts and unresolved issues is definitely needed. While the author has made use of critical tools to analyse the manifestations presented in this chapter, a series of separate works focussing on some specific topics (e.g., Vittorio Feltri’s anti-Southern tirades or the summoning of Unification-inspired imagery in commentaries on the geographical distribution of deaths and contagions) may help develop deeper and broader critical analyses.

DISCUSSION AND CONCLUSION Racist manifestations against Southern Italians were not circumscribed to the Italian national space. In early March, at the beginning of the emergency in Italy, French TV Canal+ released a humoristic video titled Corona Pizza. In the video, an Italian pizza chef sneezed, coughed and spat green phlegm from his nose on a pizza he had just baked. Italian authorities, politicians and professional associations voiced their indignation at the video, soliciting formal explanations from the French Consular authorities (Custodero, 2020). Southern commentators, like the Napoli Council collaborator Flavia Sorrentino, argued that the pizza and the pizzaiuolo are part of the traditional Neapolitan heritage and that therefore the humoristic video was prevalently an attack against Naples and Southern Italy (Vasso, 2020). Again, achieving distance from the perceived source of infection via what Roberto Esposito describes as “the ‘nonbeing’ or the ‘not-having’ anything in common” (2006, p. 28) emerges as a characteristic strategy subsumable to the broad concepts of immunitas and immunisation. By making fun of the tragedy that was taking place in Italy, the Canal+ video symbolically attempted to trace a separation from the pandemic and achieve immunitas from it. Indeed, in the Canal+ video, as much as in all the above manifestations from Italy, it seems that a set of specific ethnic and territorial characteristics needed to be given to the COVID-19 emergency, and in the most predictable fashion Southern Italy was immediately identified as a “natural repository” of the negative features associated to the infection, in a way that it was expected that the harmful, repugnant and undesirable features associated to the infection were already there in the most stereotyped representation of the region and its inhabitants. Indeed, it needs to be reiterated that the worst set of COVID-19-related prejudices and stereotypes, in this respect, was assigned to Chinese and East Asian people. Furthermore, once the first wave had ended in Europe, it kicked in elsewhere in the world, in developing countries like Brazil, India, Mexico, Iran and Peru. Suddenly, as the chapter author has noted elsewhere (Mes175

 The Southernification of the Pandemic in Italy

sina, 2020a), the global lament about the European COVID-19 “tragedy” was replaced by the label of “threats” given to the developing countries that were now experiencing thousands of daily deaths (Pittet cit. Chade, 2020). In a nutshell, European deaths are a tragedy, whereas non-European deaths are a threat. After all, during the following pandemic waves, when the figures between the North and the South of Italy became less imbalanced, it was not uncommon to witness overt manifestations of sarcasm against Southern regions either experiencing peaks of contagions and death or lagging with the vaccination. At the end of August 2021, after a thriving post-pandemic tourist season, Sicily experienced an increase in contagions and entered a yellow zone of alert, the only one in Italy. On Aug 30, the satirical Facebook page I vaccini e altri complotti leggendari [“The vaccines and other legendary conspiracies”] posted a sarcastic banter at the expense of Sicilians: “A thought for Sicilians in the yellow zone: bye-bye, losers.”34 The natural order of things was now re-established: Northern Italy experienced low levels of infection as a result of its “self-evident virtuosity”, while Sicily seemed to be sinking into a regime of restrictions again after a summer (2021) when many people thought the pandemic was gone for good. In this context, a prominent pandemic-related Facebook page thought it was all right and good to make fun of Sicilians because of that. In conclusion, it is possible to confirm here the widespread impression that the pandemic in Italy, rather than an occasion to rediscover national solidarity across social groups regardless of age, gender, ethnicity and territorial origin, has been yet another field where existing divisions have been exacerbated or reopened. In this general context, the North-centric view of the South as an internal exception, functioning as a national repository of filth, illness, and illegal behaviour, has been abundantly renovated and reinforced to the point that the pandemic has been Southernified, even at moments when its Southernification was extremely counterintuitive and utterly absurd.

REFERENCES Adnkronos. (2020). Coronavirus, presidente Calabria: “Governo blocchi partenze”. Adnkronos. Aprile, P. (2011). Terroni. Tutto quello che è stato fatto perché gli italiani del Sud diventassero «meridionali». Edizioni Piemme. Balasco, N., D’Alessandro, V., Ferrara, P., Smaldone, G., & Vitagliano, L. (2021). Analysis of the time evolution of COVID-19 lethality during the first epidemic wave in Italy. Acta Biomedica, 92(2), e2021171. doi:10.23750/abm.v92i2.11149 PMID:33988144 Banfield, E. C. (1958). The Moral Basis of a Backward Society. Free Press. Barcella, P. (2018). Percorsi leghisti: Dall’antimeridionalismo alla xenofobia. Meridiana, 91, 95–119. Bufale.net. (2020). Audio WhatsApp e caso di Coronavirus in Italia, all’Ospedale Vito Fazzi di Lecce: solo una bufala. Bufale.net. https://www.bufale.net/audio-whatsapp-e-caso-di-coronavirus-in-italiaallospedale-vito-fazzi-di-lecce-solo-una-bufala/ Bull, M. (2021). The Italian government response to Covid-19 and the making of a prime minister. Contemporary Italian Politics, 13(2), 149–165. doi:10.1080/23248823.2021.1914453

176

 The Southernification of the Pandemic in Italy

Burioni, R. (2020). Coronavirus, Roberto Burioni: “Il virus da noi non c’è, niente panico”. L’aria che tira. https://www.la7.it/laria-che-tira/video/coronavirus-roberto-burioni-il-virus-da-noi-non-ce-nientepanico-03-02-2020-305450 Capogreco, S. (2014). Italianness, Interrupted: (De)centring “self” at the transgressive boundaries and internal fissures of nation. University of Melbourne. Chade, J. (2020). ‘Covid-19: Brasil, EUA e México são “ameaças” ao mundo, diz cientista suíço.’ UOL. https://noticias.uol.com.br/colunas/jamil-chade/2020/08/03/brasil-eua-e-mexico-sao-ameacas-para-omundo-diz-cientista-suico.htm Chiaro, D. (2017). “Vivi Pericolosamente”: Christie Davies, Italians and dangerous things. The European Journal of Humour Research, 5(4), 41–50. doi:10.7592/EJHR2017.5.4.chiaro Ciardi, L. (2022, Feb 9). Covid, domande e risposte: quarantene, green pass, terza dose e positivi fuori regione. La Nazione. https://www.lanazione.it/cronaca/domande-covid-1.7301019 Conticini, E., Frediani, B., & Caro, D. (2020). Can atmospheric pollution be considered a co-factor in extremely high level of SARS-CoV-2 lethality in Northern Italy? Environmental Pollution, 261, 114465. doi:10.1016/j.envpol.2020.114465 PMID:32268945 Custodero, A. (2020, Mar 3). Francia, rimosso il video sulla pizza al coronavirus. Le scuse di Canal+. La Repubblica. https://www.repubblica.it/politica/2020/03/03/news/francia_spot_satirico_corona_ pizza_coronavirus_virus_wuhan_italia_influenza_bellanova_canal_video_vergognoso_e_raccapriccia-250109103/ D’Alessandro, J. (2020, Apr 23). Coronavirus, l’illusione della grande fuga da Milano. Ecco i veri numeri degli spostamenti verso sud. La Repubblica. https://www.repubblica.it/tecnologia/2020/04/23/news/ coronavirus_l_illusione_della_grande_fuga_da_milano_e_i_veri_numeri_degli_spostamenti_verso_sud254722355/ De Riccardis, S., & Pisa, M. (2020). Coronavirus, la Lombardia diventa “zona rossa”: fuga da Milano sui treni notturni, poi in stazione torna la calma. La Repubblica. https://milano.repubblica.it/cronaca/2020/03/08/ news/coronavirus_la_lombardia_sara_zona_rossa_fuga_da_milano_in_treno_e_in_auto-250603359/ Del Boca, A. (2011). Italiani, brava gente? Neri Pozza Editore. Dickie, J. (1994). The South as Other: From Liberal Italy to the Lega Nord. The Italianist, 14, 124–140. Dines, N. (2013). Bad news from an aberrant city: A critical analysis of the British press’s portrayal of organised crime and the refuse crisis in Naples. Modern Italy, 18(4), 409–422. doi:10.1080/13532944 .2013.801677 Esposito, M. (2013). Separiamoci. Magenes. Esposito, R. (2006). The immunisation paradigm. Campbell, T. (transl.). diacritics, 36(2), 23-48. Fanon, F. (1986). Black Skin, White Masks. Pluto Press.

177

 The Southernification of the Pandemic in Italy

Fanpage (2020). Che cosa significa infodemia, la malattia che secondo l’Oms fa più male del Coronavirus. Fanpage. https://www.fanpage.it/cultura/che-cosa-significa-infodemia-la-malattia-che-secondoloms-fa-piu-male-del-coronavirus/ Fausti, S., & Forti, G. (2019). Da dove vengono i ministri della Seconda Repubblica? YouTrend. https:// www.youtrend.it/2019/10/22/da-dove-vengono-i-ministri-della-seconda-repubblica/ Feldman, A. (1997). Violence and Vision: The Prosthetics and Aesthetics of Terror. Public Culture, 10(1), 24–60. doi:10.1215/08992363-10-1-24 Feltri, V. (2020) Vittorio Feltri: I meridionali? Credo che in molti casi siano inferiori. Fuori dal coro, Rete 4. https://mediasetinfinity.mediaset.it/article/vittorio-feltri-credo-che-i-meridionali-in-molti-casisiano-inferiori_a11749 Finkelstein, M. S. (1999). Separatism, the Allies and the Mafia: The struggle for Sicilian independence, 1943-1948. Lehigh University Press. Francioni, P. (2020). La gaffe l’inviata di Agorà: “Non siamo fortunati, non c’è nessuno”. Il Giornale. https://www.ilgiornale.it/news/spettacoli/coronavirus-linviata-agor-delusa-virtuosismo-dei-napoletani-1854418.html Francklin, E. (2020). Coronavirus Bari, ammassati con la mascherina sotto al mento: fase 8 fuori dalla posta al San Paolo. Il Quotidiano Italiano — Bari. https://bari.ilquotidianoitaliano.com/attualita/2020/04/ news/coronavirus-bari-ammassati-con-la-mascherina-sotto-al-mento-fase-8-fuori-dalla-posta-al-sanpaolo-271075.html/ Gazzetta del Sud. (2020). Reggio, la donna di Taurianova ricoverata in isolamento non sarebbe colpita da coronavirus. Gazzetta del Sud. https://reggio.gazzettadelsud.it/articoli/cronaca/2020/01/31/reggiola-donna-di-taurianova-ricoverata-in-isolamento-non-sarebbe-colpita-da-coronavirus-517acac6-be854eb3-a551-d2e7d1e3fcce/ Giovanetti, M., Benvenuto, D., Angeletti, S., & Ciccozzi, M. (2020). The first two cases of 2019‐nCoV in Italy: Where do they come from? Journal of Medical Virology, 92(5), 518–521. doi:10.1002/jmv.25699 PMID:32022275 Gramsci, A. (1966). La Questione Meridionale. Editori Riuniti. Gribaudi, G. (1997). Images of the South: The Mezzogiorno as seen by Insiders and Outsiders. The new history of the Italian South: The Mezzogiorno revisited, 83-113. Huysseune, M. (2006). Modernity and secession: The social sciences and the political discourse of the Lega Nord in Italy. Berghahn Books. doi:10.2307/j.ctv287sgzd Il Fatto Quotidiano. (2020, Mar 9). Coronavirus, l’appello di De Magistris a chi è arrivato a Napoli dalle zone rosse: “Mettetevi in quarantena. Non è il momento di fare gli egoisti”. Il Fatto Quotidiano. https:// www.ilfattoquotidiano.it/2020/03/09/coronavirus-lappello-di-de-magistris-a-chi-e-arrivato-a-napolidalle-zone-rosse-mettetevi-in-quarantena-non-e-il-momento-di-fare-gli-egoisti/5730147/ Il Gazzettino. (2020, Mar 8). Coronavirus, l’appello di Emiliano: «Fermatevi e tornate indietro, non portate in Puglia l’epidemia». Il Gazzettino. https://www.ilgazzettino.it/viaggi/news-5098641.html

178

 The Southernification of the Pandemic in Italy

Il Messaggero. (2020°, Mar 25). Coronavirus, la “vendetta” del Sud: «Non si affitta ai settentrionali». Sui social è boom. Il Messaggero. https://www.ilmessaggero.it/italia/coronavirus_sud_nord_non_si_affitta_meridionali_25_febbraio_2020-5073714.html Il Messaggero. (2020b, Mar 8). Coronavirus, a Milano la fuga dalla “zona rossa”: folla alla stazione di Porta Garibaldi. Il Messaggero. https://www.ilmessaggero.it/italia/coronavirus_milano_fuga_milano_treno_romani-5097472.html Jacobson, M. F. (1999). Whiteness of a different color. Harvard University Press. doi:10.2307/j.ctvjk2w15 Keller, D., Costalonga, L., & Messina, M. (2020, November). Ubiquitous music making in COVID-19 times. In Proceedings of the 10th Workshop on Ubiquitous Music (UbiMus 2020), 3-16. La Repubblica. (2020). “Torni a casa chi non è toscano”. Rossi sceglie la linea dura. La Repubblica (Firenze). https://bit.ly/3FkiWE9 Libero (2020, Mar 4). L’infezione crea l’Unità d’Italia: Virus alla conquista del Sud. Libero. https:// www.vesuviolive.it/ultime-notizie/327529-coronavirus-sud-libero/ Librandi, G. (2020, Dec 11). Gianfranco Librandi e la superiorità dei migranti. La Zanzara. Rai Radio 2. https://youtu.be/bAlbIiISkmM Librandi, G. (n.d.). Personal homepage. Librandi. http://www.gianfrancolibrandi.it/ Lombroso, C. (1896). Il cervello del brigante Tiburzi. Nuova Antologia, 150, 591–600. Lombroso, C. (1900). L’ultimo brigante: Giuseppe Musolino. Nuova Antologia, 181, 508–516. Maslova, K., & Savino, G. (2020). Pandemic all’italiana: Stress-test for the political system. Contemporary Europe-Sovremennaya Evropa, 97(4), 37–49. doi:10.15211oveurope420203749 Mautone, E. (2020, Jan 20). Coronavirus, caso sospetto a Napoli: al Cotugno coppia di sposini cinesi in viaggio di nozze. De Luca convoca i manager: kit diagnostici al Cotugno. Il Mattino. https://www. ilmattino.it/napoli/cronaca/coronavirus_napoli_ricoverato_cotugno-5013206.html Melley, M. (2021, Dec 29). In migliaia non riescono a prenotare la terza dose. Milano Today. https:// www.milanotoday.it/attualita/coronavirus/terza-dose.html Messina, M. (2016). Narrativas pós-italianas: A re-imaginação da unidade nacional nas canções do sul da Itália. Muiraquitã: Revista de Letras e Humanidades, 4(1), 113–125. doi:10.29327/216341.4.1-10 Messina, M. (2018a). The Demonization of the South and the Southernification of Evil in Contemporary Italian Cinema: Belluscone and Qualunquemente. Journal of Italian Cinema & Media Studies, 6(2), 193–207. doi:10.1386/jicms.6.2.193_1 Messina, M. (2018b). Definire il post-italiano: Tentativi di superamento dell’orizzonte nazionale italiano nel Mezzogiorno. Jamaxi, 2(2), 107–140. Messina, M. (2018c). La canzone meridionale “post-italiana” e il Mediterraneo. In Paci, D., Perri, P. & Zantedeschi, F. (eds), Paesaggi mediterranei: storie, rappresentazioni, narrazioni. Canterano (RM): Aracne Editrice, 153-170.

179

 The Southernification of the Pandemic in Italy

Messina, M. (2019). CompraSud: A Collective Attempt to Overturn the National Italian Consumer Goods Market. História e Economia: Revista interdisciplinar, 23(2), 60-76. Messina, M. (2020a). 107 days and counting.... Portal: Journal of Multidisciplinary International Studies, 17(1/2), 104–109. Messina, M. (2020b). Southernness, Disability and the Construction of the “Other” in Italian Cinema: Desire, Masculinities, Disfigurations and Medicalisations. Simultanea: Simultanea: A Journal of Italian Media and Pop Culture 1(1), 1-10. Messina, M., & Capogreco, S. (2019). The Silent Return of the South to the South: Verdone’s Pasquale Ametrano as a Counter-Chaplin. Tropos: Comunicação. Society and Culture, 8(1), 1–17. Messina, M., Costa, V. F., & Scarassatti, M. (2020). Cartridge Music in the Quarantine: Presence, Absence, Contingency Setups and (De-) territorialised Performances. INSAM Journal of Contemporary Music. Art and Technology, (5), 28–45. Messina, M., & Di Somma, T. (2017). Unified Italy, Southern Women and Sexual Violence: Situating the Sexual Assault TV “prank” Against Emma Marrone Within the Dynamics of Contemporary Italy as a Scopic Regime. Revista Tropos: Comunicação. Society and Culture, 6(1), 1–18. Minuz, A. (2020, Feb 16). Il virus da talk-show. Il Foglio. https://www.ilfoglio.it/televisione/2020/02/10/ news/il-virus-da-talk-show-301103/ Misuraca, R., Teuscher, U., Scaffidi Abbate, C., Ceresia, F., Roccella, M., Parisi, L., Vetri, L., & Miceli, S. (2022). Can We Do Better Next Time? Italians’ Response to the COVID-19 Emergency through a Heuristics and Biases Lens. Behavioral Sciences (Basel, Switzerland), 12(2), 39. doi:10.3390/bs12020039 PMID:35200290 Open (2020, Mar 8). Coronavirus, fuga dal nord: folle alle stazioni di Milano e treni presi d’assalto – Le immagini. Open. https://www.open.online/2020/03/08/coronavirus-fuga-dal-nord-folle-alle-stazioni-dimilano-e-treni-presi-dassalto-foto/ Paci, D., & Pietrancosta, F. (2010). Il separatismo siciliano (1943-1947). Diacronie: Studi di Storia Contemporanea, (3), 2. Palermo Today. (2020, Apr 12). VIDEO: “Arrustute” proibite allo Sperone, sui tetti spunta la polizia e scatta il fuggi fuggi. Palermo Today. https://www.palermotoday.it/cronaca/video-grigliata-tetti-speronepolizia-fuggi-fuggi.html Palombelli, B. (2020). Stasera Italia: 20 marzo 2020 VIDEO. Mediaset. https://bit.ly/424eqDx Palombo, L. (2010). The drawing of the sovereign line. In J. Pugliese (Ed.), Transmediterranean: diaspora, histories, geopolitical spaces (pp. 39–58). Peter Lang Publishing. Pardalis, S. (2009). Terroni and Polentoni: Where Does the Truth Lie? An Anthropology of Social Networks And Ethnicity in Palermo (Sicily). Duhram University. Perera, S. (2014). Dead exposures: Trophy bodies and violent visibilities of the nonhuman. Borderlandse-journal, 13(1), 1–26.

180

 The Southernification of the Pandemic in Italy

Pierini, S. (2020, Jan 24). Coronavirus, primo caso sospetto in Italia: è una cantante barese, tornava da Wuhan. Il Messaggero. https://www.leggo.it/italia/cronache/coronavirus_primo_caso_italia_ultime_notizie_oggi_23_gennaio-5002825.html Pizzo, C. (2009). L’impossibile spazio del bando. Annali-Università degli Studi Suor Orsola Benincasa, 7(2). Positano News. (2020, Feb 7). Napoli. Sospetto contagio da Coronavirus per un assistente di volo, ricoverato al Cotugno per la prevista profilassi. Positano News. https://www.positanonews.it/2020/02/napolisospetto-contagio-coronavirus-un-assistente-volo-ricoverato-al-cotugno-la-prevista-profilassi/3361402/ Prisco, F. (2020, Feb 24). Malvoluti al Sud: il Lombardo-Veneto messo in quarantena dal resto d’Italia. Il Sole 24 Ore. https://www.ilsole24ore.com/art/malvoluti-sud-veneti-e-lombardi-messi-quarantenaresto-d-italia-ACNe4aLB Puente, D. (2020, Jan 29). Due audio bufala circolano via Whatsapp: «A Lecce un cinese è stato ricoverato per il Coronavirus! State alla larga!». Open. https://www.open.online/2020/01/29/due-audio-bufalacircolano-via-whatsapp-a-lecce-un-cinese-e-stato-ricoverato-per-il-coronavirus-state-alla-larga/ Pugliese, J. (2008). Whiteness and the blackening of Italy: La guerra cafona, Extracomunitari and Provisional Street Justice. Portal: Journal of Multidisciplinary International Studies, 5(2), 1–35. Pugliese, J. (2017). The Souths of “The West”: Geocorpographical Assemblages of Plants, Colonialism and Race in Sergio Leone’s Metafigural Spaghetti Westerns. Muiraquitã: Revista de Letras e Humanidades, 5(2), 11–59. doi:10.29327/216344.5.2-3 Putnam, R. D. (1993). Making Democracy Work: Civic Traditions in Modern Italy. Princeton University Press. Regione Campania. (2020). Coronavirus, discorso del presidente De Luca. Regione Campania. https:// www.regione.campania.it/regione/it/multimedia/web-tv/coronavirus-discorso-del-presidente-de-luca Regione Puglia. (2020). Il presidente Emiliano spiega in video l’ordinanza dell’8 marzo 2020. Regione Puglia. https://press.regione.puglia.it/-/il-presidente-emiliano-spiega-in-video-l-ordinanza-dell-08.03.2020 Reuters. (2020). Two first coronavirus cases confirmed in Italy: prime minister. Reuters. https://www. reuters.com/article/us-china-health-italy-idUSKBN1ZT31H Riccio, I. (2020, Feb 2). “Napoletani figli del colera vi mettiamo in quarantena”: lo striscione choc dei tifosi dell’Inter. Il Giornale.it. https://www.ilgiornale.it/news/napoli/napoletani-figli-colera-vi-mettiamoquarantena-polemiche-1820626.html Rossano, M. (2011). La construcción del héroe y del antagonista a través de las imágenes: Garibaldi y los bandoleros del sur de Italia 1861-2011. Quaderns-e de l’Institut Català d’Antropologia, (16 (1-2)), 189–200. Roviello, V., & Roviello, G. N. (2021). Less COVID-19 deaths in southern and insular Italy explained by forest bathing, Mediterranean environment, and antiviral plant volatile organic compounds. Environmental Chemistry Letters, 20(1), 7–17. doi:10.100710311-021-01309-5 PMID:34483793

181

 The Southernification of the Pandemic in Italy

Rullo, L. (2021). The COVID-19 pandemic crisis and the personalisation of the Government in Italy. International Journal of Public Leadership, 17(2), 196–207. doi:10.1108/IJPL-08-2020-0083 Salvetti, P. (2017). Rope and soap: Lynchings of Italians in the United States. Bordighera Press. Sayad, A. (1999). La double absence. Seuil. Scognamiglio, R. (2020). “Chiudono tutta la Lombardia”. Assalto alla stazione di Milano. Il Giornale. https://www.ilgiornale.it/news/milano/pasticcio-governo-sul-decreto-scatta-fuga-milano-1837402.html Sebastiani, G., Massa, M., & Riboli, E. (2020). Covid-19 epidemic in Italy: Evolution, projections and impact of government measures. European Journal of Epidemiology, 35(4), 341–345. doi:10.100710654020-00631-6 PMID:32306149 Soscia, M. (2014). Tra storia e letteratura. Il colera in Italia e a Napoli. Sinestesieonline, 10, 7–28. Spagnolo, P. (2020). Contro il calcio moderno. Odoya. Spena, A. (2020). Coronavirus, perché nasce la psicosi. Vita. https://www.vita.it/it/article/2020/01/31/ coronavirus-perche-nasce-la-psicosi/153939/ Staples, B. (2019). How Italians Became ‘White’. The New York Times. https://www.nytimes.com/ interactive/2019/10/12/opinion/columbus-day-italian-american-racism.html Teti, V. (1993). La razza maledetta. Origini del pregiudizio antimeridionale. Manifestolibri. Van Dijk, T. A. (1991). Media contents: The interdisciplinary study of news as discourse. In A handbook of qualitative methodologies for mass communication research (pp. 108–120). Routledge. VareseNoi. (2020, Mar 9). FOTO. «Un grazie non indifferente a chi salva la nostra gente»: striscioni della curva al Circolo e in città. VareseNoi. https://www.varesenoi.it/2020/03/09/leggi-notizia/argomenti/varese/ articolo/foto-un-grazie-non-indifferente-a-chi-salva-la-nostra-gente-striscioni-della-curva-al-circolo.html Vasso, G. (2020, Mar 4). La pizza al coronavirus fa infuriare il Comune di Napoli. Il Giornale. https:// www.ilgiornale.it/news/napoli/pizza-coronavirus-fa-infuriare-comune-napoli-1835258.html VeraLeaks. (2020, Apr 20). Il coronavirus non ferma le attività illecite nel mercato ittico di Taranto davanti all’ufficio dei Vigili Urbani. VeraLeaks. https://veraleaks.org/2020/04/19/il-coronavirus-nonferma-le-attivita-illecite-nel-mercato-ittico-di-taranto-davanti-allufficio-dei-vigili-urbani/ Verdú, D. (2020, Apr 21). El virus invierte los papeles históricos del norte y el sur en Italia. El País. https://elpais.com/internacional/2020-04-21/el-virus-invierte-los-papeles-historicos-del-norte-y-el-suren-italia.html VirgilioNotizie. (2020, Apr 8). Coronavirus, a Napoli folla in strada nei Quartieri Spagnoli. VirgilioNotizie https://notizie.virgilio.it/coronavirus-napoli-folla-strada-quartieri-spagnoli-1358953 Zitara, N. (1971). L’Unità d’Italia: Nascita di una colonia. Jaca Book. Zitara, N. (1973). Il proletariato esterno: Mezzogiorno d’Italia e le sue classi. Jaca Book.

182

 The Southernification of the Pandemic in Italy

ENDNOTES 1 2

3



6 4 5

7



8



9



10



11



12



On the circumstances of the cholera epidemics in Naples throughout history, cf. Soscia (2014) The first COVID-19 death had been registered in Italy right on the very same day of the Brescia banner. Obviously, the virulent racism against East Asians conflates, in many aspects, with that against Southern Italians. “Fa più male del Coronavirus” “The talk-show virus” “In questo momento il virus in Italia non c’è, e quindi è più giusto preoccuparsi dei fulmini, delle alluvioni, che invece ci sono. [...] Oggi uno che volesse prenderlo [il coronavirus] in Italia, che si alza la mattina e dice: “io mi voglio contagiare”, non può farlo, quindi di quello è inutile parlarne” “Ieri si è determinato un punto di svolta per quanto riguarda il contagio nella nostra regione. Nel senso che ieri abbiamo registrato centinaia di arrivi di nostri concittadini dal Nord Italia in Campania. Quindi c’è un prima e un poi. [...] Si è fatto un lavoro di prevenzione davvero straordinario. Ora, con l’arrivo di centinaia di persone, non tutte controllate, ovviamente, dal Nord, noi andremo in difficoltà. [...] E dunque noi dobbiamo aspettarci per il mese di marzo un picco di contagi nella nostra regione. [...] Siamo entrati in una seconda fase perché abbiamo avuto un’ondata di ritorni in Campania. Non dipende da noi.” This state of things is part of Italians’ daily life, one that eventually created a big number of organisational issue once vaccines were ready to be injected to the population, and all Italian regions had to autonomously organise special provisions for those citizens who had their official residence status elsewhere but were temporarily located in their territories and did not want to travel in order to get vaccinated (Melley, 2021). On a personal note, the chapter author also needed to get his shots outside of his region of residence, and this resulted in not being able to book the vaccination online, and in having to literally beg the medical and paramedical staff at various vaccination hubs, until someone finally accepted to authorise his vaccination. The status of official residents vs. that of temporary domiciliati also created numerous issues when COVID-19 positives were living outside their home regions, as in many cases they were not eligible to be cured by a doctor (Ciardi, 2022). “La ‘vendetta’ del Sud”. “Por primera vez desde la unificación del país, desde los tiempos del crecimiento de la FIAT en Turín y la inmigración masiva, el éxodo se producía en dirección contraria” C’erano una volta le due «Italie», il Nord produttivo e il Sud orfano del Regno che fu, la questione meridionale (Billia) e «il guaio l’ha fatto Garibaldi» (Totò), il «non si affitta ai meridionali» degli anni Cinquanta e la grande tolleranza del Mezzogiorno, risorsa ma certe volte persino problema. Perché di troppa tolleranza – nei confronti dell’illegalità della porta accanto – si può morire. Ci sono eventi che cambiano la percezione, ribaltano i luoghi comuni, sovvertono ogni certezza, buona o cattiva che sia. Non conosciamo ancora con precisione l’impatto che l’ emergenza coronavirus potrà avere sulle nostre vite nel medio-lungo periodo. Non sappiamo se è un evento di portata storica e speriamo di no, ma di sicuro sta uccidendo le sicurezze di molti, su e giù per lo Stivale: in una manciata di ore lombardi e veneti, fino all’altro-ieri idea platonica di produttività applicata al territorio italico e al tempo stesso «big spender» coccolatissimi dall’industria turistica, non sono più i benvenuti. “Un grazie a chi è partito, la Lombardia avete ripulito”

183

 The Southernification of the Pandemic in Italy

13 14



15



16



17



18

19



20



21

22 23

24



27 25 26

28



184

Giuseppe Conte was Italy’s Prime Minister between 2018 and 2021 Umberto Bossi was the first leader of the xenophobic Northernist party Northern League (Lega Nord), and famously advocated for the separation of Northern Italy (dubbed Padania) from Central and Southern Italy, with the expulsion of Southern immigrants (cf. Dickie, 1994; Huysseune, 2006). “Terroni” is a derogatory, racist term by which Southerners are often referred to (cf. Pugliese, 2008; Pardalis, 2009). The post circulated abundantly on social media, for example on [Accessed 02/03/2023]. Original Italian: “Comunque Conte è riuscito a fare in una notte quello che Bossi non è riuscito a fare in trent’anni: chiudere la Lombardia e liberarla dai terroni che si son pure pagati il biglietto!!!” “Torni a casa chi non è toscano” As Stefania Capogreco has argued in her solo work (Capogreco, 2014) and in co-authorship with the author of this chapter (Messina and Capogreco, 2019), the return for the Southern immigrant is, in many respects, “returning elsewhere”. Messina and Capogreco (2019) illustrate the eternal and unresolvable otherness of the returning Southern immigrant referring to the filmic trajectory of Carlo Verdone’s character Pasquale Ametrano. “Se lo chiedono tutti, oggi è venuto fuori che il 90% dei morti è nelle regioni… praticamente nelle regioni del Nord. Che cosa ci può essere stato di più? Qualcosa di diverso? Comportamenti… persone più ligie, che quindi vanno tutte a lavorare? Come mai c’è questa esplosione al Nord?” “Geneticamente resistono di più” “Io, che sono un calabrese” (Librandi, 2020). In fact, Librandi was born in the Northern town of Saronno from a Calabrian (Southern) father and a Romagnola (Northern) mother (Librandi n.d.) “Noi siamo africani bianchi” “Non siamo fortunati in realtà, in questo momento si stanno comportando... Non c’è nessuno, ma fino a pochi minuti fa c’era un passaggio intenso” “Se chiedessimo spiegazioni probabilmente la giornalista direbbe che è stata fraintesa. Lo scoop non c’è e per sfortuna di chi ci osserva attraverso uno “sguardo nordico” i napoletani stanno dimostrando che il pregiudizio è nemico giurato della verità. Non si tratta di cercare a tutti i costi un motivo per indignarsi, ma di constatare ogni volta quanto sia interiorizzato il preconcetto culturale verso di noi. Le frasi pronunciate a metà, l’imbarazzo frammisto al dispiacere per non aver potuto cogliere in flagranza di inciviltà i napoletani, rappresentano moduli di un registro linguistico e gestuale che conosciamo bene e che combattiamo [...] Se c’è una cosa che questa emergenza ha portato alla luce - con le migliaia di vite spezzate e le inchieste di Bergamo e Milano per epidemia colposa - è che non possedete nessuna titolarità di cattedra etica da cui potete darci lezioni di civiltà. Probabilmente non ci dormirete la notte, ma è bene che cominciate a farvene una ragione” See Finkelstein (1999) and Paci and Pietrancosta (2010) on Sicilian Separatism. See for example the collection of writings La Questione Meridionale (1966) See Zitara’s polemical volume L’Unità d’Italia: Nascita di una colonia (1971) and his subsequent work Il proletariato esterno (1973). Cf. note 15 for an explanation of the derogatory meaning of the word terroni in Italian. Obviously, Aprile’s title conveys both a polemical remark against anti-Southern slurs and the critical proposal of reappropriating the label of terroni for identitarian and emancipatory purposes.

 The Southernification of the Pandemic in Italy

29



32 33 30 31

34



While the separatist provocation is undeniably the most important aspect of the book, Marco Esposito concludes his work by declaring that this would be an extreme move and that other solutions are possible and desirable in a united country (Esposito, 2014, pp. 153-181) Cf. note 14 for a brief introduction on the Northern League. Cfr. notes 15 and 28 for an explanation of the term. “Lasciamo a Conte il suo zoo di terroni ostili al Nord che li mantiene tutti” “Il fatto che la Lombardia sia andata in disgrazia per via del coronavirus ha eccitato gli animi di molta gente che naturalmente è nutrita da un sentimento di invidia o di rabbia nei nostri confronti perché subisce una sorta di complesso d’inferiorità. Io non credo ai complessi d’inferiorità, io credo che i meridionali in molti casi siano inferiori” “Un pensiero ai siciliani in zona gialla: Ciao, sfigati.” https://bit.ly/3limbVN.

185

186

Chapter 10

Are There Hate Speeches on Spanish Television?

Methodological Proposal and Content Analysis Over the 2020 Year Sandra Martínez Costa https://orcid.org/0000-0002-3052-736X University of A Coruña, Spain Teresa Nozal Cantarero https://orcid.org/0000-0002-2652-5898 University of A Coruña, Spain Antonio Sanjuán Pérez https://orcid.org/0000-0001-7612-2838 University of A Coruña, Spain José Juan Videla Rodríguez https://orcid.org/0000-0001-8656-9297 University of A Coruña, Spain

ABSTRACT Few academic studies focus on hate speech on television. That is partly due to the difficulty of obtaining and analyzing broadcast content. However, the spread of those hate messages implies a high reach. For this study, the researchers propose an experimental methodology to analyze the content broadcast on the 24 hours of the five Spanish free-to-air television channels over one year (2020). The authors examined the presence of abusive or hurtful vocabulary and quantified the insults aired. They extracted and studied a sample through content analysis to detect if those insults were accompanied, in any way, by expressions of hatred. Although the messages the researchers studied for this article cannot be considered speeches or hate crimes, there are some offensive comments related to gender, race, or religion, mainly on fictional products. DOI: 10.4018/978-1-6684-8427-2.ch010

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Are There Hate Speeches on Spanish Television?

BACKGROUND Defining hate speech is complex and depends on many factors, some of which are subjective. According to legal definitions, the Council of Europe describes hate speech as All types of expression that incite, promote, spread, or justify violence, hatred or discrimination against a person or group of persons or that denigrates them because of their actual or attributed personal characteristics or status such as ‘race’, colour, language, religion, nationality, national or ethnic origin, age, disability, sex, gender identity and sexual orientation. In Spain, the 510 article of the Penal Code (Boletín Oficial del Estado, 1995) identifies hate crimes as The incitement, direct or indirect, to hatred, hostility, discrimination or violence against a group, part of it or a specific person because they belong to that group, for racism, antisemitism or other reasons related to ideology, religion or beliefs, family situation, membership of their members to ethnicity, race or nation, their national origin, gender, sexual orientation or identity, gender reasons, illness or disability. The 510 legal articles also describe a hate crime as “the possession and distribution of material that directly or indirectly promotes hostility against the groups or individuals previously defined and the denial or glorification of crimes against humanity committed during an armed conflict.” Therefore, in summary, the researchers define for this study, hate speech as any message that discriminates or encourages discrimination, humiliation, harassment, stigmatization or contempt for individuals or social groups with the characteristics or attributes defined by the laws. Within this frame of reference, the research on hate speech has increased significantly in the last decade, as the emergence and consolidation of social networks have multiplied its spread and reach, promoting cyber hate and cyber racism that is becoming more and more explicit (Bustos Martínez et al., 2019). Nevertheless, hate speech is restricted on television because the editorial control over the broadcasted content is higher than in social networks. In Spain, the Union of Commercial Open Television (UTECA) maintains its commitment to the 2030 agenda promoted by the United Nations and has increased its non-tolerance policy about messages that encourage hate or inequality. In the First Barometer on the Perception of the Contribution of TV to Sustainable Development Goals (Deloitte, 2020), 58% of the surveyed population considers that “in free-to-air television, there is a greater control over the spread of content that incites violence and hate, than that at the Internet.” That is not a casual perception. In social networks, the user’s speech is uncontrolled, and any person can take part anonymously, without filtering the tone or content, and with the freedom to hold their opinions on any subject. In television, otherwise, there are journalistic controls, and the audiovisual content is selected within a schedule and a programming grid that can restrict anonymous or violent interventions. Free-to-air television remains a relevant media for sharing ideas. It reaches the most dispersed and educationally disadvantaged populations and creates a social imagination through a large number of people (Scheufele, 1999). Likewise, many of these viewers are part of “social TV”. So, they are social network users who search for content or share their opinions about what they see on television and online platforms, spreading their statements about hate speech. (Odunaiya et al., 2020). Even so, much of the hate speech forged on social networks is only known among small groups of users until other broadcasters communicate it. As a result, popular opinion is usually generated around the hate speech issue when it 187

 Are There Hate Speeches on Spanish Television?

is discussed on television or in other mass media. For this reason, the social responsibility of television is even much higher (Roy, 2019). That adds to the political polarization that has increased its exposition on television in recent years (Barreda, 2021; Pérez-Escolar & Noguera-Vivo, 2022; Baghel, 2020), which has contributed to the distrust of their audiences, that consider they spread false news and disinformation (Masip et al., 2020). Polarization and politicization are also present in debates and sports (Rojas-Torrijos & Guerrero, 2021; Mauro & Martínez-Corcuera, 2020) or other sensationalist products (De-Casas et al., 2020; Elias, 2020) as talk shows (Sakki & Hakoköngäs, 2022). That is why it is essential to check to what extent hate speech is allowed or communicated on television and what contents and discourse formulas are used to serve this purpose. Only this way will it be possible to combat their effects.

THEORETICAL FRAMEWORK A part of the research on hate speech in recent years focuses on the Internet and social networks. This is entirely because of the previous editorial work of the sender, which means it is easier for television broadcasters to avoid this type of content, resulting in a lower perception of hate speech among viewers (Deloitte, 2020). However, content analysis on a social network or in the online press is more accessible to researchers since numerous databases and tools analyze the content published on them (Tontodimamma et al., 2021; Paz et al., 2020). On the contrary, research on television content is expensive and laborious for the academic community and limits its results since it requires watching many hours of broadcast content. That is a substantial restriction when it comes to managing the available resources. Hence the contribution of this research, which scans 24 hours of television broadcast over 12 months. The research on audiovisual content can be done by watching online databases and television platforms, but not all broadcasters have everything of their content available to the audience or even have digitized their video and audio files. In recent years, initiatives such as Verba Volant (a web containing the News from La 1 de Televisión Española and its corresponding transcripts) have emerged. These databases are just the ones that allow research on television content, as they do on social networks or the Internet. Most of the studies in those fields are from a linguistic and semantic point of view, which means searching the Internet for an issue and some related words to that topic. From there, the researchers analyze the semantic structure around that subject to understand their meanings and if there is a positive or negative point of view. However, some more difficulties concern the analysis of hate speech on television. It relies on a visual symbolic code consisting of elements that suggest the topic not only by the sum of words or its meanings but by conventions such as image, tone, and voice volume, among others. That affects the perception the viewer may have about it. Thus, the process that guides the detection of an issue on television and the measurement of the time spent on the screen is relevant to the fundamental research on the topic. Many academic studies on hate speech on television focus on political debates and election campaigns (Ezeibe, 2021); fictional content (Martínez de Bartolomé & Rivera-Martín, 2022); News (Arévalo et al., 2021; Caldevilla‐Domínguez et al., 2023); how that hate speech is perceived based on the political ideology of the receiver (Abuín-Vences et al., 2022) and the limits of freedom of expression (Pavlides, 2019). They also aim to define what hate speech is (Gelber, 2019) and report politicians and influencers as propagators of it (Gelgel et al., 2023). Other authors’ research highlights the differences between

188

 Are There Hate Speeches on Spanish Television?

cyber hate and hate speech in other media (Castaño-Pulgarín et al., 2021); and reinforces the importance of control and editorial policy in television broadcasts (Brown, 2017). Furthermore, the academic works that identify the elements of hate speech against specific groups are relevant to this research. Special attention is given to racist language and stereotype detection (Kroskrity, 2021; Idevall, 2019). The studies that focus on the knowledge of the discourse against migrants (Arcila et al., 2022; Valdez-Apolo, 2019; Paasch-Colberg et al., 2021), women (Piñeiro-Otero & Martínez-Rolán, 2021) and the LGTBIQ collective (Carratalá & Herrero-Jiménez, 2019; Heim, 2020) are also numerous. Related to the forms and strategies that define hate speech, other research works focus on detecting those linguistic elements that generate prejudices and hostility towards some social groups (Mullen & Leader, 2005; Amores et al., 2021; Istaiteh et al., 2020). Also, once the linguistic elements are defined, the researchers work on automated detection mainly on the Internet and social networks (Pervez Akhter et al., 2021; Pariyani et al., 2021; Arcila et al., 2020). However, the main difficulty in hate speech research is not only detecting offensive words but as mentioned above, acknowledging those elements that affect the viewer’s perception of it. That is why it is essential to recognize and measure the tone of the speech and other metaphorical terms that can be hate speech (Neitsch et al., 2021). For that, most studies use content analysis as a research tool (Valdez-Apolo et al., 2019) to approach both the language and the rhetorical strategies (Paz et al., 2020). Following the agenda-setting theories (Edelstein, 1993; McCombs et al., 2014), the study of hate speech on television is also relevant because of the notable influence of this and other mass media. According to their theory, mass media decide the topics that will become the object of public interest and the significance assigned to each of them. Therefore, this is important for validating policies against hate speech since mass media help draw attention to a newsworthy topic (Colombini et al., 2016). For this study, the researchers consider that the perception of hate speech as a public interest problem depends fundamentally on the intensity of its media coverage (Spies, 2020). Mass media can transform or create a state of mind among viewers (Tewsbury & Scheufele, 2019) by focusing on some aspects of the situation, and others are left out (Goffman, 2006). Following the studies of Navarro and Olmo (2018), the researchers can say that the public visibility of a topic is not enough to endow it with value, but it is nevertheless essential since content that does not appear in the media does not exist for the audiences. With the analysis of the presence (or absence) of hate speech on television, it is necessary to consider how it is contextualized and framed. Despite having evolved since its inception (Ardévol-Abreu et al., 2020; Lazarsfeld et al., 2021; Deuze & McQuail, 2020; Noelle-Neumann, 2013), the framing theory remains valid when applied the hate speech in media studies (Cacciatore et al., 2016; Tewsbury & Scheufele, 2019). Framing analyses how media presents topics of public interest and their interpretation (Binderkrantz, 2017; Crow & Lawlor, 2016), making it relevant to the present study. Also, those frames help address stereotypes about social groups, which helps in political and media speeches to speak about those groups (Sánchez-Junquera et al., 2021b). Based on those and other previous research on hate speech (Matamoros-Fernández & Farkas, 2021; Paz et al., 2020), the authors developed for this work an experimental methodology to measure the presence and relevance of offensive and abusive language on television. That experimental and quantitative method was complemented by qualitative content analysis, as detailed in the following sections.

189

 Are There Hate Speeches on Spanish Television?

QUESTIONS In this context, the main goals of this research are twofold: first, to create and validate a research methodology that helps to detect hate speech on television; second, to identify if there is hate speech in the television broadcast by the five free-to-air Spanish channels throughout the year 2020. Based on this, the authors lead to the following research questions (Q): Q1. Can a clear and reliable research methodology be established to identify hate speech on television? And also: Q2. Is there hate speech spread on Spanish television? Q3. If there is hate speech, are there relevant differences between the public ownership television channel (La 1) and the others in the study? Finally, it would be interesting to know: Q4. Which are the groups or individuals that are the target of this hate speech?

METHOD Measuring hate speech on television is not easy. The number of hours of audiovisual content and the diversity of language can indicate attitudes of hatred towards a minority, or a group can be broad and have lots of shading. Consequently, it is hard to define a stable corpus that facilitates the work of researchers and makes it necessary to create an appropriate research methodology. For this research, the authors present an experimental methodology that helps to study the content broadcasted in the 24-hour programming of the five Spanish free-to-air television channels with the highest share over a year (Barlovento Comunicación, 2021): La1, La Sexta, Cuatro, Antena 3, and Tele 5. The researchers reviewed its profanity or hurtful vocabulary and measured the insults aired. Then they extracted and studied a sample through content analysis to detect if those insults were included, in any way, with expressions of hatred. For this work, the authors developed an experimental research technique, for which they have, thanks to a collaboration with the TV service provider company Cinfo S.L., subtitles of a total of 24h of the five TV channels that are the subject of the study. These subtitles are intended to integrate the population with hearing difficulties and are generated in an automated way in some channels and semi-automated in others (speech-to-text detection plus human review). This means that, for this study, the researchers analyzed 43,200 hours of broadcast content and approximately 363,175,000 words divided into text files with the subtitles of the TV channels (The average is 199,000 words per channel and day). Therefore, they observed both nonfictional and fictional content, which is interesting to know about a complete framework for hate speech on TV. The complex and enormous volume of data generates a significant amount of work for the researchers. Thus, the first measure to detect hate speech is to check the growth in the television medium of those words that are insulting or offensive since they denote hostile attitudes (Hayaty et al., 2020; Jay, 2023) that can be indicative of expressions of hate. This detection system through improper words has been previously developed by researchers such as Lee and Cheng (2020) and used in deep learning tools, although with relatively low precision in the advanced phases. So, for this study, the researchers considered content analysis essential to confirm the presence of hate speech in offensive expressions. To create a rude vocabulary corpus and with the help of the NVivo software, the researchers detected all the words said on the five TV channels over 30 days. Although 2020 was different in media program190

 Are There Hate Speeches on Spanish Television?

ming due to the COVID-19 pandemic, June was randomly chosen to collect that vocabulary. The 8,000 most said words were listed, but the researchers reduced them to 5,000 by eliminating the derivatives (feminine, plural, and verbal conjugations, as well as numbers). Then they chose those that could be offensive, violent, or rude and evaluated their offence degree on a scale of 1 to 5. Despite the prominent presence of offensive language, only insults and other hurtful words were chosen. For example, the word “shit” occupies position 362 that month (2,813 times), but despite its negative connotation, it is eliminated for not being a personal offence. Finally, the selected words for this study were: whore and fag (puta y puto) and their plurals; bastard (cabrón/a), bitch (perra) and their plurals; racist(s); idiot(s); fucktard(s) and silly(s) (imbécil-es). At this point, it is necessary to clarify that the word “puto” (fag) is used in Spanish as male-whore and as the word “fucking” (for example, the sentence “drive the fucking car” could be in Spanish “conduce el puto coche”). The word “ass” (burro) was eliminated from the list after verifying that the mentions of that month mostly referred to the animal (donkey), and “fat” and “fatty” were also eliminated because they were not used as insults. It happens the same with the words “cow” (vaca), “pig” (cerdo/a) and “dog” (perro). However, “bitch” (perra) is kept on the list because the feminine has a different and more offensive connotation than its masculine (dog/perro). That offensive use of the word was detected in some expressions on television during June. “Gay” was also eliminated from the list since the researchers agree it is not considered an aggressive or degrading word in Spain. “Niger/black” (Negro) also appears in position 736 of the most mentioned. However, it is discarded for the study due to its double meaning in Spanish (Negro is used interchangeably for the color and the Black people) and because the authors considered that if there is hate speech towards the black race, it will be accompanied by other insults that could be perceived during the content analysis. Therefore, if hate messages usually contain specific insulting vocabulary, the researchers choose the corpus of offensive words. Then, they validated the analysis methodology. To do this, they reviewed the entire television content of the five TV channels under study for one day to measure the agreement between the offensive words detected using the nVivo software and the units of analysis defined by the researchers manually. The day chosen was June 19 because it was one of the days with the highest number of insults. Therefore, the validation was done with 120 hours of broadcast content, which represented 922,370 words. Previously, an intercoder test verified the reliability of the qualitative analysis method, with a Fleiss kappa value of 0.71, a reasonable figure due to the complex definition of hate speech. The manual review of the television content allowed the identification of 19 units of analysis that had not been detected by simply searching insults. Their duration was 36 minutes and 45 seconds. Therefore, this time adds to the 2 hours 28 minutes 06 seconds detected by searching offensive words. Both reviews show that the coincidence degree between the two methods is 80.06%. Therefore, it is relatively high (Lee & Chen, 2020). So, in response to the question of whether it is possible to establish a methodology with apparent reliability that serves to recognize hate speech on television (Q1), the authors can say that, far from being perfect, the proposed method allows identifying quite precisely expressions that are susceptible to be measured as hate speech. Thus, the researchers explored all the broadcasted content of the day of each month with the highest number of insults. They jointly evaluated, on a scale of 1 to 5 degrees, if there was hate speech in all those 12 days, if those hate messages provoked action against an individual or a group, or if it was only a denigrating expression. Those that exceeded an average evaluation of 4 were considered constitutive

191

 Are There Hate Speeches on Spanish Television?

of hate speech. They also identified if this was for reasons of sex, race, religion, gender, belonging to a minority, a political or cultural group or for social or economic causes. In short, the proposed research methodology combines quantitative techniques for data collection and qualitative analysis, essential for measuring hate speech.

RESULTS They are divided into two subsections to enable the analysis of the results below. The first is dedicated to the frequency and trends of insults aired on the five free-to-air television channels during 2020. The second focuses on the analysis content of those items that could be considered hate speech.

The Overall Frequency of Abusive Words on Television In the first stage of this study, the researchers could draft the trend in the presence of insults aired on Spanish television broadcasts. Although the coronavirus pandemic was not a subject of analysis for this research, its outbreak, and the announcement of confinement in Spain on March 13, 2020, caused relevant changes in the TV programming grids that affected the study. For that reason, there was a notable decrease in the number of insults broadcasted on television from March 2020 (Figure 1). This drop occurred among all broadcasters, although Cuatro showed a more stable trend on the insults aired. Most TV channels increased their news programs, specials and other live broadcasts in March, April, May, and June 2020, which could justify this collapse, as the detected insults were said mostly in fictional or entertainment products, as discussed below. Figure 1. Total offensive words aired by the channel (2020)

192

 Are There Hate Speeches on Spanish Television?

Cuatro was the television network of the study that used the highest number of offensive words throughout the year, with 24.22% of the total detected (Table 1), followed by La Sexta with 21.91%. It can be noticed that Tele 5 is the one that uses them the least (only 13.73% of the total identified in 2020). Table 1. Total insults, by channel (2020) TV channel

Total insults

Percentage of the insults detected

Antena 3

6,446

19.75%

Cuatro

7,907

24.22%

La 1

6,649

20.37%

La Sexta

7,153

21.91%

Tele 5

4,481

13.73%

Total

32,636

100%

Of these results, the case of La 1 stands out, as it has public ownership and therefore has a greater responsibility in enforcing social welfare policies (Túñez-López et al., 2020). For this reason, it could be expected that the use of language on La 1 is taken care of, avoiding, as much as possible, offensive, or hurtful vocabulary and the spread of hate speech. This study highlights that La 1 is not the channel with the least hurtful words, but it is in the middle of the ranking of TV broadcasters. This answers, in part, question Q3 since we cannot clearly say that La 1 stands out for the use of more careful language than its rivals in this study.

Frequency and Categorization of Hate Speech on Public and Private Television Once the researchers reviewed the offensive words for the year 2020, they made the content analysis of the day of each month with the highest number of insults. Thus, the researchers studied the following 12 days: January 28, February 2, March 21, April 19, May 31, June 19, July 25, August 9, September 19, October 17, November 22, and December 5. A total of 151 units of analysis in which there was some approach to hate speech were extracted. Of the 151 units, those unrelated to possible hate crimes were analyzed separately. A total of 78 units of content had negative connotations towards some group, ethnicity, or due to race, sex, gender and physical or religious causes. Therefore, they were susceptible to constituting hate speech. There could be one or more insults in those analysis units, so the total number of offensive words present in those 78 content units was 209 out of the total of 2,121 insults detected in those 12 days. Hence, answering the question about the presence of hate speech on content television (Q2), the researchers can say that 9.8% of the insults analyzed on those 12 days could somehow induce hatred. The other offensive words located through the nVivo program referred to abusive terms with no other connotations were insults to oneself (“I feel stupid”), were not insults (for example, in the case of the word “bitch” it was used offensively only six times); or were used to emphasize an expression (“motherfucking / de puta madre”), (“it’s a fucking mess / es un puto desastre, among others.”). Once the authors identified the units of analysis with the potential of being hate messages, four researchers gave a value ranging from -1 to 1 to each to define if there was an offence. Of the 78 units of

193

 Are There Hate Speeches on Spanish Television?

analysis, 15 were eliminated because the researchers could not reach an agreement. The total broadcast time of the 63 remaining units was 2h 22m 42”, or 142 minutes and 42” out of 86,400 examined (1,440 minutes x 12 days x 5 channels). Therefore, only 0.16% of the studied time was considered offensive or violent speech, which averages 0.03% of each channel’s broadcast. The 63 units of analysis were categorized into five types of offences: racial and ethnic hate; religious causes; gender or sexual orientation; belonging to a social, political, or ideological group; and physical appearance. In addition, messages inciting violent action and those that did not were coded. It should be noticed that only 2 of the 63 were in entertainment programs, while the remaining 61 belonged to fictional content. Table 2. Total insults, by channel (2020) Hate speeches urging violent actions (20)

Hate speeches

Hate speeches not urging violent actions (43)

Racial and ethnic hate

14

12

Religious causes hate

4

3

Gender or sexual orientation hate

0

24

Hate for belonging to a social, political, or ideological group

0

4

Physical appearance hate

2

0

Note. The 15 units of analysis in which there was no consensus among the researchers are not included in this table.

The research reveals that ethnic or racial groups were the ones that received the most hate speeches urging violence (Table 2). Of these, 14 were expressions such as “I’m going to kill you, stinking Arab” or “One day we’re going to kill all the blacks and Jews. And then the world will be okay.” All these expressions were fictional content. The rest of the hate speeches were due to behavior or sexual orientation. There are also seven units of analysis with violent content for religious causes, all against Jews, and 2 for physical appearance reasons. Responding to question Q4, the groups that are the object of hate are Black people (8), Chinese (5), Russians (5), Arabs (4), Mexicans (3) and Somalis (1); Jews were insulted for religious reasons (7); homosexuals (5) and women (19) for their sexual behaviour; Twitter users (2), talk show speakers (1) and communists (1) for their belonging to a political or social group; and finally, fat people for their physical appearance (2). Although data are small to confirm a global trend, the presence of the units considered by the researchers as hatred inciters are slightly increasing as the importance of the coronavirus in television broadcasts decreases towards the end of 2020, as shown in Figure 2. That highlights once again the relevance of the COVID-19 pandemic outbreak that caused the changes in the program scheduling of the TV channels. As shown in Figure 2, when the population came out of confinement and the habitual agenda of the channels recovered, the trend to expose messages that incite hatred increased steadily.

194

 Are There Hate Speeches on Spanish Television?

Figure 2. Trends of messages inciting hate (2020)

By channels, the authors can say that most of the expressions of hate are given on La 1 (24), followed by Cuatro (18), Antena 3 (10), La Sexta (9) and Tele 5 (2). In the content study, the researchers also detected 73 units of analysis with information on hate crimes and mentions of complaints or trials about situations that could be constitutive of crime. The researchers gave these units of analysis a particular review, as they indicate that the broadcasters are sensitive to information related to hate speech and because, in this case, it could be considered a tool for fighting it. Of those information units, 30 were broadcasted on La Sexta, the channel that dedicated more time to talk about hate speech and hate crimes, followed by La 1 (13), Cuatro (11), Tele 5 (11) and Antena 3 (8). Therefore, except for La Sexta and Tele 5, the channels broadcasted more offensive messages than information on hate crimes. This data can help to answer the Q3 question about the differences between public (La 1) and private ownership channels because they indicate that La 1 is not the broadcaster that spends the most time combating hate speech. The total duration of those 73 units of analysis was 4h00m39”, so almost twice the time dedicated to expressions inciting hatred. It is to be noted, about the time of that information, that it is more complex to explain the forms, causes and consequences of a crime, even more, something as complex as hate speech, than to broadcast an insult related to issues of race, ethnicity, religion, or sex in a fictional content. Therefore, it can also explain the long time dedicated to those information units. Additionally, some paradoxes stand out. The public TV channel La 1 is the one that uses insults that could generate hatred the most, despite the assumed fact of its public service usefulness. Also, reviewing the content broadcast on the same channel throughout one day indicates that unique situations can occur. For example, La Sexta on February 8 reported on “Chinophobia” due to the fear of COVID-19 transmission, which was beginning to spread in cities like Madrid, and also on the same day, broadcasted a film in which there were racist insults towards Chinese people.

195

 Are There Hate Speeches on Spanish Television?

CONCLUSION From this research on the presence of hate speech on TV during 2020, the following conclusions highlight: First, the researchers can determine that hate speech on Spanish television is low. Insults and messages that may be offensive or hurtful hesitate and depend on the television programming grid. They are mainly fictional content, which can be modified in exceptional circumstances, such as, for example, those derived from the COVID-19 pandemic. The researchers made content analysis of 3.2% of television programming for a year, and only 9.8% of the insulting messages detected were considered hate speech. And of these, only 25% were competent in inciting violent actions. However, those are specific expressions that are part of a broader discourse marked by stereotypes that are racist and offensive. Those stereotypes are functional and help to economize on fictional audiovisual language and to explain plots efficiently. But there is a problem if they are associated with negative issues that build a polarized and racist message (Sánchez-Junquera et al., 2021a; Mostert, 2019). Moreover, the construction of stereotypes touches on several sensitive topics. The authors cannot consider that there is a hate crime in the messages sent from one fictional character to another. There is no crime, even though the insults can go against individuals or groups: Moors, Chinese, Russians, homosexuals, Jews, etc. Also, those stereotypes can generate negative actions that lead to hatred because they are often associated with race, ethnicity, sex, or gender issues. For example, Moors and Mexicans are frequently presented as drug dealers. The Chinese and Russians as human traffickers. In the analyzed cases, they were insulted by mentioning their races or origins. That is why, although there is no explicit hate speech, there is a construction in the collective imagination in which some races or ethnicities are supposed to be violent or criminal, and therefore, they are “demonized” by others. On television, those messages oppose information that analyses hate speeches and crimes of hate, which indicates the importance that UTECA and the free-to-air televisions give to the goals set for 2030. This effort to avoid offensive expressions shows the two faces of hate on television and the attempt of public institutions to hold back a problem that could grow indiscriminately through the Internet and social networks. Television helps, in this sense, to act as a “firewall”, constructing an educational attitude towards its audience. Although, as mentioned at the beginning of this article, it is often perceived that there is hate speech in politics, sports, or sensationalist talk shows. Although offences of some kind may occur in these programs, the authors did not detect them with the research methodology proposed here. Also, it should be considered that sometimes the appreciation of verbal violence is subjective and may be due to physical or non-verbal vehemence (pitched voices, gesticulations, or disagreement) that the viewer sometimes perceives less consciously than with the verbal language. Finally, the authors can say that despite the positive results of the methodology used for this study, there are ways to generate hatred without using offensive words (Miller, 2021). That is why the researchers understand that this method has its limitations. For example, it is hard to detect hate speech with low levels of offence (there must be at least an insult in the speech to be detected) and does not consider other subtle aspects of language or the context of a particular expression. The error-index derived from the validation of the methodology, for example, is due in part to the lack of that context in language and to the fact that a reduced corpus was used to facilitate the analysis work since only seven insults (and their feminine and plural derivatives) were sought. Being the most used on television, the detection of 196

 Are There Hate Speeches on Spanish Television?

expressions of hatred associated with them was high, but a complete search of offensive words would help find more units of analysis susceptible to incite hatred. Moreover, it is necessary to consider the particularities of each language and culture, which affect the meaning or the implicit hate in some words and the perception of verbal and non-verbal violence. For example, the word “black” has been a problem in this research because of its double meaning in Spanish, as explained before. The researchers had to study the insults to Black people qualitatively, but the word “negro” (black, in Spanish) may have been used offensively on more occasions than the ones detected by the authors. However, this double meaning does not exist in English since the word for the color black differs from the word used to insult Black people. Something similar occurs with the term “puta” (whore or fucking, in English) or with the perception of these words as an insult in Spanish culture. For this reason, it is essential to consider the particularities of the language of each country when studying and defining hate speech. Consequently, and despite the effectiveness of the experimental methodology presented here, the researchers consider that the detection of violent or hate speeches towards a particular social group can have many gradations that are difficult to detect using only verbal language or analyzing decontextualized phrases, which is a handicap to overcome for this and other future research on hate crimes in the media.

FUTURE RESEARCH DIRECTIONS According to the study, the researchers have detected some possible improvements in the proposed method. In addition to the weaknesses already mentioned in the conclusions section, a more exhaustive test of the research process is necessary. The hate speech detection method must be tested in more research works and different media to validate its application and correct other possible mistakes. As an extension of this work, it would be proper to compare the media coverage approach to such content from 2020 to that after January 2021, when fiction content increased on the television programming schedule. It would also be convenient to carry out a more exhaustive qualitative analysis since it is essential to measure the presence of hate speech in visual and sound elements that were not considered for this study. Likewise, as other authors have done before, it could be interesting to define the effects between live television content in which hate speech is being broadcasted and the conversation generated around that issue on social networks simultaneously.

REFERENCES Abuín-Vences, N., Cuesta-Cambra, U., Niño-González, J. I., & Bengochea-González, C. (2022). Hate speech analysis as a function of ideology: Emotional and cognitive effects. Comunicar: revista científica iberoamericana de comunicación y educación, XXX(71), 37-48. doi:10.3916/C71-2022-03 Amores, J., Blanco-Herrero, D., Sánchez-Holgado, P., & Frías-Vázquez, M. (2021). Detectando el odio ideológico en Twitter. Desarrollo y evaluación de un detector de discurso de odio por ideología política en tuits en español. Cuadernos.Info, 49(49), 98–124. doi:10.7764/cdi.49.27817

197

 Are There Hate Speeches on Spanish Television?

Arcila, C., Blanco-Herrero, D., & Valdés, M. B. (2020). Rejection and hate speech on Twitter: Content analysis of tweets about migrants and refugees. Revista Española de Investigaciones Sociológicas, 172, 21–40. doi:10.5477/cis/reis.172.21 Arcila, C., Sánchez Holgado, P., Quintana Moreno, C., Amores, J. J., & Blanco Herrero, D. (2022). Discurso de odio y aceptación social hacia migrantes en Europa: análisis de tuits con geolocalización. Comunicar: revista científica iberoamericana de comunicación y educación, XXX(71), 21-35. doi:10.3916/ C71-2022-02 Ardèvol-Abreu, A., De Zúñiga, H. G., & McCombs, M. E. (2020). Orígenes y desarrollo de la teoría de la agenda setting en Comunicación. Tendencias en España (2014-2019). El Profesional de la Información, 29(4), e290414. doi:10.3145/epi.2020.jul.14 Arévalo-Salinas, A. I., Al-Najjar Trujillo, T., & Aidar-Abib, T. (2021). La cobertura informativa de la inmigración en Televisión Española. El caso del barco Aquarius. Estudios sobre el Mensaje Periodístico, 27(1), 13–25. doi:10.5209/esmp.73744 Baghel, S. S. (2020). Ethics in Journalism: A Forgotten Pillar. MediaSpace: DME Journal of Communication, 1, 92–107. Barreda, M. (2021). Political polarisation in current democracies: Expressions and consequences. Manuel Giménez Abad Notebooks, 21. doi:10.47919/FMGA.CM21.0109 Binderkrantz, A. S. (2017). Interest group representation and framing in the media: A policy area perspective. Journal of Public Affairs, 20(1), e2009. doi:10.1002/pa.2009 Brown, A. (2017). What is so special about online (as compared to offline) hate speech? Ethnicities, 18(3), 297–326. doi:10.1177/1468796817709846 Bustos Martínez, L., De Santiago Ortega, P. P., Martínez Miró, M. A., & Rengifo Hidalgo, M. S. (2019). Hate speeches: An epidemic that spreads on the web. State of the art on racism and xenophobia on social networks. Social Mediations, 18, 25–42. doi:10.5209/meso.64527 Cacciatore, M. A., Scheufele, D. A., & Iyengar, S. (2016). The end of framing as we know it. and the future of media effects. Mass Communication & Society, 19(1), 7–23. doi:10.1080/15205436.2015.1068811 Caldevilla‐Domínguez, D., Barrientos‐Báez, A., & Padilla‐Castillo, G. (2023). Dilemmas Between Freedom of Speech and Hate Speech: Russophobia on Facebook and Instagram in the Spanish Media. Politics and Governance, 11(2), 1–13. doi:10.17645/pag.v11i2.6330 Carratalá, A., & Herrero-Jiménez, B. (2021). La regulación contra el discurso de odio hacia el colectivo LGTBI en los medios: Análisis comparado de diez leyes autonómicas. RAEIC. Revista de la Asociación Española de Investigación de la Comunicación, 6(12), 58–80. doi:10.24137/raeic.6.12.3 Castaño-Pulgarín, S., Suárez-Betancur, N., Tilano Vega, L., & Herrera López, H. (2021). Internet, social media and online hate speech. Systematic review. Aggression and Violent Behavior, 58, 101608. doi:10.1016/j.avb.2021.101608

198

 Are There Hate Speeches on Spanish Television?

Colombini, M., Mayhew, S. H., Hawkins, B., Bista, M., Kumar Joshi, S., Schei, B., & Watts, Ch. (2016). Agenda setting and framing of gender-based violence in Nepal: How it became a health issue. Health Policy and Planning, 31(4), 493–503. doi:10.1093/heapol/czv091 PMID:26412857 Crow, D. A., & Lawlor, A. (2016). Media in the policy process: Using framing and narratives to understand policy influences. The Review of Policy Research, 33(5), 472–491. doi:10.1111/ropr.12187 De-Casas, P., Vizcaíno-Verdú, A., & Aguaded, I. (2020). La televisión sensacionalista y hábitos de consumo en España e Italia. Estudios sobre el Mensaje Periodístico, 26(2), 483–496. doi:10.5209/esmp.67582 Deloitte. (2020). First Barometer on the Perception of the Contribution of TV to Sustainable Development Goals (SDG). Deloitte. https://bit.ly/3wNbZV2 Deuze, M., & McQuail, D. (2020). McQuail’s Media and Mass Communication Theory. Sage Publications. Edelstein, A. S. (1993). Thinking about the criterion variable in agenda setting research. Journal of Communication, 43(2), 85–99. doi:10.1111/j.1460-2466.1993.tb01264.x Elias, C. (2020). Hate speech in Egyptian television talk shows: a qualitative study. [Thesis, The American University in Cairo, School of Global Affairs and Public Policy]. https://fount.aucegypt.edu/etds/813/ Ezeibe, C. (2021). Hate Speech and Election Violence in Nigeria. Journal of Asian and African Studies, 56(4), 919–935. doi:10.1177/0021909620951208 Gelber, K. (2019). Differentiating hate speech: A systemic discrimination approach. Critical Review of International Social and Political Philosophy, 24(4), 393–414. doi:10.1080/13698230.2019.1576006 Gelgel, N. M. R. A., Limilia, P., & Ratri Rahmiadji, L. (2023). Politicians and Celebrities: Propagators of Hate Speech in Indonesia. Proceedings of Indonesia Focus, 1(1), 6. Goffman, E. (2006). Frame Analysis. Los marcos de la experiencia. Siglo XXI Editores. Hayaty, M., Adi, S., & Hartanto, A. (2020). Lexicon-Based Indonesian Local Language Abusive Words Dictionary to Detect Hate Speech in Social Media. Journal of Information Systems Engineering and Business Intelligence, 6(1), 9–17. doi:10.20473/jisebi.6.1.9-17 Heim, J. (2020). Italian LGBTQ representation in transnational television. Journal of Italian Cinema & Media Studies, 8(2), 189–203. doi:10.1386/jicms_00016_1 Idevall, K. (2019). “She has promised never to use the N-word again”: Discourses of racism in a Swedish media debate. Discourse, Context & Media, 31, 1–8. doi:10.1016/j.dcm.2019.100322 Istaiteh, O., Al-Omoush, R., & Tedmori, S. (October 19TH - 22ND, 2020). Racist and sexist hate speech detection: literature review. 2020 International Conference on Intelligent Data Science Technologies and Applications (IDSTA). IEEE. Jay, T. (2023). 86 Hurtful words and hate speech. In G. Schiewer, J. Altarriba, & B. Ng (Eds.), Language and Emotion. An international handbook (pp. 1794–1815). De Gruyter Mouton. Kroskrity, P. (2021). Covert Linguistic Racisms and the (Re-)Production of White Supremacy. Journal of Linguistic Anthropology, 31(2), 180–193. doi:10.1111/jola.12307

199

 Are There Hate Speeches on Spanish Television?

Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (2021). The People’s Choice. How the Voter Makes up his Mind in a Presidential Campaign. Columbia University Press. Lee, P., & Cheng, C. (2020). Profanity and Hate Speech Detection. International Journal of Information and Management Sciences., 31(3), 227–246. doi:10.6186/IJIMS.202009_31(3).0002 Martínez de Bartolomé, I., & Rivera-Martín, B. (2022). Hate speech against the LGTBIQ+ community and its representation in Veneno series. VISUAL REVIEW International Visual Culture Review, 12(1), 1–12. doi:10.37467/revvisual.v9.3718 Masip, P., Suau, J., & Ruiz-Caballero, C. (2020). Percepciones sobre medios de comunicación y desinformación: Ideología y polarización en el sistema mediático español. El Profesional de la Información, 29(5), e290527. doi:10.3145/epi.2020.sep.27 Matamoros-Fernández, A., & Farkas, J. (2021). Racism, Hate Speech, and Social Media: A Systematic Review and Critique. Television & New Media, 22(2), 205–224. doi:10.1177/1527476420982230 Mauro, M., & Martínez-Corcuera, R. (2020). Hypermasculinity and racist discourses in the Spanish media. The spectacularisation of radio sport programming. In J. O’Brien, R. Holden, & X. Ginesta (Eds.), Sport, Globalisation and Identity. New Perspectives on Regions and Nations. Routledge. doi:10.4324/9781003007104-9 McCombs, M. E., Shaw, D. L., & Weaver, D. H. (2014). New directions in agenda-setting theory and research. Mass Communication & Society, 17(6), 781–802. doi:10.1080/15205436.2014.964871 Miller, K. (2021). Hostility Toward the Press: A Synthesis of Terms, Research, and Future Directions in Examining Harassment of Journalists. Digital Journalism (Abingdon, England), (oct), 1–20. doi:10 .1080/21670811.2021.1991824 Mostert, M. (2019). Stereotypes fomenting hate. Perceptions, stigma, and real-world consequences for Africans with disabilities. In M. Sherry, T. Olsen, J. Solstad Vedeler & J. Eriksen, Disability Hate Speech. Routledge. Mullen, B., & Leader, T. (2005). Linguistic factors: antilocutions, ethonyms, ethnophaulisms, and other varieties of hate speech. In J. Dovidio, P. Glick, & L. Rudman (Eds.), On The Nature of Prejudice: Fifty Years After Allport. Wiley Online. doi:10.1002/9780470773963.ch12 Navarro, J. A., & Olmo, A. (2018). Political presence on Spanish Television (La1) in the previous electoral processes [in Spanish]. Studies in Journalism Message, 24, 1473–1488. doi:10.5209/ESMP.62229 Neitsch, J., Niebuhr, O., & Kleene, A. (2021). What if Hate Speech Really Was Speech? Towards Explaining Hate Speech in a Cross-Modal Approach. In S. Wachs, B. Koch-Priewe, & A. Zick (Eds.), Hate Speech - Multidisziplinäre Analysen und Handlungsoptionen. Springer VS. doi:10.1007/978-3-658-31793-5_6 Noelle-Neumann, E. (2013). The Spiral of Silence. Routledge Oxforshire. Odunaiya, O., Agoyi, M., & Osemeahon, O. S. (2020). Social TV Engagement for Increasing and Sustaining Social TV Viewers. Sustainability (Basel), 12(12), 4906. doi:10.3390u12124906

200

 Are There Hate Speeches on Spanish Television?

Paasch-Colberg, S., Strippel, C., Trebbe, C., & Emmer, M. (2021). From Insult to Hate Speech: Mapping Offensive Language in German User Comments on Immigration. Media and Communication, 9(1), 171–180. doi:10.17645/mac.v9i1.3399 Pariyani, B., Shah, K., Shah, M., Vyas, T., & Degadwala, S. (2021). Hate Speech Detection in Twitter using Natural Language Processing. In Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV ’21). Institute of Electrical and Electronics Engineers (IEEE) 10.1109/ICICV50876.2021.9388496 Pavlides, G. (2019). The right of freedom of expression and its limits: Hate speech in Cyprus public television (CyBC). Interview given by the singer Notis Sfakianakis on the CyBC programme ‘Tête-àTête’. The Cyprus Review, 31(2). https://cyprusreview.org/index.php/cr/article/download/655/552/996 Paz, M. A., Montero-Díaz, J., & Moreno-Delgado, A. (2020). Hate Speech: A Systematised Review. SAGE Open, 10(October-December), 1–12. doi:10.1177/2158244020973022 Pérez-Escolar, M., & Noguera-Vivo, J. M. (2022). Hate Speech and Polarisation in Participatory Society. Knowledge Unlatched. Pervez Akhter, M., Jiangbin, Z., Raza Naqvi, I., AbdelMajeed, M., & Zia, T. (2021). Abusive language detection from social media comments using conventional machine learning and deep learning approaches. Multimedia Systems, 28(6), 1925–1940. doi:10.100700530-021-00784-8 Piñeiro-Otero, M. T., & Martínez-Rolán, X. (2021). Step outside and say that: Analysis of hate speech against women on Twitter. El Profesional de la Información, 30(5), e300502. doi:10.3145/epi.2021.sep.02 Rojas-Torrijos, J. L., & Guerrero Caballero, M. (2021). ¿Un periodismo deportivo apolítico?: Estudio de las referencias al ‘procés’ catalán en la prensa deportiva digital en España. Estudios sobre el mensaje meriodístico, 27(3), 947-960. doi:10.5209/esmp.72554 Roy, S. (2019). Role of Nonviolent Communication in our Daily Life. Pragyaan: Journal of Mass Communication, 17(1-2), 24–27. Sakki, I., & Hakoköngäs, E. (2022). A Critical Discursive Psychological Study of Dialogical Constructions of Hate-Speech in Established Media and Online Discussions. In K. Pettersson & E. Nortio (Eds.), The Far-Right Discourse of Multiculturalism in Intergroup Interactions. Palgrave Studies in Discursive Psychology (pp. 85–111). Palgrave Macmillan. doi:10.1007/978-3-030-89066-7_4 Sánchez-Junquera, J., Chulvi, B., Rosso, P., & Ponzetto, S. P. (2021b). How Do You Speak about Immigrants? Taxonomy and StereoImmigrants Dataset for Identifying Stereotypes about Immigrants. Applied Sciences (Basel, Switzerland), 11(8), 3610. doi:10.3390/app11083610 Sánchez-Junquera, J., Rosso, P., Montes y Gómez, M., & Chulvi, B. (2021a). Masking and BERT-based Models for Stereotype Identification. Procesamiento del Lenguaje Natural, 67, 83–94. Scheufele, D. (1999). Framing as a theory of media effects. Journal of Communication, 49(1), 103–122. doi:10.1111/j.1460-2466.1999.tb02784.x Spies, A. (2020). The portrayal of victims of intimate femicide in the South African media. Journal of African Media Studies, 12(1), 41–59. doi:10.1386/jams_00010_1

201

 Are There Hate Speeches on Spanish Television?

Tewsbury, D., & Scheufele, D. (2019). News Framing Theory and Research. In M. B. Oliver, A. Raney, & J. Bryant (Eds.), News Framing Theory and Research (pp. 1–18). Routledge. Tontodimamma, A., Nissi, E., Sarra, A., & Fontanela, A. (2021). Thirty years of research into hate speech: Topics of interest and their evolution. Scientometrics, 126(1), 157–179. doi:10.100711192-020-03737-6 Túñez-López, M., Vaz-Álvarez, M., & Fieiras-Ceide, C. (2020). COVID-19 and public service media; the impact of the pandemic on public television in Europe. El Profesional de la Información, 29(5), e290518. doi:10.3145/epi.2020.sep.18 Valdez-Apolo, M. B., Arcila-Calderón, C., & Jiménez Amores, J. (2019). El discurso del odio hacia migrantes y refugiados a través del tono y los marcos de los mensajes en Twitter. RAEIC. Revista de la Asociación Española de Investigación de la Comunicación, 6(12), 361–384. doi:10.24137/raeic.6.12.2

202

203

Chapter 11

Hate Speech or Hate Shot? Finding Patterns of the AntiMuslim Narratives in Italy Alessandra Vitullo https://orcid.org/0000-0002-9836-0156 Sapienza University of Rome, Italy

ABSTRACT Nowadays, the Muslim community is one of the most discriminated groups in Europe. Anti-Islam hate speeches circulate online and offline especially through the intense use social media, fake news, bots, and click-baiting practices. However even if Muslim discriminatory practices have been gaining more media relevance in the recent years, anti-Muslim stereotypes date back far beyond our times. Using the theoretical frameworks developed by Said and Moscovici this research aims to analyze 31 semi-structured interviews conducted with the volunteers of the Amnesty International’s Hate Speech Task Force, to investigate which are the most persistent anti-Muslim representations in Europe and Italy today. In bringing the theory into practice this work will explore the dynamics occurring on Facebook among users which show polarized and intolerant positions while engaging an Islam-related conversation. This specific case study will allow to show how and why old anti-Islam stereotypes persist almost unchanged from an offline to an online world.

INTRODUCTION ‘Islamophobia’ is generally described as the unfounded hostility towards Islam and its resulting discriminatory practices against Muslim individuals and communities (Runnymede Trust, 1997; Allen, 2010; Poole & Richardson 2006; Bakali, 2016; Beydoun, 2018). In the recent years, several scholars have debated on the use of this term and in particular about how the word ‘phobia’ improperly describes a phenomenon which actually is not based on an irrational and unconscious fear of the subject experiencing it, but rather on a systematic recurrence of discriminatory practices (Richardson, 2009; Bunzl, 2005). In line with this understanding, this Chapter aims to reconstruct both from an historical and a sociological perspective, the narratives that persist nowadays in the anti-Islam representations in western societies. DOI: 10.4018/978-1-6684-8427-2.ch011

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Hate Speech or Hate Shot?

A special focus will be dedicated to the circulation of anti-Muslim sentiments in Europe and especially in Italy through the analysis of the characteristics of the anti-Islam hate speech online. In doing so, the initial paragraphs of this work will rely on two theoretical frameworks. The first is the one developed by Edward Said, in his book Orientalism (1978) which retraces the historical, cultural and political events that over the centuries, contributed to shape the image of “the Arab” sedimented in western cultures. The second one is the conceptual framework formulated by the social psychologist Serge Moscovici (2001), who through his analysis of the social representations, highlighted the roles of the stereotypes in the societies and the reasons why they persist. In bringing these two authors into dialogue, the succeeding paragraphs aim to highlight which negative narratives about the Arab/Islamic culture still pervade western cultures today and how the evolution of ICTs has increased their circulation within the public and political debate. In order to develop this analysis, the last part of this contribution aims to discuss some recent quantitative research which offer an extensive picture of the presence of negative representations of the Muslim community circulating online and offline in Europe, and in particular in Italy, integrating them with an original qualitative research that investigates deeply some patterns of the anti-Islam communication which occurs among users on social media platforms. Through this qualitative analysis, this work aims to address two main research questions: 1.) which are the most persistent anti-Muslim stereotypes recurring in Europe and Italy today and why do they persist; 2) which patterns characterize the online anti-Islam communication in the Italian context; recurring from an offline to an online world almost unchanged along the centuries.

FRAMING THE ANTI-ISLAM HATRED IN EUROPE In 2016, the study carried out by the Islamic Human Rights Commission1, recorded the growth and proliferation of the anti-Islam narratives in all European countries (Islamic Human Right Commission [IHRC], 2019). The research - comparing data on Islamophobia between 2010 and 2014 in the United Kingdom - had noted an increase in physical aggression against Muslims from 13.9% to 17.8% and the incidence of verbal abuse from 39.8% to 66%. A worsening of the Islamophobic environment was also revealed with regard to media content and political speeches, noting that hostile and discriminatory languages against Muslims have become more and more acceptable in many spheres of daily life (Law et al., 2018). A year later, the report of the European Fundamental Rights Agency2, which collected 10.527 interviews to Muslim people living in Europe, revealed that 31% of Muslim job seekers had been discriminated more than once in the last five years, and that only 12% of them had reported this discrimination. In 2018, the European project Counter-Islamophobia Kit (Law et al., 2018) collected data on Islamophobia in 8 European countries (Belgium, Czech Republic, France, Germany, Greece, Hungary, Portugal, United Kingdom), identifying the 10 most recurring themes in the anti-Islam narratives. The data show that Muslims are perceived as: 1) a threat to national security; 2) not integrated into the European culture and way of life; 3) dangerous for the uncontrolled growing of their population; 4) a theocracy, which refers exclusively to religious norms and values while denying democratic and secular values; 5) a threat to Western traditions and identity; 6) retrograde for everything concerning individual freedoms, especially regarding gender equality; 7) ontologically different from Western culture and values, especially as regards the development of science and human rights; 8) possessor of a violent culture, dictated by their religion; 9) out of place in the European context; 10) dangerously homophobic. 204

 Hate Speech or Hate Shot?

As Gattinara and Froio (2019) point out, in recent years the enormous and constant presence of content and news concerning the Muslim community circulating on media and in the political debate has contributed to worsening this situation. Media and political propaganda of conservative and far-right groups have pictured a narrative of the “state of siege” which has altered the perception that European citizens have about the real presence of Muslim in Europe (Gattinara, 2017). As Gattinara (2017) recalls nowadays the rhetoric of the “dangerous Islamic invasion” is fostered by the images and the rhetoric of the “migratory crisis” which Europe is living. The issue of “securization” and the “preservation of Christian values” from the wave of the Muslim migration have repeatedly been the key-issues of the populist and far-right propaganda (Vitullo, 2021). An example of their impact is the data released by IPSOS/Cdec3 that show how the majority of Italians believe that the percentage of immigrants residing in Italy is around 30% of the total of the population, when the real percentage is 7% (IPSOS/Cdec, 2017). Among this perceived 30% the presence of Muslim people is estimated around the 20%, when the real number is 4%. Narratives which have been dramatically reinforced by the images of the terrorist attacks which hit Europe in the last decades, and which contributed to validate the representation of a violent and menacing Islam (Farwell, 2014). Prejudices and stereotypes which are alimented by an isolated minority of Islamic fundamentalist groups but that are consequently extended and generalized to the majority of Muslim people living peacefully in Europe causing the detriment of their quality of life4.

RESEARCH QUESTIONS AND METHODOLOGY This research aims to explore RQ1) which are the most persistent anti-Muslim stereotypes recurring in Europe and Italy today and why do they persist. A special focus will be dedicated also to RQ2) the exploration of the patterns which characterize the online anti-Islam communication in the Italian context. In answering at these research questions, this work will proceed by setting three different levels of analysis: 1. The first level provides a general theoretical introduction which relies on the studies developed both by Said (1978) and Moscovici (2001) to underline and isolate mechanisms and topics which lay at the formulation of stereotypes in general, and of the anti-Islam narratives in particular. 2. The second level of analysis frames the contemporary anti-Islam scenario in Europe. At this stage some of the most extensive European research which contribute with updated information to outline the characteristics of the anti-Muslim narrative in Europe between the online and offline setting of communication will be introduced. 3. Then the last level of investigation integrates the previous theoretical ones by presenting an empirical and original research conducted by the author between 2021 and 2022. During this year, I have been conducting 31 interviews with the volunteers of the Task Force Hate Speech (TFHS) of Amnesty International Italy to find which patterns occur in an online conversation among users when they talk about Islam. Reflecting in detail on which are the expressions, languages ​​and stereotypes which characterize and trigger a hatred conversation with respect to this topic. ◦◦ The interviews

205

 Hate Speech or Hate Shot?

The interviews have been realized thanks to collaboration of Amnesty International Italy. The TFHS5 was constituted by the Italian section of the international NGO in 2017 and it is composed by a hundred of volunteers which have the task to respond to or report the hatred content circulating on the Facebook pages of national and local newspapers chosen randomly from a leftist to a rightist political orientation.6 In particular the Task Force monitors intolerant content which target groups which have been already indicated by previous research7 as some of the most targeted communities by the hatred narratives circulating online (migrants, Muslims, women, LGBTQ+ communities, disabled people, Jews, and Roma).8 The mission of the Task Force is precisely that of decreasing the tones of the hatred conversations or reporting to Meta Platforms those comments which do not respect the community standards.9 The 31 respondents were selected (according to their availability) among the total amount of the volunteers composing the Task Force because they resulted to be the most active online as they registered at least two online counter-narrative activations per week on Facebook10 and for spending an average of four hour per week in monitoring hate speech on the newspapers’ Facebook pages. The selected group is composed of adults aged between 24 and 70 years old. In doing their counter-narrative activities the volunteers did not use an Amnesty International Facebook account, but by using either a personal profile, or a fake profile created ad hoc. In general, they are not immediately recognizable as Amnesty volunteers. As it is understandable the delicate issue of exposing activists’ identity prompted this research to protect them with anonymity. The semi-structured interviews have been conducted between May 2021 and January 2022 on a video call platform as the activists live scattered in different cities from southern to northern Italy. The interviews have been analyzed through a manual hermeneutic approach (Wernet, 2014) developed through the isolation of the thematic nodes relevant for the purposes of this Chapter. Considering the experience that the TFHS has accumulated in observing and monitoring hatred communication online the interviews aimed to collect information regarding the languages and expressions which occur among users which express and promote intolerant and violent communication online. The thematic nodes covered by the interviews have concerned: 1. The prejudices and stereotypes most used to target the above-mentioned discriminated groups; 2. the communicative patterns presented by the hater which engage a hatred conversation with other users (which support or contest her or his position); 3. the epilogue of these conversations (if they produce an escalation or a de-escalation of the hatred communication). Finally, it’s necessary to clarify, that the data presented here should be considered as a partial analysis of a larger dataset that includes information which go beyond the very subject of this paper, precisely because, as indicated before, the TFHS does not have the exclusive task of observing and contrasting anti-Islam stereotypes.

AN HISTORICAL PERSPECTIVE OF ANTI-ISLAM NARRATIVES: THE STUDIES OF SAID AND MOSCOVICI The phenomenon of anti-Islam hatred is gaining relevance in the recent years especially because of the enormous flow of intolerant messages circulating through digital platforms. The Internet and social networks have multiplied the places and forms through which hate speeches circulate. However, the anti-Muslim stereotypes and prejudices present in western societies date back far beyond our times.

206

 Hate Speech or Hate Shot?

They are the result of a stratified cultural baggage inherited from a long tradition of thought that ended up with shaping our common sense about Islam (Moscovici, 2001). In his book Orientalism (1978) Edward Said, was one of the first scholars describing in a chronological and in a cause-effect order the events that contributed to the sedimentation of these hostile representations of the Arab world. Starting with the Crusade and arriving till the Oil crisis of the 1973, Said clearly isolated the historical benchmarks which contribute to shape the image of a violent, irrational, and oppressive Arab world (Donini, 2014; Lewis, 2016).11 The scholar recalls how in the pre-modern world, the eastern borders represented for the West a land of fascination and mystery, but also of terror and violence. In this period artistic and literary production were fulfilled with frightening allegories of a menacing Islamic world portrayed by its most representative characters, such as the bloodthirsty Saladin or Mohammed the “heretic”. 12 The medieval Christian prejudices about Islam were even reinforced during the Renaissance by the emerging scientific production. Said recalls how, in 1697, in the Bibliothéque Orientale of Barthélemy d’Herbelot - which is considered the first systematization of the oriental studies - Mohammed was described as “the famous imposter, Author and Founder of a heresy. which has taken on the name of religion, which we call Mohammedan. See entry under Islam” (Said, 1978:66). From this time onwards the medieval religious narrative which opposed Christianity to Islam (Masuzawa, 2005) gave the way to the historical, artistic, and anthropological contraposition. The religious approach to the study of Islam was replaced by the academic one and precisely this scientific effort corroborates the idea of “the Arab” as the incarnation of a primitive and violent culture. This interpretation will converge in the late XIX century also in social Darwinism (Tort & Chiesura, 2000). As Said points out, in the XX century, the orientalist approach to the study of Islam derived precisely from this pretentious scientific perspective to the study of the Middle East. This supposed western superiority did not belong only to the European governments but also to those scholars who were studying the “Arab culture” by trying to look at the reasons for the “underdevelopment” of these societies. After the Second World War and the Arab-Israeli conflict, France and Great Britain were no longer the only hegemonic powers in Europe. The emergence of US influence on the international scenario changed the European balances. During the Oil crisis in 1973, news from the Middle East filled the pages of the newspaper with negative representations of “the Arab” world. In this way anti-Muslim stereotypes assumed the same connotations in two different parts of the globe. In the films and television, the Arab is associated either with lechery or bloodthirsty dishonesty. He appears as an oversexed degenerate, capable, it is true, of cleverly devious intrigues, but essentially sadistic, treacherous, low. Slave trader, camel driver, moneychanger, colorful scoundrel: these are some traditional Arab roles in the cinema (Cardini, 2015). The Arab leader (of marauders, pirates, “native” insurgents) can often be seen snarling at the captured Western hero and the blond girl (both of them steeped in wholesomeness), “My men are going to kill you, but-they like to amuse themselves before.” He leers suggestively as he speaks: this is a current debasement of Valentino’s Sheik. In newsreels or news photos, the Arab is always shown in large numbers. No individuality, no personal characteristics or experiences. Most of the pictures represent mass rage and misery, or irrational (hence hopelessly eccentric) gestures. Lurking behind all of these images is the menace of jihad (Dante, 1983). Consequence: a fear that the Muslims (or Arabs) will take over the world. (Said 1978:287) In analyzing the anti-Islam narratives which characterize European and Italian popular culture today, it is not surprising to note how after almost forty years Said’s work maintained its validity (Huntington, 1996)13. To fully understand how and why these anti-Islam representations endure the centuries, a short 207

 Hate Speech or Hate Shot?

reference to the work of Serge Moscovici is forcibly needed. Indeed, even if the author never focused his work precisely on the anti-Islam narratives his study about social representations Moscovici (2001) exhaustively describes how the stereotyping process works. According to Moscovici’s definition of social representations, these help to make something unusual or unknown, more familiar. The representations conventionalize people, objects and events, which are unknown to us but with which we are forced to relate (Moscovici, 2001). These categories or representations, however, are not the result of our thinking, they are prescribed to us by a set of previous collective elaborations, which are settled and sedimented in our cultural background. As Moscovici underlines the origin or evolution of these representations is not remembered, it is passively and unconsciously accepted and repeated from society to society, until they become part of the common heritage of the same social group. However, this does not mean that the representations are static, in fact they are modeled according to the changes and needs of the social structures in which they reside. According to Moscovici (2001), the character of social representations is revealed in times of crisis and turmoil, when a group or its identity undergoes to a change and therefore the community is forced to reformulate new categories to give new meaning to these external elements. Today, according to Moscovici (2001), the contemporary agoràs in which these new representations are elaborated are bars, clubs, political circles, etc. However, since neither Said or Moscovici had the chance to observe deeply which is today the impact of digital communication on these social representations, this research aims to put into dialogue these two theoretical frameworks by analyzing the specific case of Muslim discrimination in Europe and in Italy, especially by looking at the hatred communicative dynamics that take place on social media platforms.

FROM THE OFFLINE TO THE ONLINE: ANTI-ISLAM STEREOTYPES AND NARRATIVES IN ITALY In the last decade Internet had proven to be one of the most fertile grounds for the diffusion of antiIslam discourses between Europe and the United States, especially immediately after the 2001 (Larsson, 2007, Campbell & Connelly, 2012). With the use of social media this violent rhetoric has reached its widest echo (Aguilera-Carnerero & Azeez, 2016). With the use of social media this violent rhetoric has reached its widest echo especially through the intense use of fake news, bots, and click-baiting practices (Douglas, 2018; Uyheng et al., 2022). At this regard, the organization Hope not Hate (2017) indicates that the online diffusion of anti-Islam speeches took place on social media precisely through the systematic use of bots. The research conducted by this organization highlights how between March and November 2017, in the US and in UK, because of bots’ implementation, some of the most famous anti-Islam Twitter profiles had increased their followers by 117% compared to the previous year. 14 The report The State of Islamophobia in Europe highlights how the spreading of the anti-Islam propaganda on the web is one of the main reasons for the radicalization of right-wing groups (Bayrakli & Hafez, 2018). At this respect, Ranieri (2016), analyzing the online communication of some far-right movements in Europe15 points out how anti-Islam messages are propagated by these groups by using the “strategy of denial” of their racist standpoints. In their public communication these groups usually claim to act in the interest of their communities (using expressions such as: “I’m not racist, but ...”) and by claiming to respect the democratic institutions (Van Dijk, 1992).

208

 Hate Speech or Hate Shot?

In this context, several pieces of research place Italy among the most intolerant European countries to the presence of the Muslim community 16, with the anti-Islam propaganda spreading both in the offline and online context.17 In 2018 the report by Amnesty International Italy, Barometro dell’odio18, which collected for three weeks posts and interactions published on the Facebook and Twitter accounts of the 1.419 candidates to the ongoing electoral campaign19, showed that hate speeches were constantly conveyed. In 23 days, the monitoring has reported 787 hatred contents, detecting more than one offensive, racist and discriminatory message per hour.20 In general, 91% of the discriminatory messages posted by the political candidates21 have targeted migrants and among those, the 11% of the messages presented anti-Muslim expressions. The analysis of these messages pointed out that the candidates link migration to Islam especially by using the topics of the “Islamization of the country”, the “dangerous” presence of extremist Islamic associations, the Muslim “invasion”, the excessive presence of mosques on the national territory, the incompatibility of Muslims with the values of the Italian constitution, and the anti-democratic nature of Islam. Furthermore, stereotypes about Islam are especially associated with the figure of the Muslim woman, considered as the symbol of the religious oppression, especially for wearing the veil.22 Another extensive data collection of the Italian anti-Islam hatred online is provided by the association VOX – Osservatorio Diritti, which updates on an annual basis the Mappa dell’Intolleranza, a database which gathers hate messages spread on Twitter by Italian accounts. Also in this case, Muslims together with women, LGBTQ+ community, people with disabilities, Jews and migrants, are the most targeted groups by the Italian haters.23 In 2019, for example, among the 30.000 tweets extracted by using keywords and hashtag related to Islam, 22.000 tweets contained messages against Muslim people. When compared to 2018, hatred against Islam increased by the 6.9%.24 In the 2021, Muslims were the second most hated category on Twitter, after women (Sulis & Gheno, 2022). Hate against Muslims reached its hype during 9/11 anniversary and during the return of the Taliban in Afghanistan, in August 2021.25 The project of the Bruno Kessler Foundation (Italy) Hatemeter26 stressed out how in 2018 hate speech on Twitter against Islam was often found in combination with keywords and hashtags which in many cases involved political parties, or their leaders, such as: #Salvini; #SalviniPremier; #iostoconsalvini; #SalviniNonMollare; #League; #housepound; #center-right; #fratelliditalia (Di Nicola et al., 2020).

PATTERNS OF THE ANTI-ISLAM ONLINE COMMUNICATION IN ITALY. A QUALITATIVE INVESTIGATION As seen so far, in the most recent years, several organizations and institutions have been involved in measuring the phenomenon of hate speech online, focusing on the numbers and topics which characterize this alarming phenomenon. However, there’s still a lack of studies which deeply investigate the specific dynamics that occur in an online hatred communication among users. Considering this lack of data, the last part of this contribution aims to integrate the quantitative research presented so far with some qualitative evidence emerged during the interviews conducted with the 31 volunteers of the Amnesty International’s Hate Speech Task Force, which for six years now have been observing intolerant and hatred conversations on Facebook platform. Finally, in accordance with the main topic addressed in this Chapter, the outcomes presented will be focused only on the information emerged especially in relation to the anti-Islam online conversations.

209

 Hate Speech or Hate Shot?

As already highlighted in the previous paragraphs, Amnesty International is one of the most engaged associations in Italy in raising awareness on the dangerousness of hate speech online. Since 2018, the organization annually publishes the Barometro dell’odio, a useful tool for understanding how hatred communication circulates on Facebook in Italy. In the various Barometro‘s editions, migrants, Muslims, women, LGBTQ+ communities, Roma, resulted to be the groups that are most affected by hate speech and for that reason the TFHS was created to respond at these intolerant messages by promoting positive counter-narratives, debunking stereotypes and fake news, and de-escalating the tones of these hatred content. The volunteers are called to take action on the posts and comments posted on the social pages of some Italian national and local newspapers. From the information emerged during the interviews, Islam is presented as an intersectional theme to the online hate speech. Islam is used to reinforce discriminatory stereotypes against groups or people who are already victims of prejudices. In a pre-existing discriminatory condition, being Muslim seems to represent an “aggravating circumstance”. F.: For example, hate speeches against migrants are exacerbated when they are associated with the Muslim faith and therefore with the risk of being potential terrorist. Hate speeches against women get worse if they are wearing a headscarf, as happened for Silvia Romano.27 Seemingly the hate speech against the LGBTQ community becomes even more violent when they stand for other Muslim victims of discrimination. Usually you can read messages like: ‘If you were born gay in their countries you would have ended up badly’”. A: When the news published on Facebook tell successful stories of Muslim women using images of them wearing a veil, these provoke sexist reactions or insulting messages which address their religion. The haters immediately relate the veil to a condition of submission and inferiority of the Muslim women. They do not even consider the possibility that wearing the veil was a free choice of these women. They are only accused of being in favor of a retrograde and conservative religion which would never be in favor of women’s rights or in favor of their and empowerment. The interviews also confirm that the topic of the “migratory crisis” amplifies the anti-Islam narratives online. In fact, migrants are usually associated with the Islamic faith. In these cases, stereotypes and hatred narratives overlap the ethnic belonging with the religious belief (Tibi 2010; Williams, et al., 2020). The majority of the interviews have shown how hate speech against migrants relies on the same stereotypes used against Muslim people. Migrants and/or Muslims are normally addressed by using terms which indicate a homogeneous community, with no reference to individualities or specific characteristics (Faloppa, 2020). Migrants/Muslims are a uniform and indistinct group, which can be easily dehumanize by using violent association with herds of animals or diseases. D.: Every time images of a new migrants’ boat landing on the Italian shores is a good opportunity for haters to describe these people as a bunch of animals or terrorists carrying disease and criminality. We read a lot of comments which spread fake news, always sharing the same link with altered and overestimated numbers and information about the migratory flows. It seems like a systematic and well-organized hate machine gets into action, following the same procedures, every time a new landing happens. M: During the pandemic period the hatred narrative against migrants changes from considering them as a danger because they were all Muslim terrorists to considering them as vehicle for the infections. Sometimes you could read comments of the same user who claimed that the Covid did not really exist and that at the same time, it was brought by migrants who were then accused of being vehicle of the same virus.

210

 Hate Speech or Hate Shot?

Another topic underlined by the interviewees which relate migration to Islam is the “Kalergi plan” conspiracy theory.28 According to the activist the Kalergi, conspiracy occurs in those argumentations which condemn Islam and migration because they are causing the “Islamization of Europe” and the contamination of the “Christian values”. P: The Kalergi conspiracy is the most used refrain by those users who pretend to be more informed and aware about the topic. So when you try to provide fact-checked information they always accuse you of being too naïve to not see what’s going on with these migratory flows coming from Muslim countries. Basically that “they” are intentionally replacing us. The interviews also confirmed something already highlighted by previous research: hatred comments against Islam are most concentrated on the newspaper with a conservative/rightist political orientation. However, the outcomes emerged from the interviews enrich this data with a further observation regarding the stereotypes circulating on national and local newspapers’ Facebook pages. E.: On national newspapers the stereotypes replicate the more general argumentation of the political propaganda; in the local ones the hate against Muslims is manifested through the narration of experiences lived by the citizens: situations of degradation in the city parks, nearby the city stations, or in some specific moments of aggregation due to the celebrations of some Islamic festivities. In these occasions the anti-Islam rhetoric seems to strengthen the users’ feeling of belonging to a threatened group, which immediately invokes the issue of securitization against crime and terrorism – and as we saw during the Covid-19 even against the spread of the infection - to justified discriminatory actions against the Muslims/migrants in Italy. M: On these occasions news published online always tends to indicate the nationality of the person who presumably committed the crime and sometimes this information ends up to be wrong or biased. Newspapers on Facebook rely on the fact that most of their users will not read the full article, first of all because you need to pay a subscription to read them, and secondly users are mostly attracted only by the images and by the titles of these articles. That’s enough for them to get a general idea of what happened and to validate all their prejudices about immigrants and Muslims. The majority of the interviews show that online hate speech mainly consists of short and repetitive mottos that make impossible a conversation between users with polarized positions. When the volunteers tell their experience of responding to these anti-Islam messages they usually got two types of reactions: L: Normally the person who made the hatred comment does not reply at all, or if s/he replies they did not provide a funded and analytical argumentation to their opinion. They just try to discredit the interlocutor normally addressing her or him with the insult piddioti.29 During one year of observation only 4 interviewees reported a “successfully story”, indicating with this definition those conversations which ended up with a situation of hate de-escalation: MR: Sometimes who wrote a hateful comment realizes how violent was her or his expression, so eventually they try to renegotiate their position by explaining it differently without using hatred languages. Other times, after our response that provides counter-narratives, they simply decide to delete the hatred comment. Or in some very rare cases, after debunking the fake news published some users thanked us for correcting their false convictions. N: Even if you politely and kindly respond to their comments, they immediately attack you either by insulting you personally or by repeating again what they have already said. Somebody makes fun of you, by discrediting you with irony or just by putting the laughing emoji below your comment. But in general, they usually do not respond to you, they only reply to those comments that endorse or reinforce their intolerant positions. 211

 Hate Speech or Hate Shot?

DISCUSSION AND CONCLUSION In the recent years because of the spreading of digital communication, the term hate speech has become familiar even to non-English speakers. In 1997, the Committee of Ministers of the Council of Ministers of Europe defined hate speech as: “the forms of expression which spread, incite, develop or justify racial hatred, xenophobia, anti-Semitism and other forms of hatred based on intolerance and which include intolerance expressed through aggressive nationalism and ethnocentrism, discrimination and hostility against minorities, migrants and people of immigrant origin” (Council of Europe, 2017). 30 About twenty years later31 the definition has been modified by the European Commission against Racism and Intolerance (ECRI), expanding the circumstances in which hate speech can occur: “Hate speech is to be understood (…) [as] any form, of the denigration, hatred or vilification of a person or group of persons, as well as any harassment, insult, negative stereotyping, stigmatization or threat in respect of such a person or group of persons and the justification of all the preceding types of expression, on the ground of ‘race’, color, descent, national or ethnic origin, age, disability, language, religion or belief, sex, gender, gender identity, sexual orientation and other personal characteristics or status” (ECRI, 2015).32 Opening any dictionary of the English language under the entry “speech” it is possible to read that it “is an expression of or the ability to express thoughts and feelings “ or “ is a formal address or discourse delivered to an audience spoken” or even “is intended to teach or explain something” etc. Anyway, the evidence emerged from the interviews brings up a first critical issue connected precisely to this definition, namely the impossibility to find a discursive communication online on the topics that are likely to become controversial or sensitive, such as migration or religious diversity, as happens for the case of Islam in Europe. As the majority of the interviews show, hate speech on Facebook is mainly a constant repetition of mottos that make impossible to establish a real conversation among users. People engaged in a hatred conversation online do not read each other, they only respond to reinforce their polarized positions. In this situation of aberrant decoding or reading (Eco, 1975) the work of the TFHS becomes that of constructing a positive counter-narrative that reaches, informs and raises awareness among the “silent readers”, indicating with this definition those users who read these hatred conversations without reacting. The purpose of the TFHS in this case is hoping that these positive counter-narrative can help the silent readers to expand her/his vision about the topic by encouraging them to make a positive intervention. P. Maybe somebody who is reading the hatred comments feels the exact frustration that I have felt that very moment. So I hope that my response may encourage him or her to stand for my position by answering to the hater or maybe by just putting a “like” or a “heart” below my comment. Other times I want to just believe that my response would help the silent reader feel less alone. Basically, the messages against the Islamic community consist in this constant repetition of “slogans that only want to hit and sink their victims” as K. states. In this context - as anticipated by the title of this chapter - the hatred communication against the Muslim community could be better described by using the words “hate shots” rather than “hate speeches”. Certainly, a broader reflection should be opened about how digital devices (especially smartphones) and digital communication (especially on social platforms) affect this lack of argumentative communication. However here, to properly accomplish the very objective of this contribution, we would only briefly refer to how these new communicative settings - based on quick reactions (Keib et al., 2021), echo chambers and filter bubbles (Flaxman et al., 2016; Zuiderveen et al., 2016) - has enormously increased the diffusion of these stereotypes (Castaño-Pulgarín et al., 2021).

212

 Hate Speech or Hate Shot?

Here the linguistic ploy “hate shot” - used to describe this kind of communication - only aims to describe a situation in which it is impossible to find a suitable space for an “ideal speech communication” online especially when the public opinion is called to reflect on more complex issues, such as diversity, solidarity, pluralism, etc. which noticeably can inflame more polarized positions (Habermas, 2008). In this “digital swarm” the anti-Islam narrative persists almost unchanged from what we inherited from an offline past world (Han, 2015). The topics of the “Muslim invasion”, the “Islam as a violent religion”, the Muslim community perceived as a “bunch of terrorists”, retrace the orientalist tradition described by the work of Said, filling today the languages of the public and political debates when provoked by the news about migration. As Moscovici recalled the approaching of these sudden and new events trigger in the social group the perception of themselves as a threatened community (Moscovici, 2001). The mechanism of preservation activated by the “menaced group” awakens old categories and stereotypes which are reformulated in the present and readapted by the social group to create its common sense and identity. It is precisely in this movement of remoteness and nearness to what is alien or stranger to the community of origin that is rooted today the “western conception of the Orient” (Said, 1978; Simmel, 1971).

REFERENCES Aguilera-Carnerero, & C., Azeez, A. H. (2016). ‘Islamonausea, not Islamophobia’: The many faces of cyber hate speech. Journal of Arab & Muslim media research, 9(1), 21–40. https://doi.org/ doi:10.1386/ jammr.9.1.21_1 Allen, C. (2010). Islamophobia. Ashgate Publishing. Amnesty International Italia. (2018). Conta fino a 10. Barometro dell’odio in campagna elettorale. Amnesty International. https://www.amnesty.it/barometro-odio/ Bakali, N. (2016). Islamophobia: Understanding Anti-Muslim Racism through the Lived Experiences of Muslim Youth. Sense Publishers. doi:10.1007/978-94-6300-779-5 Bayrakli, E., & Hafez, F. (2018). The State of Islamophobia in Europe in 2018, in Islamophobia Report. Islamophobia Europe. http://www.islamophobiaeurope.com/wp-content/uploads/2018/07 /EIR_2017.pdf Beydoun, K. A. (2018). American Islamophobia: Understanding the Roots and Rise of Fear. University of California Press. doi:10.1525/9780520970007 Bunzl, M. (2005). Between anti‐Semitism and Islamophobia: Some thoughts on the new Europe. American Ethnologist, 32(4), 499–508. doi:10.1525/ae.2005.32.4.499 Campbell, H., & Connelly, L. (2012). Cyber behavior and religious practices on the Internet. In Z. Yan (Ed.), Encyclopedia of Cyber Behavior (Vol. 1). IGI Global. doi:10.4018/978-1-4666-0315-8.ch037 Cardini, F. (2015). Europa e Islam: storia di un malinteso. Editori Laterza.

213

 Hate Speech or Hate Shot?

Castaño-Pulgarín, S. A., Suárez-Betancur, N., Vega, L. M. T., & López, H. M. H. (2021). Internet, social media and online hate speech. Systematic review. Aggression and Violent Behavior, 58, 101608. doi:10.1016/j.avb.2021.101608 Council of Europe. (2017). Recommendation N. R (97) 20. https://rm.coe.int/1680505d5b Dante, A., (1983). La divina Commedia, Inferno. Angelo Signorelli editore. Di Nicola, A., Andreatta, D., Martini, E., Antonopoulos, G. A., Baratto, G., Bonino, S., & Ferret, J. (2020). HATEMETER: Hate speech tool for monitoring, analysing and tackling Anti-Muslim hatred online. eCrime. Diritti, V. La mappa dell’intolleranza. Anno 4. https://www.fishonlus.it/allegati/mappa_intolleranza4.pdf Donini, P. G. (2014). Il mondo islamico: breve storia dal Cinquecento a oggi. Editori Laterza. Douglas, C. (2018). Religion and Fake News: Faith-Based Alternative Information Ecosystems in the US and Europe. The Review of Faith & International Affairs, 16(1), 61–73. doi:10.1080/15570274.20 18.1433522 Eco, U. (1975). Trattato di semiotica generale. Bompiani. ECRI. (2015). General Policy Recommendation NO. 15 On Combating Hate Speech. ECRI. https:// rm.coe.int/ecri-general-policy-recommendation-no-15-on-combating-hate-speech/16808b5b01 Faloppa, F. (2020). Odio: manuale di resistenza alla violenza delle parole. Utet. Farwell, J. P. (2014). The media strategy of ISIS. Survival, 56(6), 49–55. doi:10.1080/00396338.2014 .985436 Flaxman, S., Goel, S., & Rao, J. M. (2016). Filter bubbles, echo chambers, and online news consumption. Public Opinion Quarterly, 80(1), 298–320. doi:10.1093/poq/nfw006 Gallup (2016), Islamophobia: Understanding Anti-Muslim Sentiment in the West. https://news.gallup. com/poll/157082/islamophobia-understanding-anti-muslim-sentiment-west.aspx Gattinara, P. C. (2017). The “Refugee Crisis” in Italy as a Crisis of Legitimacy. Contemporary Italian Politics, 9(3), 318–331. doi:10.1080/23248823.2017.1388639 Gattinara, P. C., & Froio, C. (2019). Getting “Right” into the News: Grassroots Far-right Mobilization and Media Coverage in Italy and France. Comparative European Politics, 17(5), 738–758. doi:10.105741295018-0123-4 Habermas, J. (2008). Teoria dell’agire comunicativo. Il Mulino. Han, B. (2015). Nello sciame. Visioni del digitale. Nottetempo. Hope not Hate. (2017). Bots, Fake News And The Anti-Muslim Message On Social Media. Hope Not Hate. https://www.hopenothate.org.uk/bots-fake-news-anti-muslim-message-social-media/ Huntington, S. (1996). The Clash of Civilizations. Simon & Schuster.

214

 Hate Speech or Hate Shot?

IPSOS/Cdec. (2017). Stereotipi e pregiudizi degli Italiani: dagli immigrati agli ebrei. CDEC. http://old. cdec.it/public/Stereotipi_Opinioni_sintesi.pdf Islamic Human Right Commission. (2016). Annual report. IHRC. https://www.ihrc.org.uk/wp-content/ uploads/2019/04/Annual-Report-2016-2017-04-FV-FB.pdf Keib, K., Wojdynski, B. W., Espina, C., Malson, J., Jefferson, B., & Lee, Y. I. (2021). Living at the speed of mobile: How users evaluate social media news posts on smartphones. Communication Research, 1–17. Larsson, G. (2007). Cyber-islamophobia? The case of WikiIslam. Contemporary Islam, 1(1), 53–67. doi:10.100711562-007-0002-2 Law, I., Easat-Daas, A., & Sayyid. (2018). Countering Islamophobia through the Development of Best Practice in the Use of Counter-Narratives in EU Member States. https://cik.leeds.ac.uk/ Lewis, B. (2016). Gli arabi nella storia. Editori Laterza. Masuzawa, T. (2005). The invention of world religions: or how European universalism was preserved in the language of pluralism. University of Chicago Press. doi:10.7208/chicago/9780226922621.001.0001 Moscovici, S. (2001). Social representations: Essays in social psychology. Nyu Press. Poole, E., & Richardson, J. (2006). Muslims and the News Media. I.B. Tauris. Ranieri, M. (2016). Populism, Media and Education: Challenging discrimination in contemporary digital societies. Routledge. doi:10.4324/9781315680903 Richardson, R. (2009). Islamophobia or anti-Muslim racism–or what?–concepts and terms revisited. Insted. http://www.insted.co.uk/anti-muslim-racism.pdf Said, E. (1978). Orientalism. Pantheon Book. Simmel, G. (1971). On Individuality and Social Forms. Ed by Janowitz, M., The Heritage of Sociology. University of Chicago Press. Sulis, G., & Gheno, V. (2022). The Debate on Language and Gender in Italy, from the Visibility of Women to Inclusive Language (1980s–2020s). The Italianist, 42(1), 153–183. doi:10.1080/02614340. 2022.2125707 Tibi, B. (2010). Ethnicity of fear? Islamic migration and the ethnicization of Islam in Europe. Studies in Ethnicity and Nationalism, 10(1), 126–157. doi:10.1111/j.1754-9469.2010.01038.x Tort, P., & Chiesura, G. (2000). L’antropologia di Darwin: la laicizzazione del discorso sull’uomo. Manifestolibri. Trust, R. (1997), Islamophobia: A Challenge for Us All. Runny Med Trust. https://www.runnymedetrust. org/publications/islamophobia-a-challenge-for-us-all Uyheng, J., Bellutta, D., & Carley, K. M. (2022). Bots Amplify and Redirect Hate Speech in Online Discourse About Racism During the COVID-19 Pandemic. Social Media + Society, 8(3). doi:10.1177/20563051221104749

215

 Hate Speech or Hate Shot?

Van Dijk, T. A. (1992). Discourse and the denial of racism. Discourse & Society, 3(1), 87–118. doi:10.1177/0957926592003001005 Vitullo, A. (2021). The Online Intersection among Islamophobia, Populism, and Hate Speech: An Italian Perspective. Journal of Religion. Media and Digital Culture, 10(1), 95–114. doi:10.1163/21659214bja10028 Wernet, A. (2014). Hermeneutics and Objective Hermeneutics. Ed by Flick, U., The SAGE Handbook of Qualitative Data Analysis. Los Angeles: SAGE Publications, 235–246. doi:10.4135/9781446282243.n16 Williams, M. L., Burnap, P., Javed, A., Liu, H., & Ozalp, S. (2020). Hate in the machine: Anti-Black and anti-Muslim social media posts as predictors of offline racially and religiously aggravated crime. British Journal of Criminology, 60(1), 93–117. doi:10.1093/bjc/azz064 Zuiderveen, B. F., Trilling, D., Möller, J., Bodó, B., De Vreese, C. H., & Helberger, N. (2016). Should we worry about filter bubbles? Internet Policy Review. Journal on Internet Regulation, 5(1). doi:10.14763/2016.1.401

ADDITIONAL READING Appadurai, A. (1996). Modernity at large: Cultural dimensions of globalization. University of Minnesota Press. Appadurai, A. (2006). Fear of small numbers: An essay on the geography of anger. Duke University Press. Van Dijk, J. A. (2006). The Network Society. Sage Publications. Waldron, J. (2012). The Harm in Hate Speech. Harvard University Press. doi:10.4159/harvard.9780674065086 Said, E. W. (2008). Covering Islam: How the media and the experts determine how we see the rest of the world. Random House.

KEY TERMS AND DEFINITIONS Narratives: In sociology are conceptualizations and understandings through which people construct meaningful significance to the stories that define their and other identity. Counter-narratives: In sociology, are the processes of deconstructing and debunking the themes of dominant narratives by undermining their logic and offering alternative significances. Hate Speech: Any verbal expression which aims to attack a person or a group on the basis of ethnicity, religion, gender, or sexual orientation. Orientalism: The study, interpretation and depiction of aspects in Eastern cultures, usually done by western scholars and intellectuals from a dominant and biased perspective.

216

 Hate Speech or Hate Shot?

Populism: In the political discourse is the creation of a popular hegemonic bloc such as «the people», that strives to appeal to ordinary people who feel that their concerns are disregarded by established elite groups. Shitstorm: The massive spread of insults and negative comments towards people, pages or groups on social media. The characteristics that distinguish it from a simple negative comment concern the vulgarity and ferocity with which criticism and aggressive and offensive language are expressed. Stereotypes: Fixed general images and a set of characteristics that a lot of people passively use or reuse to represent or conceive particular type of person, social groups or things.

ENDNOTES 1



2



3



4



5



6



7



8



9



10



11



12

13



See IHRC. https://www.ihrc.org.uk/wp-content/uploads/2019/04/Annual-Report-2016-2017-04FV-FB.pdf. The data was released during the press conference of FRA: “Muslims in the EU: High Levels of Trust Despite Pervasive Discrimination”, Vienna, September 21, 2017. They were also published in the European Islamophobia Report 2017, which collected data on Islamophobia detected in 33 different European countries which included also non-member countries such as Russia and Norway. See: http://www.islamophobiaeurope.com/wp-content/uploads/2018/07/EIR_2017.pdf. See CDEC, Stereotipi e pregiudizi degli italiani. http://old.cdec.it/public/Stereotipi_Opinioni_sintesi.pdf. See the report of the European Council on Foreign Relations (https://ecfr.eu/special/the_2019_european_election/); or the European Islamophobia Report 2020 (https://islamophobiareport.com/ islamophobiareport.pdf). For more information about the Task Force see: https://www.amnesty.it/entra-in-azione/task-forceattivismo/. Amnesty International defines this kind of volunteer activity as «specialized activism». The volunteers monitor posts and comments published on the Facebook pages of national and local newspapers such as: «La Repubblica», «Il Corriere», «Il Giornale», «La Stampa», «Libero», «Milano Today», «Roma Today», «La Gazzetta di Parma», etc. See Conta fino a 10. Barometro dell’odio, URL:https://www.amnesty.it/barometro-odio/ (Accessed on 03/2023). See Words Matter report. https://eeagrants.org/sites/default/files/resources/KAR-FMOHateFolder_November%2B2014_V03-WEB.pdf. See Meta’s Community’s Standards. https://transparency.fb.com/policies/community-standards/. These activations consist in responding to public comments or posts which present hatred languages. In doing so the TFHS follow precise guidelines which aim to decrease the hatred communication or debunking stereotypes and fake news. For further readings see: Lewis (2016); Cardini (2015); Donini (2014). Among the several examples Said describes the representation of Mohammed in the Divina Commedia of Dante Alighieri. Mohammed appears in the XXVIII Canto, located nearby the very seat of the Devil. Here he is torn in two halves, which allegorical represent the schisma that he caused in the Christendom. Same fate belongs to his cousin Ali. In 1993, an article by Samuel Huntington, The Clash of Civilizations? appeared on Foreign Affairs magazine predicting an inevitable struggle between western and eastern world. The article raised

217

 Hate Speech or Hate Shot?

14



15



16



17



18



19



20

21



22



25 23 24

26 27

28





218

such an intense debate that, in 1996, Huntington decided to dedicate an entire book to his theory, The Clash of Civilizations and the Remaking of World Order. In this book Huntington foresaw a new international geopolitical order that would have come almost fifteen years later, in 2010, due to the indigenization of the Muslim world as a consequence of the effect of colonization. This new global order would have led to a process of re-Islamization of the western societies. Five years later in response to Huntington’s book, Said wrote in the magazine The Nation the article Clash of Ignorance defining Huntington’s thesis a belligerent kind of thought prisoner of historical stereotypes and ideal types. See; Hope not Hate (2017). Bots, Fake News And The Anti-Muslim Message On Social Media. https://www.hopenothate.org.uk/bots-fake-news-anti-muslim-message-social-media/ In seven European countries: Austria, Belgium, Bulgaria, France, Italy, Slovenia, and the United Kingdom See Islamophobia in Europe and Italia. https://confrontiworld.net/2022/07/islamophobia-in-europe-an d-italy/. See Islamophobia Report 2017. http://www.islamophobiaeurope.com/wp-content/uploads/2018/07 /EIR_2017.pdf. See: Amnesty Internationl Italy, Count to 10. Campaign Hate Barometer. https://www.amnesty.it/ barometro-odio/. Precisely from February 8th to March 2nd, 2018. The data are attributed to the 129 political candidates where 77 of them were elected. Precisely the half of this hatred comments came from the cadidates of the following parties: Lega (27%), Fratelli d’Italia (13%), Forza Italia (4%), Casa Pound (3%), and Five Stars Movement (2%). The report also shows that the 6% of discriminatory comments concerned the LGBTI community, 4.8% Roma, and 1.8% gender discrimination. The report also highlights that: 7% of the hatred messages directly incited violence and the 32% of them conveyed fake news and fake data. Regarding the topic «migration»: 10% of the reports concerned « securization» and indicated migration as a «social bomb», capable of leading to «social clash». Muslim women, for example, are one of the most discriminated categories in Italy, see ENAR, European Network Against Racism. https://www.enar-eu.org/. For all the maps, see the website voxdiritti.it. See Mappa dell’intolleranza 4. https://www.fishonlus.it/allegati/mappa_intolleranza4.pdf. See Vox Diritti, Mappa Islamofobia, URL:http://www.voxdiritti.it/wp-content/uploads//2021/11/ A3_Islamofobia2021-01.jpg. See Hatemeter Project. http://hatemeter.eu/. The humanitarian aid worker Silvia Romano was kidnapped in 2018 in Kenya while she was working for an NGO. When she was released in 2020 she returned to Italy wearing a jilbab, after she converted to Islam during the kidnapping. Her conversion immediately aroused an extreme violent reaction online with thousands of users commenting the news on social media. Formulated at the beginning of the twentieth century by the Austrian philosopher from whom it takes its name, the “Kalergi plan” claims that European élites and groups of power are slowly replacing European populations with African and Asian migrants, which present more primitive characteristics and so they are easier to manipulate.

 Hate Speech or Hate Shot?

29



30



31

32



The Italian term «piddioti» derives from the contraction of the acronym of the Italian center-leftist party (Democratic Party - PD) with the word ‘idiot’. This is an offensive way to indicate people who naively promote progressive positions. See Council of Europe, URL: https://rm.coe.int/1680505d5b (Accessed on 08/22). For further information on the debate regarding the definition of hate speech, see UNESCO, Counter hate speech online https://unesdoc.unesco.org/images/0023/002332/233231e.pdf (Accessed on 03/23) See ECRI – General Policy Recommendation no.15, URL: https://rm.coe.int/ecri-general-policyrecommendation-no-15-on-combating-hate-speech/16808b5b01(Accessed on 08/22).

219

220

Chapter 12

The Expression of Hate in Portuguese Digital Media: Ethnic and Racial Discrimination

Inês Casquilho-Martins https://orcid.org/0000-0002-7407-848X Centro de Investigação e Estudos de Sociologia, Iscte-Instituto Universitário de Lisboa, Portugal David Ramalho Alves https://orcid.org/0000-0002-3664-708X Iscte-Instituto Universitário de Lisboa, Portugal Helena Belchior-Rocha https://orcid.org/0000-0002-2295-2753 Centro de Investigação e Estudos de Sociologia, Iscte-Instituto Universitário de Lisboa, Portugal

ABSTRACT This chapter aims to analyze the hate speech discourse in the Portuguese context, be it spread or fed through news media platforms and by their users in Portugal. Considering the need to share evidence and produce theoretical and empirical knowledge about this field of research, this study aims to contribute to this reflection and provide information to the international audience. By analyzing the Portuguese legal framework, statistical data and narratives about hate speech against immigrants and other minority groups in digital media (e.g. online news, Facebook) and traditional media (e.g. television, radio), the reader will become better acquainted with the policies associated with hate speech present through digital media and its detection.

DOI: 10.4018/978-1-6684-8427-2.ch012

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 The Expression of Hate in Portuguese Digital Media

INTRODUCTION The rapid digital transformation we are witnessing has posed challenges to the right to freedom of expression in democratic societies. Simultaneously, other fundamental rights are violated by online hate speech, misinformation and disinformation, which increased during the pandemic (Organisation for Economic Cooperation and Development, 2022). Paisana, et al. (2020) highlight the relevance of news media in democracy, indicating the ability of digital media to promote cultural participation or, in contrast, to cultivate highly polarised environments, where consumed contents are defined according to a selection based on algorithms that do not allow a view of diversity and social reality. Han (2022) distinguishes that while traditional television media seek to combine entertainment, showmanship and information without the aim to promote misinformation and fake news, the same does not apply when it comes to digital networks, as the author states that distortions and staging can be created according to the search for followers or likes. In Portugal, Facebook continues to be the social media most used by the Portuguese, mainly the working age (73%), and WhatsApp is the application most used for communication via instant messaging (69%) (Fialho et al., 2022). Couraceiro et al. (2019) highlight the case of WhatsApp, which has expressively increased use over the last few years, namely for news consumption, and being used by about half of the Portuguese population. For Vala (2021), the expressions of racism are still present in Portuguese society, with the practice of offences legitimised by certain groups in public and media spaces. In this domain, it is relevant to understand the positioning and role of the media, fighting the digital and exclusionary cleavages associated with social and cultural segregation (Silverstone, 2004). Online spaces have been conducive to the perpetuation of information sources globally, allowing an increase in fake news with narratives of dubious character and a decrease in traditional media’s influence (Entidade Reguladora para a Comunicação Social [ERC], 2019, 2021a, 2021b). The publication of user comments on social media websites has merited the attention of some studies given the frequent forms of discriminatory and offensive speech, violence, and sometimes hate speech that threatens fundamental rights (Bliuc et al., 2018; CasquilhoMartins et al., 2022; Daniels, 2013; Fernandes & Teles, 2021; Hughey & Daniels, 2013). Evidence for this has been presented in international reports that demonstrate the influence of digital media and its relationship to hate speech and ethnic and racial discrimination for instance European Commission against Racism and Intolerance (2020), European Union Agency for Fundamental Rights (2021) (2022), OCDE (2022); United Nations (2019). According to the European Union Agency for Fundamental Rights in the recent edition of the Fundamental Rights Report (2022), it is noteworthy that hate crimes and hate speech motivated by racism have persisted across the EU, and there is also a blaming of migrants and ethnic minorities (e.g. Roma, Muslims and Asians) associated with the pandemic (COVID-19). It is also stressed that hate crimes and racist-motivated hate speech incidents continue to persist, worsened by the increasing rate of online hate speech, especially during the pandemic, often conveyed by social media or politicians. Therefore, a political and social priority should be crafting a speech that promotes non-discrimination, tolerance, and respect. The chapter begins with a brief theoretical and conceptual systematisation of the theme to clarify the main concepts and to understand how they are interpreted in the Portuguese context. Thus, we start with a problematisation of racial hate speech in the digital sphere, with particular regard to the field of ethnic and racial discrimination, which is the main reason for complaints associated with hate speech.

221

 The Expression of Hate in Portuguese Digital Media

THE INFLUENCE OF MEDIA: CHALLENGES WITH THE USE AND CONSUMPTION OF DIGITAL MEDIA The digital transition is part of an ongoing globalisation process, allowing broader contact with others, access to different information flows, and producing a new digital way of life. Digital media are means of mass communication that can bring risks to democracy and a free, critical and informed thought because content and messages are manipulated in the function of other interests (Han, 2022). In Portugal, the myth remains that disinformation campaigns have no expression and existing ones have no political goals (Couraceiro et al., 2019). The influence of information media and digital platforms on democratic systems themselves have been demonstrated in several studies linked to the influence of social media and the production of political and disinformation content showing that we are in the presence of a global phenomenon (Innes et al. 2021; Keller et al., 2020; Liberini et al., 2020; Said-Hung & Ocarranza-Prado, 2022). In 2016, the European Commission, together with major IT companies (e.g. Facebook, Microsoft, Twitter, YouTube), presented the Code of Conduct on countering illegal hate speech online (European Commission, n.d.), which is defined by criminal law as racist and xenophobic manifestations and public incitement to violence or hatred against a group of people or its members. Given the necessary balance between collective responsibility and freedom of expression online, other companies (e.g. Instagram, Snapchat, TikTok, Linked) have joined in subscribing to this code of conduct, in which organisations assume an essential role in preventing the rise of online hate speech. However, according to the FRA (2022), online hate speech shows an upward trend that has worsened with the pandemic, notably in social media and political discourse, especially against migrants and ethnic minorities. Several European countries recorded an increase in hate speech in online media (e.g. Belgium, Spain, Albania, Cyprus, Austria, and Ireland), being registered by various public bodies and civil society organisations (FRA, 2022). Access to the Internet has made it possible to set up one’s information channels, which through digital technology, reduces the costs of producing and disseminating information, for example, through YouTube or Twitter (Han, 2022). Simultaneously, it is becoming much easier to access digital content. According to Fialho et al. (2022), 77% of the Portuguese use the Internet to obtain information, and 99.5% use their smartphone to access social media. Han (2022) tells us how networking - especially during confinement - makes digital information technology transform communication into surveillance. Elias & CatalanMatamoros (2020) warn that the pandemic brought evidence of how public trust came to be influenced by social media, namely by communication channels such as WhatsApp, in which misinformation about the pandemic was spread. Even before the context marked by COVID-19, this phenomenon was already observed with an increase in the number of people who prefer to access news on social media, especially on messaging apps like WhatsApp, while the percentage of people who choose to avoid news and who feel tired with the amount and forms of dissemination of news media is growing (Couraceiro et al., 2019). Thus, Crilley and Gillespie (2019) report that the unregulated growth of social media platforms lead to the erosion of independent journalism media and a deterioration of democratic politics. This problem goes far beyond focusing on the responsibility of users who propagate disinformation through social media. A problem here concerns how social media platforms can contribute to a toxic information medium. The important role of the media, especially news media, in pursuing the right to information and in the lives of citizens is thus highlighted, as well as the significance of engaging against the present stereotyping in some media outlets (Gomes, 2011).

222

 The Expression of Hate in Portuguese Digital Media

Zattar (2017) considers that this is not an issue that arose with the Internet and the Web because the participation of multiple actors in the production and use of information also occurred in other contexts, reinforcing the need to combat misinformation. Silverstone (2002) points out that ethical issues must be responsibly assumed, both in the narrative representation and in the vocabulary used, as well as in the visual representation, and that mechanisms to protect audiences are important. Besides this being a problem for the media, it is a problem for citizens and democratic societies; it is necessary to ensure that algorithms associated with digital platforms do not promote the circulation of racist, sexist, homophobic, and other extremist content (Crilley & Gillespie, 2019).

PORTUGUESE LEGAL FRAMEWORK AND POLICIES AGAINST HATE SPEECH The influence of Portugal’s colonising past still marks the Portuguese context, which can be felt through the asymmetric power dynamics and marginalisation against ethnic and minority groups (Lima et al., 2022). As such, Vala (2021) states that in Portugal, the social awareness that racism is incompatible with democratic values is still recent. With the end of the dictatorship period in 1974, that anti-racist thought began to conquer a plural space in Portuguese society. Still, racism today is linked to historical and remote processes but is very much alive which leads to the need for a legal framework in antidiscrimination policies(Vala, 2021). The following point allows us to expand on some key aspects of the legislative system regarding this area by systematising the Portuguese legal framework and policies against hate speech. In Portugal, hate crimes consist of any criminal act, namely against people or property, associated with discriminatory motivations such as racism, xenophobia, religious intolerance, homophobia, transphobia, and prejudice against people with disabilities, among other characteristics. According to the current legislation, the acts that appear as hate speech are only criminalised in specific situations and for the special potential damage they contain. There is a legal requirement that the speech must be disseminated by public means and suitable for its dissemination or propagation. Thus, the Portuguese Criminal Code foresees the penalisation of incitement to hatred, violence, and discrimination. In addition to the measures explained in the Penal Code as defamation (article 180) or as an insult (article 181), it is explicit in article 240 the framework of discrimination and incitement to hatred and violence, as well as the legal punishment for those who commit these acts, considering that freedom of expression cannot violate the right to equality, nor the right to honour of persons or groups of persons because of their race, colour, ethnic or national origin, religion, or other discriminatory grounds. Law No. 93/2017, of August 23, establishes the legal and normative framework for preventing, prohibiting and combating any form of discrimination based on racial and ethnic origin, colour, nationality, ancestry and territory of origin. The Commission monitors the application of this law for Equality and Against Racial Discrimination (Article 6), which among its powers must: collect all information relating to discriminatory practices and the enforcement of the individual penalties; make public, by all means at its disposal, cases of an actual violation of this law; recommend the adoption of legislative, regulatory, and administrative measures it deems appropriate to prevent, prohibit, and combat discrimination based on the factors indicated in Article 1; propose measures to suppress legislative, regulatory, and administrative provisions that are contrary to the principle of equality and non-discrimination; amid other powers (Article 8).

223

 The Expression of Hate in Portuguese Digital Media

Thus, freedom of expression cannot be safeguarded when it violates principles that must be protected, such as the principle of equality in which ‘1. All citizens have the same social dignity and are equal before the law; 2. No one shall be privileged, favoured, prejudiced, deprived of any right or exempted from any duty on account of ancestry, sex, race, language, the territory of origin, religion, political or ideological beliefs, education, economic situation, social condition or sexual orientation (Constitutional Law no. 1/2005, Article 13) protected by article 240. In 2018, the Commission for Equality and Against Racial Discrimination adopted a recommendation on the principle of non-reference to racial and ethnic origin, colour, nationality, ancestry, the territory of origin and documentary status in the news published by the media. It was recommended that media organisations and entities, which disseminate informative content on the Internet, should manage comments that may constitute hate speech, such as the propagation of racist, xenophobic and other content that is offensive to human dignity (Calado, 2018). In 2020, the Assembly of the Republic, through Resolution No. 15/2021, approved a motion calling the government to develop a national anti-racist media campaign extended to schools and universities, public services and security forces. Later, in January 2021, a work group was created to promote recommendations on public policies to combat racism and discrimination. The Resolution of the Council of Ministers No. 101/2021 approved the National Plan to Combat Racism and Discrimination 2021-2025. This plan was organised into four transversal principles and ten lines of intervention, of which one line of intervention dedicated to the media and digital can be highlighted (Table 1). The National Plan to Combat Racism and Discrimination 2021-2025 has aligned itself with the European action plan consisting of an instrument to oppose the segregation and marginalisation of citizens and structural inequalities. Regarding the Media and digital spaces, the National Plan presents eight specific measures, including the development of accessible mechanisms to administer, record and report situations of discrimination and incitement to violence and hate speech online based on the example of other international practices, such as in Spain. In addition, another of the measures includes strengthening support for the production of more knowledge about the phenomena of hate speech propagation and incitement to hatred and violence, including in the virtual space. Table 1. Intervention line: media and digital, national plan to combat racism and discrimination 2021-2025 10.1. To promote and amplify free unconditional access to television and radio services that promote knowledge and appreciation of ethno-racial diversity, the inclusion of communities of African descent in Portugal, the strengthening of the connection between Portugal and the Portuguese-speaking African countries, and a greater diversification and enrichment of our country’s cultural panorama. 10.2. Promote the use of airtime on public radio and television services by nationally representative associations. 10.3. Develop actions with the media to promote greater diversity in programming, content and protagonists, which does not segregate into specific channels or programs. 10.4. Stimulate the promotion of greater inclusion and diversity among journalists, commentators (including sports) and columnists and sources in articulation with unions, the Journalists’ Professional License Committee, training centres and Higher Education Institutions. 10.5. Promote media literacy and the development of accessible mechanisms to manage, record and report situations of discrimination, incitement to violence and hate speech online (based on international practices, e.g. Spain) 10.6. To reinforce support for producing more knowledge about the propagation of hate speech and incitement to hatred and violence, namely in the virtual space. 10.7. Devise a guide of good practices for the media and various types of cultural promoters and promote and monitor compliance with the “non-reference” principle 10.8. Support research on the development of artificial intelligence and algorithm awareness to define responses to the challenges presented by automated decision-making processes in discrimination. Source: Council of Ministers Resolution 101/2021

224

 The Expression of Hate in Portuguese Digital Media

It is relevant to understand that this initiative sought to eradicate stereotypes, hate speech, racial discrimination, xenophobia, and other displays of intolerance expressed or conveyed in public communications, given the growing importance of digital media, whether through traditional media, new media, and digital social platforms.

ETHNIC-RACIAL DISCRIMINATION IN PORTUGAL: COMPLAINTS TO THE COMMISSION FOR EQUALITY AND AGAINST RACIAL DISCRIMINATION According to Law No. 93/2017, the Commission for Equality and Against Racial Discrimination (CICDR - Comissão para a Igualdade e Contra a Discriminação Racial) - has, among its various competencies, the reception of complaints and collection of all information regarding the practice of discriminatory acts, being able to apply the respective sanctions or refer to other authorities (Comissão para a Igualdade e Contra a Discriminação Racial [CICDR], 2017). Data from the official reports of this commission point to an evolutionary trend in the number of complaints about discriminatory practices until 2020 (655), with a 37.7% decrease in 2021 but only 6.4% compared to 2019 (436) (CICDR, 2019,2020). Concerning the most identified characteristic or factor of discrimination in reported complaints, the data for 2021 shows nationality in the first place (39.2%), followed by skin colour (17.2%) and racial and ethnic origin (16.9%) (CICDR, 2021). Of the complaints received in 2021, one in five resulted in an administrative offence proceeding (73), corresponding to 17.9% of the total number of complaints, and 43.9% of the complaints were forwarded to other entities with specific competence in the matter (e.g. Public Prosecutor, Regulatory Authority for the Media). Figure 1 shows the percentage of complaints associated with alleged discriminatory practices in the context of the media or Internet and concerning the areas in which these practices occur. Concerning the area/context where the situations reported to the CICDR occurred, complaints about situations conveyed by the Mass Media or the Internet are the second largest context. Since these are not limited to a particular geographical area, they may affect the entire population with access to these mass media. However, in 2020, the year in which there were long periods of confinement due to the COVID-19 pandemic restrictions, 61.7% of complaints were about discriminatory situations in a digital or audiovisual context. According to the CICDR, these data refer to the area of discrimination/context by the total number of complaints and not by the total number of reported situations, as subsequently, one can verify a high number of complaints about the same situation filed by several complainants.

225

 The Expression of Hate in Portuguese Digital Media

Figure 1. Complaints of alleged discriminatory practices in a media or internet context in Portugal (2017-2021)

Source: CICDR (2017, 2018, 2019, 2020, 2021)

In 2021, of the situations of alleged discrimination, 14.7% were associated with Internet/social media and 6.6% with traditional media. Amongst the main complaints, according to data from CICDR, are comments in the digital space and commonly used expressions of discriminatory, racist or xenophobic nature against people or groups of people (e.g. based on their nationality, skin colour, racial or ethnic origin), with greater severity, inciting hatred or violence. The situations of discrimination which occur in traditional Media are associated with television content (e.g. news programs, debates or interviews), being their actors or specific content the central issue of contestation, and also news or opinion articles published in online newspapers, including the contents published in the commentary sections. Although this situation has steadily increased recently, 2021 saw a reversal. Comparing 2021 with previous years, the CICDR highlights that 2020 was atypical due to the pandemic context (COVID-19), the broad mediatisation of atypical phenomena, and the increased use of digital means (CICDR, 2020; 2021). This situation has grown in recent years, having increased substantially in 2020, surpassing the type of situations usually most reported related to access to goods and services (CICDR, 2020). In contrast to the other contexts of discrimination (e.g. commerce, work, health, education), the Internet/Social Media and Traditional Media cases revealed that the vast majority of situations affected entire communities or social groups - and not individuals - consisting of generalisations or generic considerations revolving around stereotypes based on ethnic and racial discrimination. For example, involving “people of Roma ethnicity”, the most commonly reported complaints are situations in which the alleged discriminatory practices were directed at the Roma community as a whole (70.1%) and occurred mainly on the Internet and social media (38.8%). These discriminatory practices are collective and do not target specific persons, so they are not susceptible to gender-based characterisation (CICDR, 2021).

226

 The Expression of Hate in Portuguese Digital Media

THE ACTION OF THE PORTUGUESE REGULATORY AUTHORITY FOR THE MEDIA IN SITUATIONS OF RACIAL OR ETHNIC DISCRIMINATION In Portugal, the actions of the Regulatory Authority for the Media (ERC - Entidade Reguladora para a Comunicação Social) are subject to some limitations and difficulties, as there is no legislation on its competence regarding the publication of content on broadcasting platforms and social media (ERC, 2019). ERC was created by Law No. 53/2005 and is an independent entity responsible for regulating and supervising all entities that engage in media activities in Portugal. According to Law No. 27/2007 and Law No. 54/2010, ERC regulates the access and exercise of television and radio activities in the national territory, including situations related to all practices that, in abstract, constitute an incitement to racial hatred or motivated by ethnic origin, colour or nationality, conveyed through media outlets. It is emphasised that media content must respect the dignity of the human person and the fundamental rights, freedoms and guarantees; that television program services may not incite racial, religious, political hatred or hatred generated by colour, ethnic origin or nationality, gender, sexual orientation or disability, and must ensure respect for a culture of tolerance, non-discrimination and inclusion and against hate speech. The ERC participates in several initiatives promoted by other national and international institutions regarding its competencies of safeguarding socio-cultural diversity (e.g. Working Group for the Prevention and Combat of Racism and Discrimination, High Commissioner for Migrations, and the Mediterranean Regulators Network) (ERC, 2021a). In 2014, ERC published Directive 2/2014 for the journalistic use of user-generated content because of the possibility of user participation in digital media content and the increasing production of information content from citizens and organisations, formulating guidelines for managing user comments. In 2021, 34 inquiry procedures were registered concerning alleged racial or ethnic discrimination, including 19 complaints/indictments received by the CICDR and subsequently referred to the ERC (CICDR, 2021). Of the entire inquiry procedures (34), all the inquiry procedures mentioned refer to situations that occurred in the Traditional Media involving made statements or information or publicly disseminated, likely to threaten, insult or demean a person or group of persons, identifying the factors of alleged discrimination based on the racial or ethnic origin of alleged discrimination (23) and the territory of origin (11), with most situations referring to groups (31). According to ERC’s annual report (2021a), we can have as an example of these cases four situations associated with a television channel in which references to nationality or ethnic belonging (3) and religion (1) of subjects involved in deviant situations were likely to contribute to the discrimination of certain social groups or communities. Other more frequent deliberations on journalistic content correspond to those that appreciate the display of physical aggression and alleged discriminatory representations of Roma and Brazilian immigrant communities in Portugal for being associated with crimes in the media (ERC, 2021a). Although some studies show the perpetuation of racist and xenophobic discriminatory comments on online media from the perspective of editorial responsibility (Carvalho et al., 2022; Casquilho-Martins et al., 2022). The ERC states that the reports of ethnic discrimination against the Roma community and against comments entered by readers in the electronic editions of media outlets are less frequent (ERC, 2021a). The ERC published 27 deliberations in 2021, 15 of which related to ethno-racial discrimination, and they referred one to the Public Prosecutor’s Office because they considered that racist and xenophobic comments in online editions go beyond the legitimate exercise of freedom of expression, particularly

227

 The Expression of Hate in Portuguese Digital Media

concerning the control of users’ comments in order to prevent the publication of racist and offensive comments to human dignity (ERC, 2021a).

The Case of Readers’ Comments Published in the Online Edition of a Newspaper One case in which ERC referred to the Public Prosecutor, considering the criminal nature of its content, started with a complaint against a newspaper because of readers’ comments published in its online edition (ERC, 2021b). The news text referred to a 17-year-old youth who murdered his uncles in Santiago do Cacém [a Portuguese municipality] and allowed for the possibility of being commented on by users (Nascer do Sol, 2020). Although in the published news, no racist or xenophobic content is assumed, the participation in question refers to comments made by users of the newspaper’s online edition. The open commentary space in the article allowed users to generate a wave of comments, with some revealing racism, xenophobia and hate speech. The ERC stated that the media could not disclaim responsibility for users’ comments on their digital platforms and media pages because the commentary tools are perceived as a service provided by the media, which must moderate them on their websites. In this deliberation, the ERC specifies that some comments are purely ideological and that the tone is of a certain degree of aggressiveness, including racist comments that aim to link the committing of crimes to the ethnicity of the perpetrators, as well as advocating death for the young man who allegedly committed the crime and for people of a specific ethnicity. As mentioned above, nowhere in the news text was the ethnicity of the young man mentioned, but some of the commenting users attributed characteristics of ancestry, ethnicity and nationality to the target through comments considered racist and violent (e.g. about the ethnicity of the alleged killer, extrapolating them to the generality of people of African origin). Table 2. Examples of comments registered by ERC in the communication against the newspaper Sol forwarded by the public prosecutor’s office regarding readers’ comments published in its online edition (Portuguese to English translation) “the monkey will let him go!” [referring to Portuguese Minister of Justice Francisca Van Dunem, of African descent] “These are “Lelos” who are up to no good. People who do not integrate do not know what civility is or how to live in society, and they respect nothing and nobody. There are no laws for them. Pure scum that should be banished from the face of the earth.” [“Lelos” is a pejorative term in Portugal to refer to Roma people]. “Marcelo went to hug the black thugs in the Jamaica neighbourhood in Seixal!!!!” [Referring to far-right MP André Ventura’s comment to Marcelo Rebelo de Sousa, president of the Portuguese republic, in the face-toface presidential election debate] “…” - online user comment […] – content explanation by the authors Source: ERC Deliberation/2021/291 - CONTJOR-NET (published October 7, 2021)

Although the newspaper’s website states that comments containing insults, incitement to hatred or violence, foul or defamatory language, racist, xenophobic, sexist, obscene or homophobic comments are not allowed, and reserves the right to delete all comments that do not respect the rules and to block users who do not comply with its comment policy, the newspaper demonstrated inaction in active moderation. Thus, the racist and xenophobic comments from users on this issue online (Table 2) go beyond

228

 The Expression of Hate in Portuguese Digital Media

the legitimate exercise of freedom of expression. All comments on the news have been deactivated, and since late 2020, the newspaper (online and print) has changed its name.

The Influence of the Far Right: From Television News Media to Discrimination in Digital Media The participation of leaders and elements of political parties in the Media has led to an extension of the discourse to digital media. In 2020, during the pandemic peak in Portugal, the far-right party “Chega” leader publicly suggested a specific confinement plan for the Roma community. Besides a wave of indignation from several public figures, Ricardo Quaresma, a Portuguese international player of Roma ethnicity, publicly spoke out against the discriminatory speech of the leader of “Chega” through the social network Facebook(Agência Lusa, 2020). This publication on May 5 received approximately 32 thousand reactions, 4 thousand comments and 7.5 thousand shares, including opinions supporting and against the player’s intervention in defence of the Roma community. Considering the analysis of 99 comments to a related news article in the digital platform of the Observador newspaper, we can observe the existence of racist and xenophobic comments from readers against the Roma community, among others, which reinforce discriminatory practices and hate speech (Table 3). Table 3. Examples of comments on the online news item “Public Figures and Associations Repudiate André Ventura’s Statements About the Roma Communities” (Portuguese to English translation) “These pseudo-public figures have in common with Gypsies the fact that they are parasites on society since both live off state subsidies. There are only rights and no duties for them because Gypsies and public figures think they are above the law.” “I agree that Mr Ventura had a xenophobic position. However, he touched on an important point, the nomadic population or those living in precarious housing conditions are highly vulnerable to contagion and need an intervention and contingency plan adapted to their social situation.” “Quaresma’s civic stance should start by saying that the Gypsies cannot have parties or weddings with 200 people, they cannot allow 13/14-year-old teenagers to be forced to get married, they should go to school, they should not deal drugs, they should not steal, they should pay the rents of the council houses that they haven’t paid for more than 30 years, as it happens in Loures, and so on... That should have been Quaresma’s intervention and not encouraging hatred. If the gipsies are Portuguese, they should be governed by the same laws and not have their laws, as happens. They attack PSP, firemen and so on, and the racist is Ventura... No more political correctness...you still haven’t understood this......” [Loures is a Portuguese city in the Lisbon Metropolitan Area with a large Roma community] “Let it be clear: I would never vote for Ventura. Whoever protects the posture of the gipsy community, have they seen the shame of today in Olaias, when they surrounded the police who were going to pull over an animal that fled a traffic stop operation and took refuge in the neighbourhood? If the politicians don’t do anything to give strength and resources to the police, we, the civilians, will have to put order in the hen house. And quickly.” [Olaias is a neighbourhood in Lisbon city] “…” - online user comment […] - content explanation by the authors Source: Agência Lusa (2020) - comments taken from Observador newspaper’s online page

Although media moderators have removed some comments in the other news, defamatory and insulting comments about people from the Roma community and the encouragement of violent actions are still evident. This idea had already been advocated by Fernandes and Teles (2021), who mentions the influence of the extreme far right on digital platforms, with André Ventura, leader of the largest far-right party in Portugal, being the second most influential figure on the channels.

229

 The Expression of Hate in Portuguese Digital Media

Another mediatised case in 2021 involving André Ventura resulted from the offences made by this politician during one of the presidential election debates broadcasted on a Portuguese television channel. The MP in question was sentenced to apologise to a family of residents of Bairro da Jamaica in Seixal [a poor urban area in a Portuguese municipality] for having offended their honour and image when he referred to people standing next to Marcelo Rebelo de Sousa [president of the republic] as “bandits” and saying that the president would be the president of drug dealers. The pronounced sentence, which the far-right party leader tried to appeal to in the supreme court but without success, would force the politician to publicly apologise to the targeted family on the party’s Twitter account. Starting from this case, we analysed the online comments of users of a social media outlet (digital version) to the news of the conviction and appeal request (Santos et al., 2021). From a total of 196 comments that were processed, more than 40 comments were removed from the system and their authors. However, most of the authors of the available comments reveal that there is solidarity with the extreme right-wing leader and popular contestation of the judicial decision, legitimising the offence rendered. In another news story about the same case (Figueiredo, 2021), but now focused on the extreme-right party leader’s response to the court decision, we analysed the 41 comments available on the news platform, in which the act continues to be legitimised. Although there is no explicit direct discrimination in the discourse, it is notable that there are comments that target people, by their racial or ethnic origin, are associated with certain acts or understood in a discriminatory way. We take this opportunity to highlight that one content of the comments presented in Table 2 is associated with this case, which as in other comments observed, confirms that even if the news does not use a racist or xenophobic discourse, there are subjects that arouse hate speech and the media should monitor that.

Hate Speech on Digital Platforms and the Comments of Elements of the Security Forces on Facebook The Associação74 consortium of journalists investigated the social media Facebook, where they analysed 3090 screenshots collected by researchers in which they found hate speech reproduced through comments and sharing posts on content that disrespect the most basic principles of the rule of law. Among the participants, they identified a total of 591 elements of the Portuguese security forces in this group of social media, among which police officers of the Public Security Police (296) and soldiers of the National Republican Guard (295) who called for violence against politicians, minorities, women and alleged criminals, as well as the rape of women, racist, xenophobic, misogynistic and homophobic comments (Teles & Coelho, 2022). As the background image for SIC Notícias (2022), the media utilised the photograph of the actor Bruno Candé, who was brutally murdered on July 25, 2020, for reasons of racial hatred and who, even before this fateful event, had already been a victim of hate speech by the killer that had already threatened to kill the actor several times (Agência Lusa, 2021). Looking at the analysis of the Twitter publication made by SIC Notícias (2022) regarding this report, the users’ comments are of indignation towards a news piece that targets elements of the security forces, considering that the content of this report is false or has been manipulated. Following this tweet, a high number of users also described the media organisation as “trash” or “sewer” (Table 4).

230

 The Expression of Hate in Portuguese Digital Media

Table 4. Examples of comments on Twitter about the report “Quando o ódio veste farda” (“When Hate Wears a Uniform”) broadcasted on SIC’s Grande Reportagem on November 16 and 17, 2022 (Portuguese to English translation) “You are a disgrace that dishonours the memory of all journalists who have given their lives in the name of their commitment to the truth and to the trust of their fellow citizens. SIC never again.” “When the journalist distils hate, false truths, lack of professionalism... and SIC is top notch.” “If Mamadou Ba is a victim, then I’ll be damned!! Shameful this partisan piece is masquerading as investigative journalism.” [Mamadou Ba is the president of SOS Racism and an anti-racism activist] “Oh SIC, when they bring Mamadou Ba, the biggest racist I know to comment on this “thing”, no more explanations needed. Surely it was not commissioned, yeah right...a consortium of “journalists”.😂” “Apart from the usual crooks, I have never seen anyone suffering at the hands of the police. However, the police suffer every day.” “The title should be “When the left-wing and extreme left-wing journalists want once again to wage a smear campaign against CHEGA by using policemen to do it” …That should be the title of this report.” [CHEGA is the extreme right-wing party in Portugal and the third political force in Parliament] “Why does this guy’s picture appear here? It’s not like the police killed him!” [referring to the actor Bruno Candé] “…” - online user comment […] - content explanation by the authors Source: SIC Notícias (2022) - comments taken from Twitter

Recently, on March 14, 2023, one author of this investigation posted the following message on Twitter announcing that ERC archived a complaint against the reports broadcast in November 2022 entitled “Quando o Ódio Veste Farda” adding that the same entity praised the rigour and public interest of the story (Coelho, 2023). Following the growing concern with the harmful use of media and digital platforms, in 2021, the Inspectorate-General of Home Affairs (IGAI) published the Plan for the Prevention of Manifestations of Discrimination in the Security Forces and Services (PPMD), dedicating a specific point to the area of interaction of members of the Security Forces with other citizens and with other members of the Security Forces and Services, including on social media. One aim of this plan is to define good practices in using social media by members of the security forces related to the prevention of manifestations of discrimination, stating that there is a growing concern with the interaction of members of the security forces in social media (Inspeção-Geral da Administração Interna, 2021). This leads to guidelines for the establishment of objective limits concerning the interventions of these elements in social media, namely in what concerns comments about specific cases or police interventions; hostile statements that incite hatred or discrimination against people or groups; offensive comments, namely against people, groups, religions, political beliefs, parties or other organisations, including offensive comments, namely based on gender, sexual orientation, ethno-racial origin, nationality/country of origin; comments about the corporation to which they belong or about other police security forces that affect their image; comments that affect the decorum and courtesy that Security Forces elements should always observe; statements against the democratic rule of law (IGAI, 2021, pp. 33-34).

231

 The Expression of Hate in Portuguese Digital Media

FUTURE RESEARCH DIRECTIONS This study highlights the need for further research and understanding of hate speech in online media. While the results demonstrate that significant progress has been made in identifying and addressing this issue, there remains a considerable gap in comprehensively analysing the various forms and contexts of hate speech. It is important to understand the patterns, contexts, and effects of hate speech on online platforms, including its relation to societal inequalities, political discourse and the amplification of harmful ideologies. Analyzing online news, user comments, and social media platforms allows for a comprehensive assessment of the presence and impact of hate speech in different online spaces. Thus, continued research and development of effective strategies to combat online hate speech has become urgent. Future research into hate speech on online platforms should utilise a combination of quantitative data and qualitative content analysis, including computer-based techniques. The use of automated tools and machine learning algorithms can provide valuable insights into the frequency and patterns of hate speech in online platforms. However, content analysis is also crucial to identify and interpret the context and meaning of the discourse and narratives associated with hate speech, including the cultural and social factors that contribute to its persistence. This approach can provide a nuanced understanding of the complexity of hate speech in digital media, leading to the development of effective countermeasures and policies for online interactions, and to the development of diversified strategies to combat online hate speech, thus ensuring the protection of individuals’ fundamental rights and promoting inclusive, respectful online communication.

DISCUSSION AND CONCLUSION Social and digital media have influenced people’s ways of acting and thinking, leading to profound behavioural changes in communication. In Portugal, these media spaces are currently one of the main contexts where reports of discriminatory practices are registered. According to data from the Commission for Equality and Against Racial Discrimination, the number of complaints that have, as a source, the Internet and social media are twice the amount when compared to the number of complaints derived from traditional media (CICDR, 2019; 2020; 2021). In this scope, some situations can be understood as discrimination conveyed online, especially on webpages, online publications, and comments on social network commentary spaces. Hence, it is necessary to use qualitative content analysis methods, which, more than accounting for the incidence of hate speech, allow us to understand and interpret the messages disseminated through digital media and the reaction of users. As we can understand through the cases presented, traditional and digital media have supported populist messages, namely in the political sphere, which leads Prior (2021) to state that mainstream journalism maintains a closer relationship with the political establishment, while the tabloid press relies more on the mass public and their standard views close to the populist discourse. Another issue is that violent and criminal events seem to have conquered greater presence in the press, ending up several times giving exacerbated attention to an event by market or popularity criteria, changing the very criteria of newsworthiness (Rodrigues, 2010). In a massively digitised world in which digital media assumes an undisputed role in a global society, new media and digital social platforms are seen as potential tools for the wide dissemination of information, to which we must be alert. Based on an analysis of digital news regarding hate crimes, racism or 232

 The Expression of Hate in Portuguese Digital Media

ethnic discrimination in Portugal, there is a seeming lack of content moderation on digital platforms by the entities responsible for publishing digital content, which does not adopt internal control and filtering procedures to prevent the permanence of hate speech in their digital spaces (e.g. online commentaries of an offensive, hateful and inciting violent content). When this type of speech is verified, there seems to be a failure in their swift removal from the platforms. Because of the data collected and analysed, it is considered that it is still necessary to promote better literacy practices regarding the forms of hate speech in the context of online platforms and the development of mechanisms to combat its dissemination. These issues become even more relevant with the increase in the use of digital content, for which the policies implemented are not yet effective enough. Thus, it is vital that the countries of the European Union, and in particular Portugal, increase the incentives for denunciations, given that these mechanisms have had little reach in the universe of hate speech that has been propagated in digital media. We believe efforts should be pursued in the framework of the National Plan to Combat Racism and Discrimination 2021-2025, which includes prevention actions and reporting crimes configured by hate speech to support victims effectively. These actions should be articulated through joint social action, either through awareness-raising and promotion of socio-educational and informative content or through the ability to train and strengthen the action of the competent entities in the regulation field with citizens and governmental and civil and civil society organisations. Finally, Portugal should consider structures that facilitate the correction of gaps in current procedures concerning how hate speech and misinformation are still concealed and devalued in Portugal, especially in digital media and on the Internet.

REFERENCES Agência Lusa. (2020). Figuras públicas e associações repudiam afirmações de André Ventura sobre ciganos. Observador. https://observador.pt/2020/05/06/figuras-publicas-e-associac oes-repudiam-afirmacoes-de-andre-ventura-sobre-ciganos/ Agência Lusa. (2021). Homem Acusado de Matar ator Bruno Candé Condenado a 22 Anos e Nove Meses de Prisão. Observador. https://observador.pt/2021/06/28/homem-acusado-de-matar-ator -bruno-cande-condenado-a-22-anos-e-nove-meses-de-prisao/ Bliuc, A.-M., Faulkner, N., Jakubowicz, A., & McGarty, C. (2018). Online networks of racial hate: A systematic review of 10 years of research on cyber-racism. Computers in Human Behavior, 87, 75–86. doi:10.1016/j.chb.2018.05.026 Calado, P. (2018). Recomendação à adesão ao Princípio de não-referência da origem racial e étnica, cor, nacionalidade, ascendência, território de origem e situação documental. https://www.cicdr.pt/ documents/57891/110180/Recomenda%C3%A7%C3%A3o+CICDR+-+Princ%C3%ADpio+da+n%C 3%A3o+refer%C3%AAncia.pdf/0cca99a1-9b0d-400b-af4c-652d5816fc6f Carvalho, P., Matos, B., Santos, R., Batista, F., & Ribeiro, R. (2022). Hate Speech Dynamics Against African descent, Roma and LGBTQ+ Communities in Portugal. Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), 2362–2370.

233

 The Expression of Hate in Portuguese Digital Media

Coelho, P. (2023, Março 14). A ERC arquivou uma queixa contra as grandes reportagens “Quando o Ódio Veste Farda” e elogiou o rigor e o interesse público da matéria. Desta vez, finalmente, sinto-me tentado a concordar com a ERC. Twitter. https://twitter.com/PedroMCoelho/status/1635743946868400129? cxt=HHwWgoCxiYCdqrMtAAAA Comissão para a Igualdade e Contra a Discriminação Racial. (2017). Relatório Anual 2017—Igualdade e não Discriminação em Razão da Origem Racial e Étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2018). Relatório Anual 2018—Igualdade e não Discriminação em Razão da Origem Racial e Étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2019). Relatório Anual 2019—Igualdade e não Discriminação em Razão da Origem Racial e étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2020). Relatório Anual 2020—Igualdade e não Discriminação em Razão da Origem Racial e étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2021). Relatório Anual 2021—Igualdade e não Discriminação em Razão da Origem Racial e étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Couraceiro, P., Paisana, M., Pais, P. C., Cardoso, G., & Baldi, V. (2019). Fake News em ano eleitoral. Portugal em linha com a UE. Publicações OberCom. Crilley, R., & Gillespie, M. (2019). What to do about social media? Politics, populism and journalism. Journalism, 20(1), 173–176. doi:10.1177/1464884918807344 Daniels, J. (2013). Race and racism in Internet Studies: A review and critique. New Media & Society, 15(5), 695–719. doi:10.1177/1461444812462849 Elías, C., & Catalan-Matamoros, D. (2020). Coronavirus in Spain: Fear of ‘Official’ Fake News Boosts WhatsApp and Alternative Sources. Media and Communication, 8(2), 462–466. doi:10.17645/mac. v8i2.3217 Entidade Reguladora para a Comunicação Social. (2019). Desinformação – Contexto Europeu e Nacional. https://www.parlamento.pt/Documents/2019/abril/desinformacao_contextoeuroeunacional-ERCabril2019.pdf Entidade Reguladora para a Comunicação Social. (2021a). Relatório da Regulação 2021. https://www. erc.pt/download.php?fd=12855&l=pt&key=bc4ddcc6d69ae573d5f148d6c3659094 Entidade Reguladora para a Comunicação Social. (2021b). ERC/2021/291 (CONTJOR-NET). https://www.erc.pt/download/YToyOntzOjg6ImZpY2hlaXJvIjtzOjM5OiJtZWRpYS9kZWNpc29lcy9vYmplY3RvX29mZmxpbmUvODI5MC5wZGYiO3M6NjoidGl0dWxvIjtzOjM0OiJkZWxpYmVyYWNhby1lcmMyMDIxMjkxLWNvbnRqb3ItbmV0Ijt9/deliberacao-erc2021291-contjor-net

234

 The Expression of Hate in Portuguese Digital Media

European Comission. (n.d.). The EU Code of conduct on countering illegal hate speech online. https:// commission.europa.eu/strategy-and-policy/policies/justice-and-fundamental-rights/combatting-discrimination/racism-and-xenophobia/eu-code-conduct-countering-illegal-hate-speech-online_en European Commission against Racism and Intolerance. (2020). Annual Report 2020 European Commission against Racism and Intolerance. https://rm.coe.int/annual-report-on-ecri-s-activities-for-20 20/1680a1cd59 European Union Agency for Fundamental Rights. (2021). Fundamental Rights Report 2021. https://fra. europa.eu/sites/default/files/fra_uploads/fra-2021-fundamental-rights-report-2021_en.pdf European Union Agency for Fundamental Rights. (2022). Relatório sobre os Direitos Fundamentais 2022 — Pareceres da FRA. https://fra.europa.eu/en/publication/2022/fundamental-rights -report-2022-fra-opinions Fernandes, R. C., & Teles, F. (2021). Country Portugal. In J. Mulhall & S. Khan-Ruf (Eds.), State of Hate: Far-Right Extremism in Europe 2021 (pp. 103-105). Hope not Hate and the Amadeu Antonio Foundation. Fialho, J., Dias, E., & Macedo, V. (2022). O uso das redes sociais e as competências digitais dos portugueses. Universidade Lusíada Editora. Figueiredo, I. A. (2021, October 29). “Peço desculpa porque tribunais me obrigaram.” Chega perde contra família a quem Ventura chamou “bandidos”. Observador. https://observador.pt/2021/10/29/pecodesculpa-porque-tribunais-me-obrigaram-chega-perde-contra-familia-a-quem-ventura-chamou-bandidos/ Gomes, S. (2011). Crime na imprensa: Representações sobre Imigrantes e Ciganos em Portugal. ComTextos, 1, 1-24. https://hdl.handle.net/1822/13674 Han, B.-C. (2022). Infocracia; a digitalização e a crise da democracia. Relógio de Água. Hughey, M. W., & Daniels, J. (2013). Racist comments at online news sites: A methodological dilemma for discourse analysis. Media Culture & Society, 35(3), 332–347. doi:10.1177/0163443712472089 Innes, M., Innes, H., Roberts, C., Harmston, D., & Grinnell, D. (2021). The normalisation and domestication of digital disinformation: On the alignment and consequences of far-right and Russian state (dis) information operations and campaigns in Europe. Journal of Cyber Policy, 6(1), 31–49. doi:10.1080/2 3738871.2021.1937252 Inspeção-Geral da Administração Interna. (2021). Plano de Prevenção de Manifestações de Discriminação nas Forças e Serviços de Segurança (PPMD). IGAI. https://www.igai.pt/pt/Publicacoes/PublicacoesIGAI/ Documents/Plano%20de%20Preven%C3%A7%C3%A3o%20de%20Manifesta%C3%A7%C3%B5es%20 de%20Discrimina%C3%A7%C3%A3o%20nas%20For%C3%A7as%20e%20Servi%C3%A7os%20de%20 Seguran%C3%A7a.pdf Keller, F. B., Schoch, D., Stier, S., & Yang, J. (2020). Political astroturfing on twitter: How to coordinate a disinformation campaign. Political Communication, 37(2), 256–280. doi:10.1080/10584609.2019.1 661888 Liberini, F., Redoano, M., Russo, A., Cuevas, Á., & Cuevas, R. (2020). Politics in the Facebook Era - Evidence from the 2016 US Presidential Elections. CESifo Working Paper, No.8235, Center for Economic Studies and ifo Institute. 235

 The Expression of Hate in Portuguese Digital Media

Lima, R., Borges, M., Salvador, J., & Freitas, M. (2022). Nothing about Us without Us: Social Worker, Harm Reduction and Anti-Racist Struggle. International Journal of Social Work Values and Ethics, 19(2), 101–123. doi:10.55521/10-019-208 Nascer do Sol. (2020, May 2). Jovem que assassinou os tios em Santiago do Cacém fica em prisão preventiva. Nascer do Sol. https://sol.sapo.pt/artigo/695165/jovem-que-assassinou-os-ti os-em-santiago-do-cacem-fica-em-prisao-preventiva Notícias, S. I. C. (2022, November 17). “Quando o ódio veste farda”: As vítimas do ódio de polícias nas redes sociais. Twitter. https://twitter.com/SICNoticias/status/1593371423258836992 Organisation for Economic Cooperation and Development. (2022). The Protection and Promotion of Civic Space: Strengthening Alignment with International Standards and Guidance. OECD Publishing. doi:10.1787/d234e975Paisana, M., Pinto-Martinho, A., & Cardoso, G. (2020). Trust and fake news: Exploratory analysis of the impact of news literacy on the relationship with news content in Portugal. Communicatio Socialis, 33(2), 105–117. Rodrigues, P. (2010). Criminalidade na imprensa: Análise do Correio da Manhã, 2000-2007. Sociologia, Problemas e Práticas, 64, 149-172. https://journals.openedition.org/spp/338#tocto2n2 Said-Hung, E., & Ocarranza-Prado, I. (2022). Actores políticos y promoción de contenidos desinformativos en Twitter. Caso España. Revista Sociedade e Estado, 37(2), 575–598. doi:10.15900102-6992202237020009 Santos, C. D., Correia, G., & Figueiredo, I. A. (2021, Maio 24). André Ventura condenado por ter chamado “bandidos” a moradores do bairro da Jamaica. Líder do Chega vai recorrer. Observador. https://observador.pt/2021/05/24/andre-ventura-condenado-por-ter-chamado-bandidos-a-moradores-dobairro-da-jamaica/ Silverstone, R. (2002). Complicity and collusion in the mediation of everyday life. New Literary History, 33(4), 761–780. doi:10.1353/nlh.2002.0045 Silverstone, R. (2004). Regulation, media literacy and media civics. Media Culture & Society, 26(3), 440–449. doi:10.1177/0163443704042557 Teles, F., & Coelho, P. (2022, November 16). Polícias sem lei: o ódio de 591 agentes de autoridade. Setenta e Quatro. https://setentaequatro.pt/investigacao-74/policias-sem-lei-o-odio-de-591-agentes-deautoridade United Nations. (2019). Promotion and protection of the right to freedom of opinion and expression. United Nations. https://documents-dds-ny.un.org/doc/UNDOC/GEN/N19/308/13/PDF/N1930813. pdf?OpenElement Vala, J. (2021). Racismo, hoje: Portugal em Contexto Europeu. Fundação Francisco Manuel dos Santos. Zattar, M. (2017). Competência em informação e desinformação: Critérios de avaliação do conteúdo das fontes de informação. Liinc Em Revista, 13(2). Advance online publication. doi:10.18617/liinc.v13i2.4075

236

237

Chapter 13

Online Hate Speech and the Representations of Refugees in #VatanimdaMülteci (#RefugeeinMyCountry) Semra Demirdis Cankiri Karatekin University, Turkey

ABSTRACT In Turkey, a dramatic event involving Syrian refugees happened because of allegations that Syrian men had harassed Turkish women. Following the case, Turkish citizens generated a popular hashtag of #VatanimdaMülteci (#RefugeeinMyCountry) to share negative opinions, feelings, and ideologies towards Syrian refugees. This study is an examination of how Twitter was used to produce and spread hate speech discourse directed at refugees and focus on the representations of refugees through the online environment to provide information about anti-refugee rhetoric for specific nations. A quantitative and qualitative content analysis was carried out of the tweets under the hashtag #VatanimdaMülteci. The results demonstrate that a significant number of tweets contained hate speech comments designed to criticise Turkish government policies regarding refugees, such as the Turkish citizenship provided to refugees and their ability to open businesses in Turkey. The study shows that the hospitality of Turkish citizens turned into hostility over time.

INTRODUCTION The term “hate speech” is used for conversations that target specific groups according to their gender, race, sexual orientation, or religion and express a particular hatred toward them. While there is no fully agreed-upon convention on what constitutes the term, several attempts have been made to define it. Hate speech can be defined as “language that is used to express hatred towards a targeted group or is intended to be derogatory, to humiliate, or to insult the members of the group” (Davidson et al., 2017). The number of refugees is increasing at an accelerated pace across the world. According to the United DOI: 10.4018/978-1-6684-8427-2.ch013

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Online Hate Speech and the Representations of Refugees

Nations High Commissioner for Refugees (2021b), approximately 103 million people have been forced to flee their homes around the world as of today. Recently, European media and public discourse have increasingly categorised newcomers by using terminology such as “refugees” and “immigrants” (De Coninck, 2020). A refugee can be defined as a person who fears being persecuted because of their race, nationality, religion, political opinion, or political affiliation with a particular social group (United Nations, 1951). Because of their perilous and dangerous situation, they cross national borders to seek safety in nearby nations. Therefore, they are becoming internationally recognised as “refugees” with assistance from states and organisations such as UNHCR (UNHCR, 2016). Immigrants are people who moved based on their free will for personal comfort without the interference of an externally compelling cause such as a natural disaster or war (UNESCO, 2017). Unlike refugees, immigrants do not face any impediments and can return home safely. If immigrants choose to return to their homes, they will receive their ‘government’s protection (UNHCR, 2016). Extensive research on the media coverage of immigrants has shown that the public discourses around migrants have become progressively simplistic, reductionist and harmful. Balch and Balabanova (2016) examined media coverage in the UK in 2006 and 2013 and found that despite the contributions of foreign nationals to the country’s economy in the pre-research years, the UK news coverage had become more hostile, dismissive, and cynical towards immigrants. The European press started to frame the refugee and migrant crisis as “diversity” and “new arrivals” and categorised it as “vulnerable from outside” and “dangers from outside” in 2015. It has been observed that there is hate speech and a hostile attitude towards refugees and immigrants in the press of Eastern countries of Europe (Georgiou & Zaborowski, 2017). Similar findings have emerged in European countries, including Bulgaria and France, where hostile racial stereotyping and hate speech are indisputable and show how the immigration problem has been widely reported in recent years for example Bosev and Cheresehva (2015) and Marthoz (2017). According to White (2015), the media is framing the migration issue as a “threat” across the globe. Many politicians and parts of the mainstream media have treated immigrants as a never-ending wave of people who can steal other people’s jobs, become a burden to the state, and ultimately threaten the indigenous population (White, 2015). Similar negative representations in the mass media toward the immigrant issue have been found in the African content. For instance, an analysis of Kenyan media coverage by Kisang (2017) revealed that immigrants are often framed as a threat to national security. Kisang (2017) found that immigrants are often represented as threats to national security in mainstream news, as they are associated with stories about terrorist attacks. Kariithi et al., (2017) showed that the South African media use broad and simple terms in general to describe immigrants and immigration. The mass media has played an essential role in building the “refugee image” among the general population in Turkey. Studies examining the Turkish news coverage of refugees have revealed similar results regarding the negative representations of the mass media e.g. Doğanay & Keneş (2016), Göker & Keskin (2015) and Narlı et al., (2019). These studies have indicated that Syrians are mainly described as victims, threatening the nation and burdening the national economy (Filibeli & Ertuna, 2021). The study of hate speech and discriminatory language in the mass media has indicated that both local and national media covered the news that lacked a rights-based understanding in Turkey (Ataman, 2014). Another research on the representation of Syrian refugees in the Turkish press showed that Syrians are either not included in the media or are mostly presented negatively (Erdoğan, 2013, 2014). It is stated that hate speech in the local media increased, possibly due to the increase in hate speech against Syrian refugees in the period covering the first four months of 2017 (Evrensel, 2017). Studies have also shown that the mass media has foregrounded and aroused potential tensions and fears by stigmatising asylum 238

 Online Hate Speech and the Representations of Refugees

seekers and refugees (Eberl et al., 2018). Indeed, the discriminatory language used by the media has also laid the groundwork for crimes. Targeting the homes and workplaces of all Syrians, even in small-scale incidents in different parts of Turkey, has become a common problem (HaberTürk, 2017; Yenigün, 2016). With the acceleration of the refugee crisis, social media platforms such as Facebook, YouTube, and Twitter are being used as platforms on which perceptions about refugees are being stated. On the other hand, users have increasingly used the platform to spread anti-refugee rhetoric for people from specific nations, such as Syrians and Afghans, by targeting them with hate speech and referring to them as “others” (Burnap & Williams, 2016; Van Dijk, 2000). Therefore, scholars have attempted to examine and understand the use of social media in hate speech conversations regarding refugees. Existing research has concentrated on the refugee crisis and anti-refugee debate in Europe as well as how the global Media has portrayed them with political hostility by emphasising negative discourses (Balabanova & Trandafoiu, 2020; Goodman et al., 2017; Krzyżanowski et al., 2018). Turkey is the country that hosts the most refugees, ahead of Colombia, Germany, and Pakistan (UNHCR, 2022). Online hate speech towards refugees in Turkey started when Syrians settled in Turkey in 2018, opened a business, and/or obtained Turkish citizenship (Ozduzen et al., 2021). In Turkey, negative discourses about refugees are increasingly being replicated and redistributed on social media. For example, Turkish citizens generated hashtags such as #SuriyelilerdenBıktıkUlan (#WeAreFuckingFedUpWithSyrians), #ülkemdesuriyeliistemiyorum (# IDon’ IDon’tWantSyriansInMyCountry) and #SuriyelilerDefoluyor (#SyriansAreGettingOut) to express openly racist attitudes towards Syrians following mundane events between October 2018 and October 2019 (Ozduzen et al., 2021). Following allegations that Syrian men sexually harassed Turkish women, Turkish citizens also created several hashtags such as #Suriyeliler (#Syrians), #SuriyelilerDefolsun (#GetOutSyrians), and #VatanimdaMülteci (#RefugeeinMyCountry) (https://trends24.in/turkey/) to demand the deportation of Syrians from the country between April 12th and 14th 2022. The hashtag #VatanimdaMülteci (#RefugeeinMyCountry) was the most popular on Twitter, with users posting many tweets under it. This study focuses on this hashtag to provide data on overall discourses related to hate speech directed at Syrians during this period.

BACKGROUND Defining Hate Speech There is no universally agreed-upon definition of hate speech. For this reason, many different definitions of hate speech are put forward by many researchers in the literature. Ross et al. (2016) argue that a precise definition of hate speech can be helpful in the detection of hate speech by making it easier to explain and thus making definitions more credible. However, the blurring of the line between hate speech and appropriate freedom of expression makes some commentators cautious of precisely defining hate speech (MacAvaney et al., 2019). For example, the American Bar Association does not provide an official definition for hate speech, claiming that if a speech contributes to a criminal act, it can be punished as part of a hate crime (Wermiel, 2018). There are prominent definitions of hate speech in various sources that make it challenging to detect hate speech. Hate speech is generally described as any communication that degrades a target audience based on colour, race, gender, ethnicity, sexual orientation, religion, nationality, or other characteristics (De Gibert et al., 2018). The United Nations (UNHCR, 2023) describes hate speech as “any kind of communication in speech, writing or behaviour, that attacks or uses pejorative 239

 Online Hate Speech and the Representations of Refugees

or discriminatory language with reference to a person or a group on the basis of who they are, in other words, based on their religion, ethnicity, nationality, race, colour, descent, gender or other identity factor” (para. 1). Fortuna and Nunes (2018) also propose a definition for hate speech as follows: Hate speech is language that attacks or diminishes, that incites violence or hate against groups, based on specific characteristics such as physical appearance, religion, descent, national or ethnic origin, sexual orientation, gender identity or other, and it can occur with different linguistic styles, even in subtle forms or when humour is used. (p. 5) In sum, it can be understood that hate speech content is created and spread to insult or humiliate an individual or a group based on such categories as ethnicity, race, religion and many others. However, it must be acknowledged that these definitions are not legal, and hate speech is broader than encouragement to discrimination, violence or hostility prohibited by international human rights (UNHCR, 2023). While hate speech is a broader term, it has three critical characteristics. Firstly, hate speech can be conveyed through any type of expression, including images, symbols, objects, cartoons, memes and gestures (UNHCR, 2023). Secondly, it is discriminatory or derogatory towards a person or group. In particular, hate speech content can be bigoted, biased, intolerant, prejudiced, demeaning, and contemptuous. Finally, hate speech refers to an individual’s or group’s actual or perceived “identity factors”, such as “religion, race, nationality, ethnicity, colour, ancestry, gender”, as well as language, social or economic origin, disability, sexual orientation, or health status, among others (UNHCR, 2023). In order to prevent or minimise the hate speech problem, social media companies have adopted policies that set specific standards and restrict certain behaviours and expressions of users (Ring, 2013). Social media companies such as Twitter and YouTube have published hateful conduct policies to prevent online content that targets harassment or expresses hate towards a specific person or group (Ullmann &Tomalin, 2020). According to ‘Twitter’s policy (2023) on hateful content, any accounts whose primary purpose is to promote violence or to threaten others based on such categories as ethnicity, race, caste, national origin, gender, sexual orientation, age, religious affiliation, and/or disability is not allowed to publish content on Twitter. YouTube (2017) also published its hate speech policy, which highlights that any content promoting violence or hateful content against individuals or groups based on such attributes as age, disability, ethnicity, caste, immigration status, sexual orientation, gender identity and/or veteran status is removed from YouTube.

The Internet, Freedom of Expression, and Hate Speech The internet and new media have become a part of our daily lives with the changes they have brought in communication technologies (Berger, 2013). The effective use of new media in many areas of life, including the financial sector and journalism, has also affected freedom of speech (Castells, 2012; Balkin, 2017; Van der Haak et al., 2012). The Internet provides a resource for learning about politics and a space for expressing political ideas, which can promote civic engagement (Shah et al., 2005). Moreover, mainstream broadcasting organisations have incorporated internet journalism, and news organisations that only produce and share news over the Internet have also emerged. These developments have radically changed the formal structure of journalism. The internet has also brought a new dimension to citizen journalism and offers unlimited possibilities for individuals (Wall, 2015). These developments allow individuals to participate in political and social conversations around a specific topic and thus expand 240

 Online Hate Speech and the Representations of Refugees

the realisation of democratic culture (Balkin, 2017). For example, digital media has played a great role for individuals, wherein they can use online platforms to spread protest discourses and mobilise others to join in social and political protests around the World (Cammaerts, 2015). In other words, in the individual’s socialisation process, new media has become a part of our daily lives as an undeniable fact with the changes in the structure of traditional media (Rajendran & Thesinghraja, 2014). Scholars have focused on the Internet’s use and its significant role in democracy. For example, Papacharissi (2010) argued that individuals in representative democracies participated in civic deliberation within the public sphere, which now participated through the use of the private sphere and private media environments. Jenkins (2006) argued that online digital media enables the individual to participate civically and creatively in a democratic culture comprised of both civic life and popular culture. Papacharissi (2015) also suggested that new media technologies invite citizens to feel their place in current events, emerging stories, news, and various types of civic mobilisation. However, the elements derived from hate speech have shown that the assertions on the democratic role of the Internet need to be questioned. Recent studies have demonstrated that new media technologies promote the spread of hate speech, fake news, and disinformation, which all undermine the democratic role of new media (Guess et al., 2019; Howard et al., 2016; Siegel, 2020). The horizontal nature of social media communications allows users to report the news through their interpersonal networks before official news organisations report it (Chadwick, 2017). Some important news is first broken online, and therefore professional journalists obsessively follow their Twitter, Facebook, and blog feeds to catch up on news stories (Chadwick, 2017). Social media platforms can be used for reporting news by both professional and non-professional journalists, but they can also be used for government censorship and control. As highlighted by Gunitsky (2015), online media is “either ineffective or marginal to the process of government contestation and is easily subject to government censorship and control” (p. 44). According to the Freedom House Report (2016), governments worldwide have shut down access to mobile phone networks or the entire Internet to prevent users from spreading information through social media. New media is neither good nor bad; sometimes, it is employed for reporting news and sharing opinions, while other times, it is used for censorship and repression. However, today’s unlimited possibilities of the internet and the difficulties of controlling it have made it easier for non-professional users to cross some legal limits, such as freedom of expression (ElSherief et al., 2018). In addition, new media has a structure that encourages users to create and share their thoughts and news. One of the ethical problems that has come to the fore with the spread and effective use of new media is the production and spread of hate speech (Jakubowicz, 2017). The source of hate speech produced by traditional media and imposed between the lines is based on society’s history, culture, ideology, customs and traditions (Vardal, 2015). With the spread of new media, users have found a medium left to individual ethical understanding. This situation has caused widespread hate speech, and it has since become difficult to prevent hate speech (Syahputra, 2019).

Hate Speech on Social Media Social media platforms such as Twitter present themselves as democratic places. However, in recent years there has been a growing interest in social media platforms old and new roles in mediating and empowering harassment, hatred and discrimination (Matamoros-Fernández & Farkas, 2021). Jessie Daniels (2013) described social network sites (SNSs) as places where race and racism emerge in interesting and sometimes disturbing ways. Since then, academics have become increasingly concerned with 241

 Online Hate Speech and the Representations of Refugees

online hate speech and racism, especially in countries such as the US, the UK, India, and Brazil, where white supremacists use digital platforms as weapons. It has caused a significant increase in the literature on this topic (Matamoros-Fernández & Farkas, 2021). Therefore, studies have focused on hate speech content on digital platforms and highlighted the differences between online hate speech and offline hate speech (Brown, 2018; Citron, 2014; Delgado & Stefancic, 2014). Researchers have examined religious hate speech content spread to promote hate and violence towards people based on their religious affiliation through digital platforms (Albadi et al., 2018; Olteanu et al., 2018). This offensive language is used for religions such as Islam, Christianity, and Hinduism. As religions involve a group of people, hate speech against it can be more harmful than hate speech against an individual (Chetty & Alathur, 2018). Studies have shown that Islam is one of the most attacked religions facing negative expressions worldwide (Olteanu et al., 2018; Poole et al., 2021). It also appears that those negative expressions are motivated by a sense of Islamophobia favoured by the processes of cultural globalisation through digital media (Horsti, 2017). Muslims are defamed and framed as demonic with negative attitudes such as discrimination, stereotypes, harassment, and physical attacks to create violence. (A. Törnberg & P. Törnberg, 2016). For instance, during the Czech initiative in 2016, hateful comments against Islam were created and spread. These online comments were mainly directed at Muslim refugees and immigrants, and also people, governments and political actors who were on their side (Hanzelka & Schmidt, 2017). In addition, online hate speech content targeting individuals based on their gender and sexual orientation is burgeoning, as new media can exacerbate patterns of genderbased violence and create new forms of abuse (Dragiewicz et al., 2018). The Italian Hate Map project analysed 2,659,879 tweets and showed that users often insulted women. The project identified about 71,006 tweets expressing hateful comments against women and 12,140 tweets containing hate towards lesbian and gay people (KhosraviNik & Esposito, 2018). Besides, following the arrival of Syrian immigrants in neighbouring countries such as Iraq, Jordan, and Turkey, and in European countries such as Germany and Spain, online hate speech content is increasing and becoming more violent and offensive towards refugees (Berecz & Devinat, 2017). Online hate speech against refugees’ results from racism, a type of hate behind the rejection of refugees, immigrants and foreigners (Sánchez-Holgado et al., 2022).

Refugee and Immigration Crisis Around the World Since the Mediterranean Refugee Crisis in 2015, countries in Southern Europe have become more important in terms of accepting refugees, migrants and asylum seekers who remain there or continue to move to wealthier countries. Initially, Italy and Greece were the destination countries for most immigrants. Since 2014, 800,000 people have migrated through unsafe channels to the European Union (EU) borders to find a better life or escape violence and conflict in their homes (Metcalfe-Hough, 2015). The final closure of the migration routes was from Turkey to Greece and from Libya to Italy and Spain (Arcila-Calderón et al., 2021). According to the International Organization for Migration, 57,250 Africans arrived on Spanish territory, making Spain the Mediterranean country with the most immigrants in 2018 (Viúdez, 2019). This wave of immigration has become the most significant and challenging immigration wave European countries face since the Second World War. The flow of immigrants, refugees and others seeking access to European countries has increased significantly. Today, the number of people forced to flee to protect their freedom or save their lives has reached 103 million and increased dramatically worldwide (UNHCR, 2022). Among the nations in need of international help and protection are Syria (6.8 million), Venezuela (5.6 million), Ukraine (5.4 mil242

 Online Hate Speech and the Representations of Refugees

lion), Afghanistan (2.8 million) and South Sudan (2.4 million). Turkey is the country hosting the highest number of refugees, with 3.7 million people. Colombia comes in second with over 2.5 million people, including others who need international protection. Other countries that receive the people most in need of international assistance are Germany (2.2 million), Pakistan (1.5 million) and Uganda (1.5 million) (UNHCR, 2022). The profile of new arrivals is also changing (Metcalfe-Hough, 2015). The majority of people seeking entry into Europe countries through irregular channels were traditionally men. However, today whole family members travel together, sometimes with disabled or elderly relatives and often with young children (Metcalfe-Hough, 2015). According to the UNHCR, 26% of new arrivals in 2021 were female and 27% male. In addition, of the 89.3 million forcibly displaced people at the end of 2021, an estimated 36.5 million were children under 18 (UNHCR, 2021a). This crisis gave rise to its crisis, embodied in a growing wave of hate speech against refugees worldwide. These problems began to emerge as the number of refugees entering the EU increased (Berecz & Devinat, 2017). The people’s reaction to these new arrivals was not harmful; they welcomed them and offered help. However, the situation has become more hostile, and the voices of extremists have become increasingly louder (Berecz & Devinat, 2017). ISIS attacks in Brussels, Paris and other cities worldwide have triggered people to change their minds about refugees (Howard, 2017). For example, the idea of Syrian refugees as a threat to Europe is becoming more adamant. Van Dijk (2017) examined the news in American newspapers and identified that the discourses in those texts against minorities and immigrants have specific patterns according to their subjects. The immigrants were identified using lines such as “they are coming in enormous numbers”. Van Dijk (2017) also found that immigrants in Spain, Germany, France and the Netherlands are framed as a threat and a people who can never be adopted into society. Another interesting finding is that immigrants are mainly represented as different, insolent and criminal (Van Dijk, 2017).

Hate Speech Against Syrian Refugees in Turkey Since the start of the Syrian civil war in 2011, Turkey has either become a country to which Syrians came to escape from the civil war or has acted as a bridge in their transition to Europe (Kavaklı, 2018). Following this forced displacement, there has been a dramatic increase in the number of Syrian refugees in Turkey. Turkey currently hosts approximately 3.6 million registered Syrian refugees and approximately 320,000 interested persons of other nationalities (UNHCR, 2021b). Turkish President Recep Tayyip Erdogan, describing the Syrian refugees as “our brothers”, underlined that the common religion of many Syrians and Turks is Islam (Erdoğan, 2019, 2020). Erdoğan emphasised that Turkey accepts Syrians in need of Allah’s (God’s) help as guests and that this aid is provided by Turkey (Koc, 2021). He also shared a tweet saying that “to my Syrian brothers and sisters, who ask when Allah’s help will come, I say this: No doubt, Allah’s help is near” (Recep Tayyip Erdogan, 2013). Syrian refugees were often introduced to Turkish citizens as “our sisters and brothers” (Atalay, 2014). However, Syrian refugees have been the target of hostility, intolerance and hate speech over time (Koc, 2021). With social media, users can participate in online media content and thus contribute to the democratisation of communities. Social media is also one of the platforms on which Syrian refugees can be the target of hate speech (Kavaklı, 2018). Filibeli and Ertuna (2021) examined user comments posted on Facebook about Syrian refugees between 2018 and 2019. They found a great number of sarcastic comments posted to criticise official policies as well as to claim superiority against refugees. These sarcastic comments can be considered part of the cycle of violence and serve as discriminatory 243

 Online Hate Speech and the Representations of Refugees

rhetoric and hate speech (Filibeli & Ertuna, 2021). Aslan (2017) focused on YouTube videos posted in 2016 to analyse hate speech content created by users about Syrian immigrants in Turkey. She found that Syrians were framed with the use of negative expressions such as “a potential threat”, “traitors”, “the sources of financial difficulties”, and “overstepping” (Aslan, 2017). In summary, it has been observed that Turkey’s marginalising and discriminatory attitude towards Syrian refugees is circulating symbolically and linguistically through “new media” (Gözde, 2019).

METHOD AND OBJECTIVES While existing research has featured an exploration of the use of social media such as YouTube and Facebook in the dissemination of hate speech conversations (Aslan, 2017; Filibeli & Ertuna, 2021), little is known about the use of Twitter to disseminate hateful messages against Syrian refugees in Turkey (e.g., Ozduzen et al., 2021). In this study, a closer look is taken at online hate speech content to provide information about the reflections on the refugee problem in Turkey on Twitter. In order to explore the contribution of Twitter to hate speech discourses towards refugees, a qualitative and quantitative analysis will be applied. This study examines how Twitter was used following the dramatic events regarding Syrians in Turkey by reviewing the literature on online hate speech, the refugee crisis around the world, and more specifically in Turkey, as well as hate speech towards refugees. In addition, this study focuses on the discourses regarding refugees to understand how refugees in Turkey are represented in online platforms. More specifically, this study aims to examine and provide detailed information about the online representation of refugees within online discussions that contained hateful and racist texts. To understand the use of Twitter in hate speech dissemination about Syrians, Twitter content under the hashtag #VatanimdaMülteci (#RefugeeinMyCountry) will be analysed. Specifically, two research questions emerged during the preceding literature review on hate speech: RQ1: How were hate speech discourses directed at refugees produced and distributed in Turkey via #VatanimdaMülteci (#RefugeeinMyCountry)? RQ2: How was the hashtag #VatanimdaMülteci (#RefugeeinMyCountry) used to express negative beliefs and feelings towards refugees in Turkey? The social media analysis includes 133,910 tweets and mentions under the hashtag #VatanimdaMülteci (#RefugeeinMyCountry) between April 12th and 14th, 2022, following the dramatic events involving Syrians. The information was gathered using the GNIP app (Twitter’s API), which allowed access to every undeleted tweet on the platform. The study demonstrates widespread user reactions to Turkey’s refugee problem through qualitative and quantitative content analyses of the collected tweets using DiscoverText. It is important to understand the racist and discriminatory discourses in the context of refugees in Turkey. Figure 1 shows the composition and circulation of the tweets via the hashtag over time.

244

 Online Hate Speech and the Representations of Refugees

Figure 1. Time track of the tweets posted under #VatanimdaMülteci

This demonstrates that many tweets were published to express ideas and emotions about Syrian refugees following the dramatic event. It also shows that the number of tweets via the hashtag peaked on 4th November 2022 and started to decrease from then on. After the data set was collected, the coding process was completed using the DiscoverText program. A codebook was created based on the online hate classification list suggested by Halminen et al. (2018). The four primary codes of ‘political issue(s)’, ‘specific nation(s)’, ‘racism and xenophobia’ and ‘accusation(s)’ were used for the tweets to analyse hate speech content. If a tweet could not be coded under these headings, it was classified as ‘not codable’. Hence, the tweets were coded based on the four primary codes as demonstrated below: •

• •

•

Political Issue(s): Tweets expressing hate towards governments, political parties, war, terrorism, and the flaw of the system. For example, “I reject the sale of Turkish passports for money, nonsense such as giving citizenship to those who buy a house! I don’t want to see people from Afghan and Syria living for free and increasing our troubles #VatanimdaMülteci”. Specific Nation(s): Tweets expressing hate towards certain countries, people, and such concepts as immigration, sovereignty, and territory. For example, “I don’t want Syrian, Afghan and Pakistani in my country#VatanimdaMülteci”. Racism and Xenophobia: Tweets expressing racist comments towards refugees and generalising the characteristics of refugees along with hateful comments. For example, “People who do not even benefit themselves, who do not know any letter of civilisation, who escape from the tribe and who are uncertain about what they are, are the cancer of our country. Let them go as soon as they came #VatanimdaMülteci”. Accusation(s): Tweets accusing of people or something without evidence to support it. For example, “I don’t want refugees in my homeland, I want those who want to go away, because they are all crime machines #VatanimdaMülteci”.

245

 Online Hate Speech and the Representations of Refugees

While the study focuses on the hashtag #VatanimdaMülteci, there were also different hashtags, such as Suriyeliler (#Syrians) and #SuriyelilerDefolsun (#GetOutSyrians), created to express hateful comments towards Syrian refugees. Therefore, a generalisation cannot be made about the online hateful content created following the dramatic event. However, #VatanimdaMülteci was the most widely used hashtag during the dramatic event, so this study can significantly provide an overview of hateful content. Throughout this study, personal data such as Twitter usernames, tweet identification numbers and location data were not shared to ensure the protection of users.

RESULTS AND DISCUSSION Sample Characteristics of #VatanimdaMülteci (#RefugeeinMyCountry) The micro-blogging service of Twitter offers the possibility to “tweet” by creating original content or “retweet” by re-sharing other users’ tweets (Majmundar et al., 2018). The number of retweets under #VatanimdaMülteci was found to be 116,440, and the original tweets numbered 17,470. This demonstrates that following the dramatic events regarding Syrians in Turkey, users mostly interacted with others by retweeting others’ messages. Many scholars have discussed the importance of the retweet function, arguing that retweeting is the most effective practise in the transition of information to Twitter users e.g. Boyd et al., (2010) and Kwak et al., (2010). Although the retweeting practise can be viewed simply as the process of copying and reposting, the practise contributes to an ecology of speech in which conversations are made up of public interactions (Boyd et al., 2010). On Twitter, most visible users choose to retweet other people’s posts, while others retweet them (Boyd et al., 2010). Kwak et al. (2010) also highlighted that the retweet mechanism allows users to spread their chosen information beyond the reach of the original tweet’s followers. This finding is similar to previous studies, in which the function of retweets was demonstrated as a popular mechanism for spreading hateful online content e.g. Gupta et al., (2021). Poole et al., (2021) examined disinformation and hate speech content related to Islam and Muslims following the 2016 Brussels Terror attack through the hashtag #StopIslam. They found that users mostly spread hate speech content by retweeting others’ messages (235,578 retweets) rather than creating unique messages (66,764 original tweets) via #StopIslam. The dataset of this study also indicates that the retweet mechanism through #VatanimdaMülteci made hate speech content towards Syrian refugees more contagious. Analysis of the tweets using #VatanimdaMülteci also shows that a significant number of tweets in the dataset (57,716) were directly generated to express hateful comments towards specific nationalities such as Syrian, Afghan and Pakistani (see Table 1). Table 1. Nation types highlighted by users      Nationality

     Total

     Syrian

     35,416

     Afghan

     15,450

     Pakistani

     6,850

246

 Online Hate Speech and the Representations of Refugees

Twitter users tweeted along with images and videos showing secretly taken footage of Turkish women shared by Syrian men on social media channels. With these accusations, users started to generate tweets expressing hate speech and demanding the deportation of Syrian refugees from the country. It was found that users often posted racist comments toward not only Syrians but also other refugees in Turkey, such as Afghans and Pakistanis. As seen in Table 1, 15,450 tweets within the corpus were designed openly to express hateful comments towards Afghan refugees in Turkey following the dramatic events related to Syrian refugees. Besides, it was observed that about 6,850 tweets were posted, along with hateful comments related to Pakistani refugees. For example, users labelled the Afghan and Pakistani refugees as “‘intruders”, “illiterates” and a “crime machine”, emphasising that they do not belong in this country and demanding that they be deported. This information is significant to understanding how Syrian refugees, Afghans, and Pakistanis in Turkey are depicted on Twitter.

Content Analysis Results for #VatanimdaMülteci (#RefugeeinMyCountry) A content analysis was accomplished on the sample dataset of 3,556 original tweets that were retweeted by other users at least once. Online hateful content under #VatanimdaMülteci was analysed using four primary codes (political issue(s), specific nation(s), racism and xenophobia and accusation(s)), and it was found that a significant number of tweets were created and spread hateful comments towards not only Syrian refugees but also Afghan and Pakistani refugees in Turkey (see Figure 2). Figure 2. Result of the content analysis for #VatanimdaMülteci

The findings show that many original tweets in the dataset were created to highlight political issues regarding refugees in Turkey (see Figure 2). It shows that the hashtag #VatanimdaMülteci articulated ideas about the government’s refugee policies. In addition, many original tweets were identified as targeting ‘specific nation(s)’, as users openly generated messages to express hateful comments about

247

 Online Hate Speech and the Representations of Refugees

Syrians, Afghans and Pakistanis. Due to the existence of information overload on Twitter, finding helpful information is becoming increasingly difficult (Li et al. 2014). In this study, this situation was observed, and it was identified in more than 25% of the original tweets as ‘not codable’, demonstrating that a significant percentage of the tweets were not designed to express hateful comments towards refugees in Turkey under the hashtag. For example, the study detected tweets that contained non-relevant information about the refugee problem and referred to different topics, such as the retirement policy in Turkey. In addition, the tweets designed for advertisements and tweets containing broken media URLs were identified as “not codable”, as those did not provide any related content for the refugee problem. •

Tweets Highlighting Political Issues: It was found that most of the tweets expressing hateful comments in the corpus were posted to highlight political issues regarding refugees in Turkey. Users frequently tweeted to criticise the government’s policies regarding refugees, such as the Turkish citizenship provided to Syrians and their ability to open businesses in Turkey. They expressed hateful comments towards Syrian refugees who obtained Turkish citizenship. They also frequently criticised government institutions such as the Directorate General of Migration Management for causing this situation. For example, users posted tweets such as:

The immigration administration should be closed, the citizenships given should be taken back, the refugees should be sent to their countries #VatanimdaMülteci. In Turkey, over 117,000 Syrian refugees have acquired Turkish citizenship (Erdogan, 2020). While this is a small number compared to the Turkish population (more than 85 million), it has caused a much-discussed public debate (Güney, 2022). The tweets within #VatanimdaMülteci demonstrated that granting citizenship to refugees has caused online hate speech against them. As highlighted by Sayimer and Derman (2017), the possibility of granting Turkish citizenship to Syrians has sparked an extreme backlash, much of which can be classified as hate speech. Users also tweeted criticising the consequences of refugees opening many businesses in Turkey. For instance: Arabic signs everywhere. I wander around my country like a tourist. I do not want Arabic signs. I do not want refugees in my country #VatanimdaMülteci. They not only criticised those results, but they also used sentences containing hate speech against refugees. According to the Turkish Code of Commerce, Syrian refugees can open businesses in Turkey (Erdogan, 2019). Syrians have opened nearly 14,000 companies in Turkey (Hardan, 2021). Therefore, users criticise this situation and demand to remove Arabic signs nationwide to protect the Turkish language. The ability of refugees to start a business in Turkey has been determined as an important factor that increases hate speech within #VatanimdaMülteci. Similarly, Ozduzen et al., (2021) highlighted that since Syrian refugees gained the ability to open a business in Turkey, online attacks towards them have emerged and increased dramatically. •

248

Tweets Highlighting Specific Nations: Following the tweets highlighting political issues, users frequently posted tweets that openly express hateful comments towards specific nations through #VatanimdaMülteci. More specifically, most of the tweets within the corpus (31%) were posted

 Online Hate Speech and the Representations of Refugees

to express hateful messages regarding Syrians, Pakistanis, and Afghans in Turkey. The findings show that users sometimes tweeted openly hateful comments towards specific nations, and they also tweeted only to emphasise that they do not want refugees in Turkey: I do not want refugees in my homeland, I do not want Afghans, I do not want Syrians. Enough is enough, the patience stone has cracked #VatanimdaMülteci. I do not just want refugees in my country #VatanimdaMülteci. More than 3.5 million Syrians live in Turkey and are referred to as guests by Turkish officials (Abdelaaty, 2021). The Turkish state introduced this welcome as hospitality and generosity (Alkan, 2021). However, the temporary nature of this hospitality surpassed the hospitality of Turkish citizens, and this hospitality turned into hostility over time (Koc, 2021). It was found that users frequently tweeted to target refugees from certain countries to indicate that they were no longer welcome. For example, I do not want refugees in my homeland! I do not want Syrian, Afghan! The hospitality is over! #VatanimdaMülteci. Users also often shared images to say they do not want refugees in their homeland (see Figure 3).

249

 Online Hate Speech and the Representations of Refugees

Figure 3. The picture was accompanied by the following text: “I do not want neither Syrian nor Afghan refugees #VatanimdaMülteci”

•

Tweets Expressing Racist Comments Towards Refugees: It was found that some users created tweets (8%) to express openly racist comments towards refugees in Turkey. This suggests that Twitter was used as a space to spread racist messages. Therefore, the platform provided greater visibility of racist texts and facilitated greater participation in racist conversations. It was observed that users often called themselves racist and criticised refugees:

I do not want refugees! Yes, I am racist. We can have 3 children at the most. However, they breed 5 to 10 each #VatanimdaMülteci. If I must be racist for the future of this country, I am racist to the core. I am racist against any refugee who is ignorant, uneducated, and harmful #VatanimdaMülteci.

250

 Online Hate Speech and the Representations of Refugees

As seen in those examples, tweeters not only called themselves racist but also expressed hateful comments about some of the characteristics of refugees. It was observed that users often called Syrian refugees “cowards”. This is because they think that it is a cowardly act for Syrians to migrate to Turkey instead of fighting in their own country. Some users also frequently compared Syrian people swimming in the sea in Turkey instead of fighting in their own country with Turkish soldiers on duty in Syria. For example, the picture below was shared by users to highlight this situation using #VatanimdaMülteci (see Figure 4). Figure 4. A picture showing the comparison of Syrians swimming in Turkey with Turkish soldiers on duty in Syria

It also appeared that users often described refugees as “ignorant”, “harmful” and “uncivilised”. This shows that refugees are exposed to many insults, slander, and racist and hostile statements on Twitter.

251

 Online Hate Speech and the Representations of Refugees

As indicated by many researchers in the literature (e.g., Kurt, 2019; Ozduzen et al., 2021), refugees have become one of the critical issues in this generation and the spread of online hate speech and online racism. •

Tweets Accusing Refugees: Within the corpus, a small number of tweets (3%) were identified as tweets that were designed to accuse refugees without evidence to support any allegations. Although the Turkish government framed Syrian refugees as sisters and brothers to the public, Turkish citizens thought the opposite. The findings show that users frequently accused Syrian refugees of being criminal outsiders. For instance, users posted tweets as follows:

Where is the incident, there is the Syrian, where is the murder there is the Syrian, where there is the harassment there is the Syrian #VatanimdaMülteci. I do not want refugees in my homeland. I want those who want to go away because they are all crime machines #VatanimdaMülteci. Users also often mentioned the independence war by referring to refugees as those who tried to overthrow the Turkish Republic established after the independence war. They described refugees as “occupiers” and defined them as a “threat” to the integrity of the state. Therefore, they need to be deported from the country for the protection of the state. I do not want refugees trying to destroy our republic founded with blood and wisdom #VatanimdaMülteci. I do not want an invader in my homeland under the name of Refugee! #VatanimdaMülteci. These examples demonstrate that the hashtag #VatanimdaMülteci expressed negative beliefs and ideologies regarding refugees. The representations of refugees through the hashtag show that the messages were mainly designed with negative and marginalising language by referring to them as “invaders”, “threats” and “crime machines”. Similarly, Kreis (2017) examined the online discussion around the European refugee crisis on Twitter. The focus of that study was on tweets under the hashtag #refugeesnotwelcome, to explore the use of this hashtag in expressing negative feelings and ideas toward refugees in Europe. This study shows that the hashtag was used to represent refugees as unwanted and criminal.

DISCUSSION AND CONCLUSION This study is an examination of the use of Twitter in the spread of hateful comments towards Syrian refugees. This study focuses on the hashtag #VatanimdaMülteci (#RefugeeinMyCountry), designed to express negative beliefs, feelings, and ideologies regarding Syrian refugees in Turkey. However, the results show that many messages were generated to articulate opinions about refugees from other countries, such as Afghanistan and Pakistan. Studies on hate speech content have examined how online hate speech has been directed at Syrian refugees (Alikılıç et al., 2021; Kurt, 2019; Sayimen & Derman, 2017). This study explores hate speech content towards Afghan and Pakistani refugees and therefore offers a comprehensive analysis of hate speech content directed at refugees from other countries within 252

 Online Hate Speech and the Representations of Refugees

online media. While previous researchers have examined hate speech texts regarding refugees through the online platforms of YouTube and Facebook e.g., Aslan’s (2017), that provides detailed information about the representations of refugees in Turkey on Twitter. The study demonstrated that online hate speech content regarding refugees is increasingly becoming negative and violent through negative feelings, beliefs, and ideologies towards refugees. Studies examining online content related to refugees have also shown that users created messages to show public empathy towards refugees via Twitter. For example, the hashtag #RefugeesWelcome was designed to support Syrian refugees who fled the war and seek to reach Northern Europe and Western countries (Barisione et al., 2019). The hashtag gained attention through social media platforms (such as Instagram, Facebook, and YouTube) and manifested the public’s empathy towards Syrian refugees in Europe. Finally, it was used as a political slogan (Refugees Welcome) in the offline public sphere, as well as in mainstream news media and in public protests (Barisione et al., 2019). The results show that many tweets were generated to express hateful comments towards Syrians who obtained Turkish citizenship. Users not only expressed hateful comments but also criticised the government institutions responsible for granting citizenship. Users also often posted tweets to criticise Syrian refugees’ ability to open a business and its consequences. The study shows that many tweets containing hate speech were related to refugee integration processes in Turkey. Moreover, it was found that users frequently posted tweets openly to express hateful comments towards Syrian, Afghan and Pakistani refugees. It was observed that those tweets were designed to highlight that refugees in Turkey are no longer welcome. It suggests that the hospitality of Turkish citizens has turned into hostility over time. Turkish citizens began to express this situation openly by using negative, hateful and marginalising expressions. The results demonstrate that users often represented refugees as “ignorant”, “harmful”, and “uncivilised”. In addition, users called refugees “cowards” because they did not fight in their own country. Some users also identified refugees using terms such as “invader”, “threat” and “crime machine”. This shows that online representations of refugees under #VatanimdaMülteci were full of negative feelings, beliefs and ideologies. Those tweets were mainly generated to say that refugees are intruders and unwanted, so they must be deported from the country.

REFERENCES Abdelaaty, L. (2021). Refugees and guesthood in Turkey. Journal of Refugee Studies, 34(3), 2827–2848. doi:10.1093/jrs/fez097 Albadi, N., Kurdi, M., & Mishra, S. (2018, August). Are they our brothers? analysis and detection of religious hate speech in the arabic twittersphere. In 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) (pp. 69-76). IEEE. 10.1109/ASONAM.2018.8508247 Alikılıç, Ö., Gökaliler, E., & Alikılıç, İ. (2021). Nefret Söylemi Üzerinden Ötekileştirme: Twitter’da Mültecilere Yönelik Nefret Tipolojisi Analizi [Marginalization Through Hate Speech: Analysis of Hate Typology Against Refugees on Twitter]. Akdeniz Üniversitesi İletişim Fakültesi Dergisi, (36), 501–520. Alkan, H. (2021). The gift of hospitality and the (un) welcoming of Syrian migrants in Turkey. American Ethnologist, 48(2), 180–191. doi:10.1111/amet.13012

253

 Online Hate Speech and the Representations of Refugees

Arcila-Calderón, C., Blanco-Herrero, D., Frías-Vázquez, M., & Seoane-Pérez, F. (2021). Refugees welcome? Online hate speech and sentiments in Twitter in Spain during the reception of the boat Aquarius. Sustainability (Basel), 13(5), 2728. doi:10.3390u13052728 Aslan, A. (2017). Online hate discourse: A study on hatred speech directed against Syrian refugees on YouTube. Journal of Media Critiques, 3(12), 227–256. doi:10.17349/jmc117413 Atalay, O. (2014). Suriyeli muhacirler bizim kardeşimizdir [Syrian refugees are our brothers]. Yeni Akit. https://www.yeniakit.com.tr/yazarlar/osman-atalay/suriyeli-muhacirler-bizim-kardesimizdir-7332.html Ataman, H. (2014). Yazılı basında ayrımcı söylem, Suriyeli mülteciler. Medyada Nefret Söylemi ve Ayrımcı Dil. [Discriminatory discourse in the print media, Syrian refugees. Hate Speech and Discriminatory Language in the Media]. https://hrantdink.org/tr/asulis/yayinlar/72-medyada-nefret-s oylemi-raporlari/394-medyada-nefret-soylemi-ve-ayrimci-dil-e ylul-aralik-2014 Balabanova, E., & Trandafoiu, R. (2020). Media, migration, and human rights: Discourse and resistance in the context of the erosion of liberal norms. Journal of Language and Politics, 19(3), 379–390. doi:10.1075/jlp.20007.bal Balch, A., & Balabanova, E. (2016). Ethics, politics, and migration: Public debates on the free movement of Romanians and Bulgarians in the UK, 2006–2013. Politics, 36(1), 19–35. doi:10.1111/1467-9256.12082 Balkin, J. M. (2017). Digital speech and democratic culture: A theory of freedom of expression for the information society. In Law and Society Approaches to Cyberspace (pp. 325–382). Routledge. doi:10.4324/9781351154161-9 Barisione, M., Michailidou, A., & Airoldi, M. (2019). Understanding a digital movement of opinion: The case of# RefugeesWelcome. Information Communication and Society, 22(8), 1145–1164. doi:10. 1080/1369118X.2017.1410204 Berecz, T., & Devinat, C. (2017). Relevance of cyber hate in Europe and current topics that shape online hate speech. In Countering Cyber Hate Phenomena. INACH. Berger, J. (2013). Beyond viral: Interpersonal communication in the internet age. Psychological Inquiry, 24(4), 293–296. doi:10.1080/1047840X.2013.842203 Boyd, D., Golder, S., & Lotan, G. (2010, January). Tweet, tweet, retweet: Conversational aspects of retweeting on twitter. In 2010 43rd Hawaii international conference on system sciences (pp. 1-10). IEEE. Brown, A. (2018). What is so special about online (as compared to offline) hate speech? Ethnicities, 18(3), 297–326. doi:10.1177/1468796817709846 Burnap, P., & Williams, M. L. (2016). Us and them: Identifying cyber hate on Twitter across multiple protected characteristics. EPJ Data Science, 5(1), 1–15. doi:10.1140/epjds13688-016-0072-6 PMID:32355598 Cammaerts, B. (2015). Social media and activism. Journalism, 1027–1034. Chadwick, A. (2017). The hybrid media system: Politics and power. Oxford University Press. doi:10.1093/ oso/9780190696726.001.0001

254

 Online Hate Speech and the Representations of Refugees

Chetty, N., & Alathur, S. (2018). Hate speech review in the context of online social networks. Aggression and Violent Behavior, 40, 108–118. doi:10.1016/j.avb.2018.05.003 Citron, D. K. (2014). Hate crimes in cyberspace. Harvard University Press. doi:10.4159/harvard.9780674735613 Daniels, J. (2013). Race and racism in Internet studies: A review and critique. New Media & Society, 15(5), 695-719. Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017, May). Automated hate speech detection and the problem of offensive language. In Proceedings of the international AAAI conference on web and social media (Vol. 11, No. 1, pp. 512-515). 10.1609/icwsm.v11i1.14955 De Coninck, D. (2020). Migrant categorisations and European public opinion: Diverging attitudes towards immigrants and refugees. Journal of Ethnic and Migration Studies, 46(9), 1667–1686. doi:10.1 080/1369183X.2019.1694406 De GibertO.PerezN.García-PablosA.CuadrosM. (2018). Hate speech dataset from a white supremacy forum. doi:10.18653/v1/W18-5102 Delgado, R., & Stefancic, J. (2014). Hate speech in cyberspace. Wake Forest Law Review, 49, 319. Doğanay, Ü., & Keneş, H. Ç. (2016). Yazılı basında Suriyeli ‘mülteciler’: Ayrımcı söylemlerin rasyonel ve duygusal gerekçelerinin inşası [Syrian ‘refugees’ in the print media: constructing rational and emotional justifications for discriminatory discourses]. Mülkiye Dergisi, 40(1), 143-184. Eberl, J. M., Meltzer, C. E., Heidenreich, T., Herrero, B., Theorin, N., Lind, F., Berganza, R., Boomgaarden, H. G., Schemer, C., & Strömbäck, J. (2018). The European media discourse on immigration and its effects: A literature review. Annals of the International Communication Association, 42(3), 207–223. do i:10.1080/23808985.2018.1497452 ElSherief, M., Kulkarni, V., Nguyen, D., Wang, W. Y., & Belding, E. (2018, June). Hate lingo: A targetbased linguistic analysis of hate speech in social media. In Proceedings of the International AAAI Conference on Web and Social Media (Vol. 12, No. 1). 10.1609/icwsm.v12i1.15041 Erdoğan, M. (2014). Türkiye’deki Suriyeliler: Toplumsal kabul ve uyum araştırması [Syrians in Turkey: Social acceptance and cohesion research]. Hacettepe Üniversitesi Göç ve Siyaset Araştırmaları Merkezi. Erdoğan, M. (2019). Syrian refugees in Turkey. Konrad–Adenauer-Stiftung Report. Erdoğan, M. M. (2020). Onuncu Yılında Türkiye’deki Suriyeliler [Syrians in Turkey in its Tenth Anniversary]. Uluslararası İlişkiler Konseyi Panorama, 30. Erdoğan, R. T. [@RTErdogan]. (2013). Allah’ın yardımı ne zaman diye soran Suriyeli kardeşlerime, şunu söylüyorum Hiç kuşkusuz Allah’ın yardımı yakındır [To my Syrian brothers and sisters, who ask when Allah’s help will come, I say this: No doubt, Allah’s help is near]. Twitter. https://twitter.com/ rterdogan/status/331045737511149568

255

 Online Hate Speech and the Representations of Refugees

Evrensel. (2017). Karapınar’da Suriyelilere Dönük Irkçı Kışkırtma [Racist Provocation Against Syrians in Karapınar]. Evrensel. https://www.evrensel.net/haber/330169/karapinardasuriyeliler e-donuk-irkci-kiskirtma Filibeli, T. E., & Ertuna, C. (2021). Sarcasm beyond hate speech: Facebook comments on Syrian refugees in Turkey. International Journal of Communication, 15, 24. Fortuna, P., & Nunes, S. (2018). A survey on automatic detection of hate speech in text. ACM Computing Surveys, 51(4), 1–30. doi:10.1145/3232676 Freedom House. (2016). Freedom on the Internet 2016. Freedom House. https://freedomhouse.org/sites/ default/files/2020-02/FOTN_2016_BOOKLET_FINAL.pdf Georgiou, M., & Zaborowski, R. (2017). Media coverage of the “refugee crisis”: A cross-European perspective. Council of Europe. Goker, G., & Keskin, S. (2015). Haber medyasi ve multeciler: Suriyeli multecilerin Turk yazili basinindaki temsili [News media and refugees: Representation of Syrian refugees in Turkish press]. Journal of Communication Theory and Research/Iletisim Kuram ve Arastirma Dergisi, 41, 229-256. Goodman, S., Sirriyeh, A., & McMahon, S. (2017). The evolving (re) categorisations of refugees throughout the “refugee/migrant crisis”. Journal of Community & Applied Social Psychology, 27(2), 105–114. doi:10.1002/casp.2302 Gözde, K. U. R. T. (2019). Yeni medyada nefret söylemi: YouTube’da Suriyeli mültecilere karşı üretilen nefret söylemi üzerine bir araştırma [Hate speech in new media: A research on hate speech against Syrian refugees on YouTube]. The Journal of International Lingual Social and Educational Sciences, 5(1), 1–20. Guess, A., Nagler, J., & Tucker, J. (2019). Less than you think: Prevalence and predictors of fake news dissemination on Facebook. Science Advances, 5(1), eaau4586. doi:10.1126ciadv.aau4586 PMID:30662946 Güney, Ü. (2022). Syrian refugees between Turkish nationalism and citizenship. Journal of Immigrant & Refugee Studies, 20(4), 519–532. doi:10.1080/15562948.2021.1950256 Gunitsky, S. (2015). Corrupting the cyber-commons: Social media as a tool of autocratic stability. Perspectives on Politics, 13(1), 42–54. doi:10.1017/S1537592714003120 Gupta, S., Nagar, S., Nanavati, A. A., Dey, K., Barbhuiya, F. A., & Mukherjea, S. (2021). Consumption of Hate Speech on Twitter: A Topical Approach to Capture Networks of Hateful Users. In ROMCIR@ ECIR (pp. 25-34). Academic Press. HaberTürk. (2017, April 17). Mersin’de Suriyelilerle Belde Halkı Arasında Gerilim [Tension Between Syrians and Local People in Mersin]. Haberturk. http://www.haberturk.com/gundem/haber/1464879- mersinde-suri yelilerle-halk-arasinda-gerilim/6 Hanzelka, J., & Schmidt, I. (2017). Dynamics of Cyber Hate in Social Media: A Comparative Analysis of Anti-Muslim Movements in the Czech Republic and Germany. International Journal of Cyber Criminology, 11(1).

256

 Online Hate Speech and the Representations of Refugees

Hardan, M. (2021). How Syrian refugees contributed to Turkish economy. Al-monitor. https://www.almonitor.com/originals/2021/05/how-syrian-refugees-contributed-turkish-economy Horsti, K. (2017). Digital Islamophobia: The Swedish woman as a figure of pure and dangerous whiteness. New Media & Society, 19(9), 1440–1457. doi:10.1177/1461444816642169 Howard, E. (2017). Freedom of expression and religious hate speech in Europe. Routledge. doi:10.4324/9781315277257 Howard, P., Kollanyi, B., & Woolley, S. C. (2016). Bots and automation over twitter during the second us presidential debate. Academic Press. Jakubowicz, A. (2017). Alt_Right White Lite: Trolling, hate speech and cyber racism on social media. Cosmopolitan Civil Societies: An Interdisciplinary Journal, 9(3), 41–60. doi:10.5130/ccs.v9i3.5655 Jenkins, H. (2006). Convergence Culture: Where Old and New Media Collide. New York University Press. Kariithi, N., Mawadza, A., & Carciotto, S. (2017). Media portrayal of immigration in the South African media 2011–2015. The Scalabrini Institute for Human Mobility in Africa-SIHMA, 1-32. Kavaklı, N. (2018). Hate Speech Towards Syrian Refugees in Turkish Online Reader Comments. In Global Perspectives on Human Migration, Asylum, and Security (pp. 127-142). IGI Global. doi:10.4018/9781-5225-2817-3.ch007 KhosraviNik, M., & Esposito, E. (2018). Online hate, digital discourse and critique: Exploring digitallymediated discursive practices of gender-based hostility. Lodz Papers in Pragmatics, 14(1), 45–68. doi:10.1515/lpp-2018-0003 Kisang, A. K. (2017). The images and rights of migrants in the Kenyan media. Current Journal of Applied Science and Technology, 22(5), 1–14. doi:10.9734/CJAST/2017/32385 Koç, G. (2021). Derrida’s Concept of Hos (ti) pitality and Hate Crime in Turkey. Journal of Business Diversity, 21(2). Kreis, R. (2017). # refugeesnotwelcome: Anti-refugee discourse on Twitter. Discourse & Communication, 11(5), 498–514. doi:10.1177/1750481317714121 Krzyżanowski, M., Triandafyllidou, A., & Wodak, R. (2018). The mediatisation and the politicisation of the “refugee crisis” in Europe. Journal of Immigrant & Refugee Studies, 16(1-2), 1–14. doi:10.1080 /15562948.2017.1353189 Kwak, H., Lee, C., Park, H., & Moon, S. (2010, April). What is Twitter, a social network or a news media? In Proceedings of the 19th international conference on World wide web (pp. 591-600). 10.1145/1772690.1772751 Li, P., Li, W., Wang, H., & Zhang, X. (2014). Modeling of information diffusion in Twitter-like social networks under information overload. TheScientificWorldJournal, 2014, 2014. doi:10.1155/2014/914907 PMID:24795541 Majmundar, A., Allem, J. P., Boley Cruz, T., & Unger, J. B. (2018). The why we retweet scale. PLoS One, 13(10), e0206076. doi:10.1371/journal.pone.0206076 PMID:30335827

257

 Online Hate Speech and the Representations of Refugees

Matamoros-Fernández, A., & Farkas, J. (2021). Racism, hate speech, and social media: A systematic review and critique. Television & New Media, 22(2), 205–224. doi:10.1177/1527476420982230 Metcalfe-Hough, V. (2015). The migration crisis? Facts, challenges and possible solutions. Briefing. Narlı, N., Özaşçılar, M., & Turkan Ipek, I. Z. (2020). Turkish Daily press framing and representation of Syrian women refugees and gender-based problems: Implications for social integration. Journal of Immigrant & Refugee Studies, 18(1), 1–21. doi:10.1080/15562948.2018.1557311 Olteanu, A., Castillo, C., Boy, J., & Varshney, K. (2018, June). The effect of extremist violence on hateful speech online. In Proceedings of the international AAAI conference on web and social media (Vol. 12, No. 1). 10.1609/icwsm.v12i1.15040 Ozduzen, O., Korkut, U., & Ozduzen, C. (2021). ‘Refugees are not welcome’: Digital racism, online placemaking and the evolving categorisation of Syrians in Turkey. New Media & Society, 23(11), 3349-3369. Papacharissi, Z. (2010). A private sphere: Democracy in a digital age. Polity. Papacharissi, Z. (2015). Affective publics: Sentiment, technology, and politics. Oxford University Press. Poole, E., Giraud, E. H., & de Quincey, E. (2021). Tactical interventions in online hate speech: The case of# stopIslam. New Media & Society, 23(6), 1415-1442. Rajendran, L., & Thesinghraja, P. (2014). The impact of new media on traditional media. Middle East Journal of Scientific Research, 22(4), 609–616. Ring, C. E. (2013). Hate speech in social media: An exploration of the problem and its proposed solutions [Doctoral dissertation]. University of Colorado at Boulder. Sánchez-Holgado, P., Amores, J. J., & Blanco-Herrero, D. (2022). Online Hate Speech and Immigration Acceptance: A Study of Spanish Provinces. Social Sciences (Basel, Switzerland), 11(11), 515. doi:10.3390ocsci11110515 Sayimer, İ., & Derman, M. R. (2017). Syrian refugees as victims of fear and danger discourse in social media: A YouTube analysis. Yeni Medya’da Korku ve Tehlike Söylemlerinin Kurbanları Suriyeli Mülteciler: Bir Youtube Analizi, 8(15), 384–403. Shah, D. V., Cho, J., Eveland, W. P. Jr, & Kwak, N. (2005). Information and expression in a digital age: Modeling Internet effects on civic participation. Communication Research, 32(5), 531–565. doi:10.1177/0093650205279209 Siegel, A. A. (2020). Online hate speech. Social media and democracy: The state of the field, prospects for reform, 56-88. Suzor, N., Dragiewicz, M., Harris, B., Gillett, R., Burgess, J., & Van Geelen, T. (2019). Human rights by design: The responsibilities of social media platforms to address gender‐based violence online. Policy and Internet, 11(1), 84–103. doi:10.1002/poi3.185 Syahputra, I. (2019). Expressions of hatred and the formation of spiral of anxiety on social media in Indonesia. SEARCH Journal of Media and Communication Research, 11(1), 95–112.

258

 Online Hate Speech and the Representations of Refugees

Törnberg, A., & Törnberg, P. (2016). Muslims in social media discourse: Combining topic modeling and critical discourse analysis. Discourse, Context & Media, 13, 132–142. doi:10.1016/j.dcm.2016.04.003 Twitter. (2023). Hateful Conduct. Twitter Help Center. https://help.twitter.com/en/rules-and-policies/ hateful-conduct-policy Ullmann, S., & Tomalin, M. (2020). Quarantining online hate speech: Technical and ethical perspectives. Ethics and Information Technology, 22(1), 69–80. doi:10.100710676-019-09516-z UNESCO. (2017). Migrant/migration. United Nations Educational Scientific and Cultural Organization. http://www.unesco.org/new/en/social-and-human-sciences/themes/international-migration/glossary/ migrant UNHCR. (1951). Convention and Protocol Relating to the Status of Refugees. The UN Refugee Agency. https://www.unhcr.org/protection/basic/3b66c2aa10/convention-protocolrelating-status-refugees.html UNHCR. (2016). ‘Refugee’ or ‘migrant’ – Which is right? The UN Refugee Agency. https://www.unhcr. org/news/stories/unhcr-viewpoint-refugee-or-migrant-which-right UNHCR. (2021a). Global Trends Forced Displacement In 2021. The UN Refugee Agency. https://www. unhcr.org/62a9d1494/global-trends-report-2021 UNHCR. (2021b). Refugees and Asylum Seekers in Turkey. The UN Refugee Agency. https://www.unhcr. org/tr/en/refugees-and-asylum-seekers-in-turkey UNHCR. (2022). Refugee Data Finder. The UN Refugee Agency. https://www.unhcr.org/refugee-statistics/ UNHCR. (2023). Understanding Hate Speech. The UN Refugee Agency. https://www.un.org/en/hatespeech/understanding-hate-speech/what-is-hate-speech Van der Haak, B., Parks, M., & Castells, M. (2012). The future of journalism: Networked journalism. International Journal of Communication, 6, 16. van Dijk, T. A. (2000). New(s) Racism: A discourse analytical approach. Ethnic minorities and the media, 33-49. Van Dijk, T. A. (2017). Discourse and power. Bloomsbury Publishing. Vardal, Z. B. (2015). Nefret söylemi ve yeni medya (Hate speech and new media). Maltepe Üniversitesi İletişim Fakültesi Dergisi, 2(1), 132–156. Viúdez, J. (2019). Récord de Pateras y de Muertes de Migrantes en 2018. El País, 16. Wall, M. (2015). Citizen Journalism: A retrospective on what we know, an agenda for what we don’t. Digital Journalism (Abingdon, England), 3(6), 797–813. doi:10.1080/21670811.2014.1002513 White, A. (Ed.). (2015). Moving stories: International review of how media cover migration. Ethical Journalism Network. Yenigün. (2016, July 10). Konya’da Tehlikeli Gerginlik [Dangerous Tension in Konya]. Yenigün. https:// www.gazeteyenigun.com.tr/konyada-tehlikeli-gerginlik/146928/ Youtube. (2017). Hate speech policy. Youtube Help Center Policies. https://support.google.com/youtube/ answer/2801939 259

260

Chapter 14

Analysis of Radicalisation Prevention Policies From the Perspective of Educommunication in Mediterranean Countries Arantxa Azqueta https://orcid.org/0000-0003-2514-5989 International University of La Rioja, Spain Ángela Martín-Gutiérrez https://orcid.org/0000-0001-9847-245X Universidad de Sevilla, Spain Angel Freddy Rodríguez-Torres https://orcid.org/0000-0001-5047-2629 Universidad Central, Ecuador

ABSTRACT The fight against radicalisation is gaining prominence on international agendas. Europe proposes multilevel actions, where “educommunication” helps to prevent hate speech, as it is a tool that contributes to the formation of a critical public opinion in the 4.0 era. The aim of this chapter is to analyse the attitudes that define interculturally competent citizenship and their presence in the radicalisation prevention policies of three Mediterranean countries: France, Portugal, and Spain. Elements related to openness, respect, civic-mindedness, self-efficacy, and tolerance are analysed. The results show that the plans analysed show differences in 1) the presence or absence of victims in the attacks committed in the territory and 2) the presence of the Muslim population in the territory over a period of time. Furthermore, the analysis has led to the conclusion that it is necessary to promote cross-cutting policies for the prevention of radicalisation that address identity aspects. DOI: 10.4018/978-1-6684-8427-2.ch014

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 Analysis of Radicalisation Prevention Policies

INTRODUCTION The fight against violent radicalisation in Europe has been enshrined in official documents since 2005, with updates in 2008 and 2014 (Ruiz-Díaz, 2017) and 2020 (European Commission, 2020). In turn, European countries have developed measures to facilitate the integration of immigrant populations, refugees, and asylum seekers (Eurydice, 2019). This objective is in line with the aim of the Council of Europe, which has among its original purposes to foster European identity to achieve greater unity among states and protect fundamental freedoms and human rights (Council of Europe, 1949). Europe faces major challenges. Lack of trust in democratic processes, political detachment of citizens, lack of intercultural dialogue in culturally diverse societies and the rise of violent extremism are some of the most relevant threats to the values of freedom, citizenship and tolerance found in Europe. A high point was the attacks of September 11 2001. Since then, the terrorist threat has been perceived as dangerous in Europe. Moreover, Europe has become a place of radicalisation and recruitment, where the number of radicalised individuals has increased. Their profile is heterogeneous. However, some common characteristics can be traced: Islamic religion and culture, a fragile socio-economic situation, and a lack of cultural belonging (Municio, 2017; Troian et al., 2019). Various factors are intertwined in the radicalisation process, both personal and structural, such as the socio-economic and cultural crisis, the need to belong in a disadvantaged environment that lacks empathy for their situation or the limited possibilities for the future (Coolsaet, 2019). In this regard, we turn again to the French example. For Kepel (2016), radicalisation results from the failure of integration policies that have led to residential segregation, employment difficulties, social and political marginalisation and ultimately, the withdrawal of the Muslim community into itself. The propaganda of radical Salafist groups employing a fundamentally religious logic has taken its toll on these second or third generations of European citizens who often find themselves stigmatised and rejected and yearn to belong to an accepting group (Lahnait, 2018). Thus, grievance or the perception of an unjust situation strengthens feelings of empathy with the attacks and has led to self-marginalisation practices. However, Roy (2017), with a cross-cutting approach, emphasises European jihadists’ affinity with the disaffected and radical youth culture in which religious motivation is irrelevant since many do not know or understand Islam in depth. According to Aly et al. (2015), radicalisation is a complex phenomenon that requires cross-cutting and comprehensive prevention measures. The attacks have provoked a strong social and media shock and a threat to traditional European values. Most attacks have been perpetrated by so-called “domestic fighters” from endogenous jihadism or “homegrown”. They are primarily European citizens belonging to the second or third generation of Muslim immigrants born and raised in Europe (Municio, 2017). At the same time, there has been a growing suspicion of Muslim immigration among the indigenous population (Bayrakli, & Hafez, 2020; Cesari, 2013). The attacks in Paris in November 2015 led to a meeting of European education ministers and the signing of the ‘Paris Declaration’ (Eurydice, 2016). This document sets out common objectives and policies favouring integration, social cohesion, and the prevention of radicalisation (Eurydice, 2019). Education systems in democratic societies face challenges related to cultural integration due, among other causes, to the number of international students enrolled in compulsory education (Trujillo & Moyano, 2008). In this regard, it is desirable to prevent some schools from becoming ghettos for certain ethnic groups. These situations can have harmful effects, become risk factors for radicalisation (Verkuyten, 2018), and generate unwanted negative side effects (Jerome et al., 2019). At the same time, Europe is witnessing the resurgence of exclusionary movements of an identitarian nature that strongly erode the values of the European Union. Alongside this, the increase in prejudice, 261

 Analysis of Radicalisation Prevention Policies

discrimination and violence in the streets are expressions of the difficulty of coexistence and the deep division of today’s society (Pérez-Latre, 2022). Some of these currents are close to Nazism; in some countries, especially in Northern Europe and the USA, they are extreme right-wing currents. Based on the guidelines set by the United Nations (General Assembly, 2016), the OECD is developing its strategy to define the learning required in societies that are changing rapidly and profoundly and in which social and cultural diversity is reshaping countries and communities (OECD, 2018b). A new multidimensional and lifelong learning objective, ‘global competence’, is introduced and assessed in the 2018 uProgramme for International Student Assessment (PISA) tests alongside the usual three core domains -reading, mathematics and science-. It seeks to assess how education systems are leading young people towards a diverse and peaceful society. Students are expected to learn to engage in dialogue, value other cultures, participate actively in social and political life and embrace the principles of solidarity (OECD, 2018a, 2020). The concept of global competence is a construct that originates in the work of Lambert (1993), who, in the worldwide context of globalisation, advocates incorporating a cosmopolitan perspective in education (Nussbaum, 1997; Hansen, 2017). This educational approach advocates reflective openness and creating an emotional bond towards the other, other cultures and diversity (López- Fuentes, 2020). In addition, different models of global education, citizenship education, democratic education, education for sustainable development and intercultural education are being developed which, from different approaches (cosmopolitanism, human rights, environmental sustainability or cultural diversity), share the objective of promoting understanding of the world and training for active and transformative participation in and for society (Sanz-Leal et al., 2022). PISA defines global competence as: ‘the ability to analyse global and intercultural issues, respect for human rights, interact with people from different cultures, take action for the common good and sustainable development’ (OECD, 2018a, p. 4). It is criticised for its marked Eurocentric approach (Auld & Morris, 2019; Grotlüschen, 2018), the lack of consensus and transparency in the design of the conceptual framework (Engel et al., 2019), the difficulty in assessing some dimensions of competence that undermine its validity (Sälzer & Roczen, 2018) or that its drive responds to partisan interests that wish to legitimise the ideas underpinning the global competence model (Ledger et al., 2019; Robertson, 2021). While Simpson and Dervin (2019) criticise the indistinct and interchangeable identification of democratic and intercultural competence, Barrett and Byram (2020) justify it. Thus, the fight against radicalisation is gaining prominence on international agendas. The United Nations promotes the Plan for the Prevention of Violent Extremism through Education (PVE-E) (General Assembly, 2016). In parallel, Western democratic societies need to respond to attacks by extremism carried out in the name of religious or ethnic affiliation (OECD, 2018a). Europe sees the need to protect the European spirit. It puts forward a strategy with multi-level actions, where “educommunication” becomes relevant to prevent radicalisation and hate speech, as it is a helpful tool that contributes to forming a critical public opinion within the 5.0 era. Educommunication is a discipline that deals with communication and education in an integrated and holistic way (Aguaded & Cabero-Almenara, 2022). It is an approach that seeks to use media and technology to improve education and, in turn, uses education to improve communication. It focuses on producing and consuming media messages and their impact on education and society. It aims to promote media literacy and the critical capacity of individuals to understand and evaluate the messages they receive through the media and thus make informed decisions (Cabero-Almenara, 2022; MartínezRodríguez & Sandoval-Esquivel, 2022). 262

 Analysis of Radicalisation Prevention Policies

In practice, educommunication involves using pedagogical strategies and tools that allow students to actively participate in the creation of media content actively and, in turn, reflect on the messages they receive from the media and their impact on society (Contreras-Pulido & García-Galera, 2022). This is achieved through various media such as television, radio, cinema, the Internet and social networks. Therefore, educommunication implies the relationship between two fields of study, education and communication (Aparici, 2010; García-Galera, 2022). It is also known as communication education, media didactics, educational communication, media literacy or communication pedagogy in the Latin American context, and media literacy or media education in the Anglo-Saxon context. According to Durán (2015), educommunicators insist on the need for communication to be participatory while allowing the use of technologies for educational and training purposes for the entire population, enabling the dissemination of information in favour of cultural diversity and not in favour of hate speech. Since the 1970s, UNESCO has been calling for the relevance of citizen training in aspects related to educommunication, with the training of teachers, as well as families and even the adult population in general, being of transcendental importance (Conde, 2017). UNESCO, as well as authors such as Stevenson (2018), have highlighted the need to empower citizens, reinforce democratic values, enabling training and critical capacity, together with the civic use of ICT (UNESCO, 2011, 2018). Communication in this field has an educational effect since, as Kaplún (1998) points out, emphasis is placed on learning content, its effects, and the capacity for social transformation in individuals and communities. 2018) or that it is driven by partisan interests that wish to legitimise the ideas underpinning the global competition model (Ledger et al., 2019; Robertson, 2021). In preventing radicalisation, it is essential to develop critical thinking skills and adequate digital literacy so that adolescents, especially those who are potentially vulnerable, acquire skills that enable them to develop their critical capacity to deal with messages of indoctrination and polarisation. This paper aims to analyse the attitudes that define interculturally competent citizenship (OECD, 2018a) and their presence in the radicalisation prevention policies of three Mediterranean countries -France, Portugal and Spain-. Its analysis will allow us to make some recommendations for the preventive policies of these countries that consider education as a key and principal element in the policies for preventing violent radicalisation. Educommunication can help eradicate radicalisation by promoting media literacy, developing critical skills, encouraging active participation and strengthening democratic values. In doing so, educommunication can contribute to preventing radicalisation by fostering a more informed, critical and participatory society.

METHODOLOGICAL FRAMEWORK The conceptual framework of global competence was defined following a lengthy process of coordination involving the ministries of education of the member states of the Council of Europe. Initially, 2085 different descriptors of global competence were analysed. After a lengthy analysis process in which each of the descriptors was scaled in an attempt to capture all cultures, it was simplified into 20 multivariables that allow for differentiation: 3 sets of values -valuing human dignity and human rights, valuing cultural diversity, valuing democracy, justice, fairness and equality and the rule of law-, 6 attitudes -openness to cultural otherness and to other beliefs, world views and practices, respect, civic-mindedness, responsibility, self-efficacy and tolerance of ambiguity-; 8 skills -autonomous learning skills, analytical and critical thinking skills, skills of listening and observing, empathy, flexibility and adaptability, linguistic, 263

 Analysis of Radicalisation Prevention Policies

communicative and plurilingual skills, cooperation skills and conflict-resolution skills- and 3 bodies of knowledge - knowledge and critical understanding of the self, knowledge and critical understanding of language and communication, knowledge and critical understanding of the world: politics, law, human rights, culture, cultures, religions, history, media, economies, environment, sustainability (Council of Europe, 2016a, 2016b). The analysis regarding attitudes is justified because, on the one hand, they are critical in adolescence and during school years and contribute to the construction of European citizenship (Santana-Vega et al., 2021). Therefore, the multivariable elements connected with attitudes are analysed: openness, respect, civic-mindedness, self-efficacy and tolerance, as conceptualised by the Council of Europe (Council of Europe, 2016a) and adopted by the OECD (OECD, 2018a). Within the European Union (EU), the Mediterranean region has 7 Member States (Spain, France, Italy, Portugal, Cyprus, Greece and Malta). In this research, we have focused on the first three, Spain, France and Portugal. The other four countries have been excluded because they do not have specific governmental plans to prevent radicalisation. Moreover, it is worth noting that none of these countries, like Portugal, have had attacks with fatalities caused by jihadist terrorism within their national territories. Table 1 lists the names of the documents analysed in their vernacular language. All the documents were issued in the 2015-2018 time frame in which the largest number of terrorist attacks occurred in Europe. Table 1. Official documentation of Mediterranean countries Year of Publication

Official Document

Country

2015

Plan Estratégico Nacional de Lucha contra la Radicalización Violenta. Un marco para el respeto y el entendimiento común

Spain

2015

Estratégia Nacional de Combate ao Terrorismo

2018

Prévenir Pour Protéger. Plan national de prévention de la radicalisation

Portugal France

Source: Elaborated by the authors.

This chapter analyses whether the attitudes of global competence (openness, world views and practices, respect, civic-mindedness, responsibility, self-efficacy and tolerance) are in government documents on preventing radicalism in three European Mediterranean countries, France, Portugal and Spain. These attitudes, assessed by PISA in 2018, define a democratic and interculturally competent citizenry. Lexicological analysis based on co-occurrence clustering or grouping techniques. This technique has been chosen because it is appropriate for the content analysis of texts and is suitable when there is a large amount of digitised documentation. The lexicographic analysis software Iramuteq (Interface de R pour les Analyses Multidimensionnelles de Textes et de Questionnaires) is used. Iramuteq is a textual statistics tool that applies multidimensional exploratory analysis methods to linguistic data. The software analyses texts with the logic of similarity search and coincidence tracking (Reinert, 1986, 1990). This methodology is appropriate because the software allows the analysis of the co-occurrence of words and lexical profiles -both words and lexemes- so that identifying networks of correlations and similarities is more exhaustive. This task allows the main lexical worlds of the text to be hierarchically grouped, arranged in hierarchical order, and statistical calculations to be performed. Textual discourse analysis templates enable the narrative’s general semantics to be identified and the main thematic clusters to be recorded (Lebart et al., 2000). However, such a large volume of relational

264

 Analysis of Radicalisation Prevention Policies

information is produced that it is illegible. The software generates approximately 800 co-occurrences (lexical units within a text corpus with lexical similarity among the included forms). Consequently, a pruning algorithm must be used to display the relevant information (Kamada & Kawai, 1989). As the project involves data mining analysis, the 80 co-occurrences with the highest value have been selected to produce colour graphs in which the results of this research project are visually represented. The resulting information is relational between forms, based on research objectives. The software generates an image of the ramifications of distinguished clusters, which unite related words given their proximity to the subject matter under study. The colours of clusters are random and distinguish standard blocks. The largest size graphically represents the most significant frequency of words, and the thickness of links shows the importance of their relationship: keywords are in the graph nodes and reflect the co-occurrence between them. Finally, textual analysis templates, including a quantitative and comprehensive vocabulary description, are produced to facilitate extracting non-explicit information from the texts.

RESULTS The analysis tool has produced the graphs shown below. Figure 1 corresponds to France, Figure 2 to Portugal and Figure 3 to Spain. Each image makes it possible to visualise the network of similarities that reflect the objectives and priorities of each country’s radicalisation prevention policies.

265

 Analysis of Radicalisation Prevention Policies

Figure 1. Network of common co-words generated on the documentation of France)

Source: Produced by the authors of this study with the methodology used in Azqueta and Merino-Arribas (2022)

France presented its plan in 2018, three years after the six simultaneous attacks in Paris in November 2015 at the Bataclan concert hall, other bars and restaurants in Paris and the Stade de France (Gouvernement Republique Française, 2018). As a result of that attack, 131 people died, and many were injured. It is considered the worst massacre in France since World War II. This was followed by the attack of significant social impact on July 14 2016, when a lorry ran over passers-by who were gathering on the Promenade des Anglais in Nice to celebrate France’s bank holidays. As a result of this attack, 86 people died. As a result of the 2015 attacks, France developed a plan characterised by its defensive approach in which coordination is critical (central word and thicker core that brings together the central red cluster). The French plan is conceived as a coordinated strategy that seeks to identify and assess the threat of terrorism. The following areas of radicalisation are identified: the family, prisons and the arrival of people from overseas or extraterritorial. It is a weak plan from a preventive point of view as it focuses prevention (green cluster) on policing, seeking information and detection. Another yellow cluster stands out, which refers to how the French

266

 Analysis of Radicalisation Prevention Policies

programme incorporates 60 measures (yellow cluster), among which repressive measures stand out. The plan’s primary focus is on detecting the threat, which is explicitly mentioned. Among the measures, schools, family relations, and social and health services are highlighted as crucial allies in the fight against radicalisation. The digital sphere is also included in the measures, as it has become one of the main areas of radicalisation. Likewise, research work and Internet surveillance are also objectives of the preventive measures, but from an eminently security-oriented perspective. The regulations aim for rapid detection, as extremists may be among their citizens. France details how to detect cases of radicalisation in schools with a prevention guide aimed at teachers and school principals. In education, it specifies how to protect against the risk of radicalisation through subjects such as digital literacy and moral and civic citizenship. Figure 2. Network of common co-words generated on the documentation of Portugal

Source: Produced by the authors of this study with the methodology used in Azqueta and Merino-Arribas (2022)

Portugal has proposed a very atomised counter-terrorism strategy (Conselho de ministros do Portugal, 2015) with a variety of themes and a lack of unity of purpose (9 globes are highlighted) with few points of connection between them. The most relevant word in the dark blue cluster is ‘European terrorism’,

267

 Analysis of Radicalisation Prevention Policies

which highlights how Portugal does not consider this strategy a priority but a concern external to the country. Portugal elaborates a strategy that is an extension of the EU’s strategic objectives and not a concern of national interest. The cluster that takes centre stage is the purple one, whose most relevant word is security. Portugal recognises the importance of the threat of jihadist terrorism, but its perception is less intense than that of European countries. The document follows in the Europeanist footsteps, focusing on security and cooperation with Europe. In this purple cluster, civil society is cited as a means of seeking protection (green cluster) and preventing radicalisation through various organisations. This cluster mentions the importance of being vigilant in recruitment spaces, including vigilance in the internet environment. The central purple cluster has points of connection with the orange cluster, which brings together several words related to cooperation, such as information, collaboration and protection. The light blue cluster is the next cluster with the strongest connection to security, which occupies the central place. The word that describes it and occupies the central node is service. Among the elements that stand out in this cluster is the explicit mention of education and other sectors of civil society in prevention. Figure 3. Network of common co-words generated on the documentation of Spain

Source: Produced by the authors of this study with the methodology used in Azqueta and Merino-Arribas (2022)

268

 Analysis of Radicalisation Prevention Policies

Concerning the Spanish plan (Government of Spain, 2015), it developed a national strategy of a markedly policing nature, focused on protecting public security and controlling terrorism. The strategy is from 2015 since Spain is part of the ‘Group of 9’ pioneers in adopting the first coordinated measures to combat radicalisation in the EU following UN Security Council Resolutions 2170 and 2178 (Security Council, 2014a, 2014b). It distinguishes three areas of action: internal, external (outside Spain) and cyberspace. While social networks are a space for the social expression of diversity and an arena for recognising differences and advocating for inclusion in all spheres of society, behind the anonymity, however, they can become a stage for the promotion and dissemination of stereotypes and prejudices, hate speech, radicalisation and violent extremism, among others (Marcos-Recio & Flores-Vivar, 2023). The plan follows in the wake proposed by the EU, which emphasises four fundamental pillars that appear among the principal words: preventing the causes of radicalisation and the recruitment of terrorists, protecting citizens and infrastructures, pursuing to reduce the capabilities of terrorists and responding to attacks in an appropriate manner. The plan seeks to prevent propaganda and explores terrorist financing networks. Spain has opted to develop a global strategy (pink cluster) whose main objective is elaborating local risk maps identifying dangerous micro-scenarios. Health, education, social services and even the neighbourhood are essential. The Spanish government’s plan does not explicitly reference the education system, although it considers it a cooperating actor. Education appears in this plan in a residual and generic way, focusing on social awareness and training in democratic values. It also incorporates the importance of monitoring cyberspace as an arena for recruitment and radicalisation. This approach is poor and paradoxical because Spain is the only European country with an agreement signed with the Islamic community in 1992 (Law 26/1992) that provides for teaching the Islamic religion in schools (Vega-Gutiérrez, 2023). After this plan, the Autonomous Communities of Catalonia and the Basque Country have made plans to prevent radicalisation. In the case of Catalonia, the terrorist attacks on the Ramblas in Barcelona and Cambrils on August 17 2017, in which 18 people died in addition to five of the terrorists and many wounded, were vital. The young men had been radicalised by an imam who acted as the cell’s ringleader. In 2017, the Basque Country drew up a Plan against international terrorism under religious pretexts for the early identification of radicalisation processes, in the drafting of which the Islamic communities of the Basque Country participated. Table 2 summarises whether the documents explicitly reference global competence attitudes. Table 2. Inclusion of global competence attitudes in official documents addressing the prevention of radicalisation in the three Mediterranean countries France Openness

X

Respect

X

Civic mindedness

X

Responsibility

X

Sef-efficacy

X

Tolerance

Portugal

Spain

X

X X X

Source: OECD (2018)

269

 Analysis of Radicalisation Prevention Policies

The analysis highlights that only respect is included in all three countries and highlights the lack of consistency of the Portuguese plan as it only includes this attitude in the development of its plan. No reference is made to any of the other attitudes. It is also surprising that the French plan does not include the attitude of tolerance.

SOLUTIONS AND RECOMMENDATIONS AND DISCUSSION The study of the presence or absence of the attitudes that define interculturally competent citizenship in the plans for the prevention of violent radicalisation in the three countries analysed leaves the following recommendations. Firstly, it is worth noting that by examining and comparing the radicalisation prevention policies of the three Mediterranean countries - France, Portugal and Spain - no unified conclusions can be drawn for the three countries, nor do they follow a common pattern. All three are countries in the Mediterranean basin and have been part of the EU for decades -France since 1958; Portugal and Spain since 1986-. Each of the countries has responded to prevention differently. The main factor differentiating them is the presence or absence of attacks on their national territory. If we collect the data for 2015-2022, when the strongest wave of jihadist violence occurred in Europe between 2015 and 2017, we can see that France and Spain have had attacks within their territory. However, Portugal has not had any attacks, and this element determines the level of priority, threat and social alarm this issue has provoked. While it is an extraterritorial problem for Portugal, the French plan considers that extremists may be among its citizens, or the Spanish plan, with the same perspective, designs a plan to detect dangerous micro-scenarios that local councils will monitor. In the same vein, as the results have shown, the Portuguese plan lacks strength and development and only addresses the need for respect among the attitudes of global competence. It does not incorporate any other of the attitudes of interculturally competent European citizenship in its development. Secondly, it is worth highlighting a general recommendation: the need to promote cross-cutting integration policies. A review of the French plan shows no reference to the need for integration or inclusion, nor is diversity mentioned as an added value. However, the figures and the reality provide other data. For example, it should be noted that in 2011, it was estimated that 30% of people under sixty were of foreign origin (Tribalat, 2015). France is also estimated to be the most significant Muslim-populated household in Europe, except Turkey (Ghosh et al., 2016). In 2019, immigrants, mostly of African origin and preferably from the Maghreb, accounted for 9.9% of the total population, making the management of cultural and religious diversity a relevant issue. The values are inherent to the idiosyncrasies of the French Republic -liberté, égalité, fraternité- which guaranteed and symbolised national cohesion have a moral aspect because they embody in practice concepts inherited from the Revolution, in addition to secularism as a requirement for coexistence. This perspective is in clear opposition and conflict with Islam. In essence, it embodies an assimilationist model of diversity management that can lead to violating citizens’ rights and generating resentment. The French model of secularism is exclusionary, unlike the Spanish, Italian or German models. The failure of French integration corroborates that this model needs to be revised and not further hardened, admitting only one type of diversity in the public sphere. In fact, among the attitudes of global competence, as the results have shown, they do not include the attitude of tolerance as an essential and indispensable element to define interculturally competent citizenship. The French government has recently enacted a law entitled “law to reinforce respect for republican principles” of 270

 Analysis of Radicalisation Prevention Policies

August 24 2021, and the government has also promoted the drafting by the French Council of Muslim Worship (CFCM) of a “charter of principles” for Islam in France (Assemblée Nationale Française, 2021). However, this proposal was rejected by four of the nine organisations that make up this federation, which according to Poyet (2021), refused to sign it, so it can be affirmed that the Republic is experiencing difficulty in asserting its principles and values. relevant (Georgeault, 2022). Cross-cutting policies must address the severe identity problem that underlies many problems. The concept of integration is bidirectional. The European Commission (2003) defines it as a two-way process based on the reciprocal rights and corresponding obligations of legally resident third-country nationals and the host society, allowing for the full participation of migrants” (p. 19). It concerns all parties involved in the new contexts of coexistence and requires that integration models assume an intercultural approach (European Commission, 2020). Integration implies a positive valuation of difference in which the other is respected and integrated for being who he or she is, a human person. Combining European identity with other cultural, national, racial, and religious identities is complex. Cross-cutting policies are required to facilitate improvement in the socio-economic, political-institutional, demographic-urban, and educational spheres. We see an essential difference in time. In southern European countries such as Spain and Portugal, immigration movements of Muslim communities are more recent, but in France, as in other EU member states, the third generation of Muslim immigrants has already reached adolescence. Bagus et al. (2022) rightly consider how Spain has addressed the need to understand islam through education to erode intolerance within the framework of democracy. Along the same lines, Vega-Gutiérrez (2023) analyses the international frameworks that address education as a tool for preventing violent extremism and contrasts them with the Islamic religion curriculum recently approved in Spanish educational legislation (Organic Law 3/2020). The balance is positive and does not include specific curricular elements for preventing violent radicalisation, but it does emphasise alternative narratives that abound in shared values and behaviours that help to prevent radicalisation. An intercultural approach is needed, which facilitates integration and builds democratic citizenship in practice. Policies to prevent radicalisation should address the root identity issues mainly affecting second and third-generation Muslims settled in Europe. The humanistic and solidarity-based orientation is at the origin and the heart of Europe’s history and sets the guidelines for the education of Europeans. These basic approaches must not be lost sight of and must govern the European project now and in the future. The EU is the fruit of a political will of understanding between the Member States that comprise the Union. This will be initially based on an economic foundation conceived as a means of union and integration, which aims to overcome enmity between peoples, prevent wars, and respect human rights and democracy. European integration begins with the easiest part, the economy, and has a gradualist projection regulated by the member states. Social, educational, and cultural integration is progressive, implemented gradually, functional, and responds to concrete goals. Each Member State brings its differentiating character, considered a positive value (Brunet, 2014). All these aspects imply that the management of European policy is guided by objectives that reflect the values that shape the EU, such as cooperation, integration, equality, and competitiveness. Democracy is supposed to create resilient societies resistant to terrorism. However, research on whether democracy prevents terrorism is inconclusive. For Jonnanson (2022), insufficient democratisation can stimulate terrorism. In this context, there are significant challenges for young people with an immigrant background, as identity formation is an important task that occurs in adolescence and influences later psychosocial development. Although it is impossible to forge a profile of radicalised individuals, they share some common characteristics, such as the Islamic religion, a complex socio-economic situation, and a loss of cultural identity (Rabasa and 271

 Analysis of Radicalisation Prevention Policies

Benard, 2015; Municio, 2017). Many of these young people, coming from second or third generations of immigrants, face dilemmas regarding their cultural identity. They need to find a balance and harmonise dilemmas that can often be contradictory (authority-freedom, autonomy-community, innovation-continuity, among others). They live as a minority in historically Christian societies that are increasingly secularised and may experience a double sense of not belonging because they do not feel part of the community of their parents and, at the same time, perceive rejection by the host society. Radicalisation arises from the perception of inter-group conflict in which relationships become charged with prejudice and in which uncertainty and lack of identity accelerate the process of radicalisation. Thirdly, it should be emphasised that for thirdly, it should be stressed that prevention requires the educational community as a critical element in counteracting extremist thinking, including in the teaching of information literacy among pupils, given that it is not enough for them to know their way around the web, as they are digital natives. However, they need this competence to be able to discriminate between the inappropriate information they receive online and to safeguard them from the risks involved in entering into radical messages. In this sense, educommunication is presented as a powerful medium or strategy that facilitates the construction and reconstruction of subjectivities based on the development of capacities or competencies that subjects need to participate, express their opinions with critical awareness and collectively create ways and strategies to transform the realities in which they live (Buitrago-Alonso et al., 2017). As Gozálvez and Contreras Pulido (2014) indicate, educommunication plays an essential role in the processes of citizen and social participation insofar as its function is to train and inform citizens about the new ways of relating and the rules established in the new social order, influenced by cyberspace, virtuality, and new technologies. One of the functions of educommunication in the processes of participation is to serve as a tool to shorten distances between people, to show their ideas about themselves as citizens, to promote critical and self-critical exchanges that nurture interactions, to respect and accept the other from their diversity and to promote the recognition of new citizens (Fedorov et al., 2016). In this way, educommunication contributes valuable strategies to the visibility, recognition and self-recognition of cultural identity, political stances, ways of understanding the world, social practices, and diverse forms of interaction, which allow some to show the existence and presence of others, who historically have not appeared as legitimate protagonists to contribute relevant elements for the construction and transformation of societies (Chib et al., 2019). Undoubtedly, it is necessary to train in educommunicative competencies, as media learning has become a necessary element in today’s society, where skills, competencies and attitudes are developed in citizens (Bonilla del Río et al., 2018). The mere presence of the media and their development through social platforms is no guarantee of the acquisition of educational-communicative competencies by citizens, so it is advisable to adapt the training of subjects according to their contexts, with the ultimate aim of improving the channels and learning around the media, social networks, platforms or other existing tools. Education and communication enable channels from which images and representations of the world are generated, so, according to Osuna et al. (2018), it is crucial that the entire population, and especially the educational community, be aware of what the media offer selective versions of the world, rather than direct access to it. In this line, it is necessary to train citizens not only as consumers but also as producers (prosumers) who possess a series of competencies that allow them to develop actions both as consumers of media and audiovisual resources and as producers and creators of critical, ethical and responsible messages and content (García-Ruiz et al., 2014).

272

 Analysis of Radicalisation Prevention Policies

The use of technologies as a means of communication and participation of the educational and social community has become relevant as these tools are ideal for developing communication channels where participation, social mobilisation and training are essential (Conde, 2017). Forming media citizens empowers citizens in the plural and democratic hyper-communicated societies, thus allowing them to combat hate speech and radicalisation (Gozálvez & Contreras Pulido, 2014). We must all make a commitment to prevent a “post-truth” world in which nothing is true, and everything is possible (Grotlüschen, 2018). For this reason, multiple pedagogical initiatives aimed at promoting respect for difference have emerged to foster the value of coexistence in diversity, not only in schools but also in the social sphere, as a complement to the institutional sphere (Carratalá & Herrero-Jiménez, 2017). These initiatives carried out as educational prevention against the development of psychosocial processes of extremist radicalisation, are proactive actions in which teachers are defined as decisive actors in promoting respect for diversity and mutual understanding, both in schools and in society. Key to this task is the training of teachers because they are confronted daily with the behaviour and ideas of students in the classroom, a true social microcosm. When it comes to extreme manifestations, whatever the origin and reason for these expressions (grievance, injustice, peer pressure), the educator alone is faced with the challenge of identifying and addressing these issues safely and openly, and this task requires specific investment and support to increase their capacity to engage and act. Additional and specific teacher training is needed to address these challenges. However, it also provides concrete approaches to help in the classroom, such as collaboration with other social actors such as NGOs, support from external education networks and the incorporation of victim testimonies. Nothing can be achieved without the collaboration of education; any European, national, or regional plan will inevitably have to rely on teachers to correct misunderstandings, dispel clichés and show the positive side. The challenge of education today is to reconcile a critical sense with human connections. Educators must go beyond transmitting knowledge about a subject and be facilitators of life skills development. Today’s school requires incorporating new cognitive content into the curriculum and values such as socio-emotional values, information education and education for peace. Changing the school does not guarantee absolute success, but it can only help prevent extremist radicalism if we do not shy away from discussion in the classroom, always ensuring an appropriate methodology and environment for debating controversial issues. Students must learn to deal with controversy and messages that can lead to violent extremism. Civiccitizenship education should help students acquire the knowledge, skills and understanding to play an active and full role in society and equip them to critically explore political and social issues and calibrate messages and information on social media to debate and develop reasoned arguments. However, to date, it can be said that the prevention of violent radicalisation from the educational sphere is a field under construction with projects that are primarily incipient and still pending evaluation of what works and how it can be improved and is preferably focused on tertiary prevention actions that are mainly aimed at the disengagement, de-radicalisation and rehabilitation of terrorists, and preventive actions are very scarce (Vallinskosky et al., 2022; Vicente & García-Calvo, 2020). In this environment, therefore, it is necessary to educate to build resilience that challenges stereotypes through interfaith and ethics education initiatives, in addition to fostering opportunities for interaction between different and culturally diverse members of the same community, for which it is essential to help them appreciate differences and complexity (McNeil-Willson et al., 2019). Other aspects to consider include inter-religious dialogue, religious education, youth civic engagement, youth political dialogue and peace education that delegitimises violent extremism as practices that are protective resources against radicalisation and 273

 Analysis of Radicalisation Prevention Policies

polarisation (Weine & Osman, 2012). In addition, family support is essential to enable the forging of a flexible cultural identity (Weine, 2017) and thus avoid cultural, religious and familial isolation. Finally, as a fourth consideration, it is crucial to show the value of friendship. At school, bonds of friendship are created, prejudices are diminished, and relationships and interactions between individuals are forged (Fleischmann & Phalet, 2018). Through the relationship between peers and fundamentally through friendship, one learns to accept and value identities that are different from one’s own, which makes the school an environment that promotes citizenship. Educommunication can be a powerful tool for preventing radicalisation, as it focuses on developing skills and values that can help individuals resist and reject extremist messages and narratives (Aguaded & Cabero-Almenara, 2022). Some ways in which educommunication can contribute to the prevention of radicalisation are: 1. Promoting Cultural Diversity: Educommunication can help promote cultural diversity by fostering understanding and respect for cultural and religious differences. This can reduce the risk of radicalisation by promoting integration and intercultural dialogue; Develop critical skills: Educommunication can help develop critical skills, such as assessing the quality and veracity of information and media. This can help individuals identify and resist extremist messages they may receive through the media. 2. Encourage Active Participation: Educommunication can encourage the active participation of individuals in the production and consumption of media content. This can help individuals develop their voice and perspective and be less susceptible to the influence of extremist messages. 3. Strengthening Democratic Values: Educommunication can help strengthen democratic values, such as tolerance, respect for human rights and freedom of expression. This can reduce the risk of radicalisation by promoting a more inclusive and just society (Contreras-Pulido & García-Galera, 2022).

FUTURE RESEARCH DIRECTION For the future, it is worth noting that the sample of countries and European regions to be analysed can be expanded of countries and European regions to be analysed, as well as incorporate new categories of analysis. In addition, a study could include European Autonomous Communities or regions with specific regulations in this area.

CONCLUSION The analysis has highlighted that the prevention plans of the countries analysed have two differentiating features. Firstly, the presence or absence of victims in the attacks committed on the territory is an important differentiating factor. Thus, while Spain and France have specific prevention plans, and this is considered a relevant problem, for Portugal, where there have been no victims in the attacks, it is considered an extraterritorial problem, and only unity with European countries leads to the development of plans at the legislative level. Secondly, the presence of the Muslim population in the territory for more

274

 Analysis of Radicalisation Prevention Policies

or less time also marks essential differences. While third generations of Muslim emigrants in France are frequent in the country, in Spain and Portugal, the migratory phenomenon is more recent in time. Furthermore, the analysis has led to the conclusion that there is a need to promote cross-cutting radicalisation prevention policies that do not focus solely and exclusively on the need for security, but that, from a holistic perspective, address the identity issues are at the heart of the problem. It is also recommended to develop preventive policies that involve the educational community to counteract extremist thinking and develop educommunicative skills. Overall, radicalisation prevention policies from the perspective of educommunication in Mediterranean countries involve a holistic approach encompassing media literacy education, intercultural dialogue, and digital literacy. By promoting these policies, Mediterranean countries can create a more informed and engaged citizenry better equipped to resist extremist ideologies and promote a more peaceful and tolerant society.

ACKNOWLEDGMENT This chapter is part of the DiverProf (B0036) project, executed with the collaboration of Universidad Internacional de La Rioja (Spain) and it is part of the I+D+I project PID2020-114584GB-I00, funded by Agencia Estatal de Investigación - Ministerio de Ciencia e Innovación (Spain).

REFERENCES Aguaded, J. I., & Cabero-Almenara, J. (Eds.). (2022). Handbook of Research on Digital Education and Digital Technologies. IGI Global. Aly, A., Balbi, A. M., & Jacques, C. (2015). Rethinking countering violent extremism: Implementing the role of civil society. Journal of Policing, Intelligence and Counter Terrorism, 10(1), 3–13. doi:10.1 080/18335330.2015.1028772 Aparici, R. (2010). Educomunicación: más allá del 2.0. Gedisa. Assemblée Nationale Française. (2021). Loi nº 2021-1109 du 24 août 2021 confortant le respect des príncipes de la République. https://cutt.ly/B8IfzMq Auld, E., & Morris, P. (2019). Science by streetlight and the OECD’s measure of global competence: A new yardstick for internationalisation? Policy Futures in Education, 17(6), 677–698. doi:10.1177/1478210318819246 Azqueta, A., & Merino-Arribas, A. (2022). The role of education in Preventing Violent Radicalisation: A review for Europe. Review of Education, 397(3), 209–238. doi:10.4438/1988-592X-RE2022-397-54 Bagus, A. M., Darajaatul, N., Azizah, M., Amal, B. K., & Purba, A. S. (2022). Deradicalisation Model Through Islamic Education Curriculum in Indonesia, Spain, And Nigeria. Res Militaris, 12(2), 428–439.

275

 Analysis of Radicalisation Prevention Policies

Barrett, M., & Byram, M. (2020). Errors by Simpson and Dervin (2019) in their description of the Council of Europe’s Reference Framework of Competences for Democratic Culture. Intercultural Communication Education, 3(2), 75–95. doi:10.29140/ice.v3n2.286 Bayrakli, E & Hafez, F. (2020). European Islamophobia Report 2019. Foundation for Political, Economic and Social Research (SETA). Bonilla del Rio, M., Valor-Rodríguez, L., & García-Ruiz, R. (2018). Alfabetización mediática y discapacidad: Análisis documental de literatura científica en web of science (WOS) y scopus. Prisma Social, 20, 1–20. https://revistaprismasocial.es/article/view/2292/2472 Brunet, F. (2014). Curso de integración europea. Alianza. Buitrago-Alonso, A., García Matilla, A., & Gutiérrez Martín, A. (2017). Perspectiva histórica y claves actuales de la diversidad terminológica aplicada a la educación mediática Historical. Revista de Educación Mediática y TIC, 6(2), 81–104. https://doi.org/https://doi.org/10.21071/edmetic.v6i2.7002 Cabero-Almenara, J. (2022). Educomunicación en tiempos de pandemia: Análisis de las prácticas educativas en la educación digital. Comunicar: Revista Científica de Comunicación y Educación, 30(69), 11–20. Carratalá, A., & Herrero-Jiménez, B. (2017). Comunicación contra el odio: Análisis del estado de la discriminación en España y propuestas de actuación política, educativa y mediática. Fundación Alternativas, 91, 2017. Cesari, J. (2013). Why the west fears Islam: An exploration of Muslims in liberal democracies. Springer. doi:10.1057/9781137121202 Chib, A., Bentley, C., & Wardoyo, R. (2019). Entornos digitales distribuidos y aprendizaje: Empoderamiento personal y transformación social en colectivos discriminados. Comunicar, 27(58), 51-61. https:// doi.org/https://doi.org/10.3916/C58-2019-05 Conde, M. A. (2017). Comunicación, movilización y participación desde las redes sociales en un proyecto social. In Re-construyendo un mundo con ojos de niñas. Entre la pobreza y la educación (165-174). GEU Editorial. Consejo de Seguridad. (2014a). Resolución 2170 (2014). Amenazas a la paz y la seguridad internacionales causadas por actos terroristas. Aprobada por el Consejo de Seguridad, 7242a sesión, 15 de agosto 2014. https://www.un.org/securitycouncil/es/s/res/2170-%282014%29 Consejo de Seguridad. (2014b). Resolución 2178 (2014). Amenazas a la paz y la seguridad internacionales causadas por actos terroristas. Aprobada por el Consejo de Seguridad, 7272a sesión, 24 de septiembre 2014. https://www.un.org/securitycouncil/es/s/res/2178-%282014%29 Conselho de ministros do Portugal (2015). Estratégia Nacional de Combate ao Terrorismo. Diário da República, 1(36). https://bit.ly/3lkWZsO Contreras-Pulido, P., & García-Galera, M. C. (Eds.). (2022). Tendencias educativas y tecnológicas: El desafío de la innovación en educación. Dykinson.

276

 Analysis of Radicalisation Prevention Policies

Coolsaet, R. (2019). Radicalisation: The origins and limits of a contested concept. In N. Fadil (Ed.), Radicalisation in Belgium and the Netherlands: Critical Perspectives on Violence and Security (pp. 29-51). IB Tauris. Council of Europe. (1949). Statute of Council of Europe, London, May 5, 1949. https://rm.coe. int/1680306052 Council of Europe. (2016a). Competences for democratic culture: Living together as equals in culturally diverse democratic societies. Council of Europe. https://rm.coe.int/16806ccc0c Council of Europe. (2016b). Reference framework of competences for democratic culture: guidance for implementation (I-III). Council of Europe. https://xurl.es/e75m5 Durán, D. (2015). Tutoría entre iguales: compartir la capacidad de enseñar con los alumnos. In Mayordomo, R. & Onrubia, J. (coord.). El aprendizaje cooperativo. Editorial UOC. Engel, L., Rutkowski, D., & Thompson, G. (2019). Toward an international measure of global ompetence? A critical look at the PISA 2018 framework. Globalisation, Societies and Education, 17(2), 117–131. doi:10.1080/14767724.2019.1642183 European Commission (2003). Communication from the Commission to the Council, the Council, the European Parliament, The European Economic and Social Committee and the Committee of the Regions, 3.06.2003. On immigration, integration and employmen. COM (2003) 336 final. European Commission (2020). Communication from the Commission to the European Parlament, the European Council, the Council, the European Economic and Social Committee and the Committee of the Regions, 9.12. 2020. A Counter-Terrorism Agenda for the EU: Anticipate, Prevent, Protect,Respond. COM (2020) 795 final. Eurydice. (2016). Promoting citizenship and the common values of freedom, tolerance and non-discrimination through education: Overview of education policy developments in Europe following the Paris Declaration of March 17 2015. Luxembourg: Publications Office EU. doi:10.2797/396908 Eurydice. (2019). Integrating students from migrant backgrounds into schools in Europe: national policies and measures. Louxembourg: Education, Audiovisual and Culture Executive Agency. doi:10.2797/819077 Fedorov, A., Cortés Montalvo, J. A., & Sandoval Romero, Y. (2016). La educación en comunicación en el mundo: currículum y ciudadanía. Comunicar, 24(49). https://www.revistacomunicar.com/ojs/index. php/comunicar/issue/view/c49 Fleischmann, F., & Phalet, K. (2018). Religion and national identification in Europe: Comparing muslim youth in Belgium, England, Germany, the Netherlands, and Sweden. Journal of Cross-Cultural Psychology, 49(1), 44–61. doi:10.1177/0022022117741988 PMID:29386688 García-Galera, M. C., & Contreras-Pulido, P. (Eds.). (2022). Comunicación y Educación: Nuevas perspectivas en investigación e intervención. Dykinson. García-Ruiz, R., & Ramírez-García, A., & Rodríguez-Rosel, M. (2014). Educación en alfabetización mediática para una nueva ciudadanía prosumidora. Comunicar, 22(43), 15–23. doi:10.3916/C43-2014-01

277

 Analysis of Radicalisation Prevention Policies

General Assembly. (2016). Resolución 70/L.41. Culture of peace. Plan of Action to Prevent Violent Extremism. Seventieth session (February 9 2016). Georgeault, L. (2022). Ley de refuerzo del respeto de los principios republicanos en Francia. ¿El regreso del asimilacionismo? Eunomía. Revista en Cultura de la Legalidad, 22(22), 79–100. doi:10.20318/ eunomia.2022.6808 Ghosh, R., Manuel, A., Chan, A., Dilimulati, M., & Babaei, M. (2016). Education and Security. A Global Literature Review on the role of education in Cuntering Violent Religious Extremism. Tony Blair Faith Foundation. Gobierno de España. (2015). Plan Estratégico Nacional de Lucha contra la Radicalización Violenta. Un marco para el respeto y el entendimiento común. Madrid: Ministerio del Interior. https://www.interior. gob.es/opencms/pdf/servicios-al-ciudadano/plan-estrategico-nacional-de-lucha-contra-la-radicalizacionviolenta/documentacion-del-plan/estrategia-interior/PLAN-ESTRATEGICO-NACIONAL.pdf Gouvernement Republique Française. (2018). Prévenir Pour Protéger. Plan national de prévention de la radicalisation. https://www.cipdr.gouv.fr/prevenir-la-radicalisation/#:~:text=d%C3%A9tecter%20et%20 signaler%2C%20via%20notamment,de%20la%20radicalisation%20(CNAPR)%3B Gozálvez, V., & Contreras Pulido, P. (2014). Empoderar a la ciudadanía mediática desde la educomunicación. Comunicar, 21(42), 129–136. doi:10.3916/C42-2014-12 Grotlüschen, A. (2018). Global competence. Does the new OECD competence domain ignore the global South? Studies in the Education of Adults, 50(2), 185–202. doi:10.1080/02660830.2018.1523100 Hansen, D. (2017). The Teacher and the World: A Study of Cosmopolitanism as Education. Routledge. doi:10.4324/9780203803325 Innerarity, D. (2006). El nuevo espacio público. Espasa-Calpe. Jerome, L., Elwick, A., & Kazim, R. (2019). The impact of the Prevent duty on schools: A review of the evidence. British Educational Research Journal, 45(4), 821–837. doi:10.1002/berj.3527 Jonasson, A. K. (2022). The double-edged sword of democracy: How democracy promotion is used in the European Union’s prevention of terrorism in the southern Mediterranean. Journal of Regional Security, 17(2), 241–266. doi:10.5937/jrs17-37878 Kamada, T. & Kawai, S. (1989). An agorithm for drawing general undirected graphs. Information Pocessing Letters, 31, 7-15. Kaplún, M. (1998). Una pedagogía de la comunicación. Ediciones de la Torre. Kepel, G. (2016). El terror entre nosotros. Una historia de la yihad en Francia. Península. Lahnait, F. (2018). Prevenir el extremism violento para combatir el yihadismo autóctono: Aprender de la experiencia. CIDOB Report. Lambert, R. (1993). Educational Exchange and Global Competence. 46th International Conference on Educational Exchange. https://files.eric.ed.gov/fulltext/ED368275.pdf

278

 Analysis of Radicalisation Prevention Policies

Lebart, L., Salem, A., Bécue, M., & Peña, D. (2000). Análisis estadístico de textos. Milenio. Ledger, S., Their, M., Baile, L. & Pitts, C. (2019). OECD’s approach to measuring Global Competency: Powerful voices shaping education. Teachers College Record, 121, 1-40. doi:10.1177/016146811912100802 Ley 26/1992, de 10 de noviembre, por la que se aprueba el Acuerdo de Cooperación del Estado con la Comisión Islámica de España. Boletín Oficial del Estado, 272, de 12 de noviembre de 1992, 3821438217. doi:10.1177/016146811912100802 Ley Orgánica 3/2020, de 29 de diciembre, por la que se modifica la Ley Orgánica 2/2006, de 3 de mayo, de Educación. Boletín Oficial del Estado, 340, de 30 de diciembre de 2020, 122868-122953. López-Fuentes, V. (2020). La educación cosmopolita como enfoque educativo para trabajar la inclusión, la diversidad y el vínculo emocional en el aula. Dilemas contemporáneos: Educación. Política y Valores, 8(1), 1–20. Marcos-Recio, J. C., & Flores-Vivar, J. M. (2023). Social Media contributes to radicalisation. How to prevent its expansion? In C. Melacarne & L. Fabbri (Eds.), Understanding Radicalization in everyday life (pp. 125-142). McGraw Hill Education. Martínez-Rodríguez, J. B., & Sandoval-Esquivel, A. (2022). Educomunicación y medios digitales: Perspectivas teóricas y prácticas en la educación superior. Dykinson. McNeil-Willson, R., Gerrand, V., Scrinzi, F. & Triandafyllidou, A. (2019). Polarisation, violent extremism and resilience in Europe today: An analytical framework. BRaVE Project. Municio, N. (2017). Evolución del perfil del yihadista en Europa. Boletín del Instituto Español de Estudios Estratégicos, 6, 556–573. Nussbaum, M. (1997). Cultivating Humanity: A Classical Defense of Liberal Education. Harvard University Press. OECD. (2018a). Preparing our youth an inclusive and sustainable world. The OECD PISA global competence framework. Paris: OECD. https://www.oecd.org/education/Global-competency-for-aninclusive-world.pdf OECD. (2018b). The Future of Education and Skills OECD Education 2030 Framework. Paris: OECD Publishing. https://www.oecd.org/education/2030-project/ OECD. (2020). PISA 2018 Results (Volume VI): Are Students Ready to Thrive in an Interconnected World? OECD Publishing. doi:10.1787/d5f68679-en Osuna, S., Frau-Meigs, D., & Lazo, C. (2018). Educación Mediática y Formación del Profesorado. Educomunicación más allá de la Alfabetización Digital. Revista Interuniversitaria de Formación del Profesorado, 91(32), 29-42. http://hdl.handle.net/11162/179663 Pérez-Latre, F. (2022). Crisis de confianza (2007-2022). Eunsa. Poyet, S. (2021). DOCUMENT – Charte des principes pour l’islam de France rédigée par le CFCM. Le Figaro. https://acortar.link/Z6Jitg

279

 Analysis of Radicalisation Prevention Policies

Rabasa, A., & Benard, C. (2015). Eurojihad. Patterns of Islamist Radicalization and Terrorism in Europe. Cambridge University Press. Reinert, M. (1986). Un logiciel d’analyse lexicale: Alceste. Les Cahiers de l’Analyse des Données, 11(4), 471–484. Reinert, M. (1990). Alceste: une méthodologie d’analyse des donne textuelles et une application: Aurélia de G. de Nerval. Bulletin de Methodologie Sociologique, 26(1), 24–54. doi:10.1177/075910639002600103 Robertson, M. (2021). Provincialising the OECD-PISA global competences project. Globalisation, Societies and Education, 19(2), 167–182. doi:10.1080/14767724.2021.1887725 Roy, O. (2017). Jihad and death. The global appeal of Islamic State. Hurst & Company. Ruiz-Díaz, L. (2017). La prevención de la radicalización en la estrategia contra el terrorismo de la Unión Europea: Entre soft law e impulso de medidas de apoyo. Revista Española de Derecho Internacional, 69(2), 257–280. doi:10.17103/redi.69.2.2017.1.10 Sälzer, C., & Roczen, N. (2018). Assessing global competence in PISA 2018: Challenges and approaches to capturing a complex construct. International Journal of Development Education and Global Learning, 10(1), 5–20. doi:10.18546/IJDEGL.10.1.02 Santana-Vega, L. E., González-Morales, O., & Feliciano-García, L. (2021). Are we Europeans? Secondary education students’ beliefs and sense of belonging to the European Union. Journal of Youth Studies, 24(10), 1358–1372. doi:10.1080/13676261.2020.1839634 Sanz-Leal, M., Orozco, M., & Toma, R. (2022). Construcción conceptual de la competencia global en educación, Teoría de la Educación. Teoría de la Educación, 34(1), 83–103. doi:10.14201/teri.25394 Simpson, A., & Dervin, F. (2019). Global and intercultural competence for whom? By whom? For what purpose? An example from the Asia Society and the OECD. Compare: A Journal of Comparative Education, 49(4), 672–677. doi:10.1080/03057925.2019.1586194 Stevenson, N. (2018). Education and the alterity of democracy. Cadernos de Pesquisa, 48(167), 150–171. doi:10.1590/198053144668 Tribalat, M. (2015). Une estimation des populations d’origine étrangère en France en 2011. Espace, Populations, Sociétés, (1-2), 1–2. doi:10.4000/eps.6073 Troian, J., Baidada, O., Arciszewski, T., Apostolidis, T., Celebi, E., & Yurtbakan, T. (2019). Evidence for indirect loss of significance effects on violent extremism: The potential mediating role of anomia. Aggressive Behavior, 45(6), 691–703. doi:10.1002/ab.21863 PMID:31452219 Trujillo, H. M., & Moyano, M. (2008). El sistema educativo español en la prevención de la radicalización yihadista. Athena Intelligence Journal, 3(2), 75–83. UNESCO. (2011). Media and information literacy curriculum for teachers. https://unesdoc.unesco.org/ ark:/48223/pf0000192971 UNESCO. (2018). Media and information literacy. https://unesdoc.unesco.org/ark:/48223/pf0000265509

280

 Analysis of Radicalisation Prevention Policies

Vallinskosky, K., Benjamin, S., & Elliot, K. (2022). By correcting disinformation and encouraging them towards tolerance: Finnish educators’ considerations of PVE-E. Journal for Deradicalization, 23(33), 181–220. Vega-Gutiérrez, A.M. (2023). La educación como herramienta de prevención del radicalismo violento: aportaciones del curriculum de religión islámica desde un enfoque competencial. Anuario de Derecho eclesiástico del Estado. Verkuyten, M. (2018). Religious fundamentalism and radicalisation among Muslim minority youth in Europe. European Psychologist, 23(1), 21–31. doi:10.1027/1016-9040/a000314 Vicente, Á., & García-Calvo, C. (2020). Evaluación de proyectos a pequeña escala en materia de prevención de la radicalización violenta: Una visión aplicada. Análisis del Real Instituto Elcano, 121(1), 1–14. Weine, S. (2017). Resilience and Countering Violent Extremism. In U. Kumar (Ed.), International Handbook of Psychosocial Resilience. Routledge. Weine, S., & Osman, A. (2012). Building Resilience to Violent Extremism Amongst Somali-Americans in Minneapolis-St. Paul. Final Report. National Consortium for the Study of Terrorism and Responses to Terrorism (START). Department of Homeland Security Science and Technology Center of Excellence. University of Maryland.

KEY TERMS AND DEFINITIONS Educommunication: Integrates education and participatory communication through digital media, enabling the training of the population as consumers and prosumers of information in favour of diversity and the eradication of discrimination and hate speech in the world. Extremist Violence: There is no agreed definition of violent extremism. It is usually identified as terrorism. The United Nations General Assembly notes that violent extremism is a broader concept than terrorism as it includes ideologically motivated forms of violence that do not amount to terrorist actions. Global Competence: Is the ability to analyse global and intercultural issues, with respect for human rights, to interact with people from different cultures, to take action for the common good and sustainable development. Integration: Is a concept that seeks to respond to diversity, taking into account that it is the system and society at all levels that adjusts to the needs of all. Media Education: Training in skills and competences that enable people to act in social media, understanding, analysing and evaluating the information consulted and rethinking the information or content that is intended to be shared. Media education enables the analysis of the meaning of social media for individuals and societies. PISA: Acronym of the Programme for International Student Assessment by which the OECD (Organisation for Economic Cooperation and Development) assesses the education of students when they reach the end of compulsory education and before their possible integration into working life, around the age of 15. The PISA report is intended as a means for countries to improve the level of their education systems. The assessment covers the areas of reading, mathematics and science proficiency, plus a different non-compulsory area that is assessed each year. In 2018, global competence was assessed.

281

 Analysis of Radicalisation Prevention Policies

Preventing Violent Extremist (PVE): Prevention is one of the four pillars of the EU security strategy in line with UN guidelines. It develops tools, programmes, networks and strategies to counter terrorism, counter radicalisation and counter violent extremism. The strategy to prevent radicalisation is considered to be an area for the sovereign authority of each EU member state. Radicalisation: Radicalisation is understood as the process through which a person sympathises with and legitimises approaches and ideologies that condone the use of violence.

282

283

Chapter 15

New Narratives to Defuse Hate Speech Maximiliano Bron https://orcid.org/0000-0002-7596-4598 Universidad Nacional de La Rioja, Argentina Óscar Javier Arango Arboleda https://orcid.org/0000-0002-4693-2212 Universidad de Barcelona, Spain Angelica María Rodríguez Ortiz https://orcid.org/0000-0002-7710-9915 Univdersidad Autónoma de Manizales, Colombia Héctor Claudio Farina Ojeda https://orcid.org/0000-0002-8476-4192 Universidad de Guadalajara, Mexico

ABSTRACT According to a 2021 report by the Spanish government, hate speech has increased by 60%, with 90% of survey respondents experiencing humiliation that amounts to “hate crimes.” UNICEF has also reported a 13% increase in hate speech among young people in Latin America. Both institutions have responded with regulations and campaigns to combat hate in educational systems and society at large. This chapter presents new narratives that have been used in the Musik Thinking (Barcelona) and CoCritic.Ar (Latin America) projects, in which, through education, mechanisms are provided to strengthen critical thinking and initiate processes to deactivate hate speech, especially those directed towards immigrants. The results of the process show how music in music thinking and critical literacy, collaborative work, argumentation, and transmedia narratives in CoCritic.Ar generate spaces for citizen discussion to respond to the increase in hate speech occurring in the Mediterranean and Latin America.

DOI: 10.4018/978-1-6684-8427-2.ch015

Copyright © 2023, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

 New Narratives to Defuse Hate Speech

INTRODUCTION Social reality has been created, instituted, and intentionally transformed through speech utterance (Searle, 1969). The uses of language can be constructive and transformative. However, they can also be destructive when they semantically invoke derogatory meanings. Given the intention underlying an expression, some expressions become pejorative. If they are said with the intent to denigrate, insult, manipulate, and encourage violence against a person or a group of people, insults and expletives of venom can be categorized as destructive language. The offensive language used in this type of discourse has a performativity that goes beyond the linguistic realm. Studies conducted by Anderson and Lepore (2013), Bolinger (2017), Butler (2004; 2021), Camp (2013; 2018), Cepollaro (2017), Cepollaro and Stojanovic (2016), Cepollaro and Thommen (2019), Hedger (2012), Hess (2018; 2019), Frápolli (2013), Nunberg (2018), Pérez (2017), among others, have shown that this language promotes actions. The distinction between “they” and “us” portrays those who don’t fit the majority as an “alterity enemy”, which in turn results in negative societal actions (Jeshion, 2013; Waldron, 2012). This issue is becoming more prevalent. Hate speech is proliferating at an accelerated rate, not only in the media but also in social networks. These media undoubtedly become a powerful tool for those who seek to manipulate their speech. The use of insult discourse has increased during the past few decades. Expressions that certain groups used initially to denigrate people or minorities who differed from in some (social, racial, economic, political, sexual, ideological, among others) aspects are now habitual in everyday speech. These expressions have expanded to such an extent that they are being used “as a joke” by speakers not part of such groups. As Sullivan (2022) suggests, these speakers often ignore the pejorative dimension with which these terms were created and used in specific discourses and contexts. This leads to a non-paradigmatic use of them (Croom, 2013). Although it has a long history in Mediterranean countries, this issue persists in Latin American countries. Even though different contexts influence the social, political, and economic reality, the identified cause of this marginalization of minority groups, through discourse, is caused by the absence of criticality that plagues our society (Tosar, 2021). It is important to start working from education to generate alternative positions from new narratives. These new narratives would lead to identifying, analyzing and deactivating hate speech. Language, if used as a tool of power to manipulate people and attack human dignity, can also be used to allow alternatives of resistance and deactivation of discourses through new meanings. In this sense, it should be noted that narratives present a series of events, characters, and settings in a determined order to communicate a story or message to the receiver. Narratives can take on various formats, such as books, movies, T.V. series, video games, and others. On the other hand, transmedia narratives expand through multiple media and platforms, creating a more immersive and complex storytelling experience. In a transmedia narrative, the story is told through various media, each with its own function in the narrative. For example, a T.V. series may have a website where articles or videos related to the plot are published, or a video game may feature events and characters that do not appear in the main story. Thus, the main difference between a traditional narrative and a transmedia narrative is that the latter involves multiple media and platforms to tell a broader and interconnected story, which the receiver can explore in different ways. Therefore, faced with the described problem and in the face of the search for solutions to the problem of hate speech has led the authors of this article to consider transmedia narratives and music, understood 284

 New Narratives to Defuse Hate Speech

as discourse, as two alternative spheres. In these spheres, language games impose rules as systems of resistance and deactivation of hate speech (Wittgenstein, 2009). Two major projects with a common goal emerge as they focus on finding new narratives to tackle disinformation and hate speech: Musik Thinking and CoCriticAr are two attempts to create cross-cultural dialogue, discourse analysis, and collaborative workspaces. The former is in Barcelona, and the latter is in Latin America. Both projects share the common goal of addressing the problems caused by media misinformation and hate speech. Musik Thinking and CoCriticAr converge in their focus on the analysis of discourses that perpetuate hate speech using the tools of critical literacy (also called critical pedagogy), Musik Thinking methodology and laboratories, the philosophy of language (Searle, 1969) and critical discourse analysis, generate spaces of interaction for the exercise of critical citizenship; citizenship that contributes to global citizenship, without ignoring local reality (Arboleda, 2021; Cassany & Castellà, 2010; Searle, 1969; Tosar, 2021; Van Dijk, 1993, 1997, 1999). This convergence creates spaces for the exercise of critical citizenship, which contributes to global citizenship while acknowledging local realities. The purpose of this paper is to show how critical literacy can be used to detect hate speech and how transmedia narratives (including podcasts, memes, videos, and reels, among others) and musical language (mediated by artificial intelligence) can create spaces for intercultural dialogue on common issues across two different continents. Now then, given that the CoCriticAr and Musik Thinking projects have two components in their methodological design - one investigative and one of social development - two objectives are proposed, one for each component. In the case of research, the central objective focuses on determining the role played by critical literacy and argumentation in forming students’ critical thinking. Thus, by providing these tools to identify and analyze hate speech and disinformation circulating on social media, the project moves on to the development phase with actions that, through collaborative work, transmedia narratives, and music, allow for the deactivation of identified discourses. This leads to the objective of social development: to jointly design (by teachers and students) web spaces with the transmedia argumentation products made by students (podcasts, videos, games, memes, and argumentative texts) that educate and inform citizens in general about the problems caused by hate speech and disinformation circulating in the media and social networks. In the case of social development with music, this is achieved through the collective creation of sound pieces on a specific issue. Participants have access to musical language through artificial intelligence tools as a means of expression, critique, and citizen participation. In other words, tools are provided through social networks, project web pages, and music labs for citizens interested in identifying and deactivating hate speech. Musik Thinking and CoCriticAr provide tools to foster a democratic culture. The narratives (or media for understanding human experience or the construction of identity and meanings) aimed at deactivating hate speech are collaboratively created among students and teachers to make them accessible to interested citizens in Barcelona, Argentina, Colombia, Mexico, and Peru. The goal is to encourage critical thinking on the selected problem and propose solutions prioritizing interculturality.

METHODOLOGY Given the two components (research and development), the selected methodology, in line with the projects’ interests, was Participatory Action Research (PAR); as Contreras (2002) and Sirvent and Rigal (2012) 285

 New Narratives to Defuse Hate Speech

argue, from the social work carried out in this type of research, a path is created for reflection and critical thinking by involving subjects as part of the problem and the solution. Thus, collective knowledge is built, and democratic spaces for discussion are generated for citizen participation. To analyze the results achieved by students regarding their levels of argumentation and the use of critical literacy tools by Tosar (2021), when examining identified hate speeches, the critical discourse analysis approach proposed by Van Dijk (1999) was chosen. It is worth clarifying that this corresponds to the discourse meta-analysis in the research objective since, at first, students and teachers analyzed the hate speeches identified in each country (see Table 1), but in the second moment, the teacher-researchers analyzed students’ discourse and their processes to account for the advances presented in the levels of argumentation, which were measured according to the levels proposed by Enduran et al. (2004)1. In the hate speech analysis exercise carried out with students, in addition to the analysis table adapted from Benesch et al. (2019), elements proposed by Van Dijk (1993, 1997) were also considered. Thus, expressions that showed racism in the media and some symbolic forms of expressing this racism in the memes analyzed on social media were studied (Figure 1). In this sense, the methodological design of these projects consisted of four phases, which are described below: Figure 1. Phases of the methodological design

HATE SPEECH: BETWEEN OFFENSIVE LANGUAGE AND VIOLENT ACTIONS Language, as a biosocial element, has a dual function. Language is not only an essential tool for communication but also plays a crucial role in creating and establishing social reality through speech (Searle, 1969). Thus, through discourse, especially in the utterance of illocutionary acts, things are done with words (Austin, 1962). Discourse, then, not only calls for further action but, in many cases, is an action itself.

286

 New Narratives to Defuse Hate Speech

The discourse pragmatics, as shown by Van Dijk (1993, 1997), allows us to reveal the political and social intentionality and the psychology underlying language use. Undoubtedly, in racist discourse, as exposed by this linguist, ideologies used in the media are directed at the masses to promote discrimination and hate, thereby generating social inequalities and attacks on human dignity and democracy. Thus, discourse goes beyond the emissions of expressions or sentences and depends on the context and intentionality of the speaker. In such a way, semantics and pragmatics are closely linked. The performativity of language and the intention behind the utterances are significant in the case of hate speech. In addition to communicating what the speaker intentionally exposes, this type of speech leads to the realization of social actions that can lead to violent attacks against certain minority groups. The enunciation of pejorative speech introduces serious problems of moral and political order, as Tontodimamma et al. (2021) expose. Given that, hate incites violence and ends up rising to a political position that affects democracy, violates natural rights, and attacks the dignity of those to whom it is directed (Butler, 2021). For Davidson et al. (2017), hate speech can be defined as “language that is used to express hatred toward a targeted group, or that is intended to be derogatory, humiliating, or insulting to group members” (p. 512). Seen in this way, hate speech can be differentiated from offensive speech, in which, although the language that is used as an offensive effect, in its intentional charge, the speakers do not intentionally promote violence towards a particular minority. The difference, the authors argue, between hate speech and offensive speech is that the former contains pejorative terms with performative intentionality directed at a particular person or group. This generates an echo in the interlocutors and can promote actions that undermine the morality of the individuals denigrated by the language, given the performativity of the language used. On the other hand, in the latter (offensive speech), insulting terms can be used in a text, but if it is general and not intentionally aimed at denigrating a specific population, it does not promote hatred. The Council of Europe, more flexible in its meaning, has defined hate speech for that continent as: Encouragement, promotion or instigation...of hatred, humiliation or disparagement of a person or group. These include the harassment, disparagement, dissemination of negative stereotypes, stigmatization or threat concerning such person or group of persons and the justification of such manifestations on the grounds of “race,” color, descent, national or ethnic origin, age, disability, language, religion or belief, sex, gender, gender identity, sexual orientation and other personal characteristics or status. (European Commission, 2015, p. 29) However, under these regulatory parameters, in practice, the identification of hate speech and the subsequent corresponding sanctions to those who produce it are not easy to detect since issuing a sanction may violate the right to freedom of expression. Under the principle of proportionality, it distinguishes between “punishable hate speech.” If the hate speech takes place in a public context or if it is likely to incite acts of violence, intimidation, hostility or discrimination…and intolerant speech (or hate speech that is not punishable)…that which is protected by freedom of expression…must be fought through other strategies, as it poses a serious threat to coexistence and contributes to perpetuating discrimination against certain groups. (Barcelona City Hall, n.d.)

287

 New Narratives to Defuse Hate Speech

Given this confusion between hate speech and offensive language, Davidson et al. (2017) expound as an example the comparison between statements such as i) “If you’re a tranny...go f*ck yourself” and ii) “If someone isn’t an Anglo-Saxon Protestant, they have no right to be alive in the U.S. None at all, they are foreign filth.” In the first case, hate is evident and was promulgated with the Tweet circulated in networks, while in the second, the authors state that although it becomes offensive, it does not promote insults against a specific group. It does not marginalize a population. The sentence structure, intentionality and the terms used in these statements are essential constituents when making the semantic analysis of what the expression evokes. The separation between “offensive language” and “hate speech” has complicated the search for radical legal means that deactivate the speech that promotes violence. Only in extreme cases where aggressive behavior or violent marginalization occurs have legal resources been used to curb actions beyond language. This linguistic phenomenon, associated with contempt and intolerance against certain groups, exerts increasing power in its circulation. The increase in figures surrounding hate speech and subsequent criminal acts has led jurisprudence to monitor the so-called “hate crimes” in recent years. These are considered offences mobilized by the intolerance, discrimination, and social marginalization promoted in hate speech (Chalmers & Leverick, 2017). According to Ibarra (2013), hate crimes have increased in Europe, especially in Spain. There is empirical evidence from victims whose basic rights are violated due to racist, sexist, and xenophobic speech that mostly targets immigrants who come to the country. This is also stated in the Explanatory Memorandum of the European Commission against Racism and Intolerance ECRI General Policy Recommendation No. 15 on combating hate speech, a document in which the concern to defuse hate speech before it leads to major crimes is clearly evident. Given the complexity of criminalizing this type of speech --since, when it comes to following up on them, there are few cases in which hate speech becomes a criminal offence and is submitted for consideration before the European court-- actions have been created through different projects and programs in conjunction with the United Nations to detect the hateful intent of the speech and the damage it can cause. Unfortunately, most of the cases that are examined to determine the punishment for an offensive statement involve conflicts between the defense of human dignity and freedom of expression about the punishment for an offensive statement. However, in this situation, it is obvious that: Starting from intolerance, not only is discrimination (a less favorable treatment) possible, but also hate crime (criminal assault) against the victim or their property for the simple reason that they are a part of that group, or for being classified due to their social, religious, cultural, racial, sexual orientation, or any other distinguishing circumstance. This violence is led by perpetrators who believe it is acceptable to move forward with “identity” cleansing. (Ibarra, 2013, p. 5)

Essential Criteria to Identify Hate Speech Kaufman (2015) characterizes hateful speech to clearly identify the discursive elements that must be considered when analyzing a comment and classifying it as hate speech. This exercise has been made to prevent the discrimination that has occurred in Mexico. Among the essential components of this kind of discourse are the following:

288

 New Narratives to Defuse Hate Speech

1. Group Vulnerability Criteria: The audience of the text is explicitly stated in the speech (typically, minority groups). 2. Criteria for Humiliation: Three types of humiliation that are recognized in discourse can be presented: humiliating opinions, humiliating situations or events, and humiliating characteristics (often associated with the marginalized group). 3. Malignancy Criteria: Language that encourages actions (whether stated or implicit) against vulnerable groups. 4. Criteria for Intentionality: The message is intended to marginalize, harm, or discriminate against the group’s members. Each component is connected to language performance and establishes the relationship between the offensive language used and the actions people in vulnerable groups take while interacting in social contexts. The description provided by Kaufman (2015) is beneficial for the work being done by Arboleda (Lejanías in 2021 and Musik Thinking in 2022) and CoCriticAr because it allows for the identification of hate speech in both continuous and discontinuous texts that are circulated in the media and online. As a result, actions have been taken to address the hate speech.

Discursive Violence Against Immigrants When referring to immigrants, xenophobic expressions are used frequently in Spain and the world. In football, music, educational institutions, and the workplace, among others, camouflaged statements with derogatory connotations, such as “the people without documents,” “we’re being invaded,” and “we’re full of South Asians,” appear. Pejorative expressions abound in hatred discourse and are so widely disseminated in racist advertising that, on occasion, have escalated to aggressive behavior (Cámara, 2017). The issue worsens when the state trivializes it since victims are reluctant to come forward because it “won’t do anybody any good” (Ibarra, 2013). It is important to emphasize that, less than a decade ago, “hate crimes” caused by hate speech were referred to as “culturally conditioned crimes, ” meaning that the hate discourse was not considered concerning the criminal acts it provoked. However, the rise in the number of violent crimes brought on by antipathy and the lack of focus on the performativity of antipathy discourse has led to “the dominant culture” becoming the victims of these groups’ members who are motivated by intolerance for foreign cultures (Cámara, 2017, pp. 145-146). Prior to this, as stated in Cámara (2017), legal, linguistic forms, such as the one outlined in Article 510 of the Spanish Constitution, have been adopted (PCS). Declarative speech acts that aim to curb the promotion of hate. According to the PCS, these speeches refer to the “expression of epithets, qualifiers, or expressions containing an insult communicated in a universal form. It is a criminal type structured in the form of a crime of the danger contained in ‘hate speech’ [...]” (Chalmers & Leverick, 2017, p. 147). For that reason: The penalty of imprisonment for a term of one to four years and a fine of six to twelve months will be imposed on whoever publicly encourages, promotes or incites directly or indirectly hatred, hostility, discrimination or violence against a group, a part thereof, or against a specific person because they belong to such group, for racist, anti-Semitic or other reasons referring to ideology, religion or beliefs,

289

 New Narratives to Defuse Hate Speech

family situation, the belonging of its members to an ethnic group, race or nation, their national origin, their sex, sexual orientation or identity, for reasons of gender, illness or disability. (Art. 510, PCS) However, as has been attempted to be demonstrated, despite what has been declared and established in legal discourse, legal actions are not successful in courts in either Europe or South America, regions where xenophobic discourse is prevalent. It is important to note that, while residents of European countries may express animosity toward immigrants from Latin American countries inside their regions, such animosity has grown so strong that hate crimes have been committed against immigrants from even close neighboring countries. Alternatives like those suggested by Article XIX and the U.N., which have led to the Global Campaign for Free Speech since 2009, may arise in response to such a situation. Thus, declarations were issued, in which 12 restrictive principles were recognized at the national and international level to identify the manifestation of hate speech to guarantee the rights of civilians who are victims of this denigrating speech. However, the wave of hateful discourse in Europe and the Americas goes beyond the limits constitutionally imposed on criminal law (Cámara, 2017). In the case of Spain, as Vasallo (2020) stated in his article in La Vanguardia, even when Latinos have up to 2 university degrees, knowledge of languages such as English, and at least 1 or 2 specializations, they are treated harshly and even maliciously. A similar situation occurs in Latin America with immigrants, especially in capital cities or transit cities, where the problem is most strongly manifested in discourses that range from xenophobia to aporophobia (Cortina, 2017). Vasallo (2020) states that when Latinos are ridiculed at work, with bosses telling them, “you speak like an indigenous person... learn the language,” and are belittled with references to “savages,” their rights are violated. The situation worsens because these are degrading discourses in which the term “sudaca” - which has been linguistically adopted as a slur to belittle - is not even used directly. These expressions overlap and replicate with great force without even being aware of the semantic connotation that evokes discriminatory aversion and the wave of social contempt that increases with their use. Discourses such as those presented by Vasallo (2020), for example, even when loaded with humiliating opinions, are not only present in Spain but also in countries such as Argentina, Colombia, and Mexico, where immigrants are belittled and humiliated (see Table 1). This type of recurrent statements in different areas does not have legal repercussions, even when they end up being equally degrading for the victims to whom they are directed. For the European High Court, a distinction must be made, says Cámara (2017), for example: Between hatred that incites the commission of crimes, the hatred that sows the seeds of confrontation and erodes the essential values of coexistence and hatred that are identified with animosity or resentment, some nuances cannot be overlooked by the criminal judge with the argument that everything that does not embrace freedom of expression is intolerable and, therefore, necessarily criminal. (Cámara, 2017, p. 155) As a result, the difficulty in understanding hate speech that is based on xenophobic prejudice and stereotyping prompts government, civil, and academic institutions to consider other non-judicial institutions to deactivate them before they develop into hate crimes, as suggested by Benesch (2021), Benesch et al. (2019), and Unesco & Oxford (2021).

290

 New Narratives to Defuse Hate Speech

Currently, hate speech takes on several forms and is directed at various audiences depending on the speaker’s interests. As a result, many memes with broken text that use images to symbolize violent representations of immigrants in the four countries have been widely shared on social media. These memes, which reach many citizens, have great convincing power in the target audience, increasing xenophobia and aporophobia against those who migrate and “invade” the territory and culture. The following list of memes illustrates the vitriol seen in social media that expose images and slurs such as ‘sudacas’, ‘venecos’, ‘indios’, and ‘salvajes’, among others that reveal contempt and marginalization. (Figure 2). Figure 2. Memes circulated in social media (hatred of immigrants in Spain, Argentina, Colombia, and Mexico)

The analysis matrix proposed by Benesch et al. (2019) applies in continuous texts. Through their project on “Dangerous Speech,” they let students and teachers start processes identifying this kind of speech and collectively start thinking about how social norms are changing (Benesch et al., 2019). In the face of violent actions that promote hate, it is important to generate responses with new discourses, as this Harvard professor (Benesch) explains in her most recent project supporting responses for those who are victims of these violent discourses. Therefore, Arboleda, with Musik Thinking, and Rodríguez, Bron, Farina and Terrones, with CoCriticAr, use critical literacy as an initial tool to counteract this discourse that incites violence. Later in this paper, new discursive tools are proposed to put into practice contributing potential solutions to this issue.

Discourses of Hatred in the Spanish Music Industry Discourses of hatred frequently circulate in the music industry, both in terms of its auditory dimension and the content related to it. In the Spanish case, we find examples that come from several sources, in-

291

 New Narratives to Defuse Hate Speech

cluding statements, opinions, songs, and press reports that criticize and discriminate against immigrant groups both explicitly and implicitly. The following are some examples: In an interview with music critic Diego Manrique published in the Daily Continental, Manrique was asked about the South American influence on Spanish music. In a conversation with Lenore, Manrique said: We liked (a particular part of) Latin music when no Latinos were living in Spain. By settling here, they broke the idyllic image we had about the music that belonged to them. They were hearing faster beats, more emotive lyrics, and louder speakers. (Lenore, 2013) Although in discourse analysis it is impossible to label the discourse as “hate speech” under the criteria of malignity and intentionality exposed by Kaufman (2015), it is possible to identify the criteria of humiliation and disrespectful connotations about the figure of colonial imagery of “good savage.” Alternatively, and more directly, the singer Hard Gz appeals to violence in a critique of the values of the Spanish monarchy. He writes in his song Viva España: “I dream the rifle shouting: Kneel! Put each one in line. Make sure you put yourself under cover until the last one dies. . . . Revolution is not a revolution without a lot of deaths” (Hard, 2013). Over the years, the influence of Latin American musical genres (e.g., reggaeton) in Spanish youth has become “exotic music.” As reggaeton has been the most famous music in Spain for the last few years, this genre no longer fits the “idyllic” culture Manrique refers to today. However, from a “scientific” perspective, the close relationship between Latin America and the European worldview has led to news reports that replicate “research” in which: Scientists have found that reggaeton does not stimulate our cognitive ability or our intelligence. Instead, it induces a state of drowsiness that encourages early cognitive decline. . . . Their sticky melodies and easy-to-remember lyrics take our neurons to a state where they don’t have to struggle,” in contrast, of course, to “classical music [which] in a concert, sitting on a chair, with little light or in the living room of our home trying to taste a few minutes of peace” (Raya, 2018, p. 1) They are directly or indirectly involved in the problem of hate speech and add the difficulty of finding a global consensus between what and how hate speech should be treated and typified because, in addition to each country, the definitions that “lawyers and lawmakers have formulated from all over the world, end up being vague and are subject to the criterion of whoever has the power to apply them,” as the civil rights defender Nadine Strossen (2020) states. Censorship, as an alternative to stopping hate speech, does not successfully contribute to the solution of the problem; on the contrary, driven by controversy, hate speech that is not ransform ends up being reproduced with greater intensity. This is why our projects promote dialogue and integrate messages through music and transmedia narratives as a counter-narrative to foster critical citizenship.

CRITICAL LITERACY: A TOOL TO IDENTIFY AND ANALYSE HATE SPEECH Critical literacy can be considered a necessary tool to analyze social, moral, and political issues intertwined between hate speech and violent action. It is also an essential tool to identify and analyze discourses 292

 New Narratives to Defuse Hate Speech

in different media and social networks. Blackburn and Clark (2007) state that literacy is needed when betting on social change. In its beginnings, the term critical literacy was associated with literacy practices in reading and writing to select texts and information in different formats on the Internet (Cassany, 2006, 2012). Currently, for the purpose of the research conducted by Arboleda with the Musik Thinking project and the CoCriticAr team, critical literacy has been ransformero as the social practices of analyzing the senses and meanings achieved by social discourse readings and writings that allow humans to be part of political and moral life (Lankshear & McClaren, 1993). Thus, these characteristics have become fundamental to social changes, particularly to generating mechanisms for deactivating hate speech. The discursive exercise that critical literacy provides enhances the ransformer of the structures of thought and mental states to move to decision-making and action since the discourse practice of critical literacy. There is marked political learning for participating citizens, as the researchers of the GREDICS group of the Autonoma University of Barcelona propose. As one of the essential features of hate speech lies in intentionality, in critical literacy, this element of speech is crucial to the analysis. Consequently, as a social reality is constructed through language, it is essential to ransform ideologies, ideas, senses, and inconsistencies (Searle, 1969, 2010). According to Cassany (2006), in the world of technologies with disinformation and malicious discourses, critical literacy becomes a toolbox in sociolinguistics to think and live better in the world (McDaniel, 2006; Wittgenstein, 2009). Decoding terms in hate speech and the meanings and senses achieved amid spoken contexts is critical to proposing solutions to eliminate this social issue that affects us. Words play an important role in naming and making evident social storms, but they are also crucial in initiating processes of discursive change that impact actions that transform social reality (Cortina 2017; Searle, 1995, 2010). For this reason, in response to the issue raised by xenophobic and occasionally aporophobic discourses in Europe and Latin America, currently, Arboleda and CoCriticAr propose the use of literary tools as a necessary component in the development of critical citizens of human beings capable of ransformer the issue and feeling part of it, and of subjects capable of proposing feasible solutions to defend human dignity when this is necessary.

Characteristics of Critical Literacy Once critical literacy is understood as a social practice to identify and analyze social discourses, some features that have been taken into account in the projects can be employed in new discursivities, such as Musik Thinking and transmedia narratives used by Arboleda and CoCriticAr projects, respectively to deactivate hate in discursive performativity in social networks and media. The main aspects that have been addressed by the authors in these projects are the following: 1. The analysis of discourse to identify the meanings given to Slurs and Dogwhistles in specific contexts. 2. The identification of the speaker’s intentionality in specific terms that call for hate and generate actions of contempt, ransformeron, and violence. 3. The analysis of hate speech and offensive language, both implicit and explicit in messages in networks and media. 4. The identification of and differentiation between facts and opinions in hate speech. 293

 New Narratives to Defuse Hate Speech

5. 6. 7. 8.

The evaluation of the scope of the discourse and its moral and political implications. Identifying the ideologies underlying hate speech and the population directly and indirectly affected. The verification and contrast of the accuracy of the information in hate speech. The identification of the hegemonic and counter-hegemonic powers in hate speech.

Thus, through these conceptual approaches for analysis, the projects presented here provide students and citizens with tools to address this problem and to avoid the proliferation of these discourses through social networks by generating awareness of them through critical analysis of sociopolitical discourse (Morrell, 2008). Table 1 presents a brief sample of the critical analysis of hate speech against immigrants in countries such as Spain, Mexico, Argentina, and Colombia, regions in which Musik Thinking and CoCriticAr are implemented.

294

 New Narratives to Defuse Hate Speech

Table 1. Analysis of hate speech against immigrants Fragment of the Text Diego Manrique, musical critic (Spanish). “we liked (a certain part of) Latin music when no Latinos were living in Spain. By settling here, they broke the idyllic image we had about the music that belonged to them. They heard faster beats, more emotive lyrics, and louder speakers.” Diego Falconi, University Professor in Barcelona (Ecuador): “The worst thing was not that ‘marica and sudaca’ were screamed at me; the worst thing was that the gym wanted to hide it.” Juan Manuel Gastelum, Mayor of Tijuana, Mexico “We want the 33rd article of the Constitution to be applied to them (that they are expelled from the country).” “Tijuana is a city of migrants, but we do not want them this way.” “It was different with the Haitians; they had papers, they were in order, it was not a horde, forgive me for the expression.” “Human rights are for righteous humans.” “Migrants are aggressive and rude.” “The members of the migrant caravan are unemployed, drug addicts, undesirable.” “They are still invading us by the thousands.”

The Audience to Which It Is Addressed

Message and Activities That Are Generated

Social and Historical Context

Medium of Dissemination and Year

Spanish citizens, in general.

Xenophobic speech that foments hate among Spanish citizens against Latin American immigrants and strengthens the image of the “good savage,” a colonial category that encourages contempt and humiliation of Latin American culture.

Spanish interview on the role and influence of Latin American music in Spanish culture.

Spanish Press. Diario el Continental. (2013) https://www.elconfidencial.com/ cultura/2013-04-29/en-espanagustaba-la-musica-latina-hasta-quese-instalaron-los-inmigra ntes_495090/ 2013

Comunidad de Barcelona, Judicial entities.

It shows xenophobic and homophobic speech committed by citizens. Disrespectful language in which ‘sudaca’ and ‘marica’ are used to genera‘e hatred towards Latin Americans.

Daily life in Spain, with hate speech against immigrants with disrespectful terms.

Spanish Press. El País. (2022) https://elpais.com/espana/ catalunya/2022-03-11/lo-peor-no-fu e-que-me-gritaran-marica-y-sudacasino-que-el-gimnasio-lo-qu iso-ocultar.html

Eight hundred migrants from Central America arrived in Tijuana and were insulted with racist and xenophobic expressions by the local inhabitants near the hostel. Several migrants were insulted with expressions such as “whores” and “maricones” (fags) by 30 inhabitants of Playa de Tijuana who demanded that they leave.

The hate speech of the mayor of Tijuana and the reaction of a part of the Tijuana population took place in the context of one of the caravans of migrants of Central Americans travelling through Mexican territory toward the United States. It was October 2018 when 2,000 migrants were in Tijuana, at the border with the United States, waiting for permission to enter U.S. territory.

Mexican media and the Internet. Univisión: Xenofobia in México. La Jornada: Let them apply Article 33; we don’t want them like that, says the mayor of Tijuana https:// www.jornada.com.mx/2018/11/16/ politica/014n1polhttps:// partidero.com/alcalde-detijuana-pide-expulsar-amigrantes-centroamericanos-sonindeseables/?fbclid=IwAR3hkf8UPZwxiNwnaF9tDRwsXUwgG1Pl2 H2Gm3L0xhtFbVq9nJGkxZVuqo

Local and national audience in Mexico: television, radio, written press and Internet

Continued on following page

295

 New Narratives to Defuse Hate Speech

Table 1. Continued The Audience to Which It Is Addressed

Message and Activities That Are Generated

Social and Historical Context

Medium of Dissemination and Year

Argentinian population in general. Listening to radio and other media (previously, it was reprinted by some media). Possible voters from their political sector.

It links migration with crime and drug trafficking, as well as poor people with a lack of intention to work a“d “live off the governme”t.” It drives the rejection of these collectives and generates expulsive measures expressed in proposals before the chamber and by ordinary citizens in acts of violence in public toward migrants.

The speaker is Senator Miguel Ángel Pichetto, former vice-presidential candidate for the Juntos por el Cambio party, which received 49% of the vote in the last presidential election. The speech takes place before the election and is part of a communication strategy devised by some media conglomerate referents that permanently associate Latin American foreign migrants with crime and drug trafficking.

Argentine radio broadcasting: La Red (Buenos Aires). Morning show from 9 a.m. to noon hosted by Eduardo Fienmann. 2019 https:// youtu.be/cPWDkEPwUEk

Readers of the press and media followers in Colombia. Colombia citizens in general.

Xenophobic, aporophobic, and sexist remarks that incite hate speech among citizens toward Venezuelan immigrants and violent actions against them. They intended to emphasise that their government guaranteed assistance and health care for the 400 Venezuelan mothers and children who had been born and were about to be born. The derogatory tone “s “poor little Chin”se”.

The increased rate of pregnant Venezuelan women in Bucaramanga and the governm’nt’s actions to improve their health care, but the connotations are derogatory.

Colombian Press. Press: Boyacá extra, (2019). Networks: Twitter and Facebook. Press and television: El Nuevo Siglo and RCN: (27 January 2017). Press: Semana Magazine. Venezuela Migration Project. (18 January 2021)

Fragment of the Text

Senator Miguel Ángel Pichetto: “We are sick in Argentina. I do not know which international entity arranged for this woman to return (“this Peruvian” in another part of the speech).” “A criminal, Peruvian, a drug dealer.” “This government has given everything to these people… who ’on’t work. This is the problem we ha”e.”

Mayor of Bucaramanga, Colombia, Rodolfo Hernánde“: “Venezuelan women are a factory to make poor child”en“. “How dangerous are migrant crimina”s?” Minister of Defense Vargas Llera“: “A minority of Venezuelans, deeply viol”nt” who a“e “a factor of insecuri”y“” “The houses are for the displaced people who live in Tibú. T‘e ‘vene’os’ will not be allowed to get in under any circumstance. This is not for the ‘venecos’.” -Mayor of Bogota, Colombia, Claudia Lope“: “How dangerous are migrant crimina”s?” Police office“: “We must kill the Meleán members who are in the coun”ry”.

Source: Based on Benesch et al. (2019)

The analysis of the previous fragments reveals a counter-hegemonic power in which political leaders incite citizens to commit violent acts against immigrant groups, such as Latin Americans, Venezuelans, Peruvians, and Mexicans. The images and hate speech, like the memes in Image 1, are explicit and offensive, inciting contempt for the victims. However, if the goal is to counteract and deactivate the performative power of the offensive language used in hate speech, identifying and critically analyzing such discourse is only the beginning. Once identified, music, collaborative work, and transmedia narratives become allies in promoting interculturality, tolerance, and peaceful coexistence.

296

 New Narratives to Defuse Hate Speech

MUSIK THINKING AS A METHODOLOGY FOR PROMOTING CRITICAL CITIZENSHIP Although analysis, criticism, proposals, and counterproposals against hate speech have advanced, they still have a long road ahead of them in the linguistic field; work in the music field is just getting started. Music is perceived to set the tone of the texts rather than as a text in and of itself. However, as Tagg and Clarida (2003), cited by Hernández (2016), point out, audiovisual media, more than the written word, Contains the most pervasive and persuasive messages, influencing which political candidates are elected and governments are deposed, not to mention what goods are sold, lifestyles are imposed, fashions followed, myths upheld, and ideologies embraced. (Hernández, 2016, p. 17) The persuasive power of music is not solely used as a backdrop for campaigns, cults, or trends; music is also use “, “not always consciously, to accomplish at least three tasks: 1) subjectivity configuration and ordering, 2) bodily experience configuration and ordering, and 3) social exchange configuration and order”ng” (Hernández, 2016, p. 20). Under these conditions, musician and philosopher Oscar Arbol’da’s anti-hate speech proposal is to use music, in its sonorous and corporal dimension, processes, roles, and contexts, as well as its persuasive power, to think about texts, their contents, and their intentions, and thus develop collective counter-narratives through musical language, critical thinking, and artificial intelligence. His proposal, call“d “Musik Thinki”g,” entails discourse analysis through musical language and philosophy, developing critical skills in participants actively listening to the sound content, and proposing, with the help of artificial intelligence tools, counter-narrative responses to hate speech in Spain based on musical thought and critical literacy. Musik Thinking is a discursive approach to combating hate through music and philosophy as an alternative to seeking social change. It is a way to counteract hate speech through music, philosophy, and technological tools. In this way, spaces for intercultural dialogue are created, inviting Latin and Spanish individuals interested in addressing the problem through music. Spaces for reflection and discussion are opened up between music and philosophy, and with the help of artificial intelligence, counter-narratives are created that promote the deactivation of hate speech circulating against immigrants (Figure 3).

297

 New Narratives to Defuse Hate Speech

Figure 3. Model and strategies used in Musik Thinking

CoCRITICAR: A SPACE TO THINK CRITICALLY ABOUT SOCIAL REALITY CoCriticar project (http://cocritic.ar) is another example of a proposal developed by teachers and researchers from Latin American universities using a two-way strategy. First, conducting inter-institutional collaborative learning work online has improved critical thinking skills through literacy and argumentation by discussing socially relevant problems such as disinformation in networks and the proliferation of hate speech. CoCriticAr is a commitment to helping young people in participating institutions develop critical thinking skills. Second, transmedia narratives have been developed collaboratively between teachers and students, from which development products have been created that are tools for the use of citizens in general (web pages with social content, podcasts, argumentative texts, video-format interviews created by students and the participation of experts in the field). These narratives have gone beyond work between classes and have demonstrated the successful combination of collaborative work, transmedia narratives, and inter-university links, the urgency of beginning to generate new discourses to combat discursive hatred and misinformation that harms citizens (Rodriguez, 2021). In this regard, collaborative work has been based on the didactics of Collaborative Project-based Learning (CPBL) from different conceptual definitions, authors, and theoretical perspectives that argue th“t “Project-Based Learning is a learning model in which students plan, implement, and evaluate projects that have an application or applications in the real world beyond the classr”om” (Blank, 1997; Dickinson et al., 1998). Similarly, it has been stated that” PBL is a learning strategy that aids in achieving one or more objectives by implementing a se”ies of actions, interactions, and resources (Ayuste et al., 1998). On the other hand, for the Northwest Regional Educational Laboratory (2002), it “s “a holistic education strategy rather than just a compliment. Project-based learning is an important part of the learning proc”ss” because it allows students to think about real-world problems. Similarly, Maldonado (2008) states that:

298

 New Narratives to Defuse Hate Speech

The use of PBL as a didactic strategy is considered relevant in the educational experience, considering that: (a) the project methodology is a learning strategy that allows the achievement of significant learning because it arises from relevant activities for the students and often contemplates objectives and contents that go beyond the curricular ones. (b) It enables subject integration, reinforcing the overall vision of human knowledge. (c) It enables transformerion of activities centered on a common goal, defined by the stud’nt’s interests and with commitment. (d) It promotes, among other things, creativity, individual responsibility, collaborative work, and critical capacity. (p. 161) Transmedia narratives have been used as the core of the activities developed to conduct this collaborative project simultaneously “. “Transmedia storytelling is based on the narration of a story via the creation of various products that reach the receiver via different me”ia” (Gifreu, 2012, p. 82). For this project, transmedia narratives are fiction or non-fiction stories that expand across multiple media and are subject to user intervention. (Scolari, 2012; Jenkins, 2008). Jenkins (2008) addressed this narrative modality in the same way, arguing th“t “transmediation lies in the integration of multiple texts to create a narrative of such dimensions that it cannot be confined to a single med”um” (p.101). As a resul“, “it is the creation of an atmosphere, a universe, around a st”ry” rather than defining a story with a kind of complex structure (beginning, middle, and end) (Gifreu, 2012, p. 83), and this is the direction in which the CoCriticAr project has been developed. Recent studies have demonstrated that transmedia narratives are a reality and contribute to the daily work of teachers and students in the teaching and learning processes. Thus, learning, teaching, and education, which play a key role in societal development, provide fertile ground for educational innovation experiencransforming transmedia narratives. The CoCriticAr project exemplifies these statements because it focuses on improving argumentative capacity and critical literacy while providing tools and criteria to combat hate speech. Likewise, “f “the transmedial world is always experienced through mediation, the means of departure of that world is important, but so are how it devel”ps”Rosendo Sánchez, 2016, p. 59) . The content production process that forms those worlds is not without complexity. As a result, this apparent difficulty presents a favorable opportunity for teaching digital content production, beginning with creating and developing a transmedia product, in which the students generate strategies to disarticulate hate speech using media tools. Furthermore, as previously stated, CPBL allows for the integration of different points of view, knowledge, and prior experiences, as well as stude’ts’ individual and collective strategies to build new knowledge; thus, this methodology is critical when generating intercultural spaces for dialogue; spaces in which students think about and discuss their reality and create new narratives to address the problem of hate that is promoted through discourse. As Bron (2019) argues: CPBL favors transmedia communication practices in the classroom, improving the necessary competencies for the future professional practice of digital communicators, such as teamwork, information search, generation of effective strategies, improvement of social skills, audience consumption analysis, written and oral communication, use of technologies and social networks, efficient management of interactions with project audiences, as well as the transfer and construction of knowledge. (p. 245)

299

 New Narratives to Defuse Hate Speech

As a result, projects like CoCriticAr highlight the possibilities offered to educational processes by the articulation of collaborative project-based learning to think about the problems that afflict citizens and generate answers to collaborate in the dismantling of the scourge of hate speech from educational institutions in the teaching and learning processes themselves through the use of transmedia narratives (Figure 4). Figure 4. Essential elements of the CoCriticAr project

DISCUSSION AND CONCLUSION The proposals of Arboleda and CoCriticAr are an applied form of critical thinking for the identification and dismantling of hate speech. By using the tools of critical literacy, through music and transmedia narratives, they contribute to thinking, identifying, and confronting hate speech in different contexts, considering what happens in Spain and various countries in Latin America. The propos“l “Musik Think”ng” from the Arboledas project consists of analyzing discourses through music and philosophy, aiming to develop critical thinking skills that allow for the analysis of sound content, and then making a counter-narrative proposal against the hate speech that proliferates in Spain. In the case of CoCriticAr, the use of critical thinking and its practical application in communication products corresponding to transmedia narratives allows not only to present the criteria for identifying hate speech but also to construct new narratives that highlight and counteract hate speech in Latin American contexts. Through communication products such as audio, videos, infographics, texts, and other multimedia resources, education is achieved in classrooms and extends to different areas of society. In this way, the projects connect with audiences to teach about the nature of hate speech, provide tools for its identification, and offer clues on how to counteract it. If we revisit Sea’le’s (1969) thinking in the sense that social reality has been created and institutionally transformed intentionally through speech acts, we find that projects like Musik Thinking and CoCriticar

300

 New Narratives to Defuse Hate Speech

construct a counter-narrative that allows us to confront and counteract hate speech. There is an intentional process of transformation that occurs through the use of critical thinking to identify hate speech, whether in music, media, or social networks, and from there, deactivate such discourses. Critical literality (Casany, 2010; Tosar, 2021), the methodological proposal of Musik Thinking (Arboleda, 2022), as well as the philosophy of language (Searle, 1969), and critical discourse analysis (Van Dijk, 1993, 1997, 1999) constitute tools that simplify the process of identifying hate speech. With the support of Kaufman (2015) and his criteria regarding vulnerable groups, humiliation, malignancy, and intentionality, it becomes more plausible to identify hate speech, even when it hides under the guise of freedom of expression. Based on these methodologies and criteria, hate speech against migrants and members of diverse communities could be identified: expressions such as “fag and dirty South American” “human rights are for humans with rights,” “delinquent, Peruvian, drug dealer,” “a minority of Venezuelans, deeply violent…who are a security risk,” among others, have appeared in local and national media, as well as on social networks, both in Latin American countries and in Spain. After applying these criteria and locating the narratives of hate, the proposals of Musik Thinking and CoCriticar become strategies to counteract hate speech through the use of transmedia narratives that share criteria and examples that can be adopted by individuals to determine if a discourse incites hatred towards migrants or a minority or diverse group. While this text focuses on a proposal based on music and critical thinking, applied to hate speech in the Mediterranean and Latin America, the underlying idea is that methodologies, models, and criteria can be developed to address the proliferation of hatred on social networks, in the media, and in various contexts.

REFERENCES Ajuntament de Barcelona. (n.d.). ¿Qué es el discurso de odio? https://ajuntament.barcelona.cat/bcnvsodi/ es/que-es-el-discurso-de-odio/ Anderson, L., & Lepore, E. (2013). Slurring words. Noûs (Detroit, Mich.), 47(1), 25–48. doi:10.1111/ j.1468-0068.2010.00820.x Arboleda, Ó. (2021). Lejanías, trabajo musical y literario sobre el desplazamiento en Colombia. Onces Ed. Austin, J. (1962). How to do things with words. Oxford University Press. Ayuste, A., Flecha, R., López, P., & Lleras, J. (1998). Planteamientos de la pedagogía crítica. Comunicarransformermar. Editorial Graó. Benesch, S. (2021). Dangerous Speech: A Practical Guide. The Dangerous Speech Project. https:// dangerousspeech.org/ Benesch, S., Buerger, C., Glavinic, T., & Manion, S. (2019). Dangerous Speech. A practical guide. Dangerous Speech Project. https://www.researchgate.net/publication/330684579_Dangerous _Speech_A_Practical_Guide

301

 New Narratives to Defuse Hate Speech

Blackburn, M. V., & Clark, C. T. (2007). Literacy research for political action and social change. Peter Lang. Blank, W. (1997). Authentic instruction. In W.E. Blank & S. Harwell (Eds.), Promising practices for connecting high school to the real world (pp. 15–21). University of South Florida. Bolinger, R. (2017). The Pragmatics of Slurs. Noûs (Detroit, Mich.), 51(3), 439–462. doi:10.1111/ nous.12090 Bron, M. (2019). Comunicación transmedia y educación: El Aprendizaje Basado en Proyectos Colaborativos como método de aprendizaje en comunicación digital [Tesis doctoral]. Universidad Rey Juan Carlos. Butler, J. (2004). Lenguaje, poder e identidad. Síntesis. Butler, J. (2021). La fuerza de la no violencia. La ética en lo político. Paidós. Cámara, S. (2017). El concepto de delitos de odio y su comisión a través del discurso. Especial referencia al conflicto con la libertad de expresión. Anuario de Derecho Penal y Ciencias Penales, 70, 141–225. Camp, E. (2013). Slurring perspectives. Analytic Philosophy, 54(3), 330–349. doi:10.1111/phib.12022 Camp, E. (2018). A dual act analysis of slurs. In D. Sosa (Ed.), Bad words: philosophical perspectives on slurs. Oxford University Press. Cassany, D. (2006). Tras las líneas. Sobre la lectura contemporánea. Anagrama. Cassany, D. (2012). Leer y escribir en la red. Anagrama. Cassany, D., & Castellà, J. (2010). Aproximación a la literacidad crítica. Perspectiva Florianópolis, 28(2), 353–374. doi:10.5007/2175-795X.2010v28n2p353 Cepollaro, B. (2017). When evaluation changes – an echoic account of appropriation and variability. Journal of Pragmatics, 117, 29–40. doi:10.1016/j.pragma.2017.05.009 Cepollaro, B., & Stojanovic, I. (2016). Hybrid evaluatives: In defense of a presuppositional account. Grazer Philosophische Studien, 93(3), 458–488. doi:10.1163/18756735-09303007 Cepollaro, B., & Thommen, T. (2019). What’s wrong with truth-conditional accounts of slurs. Linguistics and Philosophy, 42(4), 333–347. doi:10.100710988-018-9249-8 Chalmers, J., & Leverick, F. (2017). A Comparative Analysis of Hate Crime Legislation. A Report to the Hate Crime Legislation Review. University of Glasgow. Contreras, R. (2002). La Investigación Acción Participativa (IAP): revisando sus metodologías y sus potencialidades. In Experiencias y metodologías de la investigación participativa. Naciones Unidas, CEPAL, ECLAC. Cortina, A. (2017). Aporofobia, el rechazo al pobre. Un desafío para la democracia. Paidós. Croom, A. (2013). How to Do Things with Slurs: Studies in the Way of Derogatory Words. Language & Communication, 33(3), 177–204. doi:10.1016/j.langcom.2013.03.008

302

 New Narratives to Defuse Hate Speech

Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated Hate Speech Detection and the Problem of Offensive Language. Proc. Int. AAAI Conf. Web Soc. Media, 11, 512–515. 10.1609/icwsm. v11i1.14955 Gifreu, A. (2012). El documental interactivo como nuevo género audiovisual. Estudio de la aparición del nuevo género, aproximación a su definición y propuesta de taxonomía y de modelo de análisis a efectos de evaluación, diseño y producción [Tesis doctoral]. Universidad Pompeu Fabra. Hard, G. Z. (2013, 7 November). Hard GZ - Viva España [Videoclip]. https://www.youtube.com/ watch?v=QpFViVQG32Y Hedger, J. A. (2012). The semantics of racial slurs: Using Kaplan’s framework to provide a theory of the meaning of derogatory epithets. Linguistic and Philosophical Investigations, 11, 74–843. Hernández, O. (2016). Música y discurso: aproximaciones analíticas desde América Latina. Eduvim. Hess, L. (2018). Perspectival expressives. Journal of Pragmatics, 129, 13–33. doi:10.1016/j.pragma.2018.02.007 Hess, L. (2019). Expressive meanings and expressive commitments: a case of meaning as use. In Philosophical insights into pragmatics. De Gruyter. doi:10.1515/9783110628937-010 Ibarra, E. (2013). Víctimas y Seguridad ante los Crímenes de Odio. Cuadernos de Análisis. Jenkins, H. (2008). Convergence Culture. La cultura de la convergencia en los medios de comunicación. Paidós. Jeshion, R. (2013). Slurs and Stereotypes. Analytic Philosophy, 54(3), 314–329. doi:10.1111/phib.12021 Kaufman, G. (2015). Odium dicta. Libertad de expresión y protección de grupos discriminados en internet. Conapred. Lankshear, C., & McClaren, P. (Eds.). (1993). Critical literacy: Radical and postmodernist. State University of New York Press. Lenore, V. (2013). Entrevista a Diego Manrique. https://www.elconfidencial.com/cultura/2013-04-29/ en-espana-gustaba-la-musica-latina-hasta-que-se-instalaron-los-inmigrantes_495090/ Maldonado, M. (2008). Aprendizaje basado en proyectos colaborativos. Una experiencia en educación superior. Laurus, 14(28), 158-180. https://www.redalyc.org/pdf/761/76111716009.pdf McDaniel, C. (2006). Critical literacy: A way of thinking, a way of life. Peter Lang. Morrell, E. (2008). Critical literacy and urban youth: Pedagogies of access, dissent, and liberation. Taylor & Francis. NorthWest Regional Educational Laboratory. (2002). Aprendizaje por Proyectos (trad. Eduteka). https:// eduteka.icesi.edu.co/modulos/7/184/468/1 Nunberg, G. (2018). The social life of slurs. In D. Fogal, D. Harris, & M. Moss (Eds.), New work on speech acts. Oxford University Press.

303

 New Narratives to Defuse Hate Speech

Pérez, E. (2017). The role of the ascribee’s representations in belief ascriptions. Teorema., 36(1), 113–125. Raya, C. (2018). El reguetón afecta a tus neuronas. TecnoXplora. https://www.lasexta.com/tecnologiatecnoxplora/ciencia/divulgacion/regueton-afecta-tus-neuronas_201802065a7b860d0cf2195169bd407b. html Rodríguez, A. (2021). Proyecto Co-critic.ar: un trabajo colaborativo interinstitucional (coil) para promover en los estudiantes habilidades para pensar críticamente la realidad social. In Innovación e investigación docente en educación: experiencias prácticas. Dyckinson, S, L. Rosendo Sánchez, N. (2016). Mundos transmediales: Revisión conceptual y perspectivas teóricas del arte de crear mundos. Icono, 14(14), 49–70. doi:10.7195/ri14.v14i1.930 Scolari, C. (2012). Narrativas transmedia: cuando todos los medios cuentan. Deusto. Searle, J. (1969). Speech Acts. An essay in the philosophy of language. Cambridge at the University Press. doi:10.1017/CBO9781139173438 Searle, J. (1995). The Construction of Social Reality. The Free Press. Searle, J. (2010). Making the Social World. The Structure of Human Civilization. Oxford University Press. doi:10.1093/acprof:osobl/9780195396171.001.0001 Sirvent, M., & Rigal, L. (2012). Investigación acción participativa. Un desafío de nuestros tiempos para la construcción de una sociedad democrática. Páramos Andinos. Strossen, N. (2020). Hate: Why We Should Resist It With Free Speech, Not Censorship. Oxford University Press. Sullivan, A. (2022). Semantic Dimensions of Slurs. Philosophia, 50(3), 1479–1493. doi:10.100711406021-00427-2 Tagg, P., & Clarida, B. (2003). Ten Little Title Tunes. The Mass Media Music Scholars’ Press, Inc. Tontodimamma, A., Nissi, E., Sarra, A., & Fontanella, L. (2021). Thirty years of research into hate speech: topics of interest and their evolution. Scientometrics. Springer., doi:10.100711192-020-03737-6 Tosar, B. (2021). Literacidad crítica. Enseñar a leer el mundo para transformarlo a través del encuentro. Universitat Autònomade Barcelona. UNESCO & Oxford. (2021). Hacer frente al discurso de odio en las redes sociales: desafíos contemporáneos. Unesco. Van Dijk, T. (1993). Elite discourse and racism. SAGE Publications, Inc. doi:10.4135/9781483326184 Van Dijk, T. (1997). Racismo y análisis crítico de los medios. Paidós. Van Dijk, T. (1999). El análisis crítico del discurso. Anthropos, 186(1), 23–36. Vasallo, A. (2020). ¿Se discrimina a los inmigrantes latinoamericanos en España? La Vanguardia. https://www.lavanguardia.com/participacion/cartas/20200310/474046212445/discriminacion-maltratoinmigrantes-latinos-latinoamericanos-espana-debate.html

304

 New Narratives to Defuse Hate Speech

Waldron, J. (2012). The Harm in Hate Speech. Harvard University Press. doi:10.4159/harvard.9780674065086 Wittgenstein, L. (2009). Investigaciones filosóficas. In I. Reguera (Ed.), Obra completa. Vol. 1: Tractatus logico-philosophicus. Investigaciones filosóficas. Sobre la certeza. Gredos.

ENDNOTE 1



To know the results achieved in argumentation, please refer to Rodríguez (2021).

305

306

Compilation of References

Abdelaaty, L. (2021). Refugees and guesthood in Turkey. Journal of Refugee Studies, 34(3), 2827–2848. doi:10.1093/ jrs/fez097 Abercrombie, G., Rieser, V., & Hovy, D. (2023). Consistency is Key: Disentangling Label Variation in Natural Language Processing with Intra-Annotator Agreement. arXiv:2301.10684. https://doi.org/ doi:10.48550/arXiv.2301.10684 Abuín-Vences, N., Cuesta-Cambra, U., Niño-González, J. I., & Bengochea-González, C. (2022). Hate speech analysis as a function of ideology: Emotional and cognitive effects. Comunicar: revista científica iberoamericana de comunicación y educación, XXX(71), 37-48. doi:10.3916/C71-2022-03 Acciai, E. & del Rizzo, S. (2023). L’Italia è l’unica a garantire la protezione umanitaria? La Voce. https://lavoce.info/ archives/category/rubriche/fact-checking/ Adnkronos. (2020). Coronavirus, presidente Calabria: “Governo blocchi partenze”. Adnkronos. Agência Lusa. (2020). Figuras públicas e associações repudiam afirmações de André Ventura sobre ciganos. Observador. https://observador.pt/2020/05/06/figuras-publicas-e-associacoes-repudiam-afirmacoes-de-andre-ventura-sobre-ciganos/ Agência Lusa. (2021). Homem Acusado de Matar ator Bruno Candé Condenado a 22 Anos e Nove Meses de Prisão. Observador. https://observador.pt/2021/06/28/homem-acusado-de-matar-ator-bruno-cande-condenado-a-22-anos-enove-meses-de-prisao/ Aguaded, J. I., & Cabero-Almenara, J. (Eds.). (2022). Handbook of Research on Digital Education and Digital Technologies. IGI Global. Aguilar, M. A., Gómez, V., Marquina, M., de Rosa, M., & Tamarit, J. M. (2015). Manual práctico para la investigación y enjuiciamiento de delitos de odio. Generalitat de Catalunya. https://tinyurl.com/2hbyd9w6 Aguilera-Carnerero, C., & Azeez, A. H. (2016). ‘Islamonausea, Not Islamophobia’: The Many Faces of Cyber Hate Speech. Journal of Arab & Muslim Media Research., 9(1), 21–40. doi:10.1386/jammr.9.1.21_1 Ajuntament de Barcelona. (n.d.). ¿Qué es el discurso de odio? https://ajuntament.barcelona.cat/bcnvsodi/es/que-es-el-discu rso-de-odio/ Al Serhan, F., & Elareshi, M. (2020). New media and hate speech: A study of university students in Jordan. Opción: Revista de Ciencias Humanas y Sociales, (26), 166–184. Albadi, N., Kurdi, M., & Mishra, S. (2018, August). Are they our brothers? analysis and detection of religious hate speech in the arabic twittersphere. In 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) (pp. 69-76). IEEE. 10.1109/ASONAM.2018.8508247

 

Compilation of References

Albornoz, F., Bradley, J., & Sonderegger, S. (2018). The Brexit Referendum and the Rise in Hate Crime. mimeo. https:// cep.lse.ac.uk/seminarpapers/19-09-18-ALB_v2.pdf Alikılıç, Ö., Gökaliler, E., & Alikılıç, İ. (2021). Nefret Söylemi Üzerinden Ötekileştirme: Twitter’da Mültecilere Yönelik Nefret Tipolojisi Analizi [Marginalization Through Hate Speech: Analysis of Hate Typology Against Refugees on Twitter]. Akdeniz Üniversitesi İletişim Fakültesi Dergisi, (36), 501–520. Alirangues, M. (2018). La palabra precaria. Prolegómenos para una teoría negativa de la ficción. Escritura e imagen, 14, 149. Alkan, H. (2021). The gift of hospitality and the (un) welcoming of Syrian migrants in Turkey. American Ethnologist, 48(2), 180–191. doi:10.1111/amet.13012 Alkiviadou, N. (2018). The Legal Regulation of Hate Speech: The International and European Frameworks. Politička misao, 55(4), 203-229. doi:10.20901/pm.55.4.08 Allen, C. (2010). Islamophobia. Ashgate Publishing. Allport, G. W. (1954). The nature of prejudice. Addison-Wesley. Alonso-García, S., Gómez, G., Sanz, M., Moreno, A. J., & Rodríguez, C. (2020). The Impact of Term Fake News on the Scientific Community. Scientific Performance and Mapping in Web of Science. Social Sciences (Basel, Switzerland), 9(5), 73. doi:10.3390ocsci9050073 Aluru, S., Cingel, D. P., Miller, C. H., & Stavrositu, C. (2020). Individual and environmental factors associated with cyberbullying perpetration: A meta-analysis. Journal of Communication, 70(6), 739–758. doi:10.1093/joc/jqaa039 Aly, A., Balbi, A. M., & Jacques, C. (2015). Rethinking countering violent extremism: Implementing the role of civil society. Journal of Policing, Intelligence and Counter Terrorism, 10(1), 3–13. doi:10.1080/18335330.2015.1028772 Amnesty International Italia. (2018). Conta fino a 10. Barometro dell’odio in campagna elettorale. Amnesty International. https://www.amnesty.it/barometro-odio/ Amores, J. J., Blanco-Herrero, D., Sánchez-Holgado, P., & Frías-Vázquez, M. (2021). Detectando el odio ideológico en Twitter. Desarrollo y evaluación de un detector de discurso de odio por ideología política en tuits en español. Cuadernos. Info, (49), 98–124. doi:10.7764/cdi.49.27817 Anderson, K. L., & Meyer, D. (2016). Violence against Queer People: Race, Class, Gender, and the Persistence of AntiLGBT Discrimination. By Doug Meyer. New Brunswick, N.J.: Rutgers University Press, 2015. Pp. viii+194. $80.00 (cloth); $25.95 (paper). American Journal of Sociology, 122(2), 631–633. https://www.jstor.org/stable/26545851. doi:10.1086/687190 Anderson, L., & Lepore, E. (2013). Slurring words. Noûs (Detroit, Mich.), 47(1), 25–48. doi:10.1111/j.14680068.2010.00820.x Aparici, R. (2010). Educomunicación: más allá del 2.0. Gedisa. Aprile, P. (2011). Terroni. Tutto quello che è stato fatto perché gli italiani del Sud diventassero «meridionali». Edizioni Piemme. Arboleda, Ó. (2021). Lejanías, trabajo musical y literario sobre el desplazamiento en Colombia. Onces Ed. Arce-García, S., Said-Hung, E.; Mottareale-Calvanese, D. (2022). Astroturfing as a strategy for manipulating public opinion on Twitter during the pandemic in Spain. Profesional de la información, 31(3), e310310. doi:10.3145/epi.2022.may.10

307

Compilation of References

Arce-García, S., & Menéndez-Menéndez, M.-I. (2023). Inflaming public debate: A methodology to determine origin and characteristics of hate speech about sexual and gender diversity on Twitter. El Profesional de la Información, 32(1), e320106. doi:10.3145/epi.2023.ene.06 Arce-García, S., Said-Hung, E., & Mottareale-Calvanese, D. (2022). Tipos de campaña Astroturfing de contenidos desinformativos y polarizados en tiempos de pandemia en España. Revista ICONO, 14. doi:10.7195/ri14.v21i1.1890 Arcila Calderón, C., Sánchez Holgado, P., Quintana Moreno, C., Amores, J. J., & Blanco Herrero, D. (2022). Discurso de odio y aceptación social hacia migrantes en Europa: Análisis de tuits con geolocalización. Comunicar: revista científica iberoamericana de comunicación y educación. . doi:10.3916/C71-2022-02 Arcila, C., Blanco-Herrero, D., & Valdés, M. B. (2020). Rejection and hate speech on Twitter: Content analysis of tweets about migrants and refugees. Revista Española de Investigaciones Sociológicas, 172, 21–40. doi:10.5477/cis/reis.172.21 Arcila-Calderón, C., Blanco-Herrero, D., Frías-Vázquez, M., & Seoane-Pérez, F. (2021). Refugees welcome? Online hate speech and sentiments in Twitter in Spain during the reception of the boat Aquarius. Sustainability (Basel), 13(5), 2728. doi:10.3390u13052728 Ardèvol-Abreu, A., De Zúñiga, H. G., & McCombs, M. E. (2020). Orígenes y desarrollo de la teoría de la agenda setting en Comunicación. Tendencias en España (2014-2019). El Profesional de la Información, 29(4), e290414. doi:10.3145/ epi.2020.jul.14 Arévalo-Salinas, A. I., Al-Najjar Trujillo, T., & Aidar-Abib, T. (2021). La cobertura informativa de la inmigración en Televisión Española. El caso del barco Aquarius. Estudios sobre el Mensaje Periodístico, 27(1), 13–25. doi:10.5209/ esmp.73744 Arguedas, A. R., Robertson, C. T., Fletcher, R., & Nielsen, R. K. (2022). Echo Chambers, Filter Bubbles, and Polarisation: a Literature Review. Reuters Institute-University of Oxford. https://reutersinstitute.politics.ox.ac.uk/echo-chambers-fil ter-bubbles-and-polarisation-literature-review Aro, J. (2022). Putin’s trolls. Ig Publishing. Arrocha, W. (2019). Combating xenophobia and hate through compassionate migration: The present struggle of irregular migrants escaping fear and extreme poverty. Crime, Law, and Social Change, 71(3), 245–260. doi:10.100710611-01909833-w Arroyo-López, C., Moreno-López, R., & Flores-Martos, J. A. (2021). Dignity, equality, and human rights. En MorenoLópez., R & Arroyo-López., C (Eds.). Support and advice through health system for hate crimes victims: a socio-sanitary approach (9-20). Tirant Lo Blanch. Aruguete, N. (2015). El poder de la agenda. Política, medios y público. Biblos. Aslan, A. (2017). Online hate discourse: A study on hatred speech directed against Syrian refugees on YouTube. Journal of Media Critiques, 3(12), 227–256. doi:10.17349/jmc117413 Assemblée Nationale Française. (2021). Loi nº 2021-1109 du 24 août 2021 confortant le respect des príncipes de la République. https://cutt.ly/B8IfzMq Assimakopoulos, S., Baider, F. H., & Millar, S. (2017). Online hate speech in the European Union: a discourse-analytic perspective. Springer Nature. doi:10.1007/978-3-319-72604-5 Atalay, O. (2014). Suriyeli muhacirler bizim kardeşimizdir [Syrian refugees are our brothers]. Yeni Akit. https://www. yeniakit.com.tr/yazarlar/osman-atalay/suriyeli-muhacirler-bizim-kardesimizdir-7332.html

308

Compilation of References

Ataman, H. (2014). Yazılı basında ayrımcı söylem, Suriyeli mülteciler. Medyada Nefret Söylemi ve Ayrımcı Dil. [Discriminatory discourse in the print media, Syrian refugees. Hate Speech and Discriminatory Language in the Media]. https://hrantdink.org/tr/asulis/yayinlar/72-medyada-nefret-soylemi-raporlari/394-medyada-nefret-soylemi-ve-ayrimcidil-eylul-aralik-2014 Auld, E., & Morris, P. (2019). Science by streetlight and the OECD’s measure of global competence: A new yardstick for internationalisation? Policy Futures in Education, 17(6), 677–698. doi:10.1177/1478210318819246 Austin, J. (1962). How to do things with words. Oxford University Press. Awan, I., & Zempi, I. (2016). Islamophobia and Twitter: A typology of online hate against Muslims on social media. Policy and Internet, 8(2), 133–150. doi:10.1002/1944-2866.POI364 Ayuntamiento de Barcelona. (2017). Informe Discurso del Odio. Resumen Ejecutivo. Ayuntamiento de Barcelona. Ayuste, A., Flecha, R., López, P., & Lleras, J. (1998). Planteamientos de la pedagogía crítica. Comunicarransformermar. Editorial Graó. Azqueta, A., & Merino-Arribas, A. (2022). The role of education in Preventing Violent Radicalisation: A review for Europe. Review of Education, 397(3), 209–238. doi:10.4438/1988-592X-RE2022-397-54 Bachner, J. (2013). Predictive policing: preventing crime with data and analytics. IBM Center for the Business of Government. https://businessofgovernment.org/sites/default/files/Management%20Predictive%20Policing.pdf Baghel, S. S. (2020). Ethics in Journalism: A Forgotten Pillar. MediaSpace: DME Journal of Communication, 1, 92–107. Bagus, A. M., Darajaatul, N., Azizah, M., Amal, B. K., & Purba, A. S. (2022). Deradicalisation Model Through Islamic Education Curriculum in Indonesia, Spain, And Nigeria. Res Militaris, 12(2), 428–439. Bahlmann, C., Haasdonk, B., & Burkhardt, H. (2002). Online handwriting recognition with support vector machines— A kernel approach. Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition, (pp. 49-54). IEEE. 10.1109/IWFHR.2002.1030883 Bakali, N. (2016). Islamophobia: Understanding Anti-Muslim Racism through the Lived Experiences of Muslim Youth. Sense Publishers. doi:10.1007/978-94-6300-779-5 Baker, P., & Gabrielatos, C., KhosraviNik, M., Krzyanowski, M., McEnery, T., & Wodak, R. (2011). ¿Una sinergia metodológica útil? Combinar análisis crítico del discurso y lingüística de corpus para examinar los discursos de los refugiados y solicitantes de asilo en la prensa británica. Discurso & Sociedad, 5(2), 376–416. Bakir, V., & McStay, A. (2018). Fake news and the economy of emotions. Digital Journalism (Abingdon, England), 6(2), 154–175. doi:10.1080/21670811.2017.1345645 Balabanova, E., & Trandafoiu, R. (2020). Media, migration, and human rights: Discourse and resistance in the context of the erosion of liberal norms. Journal of Language and Politics, 19(3), 379–390. doi:10.1075/jlp.20007.bal Balasco, N., D’Alessandro, V., Ferrara, P., Smaldone, G., & Vitagliano, L. (2021). Analysis of the time evolution of COVID-19 lethality during the first epidemic wave in Italy. Acta Biomedica, 92(2), e2021171. doi:10.23750/abm. v92i2.11149 PMID:33988144 Balch, A., & Balabanova, E. (2016). Ethics, politics, and migration: Public debates on the free movement of Romanians and Bulgarians in the UK, 2006–2013. Politics, 36(1), 19–35. doi:10.1111/1467-9256.12082 Balkin, J. M. (2017). Digital speech and democratic culture: A theory of freedom of expression for the information society. In Law and Society Approaches to Cyberspace (pp. 325–382). Routledge. doi:10.4324/9781351154161-9 309

Compilation of References

Banfield, E. C. (1958). The Moral Basis of a Backward Society. Free Press. Barbieri, F., Anke, L. E., & Camacho-Collados, J. (2022). Xlm-t: Multilingual language models in twitter for sentiment analysis and beyond. In Proceedings of the Thirteenth Language Resources and Evaluation Conference (pp. 258-266). ACL. https://aclanthology.org/2022.lrec-1.27/ Barcella, P. (2018). Percorsi leghisti: Dall’antimeridionalismo alla xenofobia. Meridiana, 91, 95–119. Barisione, M., Michailidou, A., & Airoldi, M. (2019). Understanding a digital movement of opinion: The case of# RefugeesWelcome. Information Communication and Society, 22(8), 1145–1164. doi:10.1080/1369118X.2017.1410204 Barreda, M. (2021). Political polarisation in current democracies: Expressions and consequences. Manuel Giménez Abad Notebooks, 21. doi:10.47919/FMGA.CM21.0109 Barrett, M., & Byram, M. (2020). Errors by Simpson and Dervin (2019) in their description of the Council of Europe’s Reference Framework of Competences for Democratic Culture. Intercultural Communication Education, 3(2), 75–95. doi:10.29140/ice.v3n2.286 Barrios-Rubio, A. & Fajardo, G. (2022). The Media in the Construction of Reality in the Context of Colombian Social Nonconformity. Information, 13(8), 378, 1-14. https://doi.org/ doi:10.3390/info13080378 Barthes, R. (1970). L’ancienne rhétorique. Communications (Englewood), 16(1), 172–229. doi:10.3406/comm.1970.1236 Basile, V., & Cagnazzo, C. (2021). Litescale: A lightweight tool for best-worst scaling annotation. In Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021). INCOMA Ltd. 10.26615/978-954-452-072-4_015 Basile, V., Bosco, C., Fersini, E., Nozza, D., Patti, V., Pardo, F. M. R., Rosso, P., & Sanguinetti, M. (2019). SemEval-2019 task 5: Multilingual detection of hate speech against immigrants and women in Twitter. In NAACL HLT 2019 - International Workshop on Semantic Evaluation, SemEval 2019, Proceedings of the 13th Workshop (pp. 54–63). Association for Computational Linguistics (ACL). 10.18653/v1/S19-2007 Basile, V., Fell, M., Fornaciari, T., Hovy, D., Paun, S., Plank, B., Poesio, M., & Uma, A. (2021). We need to consider disagreement in evaluation. In Proceedings of the 1st Workshop on Benchmarking: Past, Present and Future. Association for Computational Linguistics. https://doi.org/10.18653/v1/2021.bppf-1.3 Bassignana, E., Basile, V., & Patti, V. (2018). Hurtlex: A multilingual lexicon of words to hurt. In Proceedings of the Fifth Italian Conference on Computational Linguistics (CLiC-it). 10.4000/books.aaccademia.3085 Battistelli, A., Cipolletta, S., & Mocini, F. (2020). The role of empathy and moral disengagement in predicting cyberbullying perpetration and victimization among adolescents. Journal of Interpersonal Violence, 35(23-24), 5411–5430. doi:10.1177/0886260518765961 Bauman, Z. (2016). Extraños llamando a la puerta. Paidós. Baumgartner, S. E., Sumter, S. R., Peter, J., & Valkenburg, P. M. (2018). Online hate and harmful speech: Psychological consequences and empirical responses. Journal of Social and Political Psychology, 6(1), 183–213. doi:10.5964/jspp. v6i1.849 Baviera, T., Sampietro, A., & García-Ull, F. J. (2019). Political conversations on Twitter in a disruptive scenario: The role of “party evangelists” during the 2015 Spanish general elections. Communication Review, 22(2), 117–138. doi:10 .1080/10714421.2019.1599642

310

Compilation of References

Bayer, J. (2021). High-impact hate speech by persons of authority: A lower threshold needed? Hungarian Journal of Legal Studies, 61(3), 269–284. doi:10.1556/2052.2020.00003 Bayrakli, E & Hafez, F. (2020). European Islamophobia Report 2019. Foundation for Political, Economic and Social Research (SETA). Bayrakli, E., & Hafez, F. (2018). The State of Islamophobia in Europe in 2018, in Islamophobia Report. Islamophobia Europe. http://www.islamophobiaeurope.com/wp-content/uploads/2018/07 /EIR_2017.pdf Ben-David, A., & Fernández, A. M. (2016). Hate speech and covert discrimination on social media: Monitoring the Facebook pages of extreme-right political parties in Spain. International Journal of Communication, 10, 1167–1193. Benesch, S. (2021). Dangerous Speech: A Practical Guide. The Dangerous Speech Project. https://dangerousspeech.org/ Benesch, S., Buerger, C., Glavinic, T., & Manion, S. (2019). Dangerous Speech. A practical guide. Dangerous Speech Project. https://www.researchgate.net/publication/330684579_Dangerous _Speech_A_Practical_Guide Benítez-Andrades, J. A., González-Jiménez, Á., López-Brea, Á., Aveleira-Mata, J., Alija-Pérez, J. M., & García-Ordás, M. T. (2022). Detecting racism and xenophobia using deep learning models on Twitter data: CNN, LSTM and BERT. PeerJ. Computer Science, 8, e906. doi:10.7717/peerj-cs.906 PMID:35494847 Berecz, T., & Devinat, C. (2017). Relevance of cyber hate in Europe and current topics that shape online hate speech. In Countering Cyber Hate Phenomena. INACH. Berger, J. M., & Morgan, J. (2016). The ISIS Twitter Census: Defining and Describing the Population of ISIS Supporters on Twitter. The Brookings Project on U.S. Relations with the Islamic World. https://www.brookings.edu/wp-content/ uploads/2016/07/The-ISIS-Twitter-Census-Berger-Morgan.pdf Berger, J. (2013). Beyond viral: Interpersonal communication in the internet age. Psychological Inquiry, 24(4), 293–296. doi:10.1080/1047840X.2013.842203 Berger, P. L., & Luckman, T. (1966). The Social Construction of Reality. Anchor Books. Bernal, P., Villadiego, M., & Urbanczyk, M. (2013). Narrativas publicitarias de la modernidad en la prensa colombiana de los años treinta. Poliantea, IX(16), 211–239. Beydoun, K. A. (2018). American Islamophobia: Understanding the Roots and Rise of Fear. University of California Press. doi:10.1525/9780520970007 Bhavsar, H., & Panchal, M. H. (2012). A review on support vector machine for data classification. [IJARCET]. International Journal of Advanced Research in Computer Engineering and Technology, 1(10), 185–189. Binderkrantz, A. S. (2017). Interest group representation and framing in the media: A policy area perspective. Journal of Public Affairs, 20(1), e2009. doi:10.1002/pa.2009 Blackburn, M. V., & Clark, C. T. (2007). Literacy research for political action and social change. Peter Lang. Blank, W. (1997). Authentic instruction. In W.E. Blank & S. Harwell (Eds.), Promising practices for connecting high school to the real world (pp. 15–21). University of South Florida. Blázquez, M. D. (2013). La Directiva 2012/29/UE ¿Un paso adelante en materia de protección a las víctimas en la Unión Europea? Revista de Derecho Comunitario Europeo, 17(46), 897–934.

311

Compilation of References

Bliuc, A.-M., Faulkner, N., Jakubowicz, A., & McGarty, C. (2018). Online networks of racial hate: A systematic review of 10 years of research on cyber-racism. Computers in Human Behavior, 87, 75–86. doi:10.1016/j.chb.2018.05.026 Bohórquez, D. (1997). Julia Kristeva: teoría, proceso e interpretación del sentido. Signa: revista de la Asociación Española de Semiótica, 6. Bolinger, R. (2017). The Pragmatics of Slurs. Noûs (Detroit, Mich.), 51(3), 439–462. doi:10.1111/nous.12090 Bonacho, M. (2021). La responsabilidad de los medios de comunicación al informar sobre la comisión de delitos de odio. Economist & Jurist, 29(255), 68–73. Bonilla del Rio, M., Valor-Rodríguez, L., & García-Ruiz, R. (2018). Alfabetización mediática y discapacidad: Análisis documental de literatura científica en web of science (WOS) y scopus. Prisma Social, 20, 1–20. https://revistaprismasocial.es/article/view/2292/2472 Borell, K. (2015). When is the time to hate? A research review on the impact of dramatic events on Islamophobia and Islamophobic hate crimes in Europe. Islam & Christian-Muslim Relations, 26(4), 409–421. doi:10.1080/09596410.20 15.1067063 Bourdieu, P. (2012). ¿Cómo se forma la “opinión pública”? Sociólogos-Blog de Actualidad y Sociología. https://goo. gl/lc5kuo Bourou, M. (2021). Diversity is a code word. And what it means is white genocide! Journal of Language Aggression and Conflict, 10(1), 219 239. https://doi.org/ doi:10.1075/jlac.00060.bou Boyd, D., Golder, S., & Lotan, G. (2010, January). Tweet, tweet, retweet: Conversational aspects of retweeting on twitter. In 2010 43rd Hawaii international conference on system sciences (pp. 1-10). IEEE. Branigan, H. P., & Feleki, E. (2020). Conceptual Accessibility and Serial Order in Greek Speech Production. In Proceedings of the Twenty-first Annual Conference of the Cognitive Science Society. Psychology Press. 10.4324/9781410603494-22 Breazu, P., & Machin, D. (2022). Racism is not just hate speech: Ethnonationalist victimhood in YouTube comments about the Roma during Covid-19. Language in Society, 1–21. doi:10.1017/S0047404522000070 Briant, E. (2020). Inside Cambridge Analytica and the Digital Influence Industry. Propaganda Machine. https://www. propagandamachine.tech/ Brindle, A. (2016). Cancer has nothing on Islam: A study of discourses by group elite and supporters of the English defense league. Critical Discourse Studies, 13(4), 444–459. doi:10.1080/17405904.2016.1169196 Bron, M. (2019). Comunicación transmedia y educación: El Aprendizaje Basado en Proyectos Colaborativos como método de aprendizaje en comunicación digital [Tesis doctoral]. Universidad Rey Juan Carlos. Brown, A. (2017). What is so special about online (as compared to offline) hate speech? Ethnicities, 18(3), 297–326. doi:10.1177/1468796817709846 Brunet, F. (2014). Curso de integración europea. Alianza. Bufale.net. (2020). Audio WhatsApp e caso di Coronavirus in Italia, all’Ospedale Vito Fazzi di Lecce: solo una bufala. Bufale.net. https://www.bufale.net/audio-whatsapp-e-caso-di-coronavirus-in-italia-allospedale-vito-fazzi-di-lecce-solouna-bufala/ Buitrago-Alonso, A., García Matilla, A., & Gutiérrez Martín, A. (2017). Perspectiva histórica y claves actuales de la diversidad terminológica aplicada a la educación mediática Historical. Revista de Educación Mediática y TIC, 6(2), 81–104. https://doi.org/https://doi.org/10.21071/edmetic.v6i2.7002 312

Compilation of References

Bull, M. (2021). The Italian government response to Covid-19 and the making of a prime minister. Contemporary Italian Politics, 13(2), 149–165. doi:10.1080/23248823.2021.1914453 Bunzl, M. (2005). Between anti‐Semitism and Islamophobia: Some thoughts on the new Europe. American Ethnologist, 32(4), 499–508. doi:10.1525/ae.2005.32.4.499 Burioni, R. (2020). Coronavirus, Roberto Burioni: “Il virus da noi non c’è, niente panico”. L’aria che tira. https://www. la7.it/laria-che-tira/video/coronavirus-roberto-burioni-il-virus-da-noi-non-ce-niente-panico-03-02-2020-305450 Burnap, P., & Williams, M. L. (2014). Hate speech, machine classification and statistical modelling of information flows on Twitter: Interpretation and communication for policy decision making. Cardiff. https://orca.cardiff.ac.uk/id/ eprint/65227/1/IPP2014-Burnap.pdf Burnap, P., & Williams, M. L. (2016). Us and them: Identifying cyber hate on Twitter across multiple protected characteristics. EPJ Data Science, 5(1), 1–15. doi:10.1140/epjds13688-016-0072-6 PMID:32355598 Burrell, J., Kahn, Z., Jonas, A., & Griffin, D. (2019). When users control the algorithms: values expressed in practices on twitter. Proceedings of the ACM on human-computer interaction, 3(CSCW), 1-20. ACM. 10.1145/3359240 Bustos Martínez, L., De Santiago Ortega, P. P., Martínez Miró, M. A., & Rengifo Hidalgo, M. S. (2019). Hate speeches: An epidemic that spreads on the web. State of the art on racism and xenophobia on social networks. Social Mediations, 18, 25–42. doi:10.5209/meso.64527 Butler, J. (2004). Lenguaje, poder e identidad. Síntesis. Butler, J. (2021). La fuerza de la no violencia. La ética en lo político. Paidós. Cabero-Almenara, J. (2022). Educomunicación en tiempos de pandemia: Análisis de las prácticas educativas en la educación digital. Comunicar: Revista Científica de Comunicación y Educación, 30(69), 11–20. Cabo Isasi, A., & Juanatey, A. (2016). El discurso de odio en las redes sociales: un estado de la cuestión. Ayuntamiento de Barcelona. Cacciatore, M. A., Scheufele, D. A., & Iyengar, S. (2016). The end of framing as we know it. and the future of media effects. Mass Communication & Society, 19(1), 7–23. doi:10.1080/15205436.2015.1068811 Calado, P.(2018). RecomendaçãoàadesãoaoPrincípiodenão-referênciadaorigemracialeétnica,cor,nacionalidade,ascendência, território de origem e situação documental. https://www.cicdr.pt/documents/57891/110180/Recomenda%C3%A7% C3%A3o+CICDR+-+Princ%C3%ADpio+da+n%C3%A3o+refer%C3%AAncia.pd f/0cca99a1-9b0d-400b-af4c-652d5816fc6f Calderón, B. G., Córdoba-Cabús, A., & López-Martín, Á. (2022). Las fake news y su percepción por parte de los jóvenes españoles: El influjo de los factores sociodemográficos. Doxa Comunicación. Revista Interdisciplinar de Estudios de Comunicación y Ciencias Sociales, 36, 19–42. doi:10.31921/doxacom.n36a1741 Caldevilla‐Domínguez, D., Barrientos‐Báez, A., & Padilla‐Castillo, G. (2023). Dilemmas Between Freedom of Speech and Hate Speech: Russophobia on Facebook and Instagram in the Spanish Media. Politics and Governance, 11(2), 1–13. doi:10.17645/pag.v11i2.6330 Camacho-Collados, M., & Liberatore, F. (2015). A decision support system for predictive police patrolling. Decision Support Systems, 75, 25–37. doi:10.1016/j.dss.2015.04.012

313

Compilation of References

Camacho-Collados, M., Liberatore, F., & Angulo, J. M. (2015). A multi-criteria police districting problem for the efficient and effective design of patrol sector. European Journal of Operational Research, 246(2), 674–684. doi:10.1016/j. ejor.2015.05.023 Cámara, S. (2017). El concepto de delitos de odio y su comisión a través del discurso. Especial referencia al conflicto con la libertad de expresión. Anuario de Derecho Penal y Ciencias Penales, 70, 141–225. Camargo, L. (2021). El nuevo orden discursivo de la extrema derecha española: De la deshumanización a los bulos en un corpus de tuits de Vox sobre la inmigración. Cultura. Lenguaje y Representación, 26, 63–82. doi:10.6035/clr.5866 Cammaerts, B. (2015). Social media and activism. Journalism, 1027–1034. Campbell, H., & Connelly, L. (2012). Cyber behavior and religious practices on the Internet. In Z. Yan (Ed.), Encyclopedia of Cyber Behavior (Vol. 1). IGI Global. doi:10.4018/978-1-4666-0315-8.ch037 Camp, E. (2013). Slurring perspectives. Analytic Philosophy, 54(3), 330–349. doi:10.1111/phib.12022 Camp, E. (2018). A dual act analysis of slurs. In D. Sosa (Ed.), Bad words: philosophical perspectives on slurs. Oxford University Press. Campos Domínguez, E. M. (2017). Twitter y la comunicación política. El Profesional de la Información, 26(5), 785–794. doi:10.3145/epi.2017.sep.01 Campos-Domínguez, E., & Calvo, D. (2017). La campaña electoral en Internet: Planificación, repercusión, y viralización en Twitter durante las elecciones españolas de 2015. Comunicación y Sociedad, 29. https://cutt.ly/jyHur81 Capogreco, S. (2014). Italianness, Interrupted: (De)centring “self” at the transgressive boundaries and internal fissures of nation. University of Melbourne. Cardini, F. (2015). Europa e Islam: storia di un malinteso. Editori Laterza. Carratalá, A., & Herrero-Jiménez, B. (2017). Comunicación contra el odio: Análisis del estado de la discriminación en España y propuestas de actuación política, educativa y mediática. Fundación Alternativas, 91, 2017. Carratalá, A., & Herrero-Jiménez, B. (2021). La regulación contra el discurso de odio hacia el colectivo LGTBI en los medios: Análisis comparado de diez leyes autonómicas. RAEIC. Revista de la Asociación Española de Investigación de la Comunicación, 6(12), 58–80. doi:10.24137/raeic.6.12.3 Carr, J., Clifton-Sprigg, J., James, J., & Vujic, S. (2020). Love thy neighbour? Brexit and hate crime (13902). IZA Discussion Papers. doi:10.2139srn.3738844 Carr, P. R., Cuervo, S. L., & Daros, M. A. (2020). Citizen Engagement in the Contemporary Era of Fake News: Hegemonic Distraction or Control of the Social Media Context? Postdigital Science and Education, 2(1), 39–60. doi:10.100742438019-00052-z Carvalho, P., Matos, B., Santos, R., Batista, F., & Ribeiro, R. (2022). Hate Speech Dynamics Against African descent, Roma and LGBTQ+ Communities in Portugal. Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), 2362–2370. Cassany, D. (2006). Tras las líneas. Sobre la lectura contemporánea. Anagrama. Cassany, D. (2012). Leer y escribir en la red. Anagrama. Cassany, D., & Castellà, J. (2010). Aproximación a la literacidad crítica. Perspectiva Florianópolis, 28(2), 353–374. doi:10.5007/2175-795X.2010v28n2p353 314

Compilation of References

Castaño-Pulgarín, S., Suárez-Betancur, N., Tilano Vega, L., & Herrera López, H. (2021). Internet, social media and online hate speech. Systematic review. Aggression and Violent Behavior, 58, 101608. doi:10.1016/j.avb.2021.101608 Castells, M. (2009). Comunicación y poder. Alianza Editorial. Castillo Esparcia, A., & Ponce, D. G. (2015). Comunicación de crisis 2.0. Fragua. Cepollaro, B. (2017). When evaluation changes – an echoic account of appropriation and variability. Journal of Pragmatics, 117, 29–40. doi:10.1016/j.pragma.2017.05.009 Cepollaro, B., & Stojanovic, I. (2016). Hybrid evaluatives: In defense of a presuppositional account. Grazer Philosophische Studien, 93(3), 458–488. doi:10.1163/18756735-09303007 Cepollaro, B., & Thommen, T. (2019). What’s wrong with truth-conditional accounts of slurs. Linguistics and Philosophy, 42(4), 333–347. doi:10.100710988-018-9249-8 Cesari, J. (2013). Why the west fears Islam: An exploration of Muslims in liberal democracies. Springer. doi:10.1057/9781137121202 Chade, J. (2020). ‘Covid-19: Brasil, EUA e México são “ameaças” ao mundo, diz cientista suíço.’ UOL. https://noticias. uol.com.br/colunas/jamil-chade/2020/08/03/brasil-eua-e-mexico-sao-ameacas-para-o-mundo-diz-cientista-suico.htm Chadwick, A. (2017). The hybrid media system: Politics and power. Oxford University Press. doi:10.1093/ oso/9780190696726.001.0001 Chalmers, J., & Leverick, F. (2017). A Comparative Analysis of Hate Crime Legislation. A Report to the Hate Crime Legislation Review. University of Glasgow. Charitidis, P., Doropoulos, S., Vologiannidis, S., Papastergiou, I., & Karakeva, S. (2020). Towards countering hate speech against journalists on social media. Online Social Networks and Media, 17, 100071. doi:10.1016/j.osnem.2020.100071 Chaudhuri, B. B., & Bhattacharya, U. (2000). Efficient training and improved performance of multilayer perceptron in pattern classification. Neurocomputing, 34(1), 11–27. doi:10.1016/S0925-2312(00)00305-2 Cheddadi, Z. (2020). Discurso político de Vox sobre los menores extranjeros no acompañados. Inguruak. Revista Vasca de Sociología y Ciencia Política, 69(69), 57–77. doi:10.18543/inguruak-69-2020-art04 Cheng, J., Danescu-Niculescu-Mizil, C., & Leskovec, J. (2018). Antisocial Behavior in Online Discussion Communities. In Proceedings of the ACM on Human-Computer Interaction, 2(CSCW), 1-27. ACM. 10.1145/3274437 Chen, Y., Zhou, Y., Zhu, S., & Xu, H. (2012). Detecting offensive language in social media to protect adolescent online safety. In 2012 International Conference on Privacy, Security, Risk and Trust and 2012 International Conference on Social Computing. IEEE. 10.1109/SocialCom-PASSAT.2012.55 Chenzi, V. (2021). Fake news, social media and xenophobia in South Africa. African Identities, 19(4), 502–521. doi:1 0.1080/14725843.2020.1804321 Chetty, N., & Alathur, S. (2018). Hate speech review in the context of online social networks. Aggression and Violent Behavior, 40, 108–118. doi:10.1016/j.avb.2018.05.003 Chiaro, D. (2017). “Vivi Pericolosamente”: Christie Davies, Italians and dangerous things. The European Journal of Humour Research, 5(4), 41–50. doi:10.7592/EJHR2017.5.4.chiaro

315

Compilation of References

Chib, A., Bentley, C., & Wardoyo, R. (2019). Entornos digitales distribuidos y aprendizaje: Empoderamiento personal y transformación social en colectivos discriminados. Comunicar, 27(58), 51-61. https://doi.org/https://doi.org/10.3916/ C58-2019-05 Christopoulou, A. (2012). A lexicological analysis of slang vocabulary of Modern Greek. [PhD dissertation, University of Patras]. Christopoulou, K., Xydopoulos, G. J., & Tsangalidis, A. (2022). Grammatical gender and offensiveness in Modern Greek slang vocabulary. In N. Knoblock (Ed.), The Grammar of Hate: Morphosyntactic Features of Hateful, Aggressive, and Dehumanizing Discourse. Cambridge University Press. doi:10.1017/9781108991841.005 Ciardi, L. (2022, Feb 9). Covid, domande e risposte: quarantene, green pass, terza dose e positivi fuori regione. La Nazione. https://www.lanazione.it/cronaca/domande-covid-1.7301019 CIS. (2022). Barómetro de diciembre de 2022. CIS. https://www.cis.es/cis/export/sites/default/Archivos/Margina les/3380_3399/3388/es3388mar.pdf Citron, D. K. (2014). Hate crimes in cyberspace. Harvard University Press. doi:10.4159/harvard.9780674735613 Clarin:el. (2021). LEXIS Computational Lexicon of modern Greek. Version 1.0.0 (automatically assigned). Institute for Language and Speech Processing - Athena Research Center. http://hdl.handle.net/11500/CLARIN-EL-0000-0000-6105-D Coelho, P. (2023, Março 14). A ERC arquivou uma queixa contra as grandes reportagens “Quando o Ódio Veste Farda” e elogiou o rigor e o interesse público da matéria. Desta vez, finalmente, sinto-me tentado a concordar com a ERC. Twitter. https://twitter.com/PedroMCoelho/status/1635743946868400129? cxt=HHwWgoCxiYCdqrMtAAAA Cohen, L. E., & Felson, M. (2018). Cambio social y tendencias en la tasa de criminalidad: un enfoque desde las actividades cotidianas. Revista de derecho penal y criminología, (20), 359-369. doi:10.5944/rdpc.20.2018.26497 Colombini, M., Mayhew, S. H., Hawkins, B., Bista, M., Kumar Joshi, S., Schei, B., & Watts, Ch. (2016). Agenda setting and framing of gender-based violence in Nepal: How it became a health issue. Health Policy and Planning, 31(4), 493–503. doi:10.1093/heapol/czv091 PMID:26412857 Comissão para a Igualdade e Contra a Discriminação Racial. (2017). Relatório Anual 2017—Igualdade e não Discriminação em Razão da Origem Racial e Étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2018). Relatório Anual 2018—Igualdade e não Discriminação em Razão da Origem Racial e Étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2019). Relatório Anual 2019—Igualdade e não Discriminação em Razão da Origem Racial e étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2020). Relatório Anual 2020—Igualdade e não Discriminação em Razão da Origem Racial e étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Comissão para a Igualdade e Contra a Discriminação Racial. (2021). Relatório Anual 2021—Igualdade e não Discriminação em Razão da Origem Racial e étnica, cor, Nacionalidade, Ascendência e Território de Origem. CICDR. Conde, M. A. (2017). Comunicación, movilización y participación desde las redes sociales en un proyecto social. In Re-construyendo un mundo con ojos de niñas. Entre la pobreza y la educación (165-174). GEU Editorial.

316

Compilation of References

Consejo de Seguridad. (2014a). Resolución 2170 (2014). Amenazas a la paz y la seguridad internacionales causadas por actos terroristas. Aprobada por el Consejo de Seguridad, 7242a sesión, 15 de agosto 2014. https://www.un.org/ securitycouncil/es/s/res/2170-%282014%29 Consejo de Seguridad. (2014b). Resolución 2178 (2014). Amenazas a la paz y la seguridad internacionales causadas por actos terroristas. Aprobada por el Consejo de Seguridad, 7272a sesión, 24 de septiembre 2014. https://www.un.org/ securitycouncil/es/s/res/2178-%282014%29 Conselho de ministros do Portugal (2015). Estratégia Nacional de Combate ao Terrorismo. Diário da República, 1(36). https://bit.ly/3lkWZsO Conticini, E., Frediani, B., & Caro, D. (2020). Can atmospheric pollution be considered a co-factor in extremely high level of SARS-CoV-2 lethality in Northern Italy? Environmental Pollution, 261, 114465. doi:10.1016/j.envpol.2020.114465 PMID:32268945 Contreras, J.M., & Sánchez, R. (dir.) (2021). El tratamiento normativo del discurso del odio. Thomson Reuters Aranzadi. Contreras, R. (2002). La Investigación Acción Participativa (IAP): revisando sus metodologías y sus potencialidades. In Experiencias y metodologías de la investigación participativa. Naciones Unidas, CEPAL, ECLAC. Contreras-Pulido, P., & García-Galera, M. C. (Eds.). (2022). Tendencias educativas y tecnológicas: El desafío de la innovación en educación. Dykinson. Cook, C., Conijn, R., Schaafsma, J., & Antheunis, M. (2019). For Whom the Gamer Trolls: A Study of Trolling Interactions in the Online Gaming Context. Journal of Computer-Mediated Communication, 24(6), 293–318. doi:10.1093/ jcmc/zmz014 Coolsaet, R. (2019). Radicalisation: The origins and limits of a contested concept. In N. Fadil (Ed.), Radicalisation in Belgium and the Netherlands: Critical Perspectives on Violence and Security (pp. 29-51). IB Tauris. Cortes, C., & Vapnik, V. (1995). Support-Vector Networks. Machine Learning, 20(3), 3. doi:10.1007/BF00994018 Cortina, A. (2017). Aporofobia, el rechazo al pobre. Un desafío para la democracia. Paidós. Council of Europe. (1949). Statute of Council of Europe, London, May 5, 1949. https://rm.coe.int/1680306052 Council of Europe. (2014). Annual report on E.C.R.I.’s activities. C.R.I., 26. https://rm.coe.int/annual-report-on-ecri-sactivities-covering-the-period-from-1-january-/16808ae6d4 Council of Europe. (2016a). Competences for democratic culture: Living together as equals in culturally diverse democratic societies. Council of Europe. https://rm.coe.int/16806ccc0c Council of Europe. (2016b). Reference framework of competences for democratic culture: guidance for implementation (I-III). Council of Europe. https://xurl.es/e75m5 Council of Europe. (2017). Recommendation N. R (97) 20. https://rm.coe.int/1680505d5b Couraceiro, P., Paisana, M., Pais, P. C., Cardoso, G., & Baldi, V. (2019). Fake News em ano eleitoral. Portugal em linha com a UE. Publicações OberCom. Crilley, R., & Gillespie, M. (2019). What to do about social media? Politics, populism and journalism. Journalism, 20(1), 173–176. doi:10.1177/1464884918807344 Croom, A. (2013). How to Do Things with Slurs: Studies in the Way of Derogatory Words. Language & Communication, 33(3), 177–204. doi:10.1016/j.langcom.2013.03.008 317

Compilation of References

Crow, D. A., & Lawlor, A. (2016). Media in the policy process: Using framing and narratives to understand policy influences. The Review of Policy Research, 33(5), 472–491. doi:10.1111/ropr.12187 Currant, K., O’Hara, K., & O’Brien, S. (2011). The role of Twitter in the world of business. International Journal of Business Data Communications and Networking, 7(3), 1–15. doi:10.4018/jbdcn.2011070101 Custodero, A. (2020, Mar 3). Francia, rimosso il video sulla pizza al coronavirus. Le scuse di Canal+. La Repubblica. https://www.repubblica.it/politica/2020/03/03/news/francia_spot_satirico_corona_pizza_coronavirus_virus_wuhan_italia_influenza_bellanova_canal_video_vergognoso_e_raccapriccia-250109103/ Cuthbertson, L. (2019, Nov 25). Women, politics and Twitter: using machine learning to change the discourse. arXiv preprint. https://arxiv.org/abs/1911.11025 D’Alessandro, J. (2020, Apr 23). Coronavirus, l’illusione della grande fuga da Milano. Ecco i veri numeri degli spostamenti verso sud. La Repubblica. https://www.repubblica.it/tecnologia/2020/04/23/news/coronavirus_l_illusione_della_ grande_fuga_da_milano_e_i_veri_numeri_degli_spostamenti_verso_sud-254722355/ Da Empoli, G. (2019). Les ingénieurs du chaos. Jean-Claude Lattès. Damcevic, K., & Rodik, F. (2018). Ready for the Homeland. Romanian Journal of Communication and Public Relations, 20(3), 31. doi:10.21018/rjcpr.2018.3.264 Daniels, J. (2013). Race and racism in Internet studies: A review and critique. New Media & Society, 15(5), 695-719. Daniels, J. (2013). Race and racism in Internet Studies: A review and critique. New Media & Society, 15(5), 695–719. doi:10.1177/1461444812462849 Dante, A., (1983). La divina Commedia, Inferno. Angelo Signorelli editore. Davidson, T., Bhattacharya, A., & Weber, I. (2018). Identifying multiple types of social influence in online hate speech. In Proceedings of the 2018 World Wide Web Conference (pp. 1591-1600). International World Wide Web Conferences Steering Committee. Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated hate speech detection and the problem of offensive language. In Proceedings of the International AAAI Conference on Web and Social Media (pp. 512-515). AAAI. https://arxiv.org/abs/1703.04009 10.1609/icwsm.v11i1.14955 De Coninck, D. (2020). Migrant categorisations and European public opinion: Diverging attitudes towards immigrants and refugees. Journal of Ethnic and Migration Studies, 46(9), 1667–1686. doi:10.1080/1369183X.2019.1694406 De GibertO.PerezN.García-PablosA.CuadrosM. (2018). Hate speech dataset from a white supremacy forum. doi:10.18653/ v1/W18-5102 De Haro, J. (2019). Redes sociales en Educación. Eduredes. http://eduredes.antoniogarrido.es/uploads/6/3/1/1/6311693/ redes_sociales_educacion.pdf De LucasA.RömerM.Izquierdo-AlegríaD.Montero-DíazJ.Said-HungE. (2022). Manual para el etiquetado de mensaje de odio en medios digitales en España. doi:10.6084/m9.figshare.18316313 De Mauro, T. (2016). Le parole per ferire. Internazionale. De Riccardis, S., & Pisa, M. (2020). Coronavirus, la Lombardia diventa “zona rossa”: fuga da Milano sui treni notturni, poi in stazione torna la calma. La Repubblica. https://milano.repubblica.it/cronaca/2020/03/08/news/coronavirus_la_lombardia_sara_zona_rossa_fuga_da_milano_in_treno_e_in_auto-250603359/

318

Compilation of References

De Swart, P., Lamers, M., & Lestrade, S. (2008). Animacy, argument structure, and argument encoding. Lingua, 118(2), 131–140. doi:10.1016/j.lingua.2007.02.009 De-Casas, P., Vizcaíno-Verdú, A., & Aguaded, I. (2020). La televisión sensacionalista y hábitos de consumo en España e Italia. Estudios sobre el Mensaje Periodístico, 26(2), 483–496. doi:10.5209/esmp.67582 Del Boca, A. (2011). Italiani, brava gente? Neri Pozza Editore. Del Valle-Cano, G., Quijano-Sánchez, L., Liberatore, F., & Gómez, J. (2023). SocialHaterBERT: A dichotomous approach for automatically detecting hate speech on Twitter through textual analysis and user profiles. Expert Systems with Applications, 216, 119446. doi:10.1016/j.eswa.2022.119446 Delgado, R., & Stefancic, J. (2014). Hate speech in cyberspace. Wake Forest Law Review, 49, 319. Deloitte. (2020). First Barometer on the Perception of the Contribution of TV to Sustainable Development Goals (SDG). Deloitte. https://bit.ly/3wNbZV2 Deloughery, K., King, R. D., & Asal, V. (2012). Close cousins or distant relatives? The relationship between terrorism and hate crime. Crime and Delinquency, 58(5), 663–688. doi:10.1177/0011128712452956 Deuze, M., & McQuail, D. (2020). McQuail’s Media and Mass Communication Theory. Sage Publications. DevineD. (2018). The UK referendum on membership of the European Union as a trigger event for hate crimes. SSRN 3118190. doi:10.2139/ssrn.3118190 Devine, D. (2021). Discrete events and hate crimes: The causal role of the Brexit Referendum. Social Science Quarterly, 102(1), 374–386. doi:10.1111squ.12896 Di Fátima, B. (Ed.). (2023). Hate speech on social media. University of Beira Interior. Di Nicola, A., Andreatta, D., Martini, E., Antonopoulos, G. A., Baratto, G., Bonino, S., & Ferret, J. (2020). HATEMETER: Hate speech tool for monitoring, analysing and tackling Anti-Muslim hatred online. eCrime. Díaz, E., Flores, J. A., Árias, E., Arroyo, C., Fernández, P., Hipólito, N., & Pulido, M. (2020). Support and advice through health system for hate crimes victims. Proyecto SHELTER, Spanish Nacional Report, Llere. https://stophatedamages.eu/ wp-content/uploads/2021/01/Spanish-National-Report-DEF-11.01.21.pdf Díaz, J. A. (2018). Informe de delimitación conceptual en materia de delitos de odio. Observatorio Español del Racismo y la Xenofobia. https://tinyurl.com/spny398s Dickie, J. (1994). The South as Other: From Liberal Italy to the Lega Nord. The Italianist, 14, 124–140. Dines, N. (2013). Bad news from an aberrant city: A critical analysis of the British press’s portrayal of organised crime and the refuse crisis in Naples. Modern Italy, 18(4), 409–422. doi:10.1080/13532944.2013.801677 Diritti, V. La mappa dell’intolleranza. Anno 4. https://www.fishonlus.it/allegati/mappa_intolleranza4.pdf DISARM Foundation. (2022). Disarm blue framework. DISARM Foundation. https://disarmframework.herokuapp.com/ Disha, I., Cavendish, J. C., & King, R. D. (2011). Historical events and spaces of hate: Hate crimes against Arabs and Muslims in post-9/11 America. Social Problems, 58(1), 21–46. doi:10.1525p.2011.58.1.21 Doğanay, Ü., & Keneş, H. Ç. (2016). Yazılı basında Suriyeli ‘mülteciler’: Ayrımcı söylemlerin rasyonel ve duygusal gerekçelerinin inşası [Syrian ‘refugees’ in the print media: constructing rational and emotional justifications for discriminatory discourses]. Mülkiye Dergisi, 40(1), 143-184.

319

Compilation of References

Donini, P. G. (2014). Il mondo islamico: breve storia dal Cinquecento a oggi. Editori Laterza. Douglas, C. (2018). Religion and Fake News: Faith-Based Alternative Information Ecosystems in the US and Europe. The Review of Faith & International Affairs, 16(1), 61–73. doi:10.1080/15570274.2018.1433522 Drid, A. B. S., Abdelhamid, D., & Taleb-Ahmed, A. (2022). Support vector machine based clustering: A review. 2022 International Symposium on iNnovative Informatics of Biskra (ISNIB), (pp. 1-6). IEEE. 10.1109/ISNIB57382.2022.10076027 Duncombe, C. (2019). The politics of Twitter: Emotions and the power of social media. International Political Sociology, 13(4), 409–429. doi:10.1093/ips/olz013 Durán, D. (2015). Tutoría entre iguales: compartir la capacidad de enseñar con los alumnos. In Mayordomo, R. & Onrubia, J. (coord.). El aprendizaje cooperativo. Editorial UOC. Eberl, J. M., Meltzer, C. E., Heidenreich, T., Herrero, B., Theorin, N., Lind, F., Berganza, R., Boomgaarden, H. G., Schemer, C., & Strömbäck, J. (2018). The European media discourse on immigration and its effects: A literature review. Annals of the International Communication Association, 42(3), 207–223. doi:10.1080/23808985.2018.1497452 Echebarría-Echabe, A., & Fernández-Guede, E. (2006). Effects of attributional content and coping strategy on emotions and on behavioral intentions towards immigrants. Social Behavior and Personality, 34(1), 1–14. doi:10.2224bp.2006.34.1.1 Eco, U. (1975). Trattato di semiotica generale. Bompiani. ECRI. (2015). General Policy Recommendation NO. 15 On Combating Hate Speech. ECRI. https://rm.coe.int/ecrigeneral-policy-recommendation-no-15-on-combating-hate-speech/16808b5b01 ECRI. (2017). Recomendación de Política General nº 7 sobre la legislación nacional para combatir el racismo y la discriminación racial, aprobada el 13 de diciembre de 2002. ECRI. https://www.coe.int/t/dghl/monitoring/ecri/activities/GPR/EN /Recommendation_N7/REC7-2003-8- ESP.pdf Edelstein, A. S. (1993). Thinking about the criterion variable in agenda setting research. Journal of Communication, 43(2), 85–99. doi:10.1111/j.1460-2466.1993.tb01264.x EdwardsG. S.RushinS. (2018). The effect of President Trump’s election on hate crimes. doi:10.2139/ssrn.3102652 Efthymiou, A., Gavriilidou, Z., & Papadopoulou, E. (2014). Labeling of Derogatory Words in Modern Greek Dictionaries. In N. Lavidas, T. Alexiou, & A. Sougari (Eds.), Major Trends in Theoretical and Applied Linguistics 2. De Gruyter Open Poland. doi:10.2478/9788376560885.p12 Ekman, M. (2019). Anti-immigration and racist discourse in social media. European Journal of Communication, 34(6), 606–618. journals.sagepub.com/home/ejc. doi:10.1177/0267323119886151 Elias, C. (2020). Hate speech in Egyptian television talk shows: a qualitative study. [Thesis, The American University in Cairo, School of Global Affairs and Public Policy]. https://fount.aucegypt.edu/etds/813/ Elias, S., & Gurbanova, N. (2018). Relocating Gender Stereotypes Online: Critical Analysis of Sexist Hate Speech in Selected Social Media. Proceedings of the International Conference on Language Phenomena in Multimodal Communication (KLUA 2018). https://www.atlantis-press.com/proceedings/klua-18/25900116 Elías, C., & Catalan-Matamoros, D. (2020). Coronavirus in Spain: Fear of ‘Official’ Fake News Boosts WhatsApp and Alternative Sources. Media and Communication, 8(2), 462–466. doi:10.17645/mac.v8i2.3217 Ellinika Hoaxes. (2023). Ellinika Hoaxes. Ellinika Hoaxes. https://www.ellinikahoaxes.gr/

320

Compilation of References

Elmas, T., Overdorf, R., Özkalay, A. F., & Aberer, K. (2021). Ephemeral astroturfing attacks: The case of fake twitter trends. In 2021 IEEE European Symposium on Security and Privacy (EuroS&P) (pp. 403-422). IEEE. 10.1109/ EuroSP51992.2021.00035 Elósegui, M. (2017). Las recomendaciones de la ECRI sobre discurso del odio y la adecuación del ordenamiento jurídico español a las mismas (No. ART-2017-103148), Revista General de Derecho Canónico y Derecho Eclesiástico del Estado, 44. ElSherief, M., Kulkarni, V., Nguyen, D., Wang, W. Y., & Belding, E. (2018, June). Hate lingo: A target-based linguistic analysis of hate speech in social media. In Proceedings of the International AAAI Conference on Web and Social Media (Vol. 12, No. 1). 10.1609/icwsm.v12i1.15041 Engel, L., Rutkowski, D., & Thompson, G. (2019). Toward an international measure of global ompetence? A critical look at the PISA 2018 framework. Globalisation, Societies and Education, 17(2), 117–131. doi:10.1080/14767724.20 19.1642183 ENISA. (2023). NIS Directive. European Union Agency for Cybersecurity [Topic]. ENISA. https://www.enisa.europa. eu/topics/cybersecurity-policy/nis-directive-new Entidade Reguladora para a Comunicação Social. (2019). Desinformação – Contexto Europeu e Nacional. https://www. parlamento.pt/Documents/2019/abril/desinformacao_contextoeuroeunacional-ERC-abril2019.pdf Entidade Reguladora para a Comunicação Social. (2021a). Relatório da Regulação 2021. https://www.erc.pt/download. php?fd=12855&l=pt&key=bc4ddcc6d69ae573d5f148d6c3659094 Entidade Reguladora para a Comunicação Social. (2021b). ERC/2021/291 (CONTJOR-NET). https://www.erc.pt/download/YToyOntzOjg6ImZpY2hlaXJvIjtzOjM5OiJtZWRpYS9kZWNpc29lcy9vYmplY3RvX29mZmxpbmUvODI5MC5wZGYiO3M6NjoidGl0dWxvIjtzOjM0OiJkZWxpYmVyYWNhby1lcmMyMDIxMjkxLWNvbnRqb3ItbmV0Ijt9/ deliberacao-erc2021291-contjor-net Erdoğan, M. (2019). Syrian refugees in Turkey. Konrad–Adenauer-Stiftung Report. Erdoğan, M. M. (2020). Onuncu Yılında Türkiye’deki Suriyeliler [Syrians in Turkey in its Tenth Anniversary]. Uluslararası İlişkiler Konseyi Panorama, 30. Erdoğan, R. T. [@RTErdogan]. (2013). Allah’ın yardımı ne zaman diye soran Suriyeli kardeşlerime, şunu söylüyorum Hiç kuşkusuz Allah’ın yardımı yakındır [To my Syrian brothers and sisters, who ask when Allah’s help will come, I say this: No doubt, Allah’s help is near]. Twitter. https://twitter.com/rterdogan/status/331045737511149568 Erdoğan, M. (2014). Türkiye’deki Suriyeliler: Toplumsal kabul ve uyum araştırması [Syrians in Turkey: Social acceptance and cohesion research]. Hacettepe Üniversitesi Göç ve Siyaset Araştırmaları Merkezi. Espino, T. (2019). Palabras extrañas a sí mismas una aproximación desde la hermenéutica intercultural al bilingüismo de Julia Kristeva. RILCE. Revista de Filología Hispánica, 1(35), 117–136. Esposito, R. (2006). The immunisation paradigm. Campbell, T. (transl.). diacritics, 36(2), 23-48. Esposito, E. (2021). The visual semiotics of digital misogyny: Female leaders in the viewfinder. Feminist Media Studies, 1–17. doi:10.1080/14680777.2022.2139279 Esposito, M. (2013). Separiamoci. Magenes.

321

Compilation of References

Estrada-Cuzcano, A., Alfaro-Mendives, K., & Saavedra-Vásquez, V. (2020). Disinformation y Misinformation, Posverdad y Fake News: Precisiones conceptuales, diferencias, similitudes y yuxtaposiciones. Información, Cultura y Sociedad, (42), 93–106. doi:10.34096/ics.i42.7427 Eurobarómetro Comisión Europea. (2022). Eurobarómetro Standard 96: Opinión pública en la Unión Europea. EC. https://cutt.ly/3GAkK5L European Comission. (n.d.). The EU Code of conduct on countering illegal hate speech online. https://commission.europa.eu/strategy-and-policy/policies/justice-and-fundamental-rights/combatting-discrimination/racism-and-xenophobia/ eu-code-conduct-countering-illegal-hate-speech-online_en European Commission (2003). Communication from the Commission to the Council, the Council, the European Parliament, The European Economic and Social Committee and the Committee of the Regions, 3.06.2003. On immigration, integration and employmen. COM (2003) 336 final. European Commission (2020). Communication from the Commission to the European Parlament, the European Council, the Council, the European Economic and Social Committee and the Committee of the Regions, 9.12. 2020. A CounterTerrorism Agenda for the EU: Anticipate, Prevent, Protect,Respond. COM (2020) 795 final. European Commission against Racism and Intolerance. (2016). Recomendación General Nº 5 relativa a la lucha contra el discurso de odio y memorándum explicativo. Unión Europea. https://rm.coe.int/ecri-general-policy-recommendation-n-15-o n-combating-hate-speech-adopt/16808b7904 European Commission against Racism and Intolerance. (2020). Annual Report 2020 European Commission against Racism and Intolerance. https://rm.coe.int/annual-report-on-ecri-s-activities-for-20 20/1680a1cd59 European Commission. (2016). Code of conduct on countering illegal hate speech online. EC. https://ec.europa.eu/ justice/fundamental-rights/files/hate_speech_code_of_conduct_en.pdf European Commission. (2021). REAL-UP project, CERV-2021-EQUA. EC. https://real-up.eu/ European Network Against Racism - E.N.A.R. (2012). 2011/12 Shadow Reports on racism in Europe. Author. European Union Agency for Fundamental Rights. (2012). Making hate crime visible in the European Union: Acknowledging victims’ rights, Luxembourg, Publications Office; European Union Agency for Fundamental Rights. (2021). Fundamental Rights Report 2021. https://fra.europa.eu/sites/ default/files/fra_uploads/fra-2021-fundamental-rights-report-2021_en.pdf European Union Agency for Fundamental Rights. (2022). Relatório sobre os Direitos Fundamentais 2022 — Pareceres da FRA. https://fra.europa.eu/en/publication/2022/fundamental-rights-report-2022-fra-opinions European Union Agency for the Fundamental Rights. (2013). F.R.A. Opinion – 02/2013 Framework Decision on Racism and Xenophobia. Author. Eurydice. (2016). Promoting citizenship and the common values of freedom, tolerance and non-discrimination through education: Overview of education policy developments in Europe following the Paris Declaration of March 17 2015. Luxembourg: Publications Office EU. doi:10.2797/396908 Eurydice. (2019). Integrating students from migrant backgrounds into schools in Europe: national policies and measures. Louxembourg: Education, Audiovisual and Culture Executive Agency. doi:10.2797/819077

322

Compilation of References

Evenden-Kenyon, E. (2020). Educating the English: The role of universities in tackling hate speech and Islamaphobia in post-EU-Referendum Britain. Papeles de Europa, 32(2), 129–144. doi:10.5209/pade.68027 Evrensel. (2017). Karapınar’da Suriyelilere Dönük Irkçı Kışkırtma [Racist Provocation Against Syrians in Karapınar]. Evrensel. https://www.evrensel.net/haber/330169/karapinardasuriyelilere-donuk-irkci-kiskirtma Ezeibe, C. (2021). Hate Speech and Election Violence in Nigeria. Journal of Asian and African Studies, 56(4), 919–935. doi:10.1177/0021909620951208 Facta News. (2023). Antibufale. Facta. https://facta.news/antibufale/. Fallis, D. (2015). What is disinformation? Library Trends, 63(3), 401–426. doi:10.1353/lib.2015.0014 Faloppa, F. (2020). Odio: manuale di resistenza alla violenza delle parole. Utet. Fanon, F. (1986). Black Skin, White Masks. Pluto Press. Fanpage (2020). Che cosa significa infodemia, la malattia che secondo l’Oms fa più male del Coronavirus. Fanpage. https://www.fanpage.it/cultura/che-cosa-significa-infodemia-la-malattia-che-secondo-loms-fa-piu-male-del-coronavirus/ Farkas, J., Schou, J., & Neumayer, C. (2018). Cloaked Facebook Pages: Exploring Fake Islamist Propaganda in Social Media. New Media & Society, 20(5), 1850–1867. doi:10.1177/1461444817707759 Farwell, J. P. (2014). The media strategy of ISIS. Survival, 56(6), 49–55. doi:10.1080/00396338.2014.985436 Fausti, S., & Forti, G. (2019). Da dove vengono i ministri della Seconda Repubblica? YouTrend. https://www.youtrend. it/2019/10/22/da-dove-vengono-i-ministri-della-seconda-repubblica/ Fedorov, A., Cortés Montalvo, J. A., & Sandoval Romero, Y. (2016). La educación en comunicación en el mundo: currículum y ciudadanía. Comunicar, 24(49). https://www.revistacomunicar.com/ojs/index.php/comunicar/issue/view/c49 Feinberg, A., Branton, R., & Martinez-Ebers, V. (2022). The Trump effect: How 2016 campaign rallies explain spikes in hate. PS, Political Science & Politics, 55(2), 257–265. doi:10.1017/S1049096521001621 Feldman, A. (1997). Violence and Vision: The Prosthetics and Aesthetics of Terror. Public Culture, 10(1), 24–60. doi:10.1215/08992363-10-1-24 Feltri, V. (2020) Vittorio Feltri: I meridionali? Credo che in molti casi siano inferiori. Fuori dal coro, Rete 4. https:// mediasetinfinity.mediaset.it/article/vittorio-feltri-credo-che-i-meridionali-in-molti-casi-siano-inferiori_a11749 Fernandes, R. C., & Teles, F. (2021). Country Portugal. In J. Mulhall & S. Khan-Ruf (Eds.), State of Hate: Far-Right Extremism in Europe 2021 (pp. 103-105). Hope not Hate and the Amadeu Antonio Foundation. Fernández, I. B. (2010). Metalenguaje interactivo. Herramientas en la red para los nativos digitales. ICONO 14, Revista de comunicación y tecnologías emergentes, 8(1), 274-288. Fernández-Poncela, A. M. (2011). Prejuicios y estereotipos. Refranes, chistes y acertijos, reproductores y transgresores. Antropología experimental, (11) 17-328. http://revista.ujaen.es/huesped/rae/articulos2011/22fernande z11.pdf Fialho, J., Dias, E., & Macedo, V. (2022). O uso das redes sociais e as competências digitais dos portugueses. Universidade Lusíada Editora.

323

Compilation of References

Figueiredo, I. A. (2021, October 29). “Peço desculpa porque tribunais me obrigaram.” Chega perde contra família a quem Ventura chamou “bandidos”. Observador. https://observador.pt/2021/10/29/peco-desculpa-porque-tribun ais-me-obrigaram-chega-perde-contra-familia-a-quem-ventura-c hamou-bandidos/ Filibeli, T. E., & Ertuna, C. (2021). Sarcasm beyond hate speech: Facebook comments on Syrian refugees in Turkey. International Journal of Communication, 15, 24. Finkelstein, M. S. (1999). Separatism, the Allies and the Mafia: The struggle for Sicilian independence, 1943-1948. Lehigh University Press. Fiscalía General del Estado. (2019). Circular 7/2019, de 14 de mayo, de la Fiscalía General del Estado, sobre pautas para interpretar los delitos de odio tipificados en el artículo 510 del Código Penal. Fiscalía General del Estado. Flaxman, S., Goel, S., & Rao, J. M. (2016). Filter bubbles, echo chambers, and online news consumption. Public Opinion Quarterly, 80(1), 298–320. doi:10.1093/poq/nfw006 Fleischmann, F., & Phalet, K. (2018). Religion and national identification in Europe: Comparing muslim youth in Belgium, England, Germany, the Netherlands, and Sweden. Journal of Cross-Cultural Psychology, 49(1), 44–61. doi:10.1177/0022022117741988 PMID:29386688 Florio, E., Strapparava, C., & Bolioli, A. (2017). Analyzing message text with most algorithms trained for other languages. [TALLIP]. ACM Transactions on Asian and Low-Resource Language Information Processing, 16(2), 1–19. doi:10.1145/3094249 Fortuna, P., & Nunes, S. (2018). A survey on automatic detection of hate speech in text. ACM Computing Surveys, 51(4), 1–30. doi:10.1145/3232676 FRA European Union Agency for Fundamental Rights. (2019). Informe sobre los Derechos Fundamentales 2019. Luxembourg, Publications Office, FRA. FRA European Union Agency for Fundamental Rights. (2021). Encouraging hate crime reporting — The role of law enforcement and other authorities. Luxembourg, Publications Office; FRA. FRA. (2013). Opinion of the european union agency for fundamental rights on the framework decision on racism and Xenophobia—with special attention to the rights of victims of crime. European Union Agency for Fundamental Rights document. http://fra.europa.eu/sites/default/files/fra-opinion-2-2013-framework-decision-racism-xenophobia_en.pdf Francescutti, L. F. (2020). La semiótica de la comunicación masiva en España: Sus avatares académicos contados por sus protagonistas. Revista Signa, 29, 391–414. doi:10.5944igna.vol29.2020.23455 Francioni, P. (2020). La gaffe l’inviata di Agorà: “Non siamo fortunati, non c’è nessuno”. Il Giornale. https://www. ilgiornale.it/news/spettacoli/coronavirus-linviata-agor-delusa-virtuosismo-dei-napoletani-1854418.html Francklin, E. (2020). Coronavirus Bari, ammassati con la mascherina sotto al mento: fase 8 fuori dalla posta al San Paolo. Il Quotidiano Italiano — Bari. https://bari.ilquotidianoitaliano.com/attualita/2020/04/news/coronavirus-bari-ammassaticon-la-mascherina-sotto-al-mento-fase-8-fuori-dalla-posta-al-san-paolo-271075.html/ Frantzi, K., Georgalidou, M., & Giakoumakis, G. (2019). Greek Parliamentary Discourse in the Years of the Economic Crisis. In E. Jakasa (Ed.), Argumentation and Appraisal in Parliamentary Discourse. Information Science Reference. doi:10.4018/978-1-5225-8094-2.ch001 Freedom House. (2016). Freedom on the Internet 2016. Freedom House. https://freedomhouse.org/sites/default/ files/2020-02/FOTN_2016_BOOKLET_FINAL.pdf 324

Compilation of References

FreedomHouse. (2020). The Pandemic’s Digital Shadow. Freedom House. https://freedomhouse.org/report/freedomnet/2020/pandemics-digital-shadow Frischlich, L. (2022). ‘Resistance!’: Collective action cues in conspiracy theory-endorsing Facebook groups. Impact of social media on social cohesion. Media and Communication, 10(2), 130–143. doi:10.17645/mac.v10i2.5182 Gaines, E. (2015). Undestanding the Codes and Assumptions of New Media. In P. P. Trifonas (Ed.), International Handbook of Semiotics (pp. 369–376). Springer., doi:10.1007/978-94-017-9404-6_16 Galar, M., Fernandez, A., Barrenechea, E., Bustince, H., & Herrera, F. (2012). A Review on Ensembles for the Class Imbalance Problem: Bagging-, Boosting-, and Hybrid-Based Approaches. IEEE Transactions on Systems, Man, and Cybernetics. Part C, Applications and Reviews, 42(4), 463–484. doi:10.1109/TSMCC.2011.2161285 Gallo, G. I. ., & Galmarini, M. R. . (2021). Reactivación del turismo rural post cuarentena 2020: TIC y adaptaciones aplicadas por emprendedores. Revista Intersección. Eventos, Turismo, gastronomía Y Moda, 2(4), 72–82. Gallup (2016), Islamophobia: Understanding Anti-Muslim Sentiment in the West. https://news.gallup.com/poll/157082/ islamophobia-understanding-anti-muslim-sentiment-west.aspx Galofaro, F. & Toffano, Z. (2022). Enunciation and topic/comment structure: the offensive replies to Pope Francis’ tweets. RIFL, 243-253. doi:10.4396/SFL2021A23 Gamir-Ríos, J., Tarullo, R., & Ibáñez-Cuquerella, M. (2021). La desinformación multimodal sobre la otredad en Internet. Difusión de bulos racistas, xenófobos e islamófobos en 2020. Anàlisi, 64(64), 49–64. doi:10.5565/rev/analisi.3398 García-De-Torres, E., Yezers’ka, L., Rost, A., Calderín, M., Edo, C., Rojano, M., Said-Hung, E., Jerónimo, P., Arcila, C., Serrano-Tellería, A., Sánchez-Badillo, J., & Corredoira, L. (2011). Uso de Twitter y Facebook por los medios iberoamericanos. El Profesional de la Información, 20(6), 611–620. doi:10.3145/epi.2011.nov.02 García, G. (2019). Quizças… Barthes y Derrida. Acta Poética, 40(1), 63–85. doi:10.19130/iifl.ap.2019.1.846 García-Galera, M. C., & Contreras-Pulido, P. (Eds.). (2022). Comunicación y Educación: Nuevas perspectivas en investigación e intervención. Dykinson. García-Ruiz, R., & Ramírez-García, A., & Rodríguez-Rosel, M. (2014). Educación en alfabetización mediática para una nueva ciudadanía prosumidora. Comunicar, 22(43), 15–23. doi:10.3916/C43-2014-01 Gattinara, P. C. (2017). The “Refugee Crisis” in Italy as a Crisis of Legitimacy. Contemporary Italian Politics, 9(3), 318–331. doi:10.1080/23248823.2017.1388639 Gattinara, P. C., & Froio, C. (2019). Getting “Right” into the News: Grassroots Far-right Mobilization and Media Coverage in Italy and France. Comparative European Politics, 17(5), 738–758. doi:10.105741295-018-0123-4 Gazzetta del Sud. (2020). Reggio, la donna di Taurianova ricoverata in isolamento non sarebbe colpita da coronavirus. Gazzetta del Sud. https://reggio.gazzettadelsud.it/articoli/cronaca/2020/01/31/reggio-la-donna-di-taurianova-ricoveratain-isolamento-non-sarebbe-colpita-da-coronavirus-517acac6-be85-4eb3-a551-d2e7d1e3fcce/ Geiger, R. S. (2016). Bot-based collective blocklists in Twitter: The counterpublic moderation of harassment in a networked public space. Information Communication and Society, 19(6), 787–803. doi:10.1080/1369118X.2016.1153700 Gelber, K. (2019). Differentiating hate speech: A systemic discrimination approach. Critical Review of International Social and Political Philosophy, 24(4), 393–414. doi:10.1080/13698230.2019.1576006 Gelgel, N. M. R. A., Limilia, P., & Ratri Rahmiadji, L. (2023). Politicians and Celebrities: Propagators of Hate Speech in Indonesia. Proceedings of Indonesia Focus, 1(1), 6. 325

Compilation of References

General Assembly. (2016). Resolución 70/L.41. Culture of peace. Plan of Action to Prevent Violent Extremism. Seventieth session (February 9 2016). Georgeault, L. (2022). Ley de refuerzo del respeto de los principios republicanos en Francia. ¿El regreso del asimilacionismo? Eunomía. Revista en Cultura de la Legalidad, 22(22), 79–100. doi:10.20318/eunomia.2022.6808 Georgiou, M., & Zaborowski, R. (2017). Media coverage of the “refugee crisis”: A cross-European perspective. Council of Europe. Ghosh, R., Manuel, A., Chan, A., Dilimulati, M., & Babaei, M. (2016). Education and Security. A Global Literature Review on the role of education in Cuntering Violent Religious Extremism. Tony Blair Faith Foundation. Gifreu, A. (2012). El documental interactivo como nuevo género audiovisual. Estudio de la aparición del nuevo género, aproximación a su definición y propuesta de taxonomía y de modelo de análisis a efectos de evaluación, diseño y producción [Tesis doctoral]. Universidad Pompeu Fabra. Gil-Borrelli, C.C, Martín-Ríos, M.D., López-Corcuera, L., Reche-Martínez, B., Torres-Santos Olmo, R., Muriel-Pati, E., & Rodríguez-Arenas, M.A. (2020). Elaboración de un cuestionario de detección de casos de violencia de odio en urgencias hospitalarias. Gac Sanit, 34(2). doi:10.1016/j.gaceta.2019.01.006 Gil-Borrelli, C. C., Martín-Ríos, M. D., & Rodríguez-Arenas, M. Á. (2018). Propuesta de actuación para la detección y la atención a víctimas de violencia de odio para profesionales de la salud. Medicina Clínica, 150(4), 155–159. doi:10.1016/j.medcli.2017.06.017 Gillespie, T. (2018). The Politics of Platforms: A Comparative Study of Hate Speech Regulation Online. Yale University Press. Ging, D., & Siapera, E. (2019). Gender hate online: understanding the new anti-feminism. Palgrave Macmillan. doi:10.1007/978-3-319-96226-9 Giovanetti, M., Benvenuto, D., Angeletti, S., & Ciccozzi, M. (2020). The first two cases of 2019‐nCoV in Italy: Where do they come from? Journal of Medical Virology, 92(5), 518–521. doi:10.1002/jmv.25699 PMID:32022275 Giraldo Pérez, S. (2022). Delitos de odio. Incidencia de la pandemia COVID en los delitos de odio en España. Sociología y tecnociencia, 12(1), 216- 240. doi:10.24197/st.1.2022.216-240 Gislason, P. O., Benediktsson, J. A., & Sveinsson, J. R. (2006). Random Forests for land cover classification. Pattern Recognition Letters, 27(4), 294–300. doi:10.1016/j.patrec.2005.08.011 Gitari, N. D., Zuping, Z., Damien, H., & Long, J. (2015). A lexicon-based approach for hate speech detection. International Journal of Multimedia and Ubiquitous Engineering, 10(4), 215–230. doi:10.14257/ijmue.2015.10.4.21 Gobierno de España. (2015). Plan Estratégico Nacional de Lucha contra la Radicalización Violenta. Un marco para el respeto y el entendimiento común. Madrid: Ministerio del Interior. https://www.interior.gob.es/opencms/pdf/serviciosal-ciudadano/plan-estrategico-nacional-de-lucha-contra-la-radicalizacion-violenta/documentacion-del-plan/estrategiainterior/PLAN-ESTRATEGICO-NACIONAL.pdf Goffman, E. (2006). Frame Analysis. Los marcos de la experiencia. Siglo XXI Editores. Goker, G., & Keskin, S. (2015). Haber medyasi ve multeciler: Suriyeli multecilerin Turk yazili basinindaki temsili [News media and refugees: Representation of Syrian refugees in Turkish press]. Journal of Communication Theory and Research/Iletisim Kuram ve Arastirma Dergisi, 41, 229-256.

326

Compilation of References

Gomes, S. (2011). Crime na imprensa: Representações sobre Imigrantes e Ciganos em Portugal. ComTextos, 1, 1-24. https://hdl.handle.net/1822/13674 González-Álvarez, J. L., Hermoso, J. S., & Camacho-Collados, M. (2020). Policía predictiva en España. Aplicación y retos futuros. Behavior & Law Journal, 6(1), 26–41. doi:10.47442/blj.v6.i1.75 González, S., & Ramos, F. (2013). El uso periodístico de Facebook y Twitter: Un análisis comparativo de la experiencia europea. Historia y Comunicación Social, 18(0), 419–433. doi:10.5209/rev_HICS.2013.v18.44253 Goodman, S., Sirriyeh, A., & McMahon, S. (2017). The evolving (re) categorisations of refugees throughout the “refugee/ migrant crisis”. Journal of Community & Applied Social Psychology, 27(2), 105–114. doi:10.1002/casp.2302 Gordon, W., & Allport, G. (1954). The Nature of Prejudice. AddisonWesley Publishing Company. https://books.google. com/books? id=u94XUyRuDl4C Gouvernement Republique Française. (2018). Prévenir Pour Protéger. Plan national de prévention de la radicalisation. https://www.cipdr.gouv.fr/prevenir-la-radicalisation/#:~:text=d%C3%A9tecter%20et%20signaler%2C%20via%20 notamment,de%20la%20radicalisation%20(CNAPR)%3B Gozálvez, V., & Contreras Pulido, P. (2014). Empoderar a la ciudadanía mediática desde la educomunicación. Comunicar, 21(42), 129–136. doi:10.3916/C42-2014-12 Gözde, K. U. R. T. (2019). Yeni medyada nefret söylemi: YouTube’da Suriyeli mültecilere karşı üretilen nefret söylemi üzerine bir araştırma [Hate speech in new media: A research on hate speech against Syrian refugees on YouTube]. The Journal of International Lingual Social and Educational Sciences, 5(1), 1–20. Gramsci, A. (1966). La Questione Meridionale. Editori Riuniti. Granovetter, M. S. (1973). The Strength of Weak Ties. American Journal of Sociology, 78(6), 1360–1680. https://bit. ly/3GnnRDD. doi:10.1086/225469 Greimas, A. J. (1976). Semántica estructural. Editorial Gredos. Greimas, A. J., & Courtés, J. (1979). Semiotics and Language. Indiana Press University. Gribaudi, G. (1997). Images of the South: The Mezzogiorno as seen by Insiders and Outsiders. The new history of the Italian South: The Mezzogiorno revisited, 83-113. Grimme, C., Assenmacher, D., & Adam, L. (2018). Changing Perspectives: Is It Sufficient to Detect Social Bots? In G. Meiselwitz (Ed.), Lecture Notes in Computer Science: Social Computing and Social Media. User Experience and Behavior (pp. 445–461). Springer. doi:10.1007/978-3-319-91521-0_32 Grotlüschen, A. (2018). Global competence. Does the new OECD competence domain ignore the global South? Studies in the Education of Adults, 50(2), 185–202. doi:10.1080/02660830.2018.1523100 Guess, A., Nagler, J., & Tucker, J. (2019). Less than you think: Prevalence and predictors of fake news dissemination on Facebook. Science Advances, 5(1), eaau4586. doi:10.1126ciadv.aau4586 PMID:30662946 Guizardi, M. L., & Mardones, P. (2020). Las configuraciones locales de odio. Discursos antimigratorios y prácticas xenofóbicas en Foz de Iguazú, Brasil. Estudios Fronterizos (México), 21. Advance online publication. doi:10.21670/ ref.2003045 Güney, Ü. (2022). Syrian refugees between Turkish nationalism and citizenship. Journal of Immigrant & Refugee Studies, 20(4), 519–532. doi:10.1080/15562948.2021.1950256

327

Compilation of References

Gunitsky, S. (2015). Corrupting the cyber-commons: Social media as a tool of autocratic stability. Perspectives on Politics, 13(1), 42–54. doi:10.1017/S1537592714003120 Gupta, S., Nagar, S., Nanavati, A. A., Dey, K., Barbhuiya, F. A., & Mukherjea, S. (2021). Consumption of Hate Speech on Twitter: A Topical Approach to Capture Networks of Hateful Users. In ROMCIR@ ECIR (pp. 25-34). Academic Press. Habermas, J. (2008). Teoria dell’agire comunicativo. Il Mulino. HaberTürk. (2017, April 17). Mersin’de Suriyelilerle Belde Halkı Arasında Gerilim [Tension Between Syrians and Local People in Mersin]. Haberturk. http://www.haberturk.com/gundem/haber/1464879- mersinde-suri yelilerle-halk-arasinda-gerilim/6 Han, B. (2015). Nello sciame. Visioni del digitale. Nottetempo. Han, B.-C. (2022). Infocracia; a digitalização e a crise da democracia. Relógio de Água. Hanes, E., & Machin, S. (2014). Hate crime in the wake of terror attacks: Evidence from 7/7 and 9/11. Journal of Contemporary Criminal Justice, 30(3), 247–267. doi:10.1177/1043986214536665 Hangartner, D., Gennaro, G., Alasiri, S., & Donnay, K. (2021). Empathy-based counterspeech can reduce racist hate speech in a social media field experiment. Proceedings of the National Academy of Sciences of the United States of America, 50(118), 1–3. doi:10.1073/pnas.2116310118 PMID:34873046 Hansen, D. (2017). The Teacher and the World: A Study of Cosmopolitanism as Education. Routledge. doi:10.4324/9780203803325 Hanzelka, J., & Schmidt, I. (2017). Dynamics of Cyber Hate in Social Media: A Comparative Analysis of Anti-Muslim Movements in the Czech Republic and Germany. International Journal of Cyber Criminology, 11(1). Hard, G. Z. (2013, 7 November). Hard GZ - Viva España [Videoclip]. https://www.youtube.com/watch?v=QpFViVQG32Y Hardan, M. (2021). How Syrian refugees contributed to Turkish economy. Al-monitor. https://www.al-monitor.com/ originals/2021/05/how-syrian-refugees-contributed-turkish-economy Hardyns, W., & Rummens, A. (2018). Analyzing the effects of social media on police legitimacy beliefs: An empirical study. Journal of Police and Criminal Psychology, 33(2), 117–129. doi:10.100711896-017-9243-7 Harenslak, B. P., & Ruiter, J. R. (2021). Data Pipelines with Apache Airflow. Manning Publications. Hasib, K. M., Iqbal, M. S., Shah, F. M., Mahmud, J. A., Popel, M. H., Showrov, M. I. H., Ahmed, S., & Rahman, O. (2020). A Survey of Methods for Managing the Classification and Solution of Data Imbalance Problem. Journal of Computational Science, 16(11), 1546–1557. doi:10.3844/jcssp.2020.1546.1557 Hayaty, M., Adi, S., & Hartanto, A. (2020). Lexicon-Based Indonesian Local Language Abusive Words Dictionary to Detect Hate Speech in Social Media. Journal of Information Systems Engineering and Business Intelligence, 6(1), 9–17. doi:10.20473/jisebi.6.1.9-17 Hedger, J. A. (2012). The semantics of racial slurs: Using Kaplan’s framework to provide a theory of the meaning of derogatory epithets. Linguistic and Philosophical Investigations, 11, 74–843. Heim, J. (2020). Italian LGBTQ representation in transnational television. Journal of Italian Cinema & Media Studies, 8(2), 189–203. doi:10.1386/jicms_00016_1 Hernández, O. (2016). Música y discurso: aproximaciones analíticas desde América Latina. Eduvim.

328

Compilation of References

Hernández-Sampieri, R., & Mendoza-Torres, C. P. (2018). Metodología de la Investigación. Las rutas cuantitativa, cualitativa y mixta. McGraw-Hill. Herrmann, C. R. (2015). The dynamics of robbery and violence hot spots. Crime Science, 4(1), 1–14. doi:10.118640163015-0042-5 Hess, L. (2018). Perspectival expressives. Journal of Pragmatics, 129, 13–33. doi:10.1016/j.pragma.2018.02.007 Hess, L. (2019). Expressive meanings and expressive commitments: a case of meaning as use. In Philosophical insights into pragmatics. De Gruyter. doi:10.1515/9783110628937-010 HLEG. (2018). A multi-dimensional approach to disinformation: report of the independent high-level group (HLEG) on fake news and online disinformation. European Commission. https://ec.europa.eu/digital-singlemarket/en/news/final-repo rt-high-level-expert-group-fake-news-andonline-disinformatio n Holcomb, L., & Beal, C. (2010). Capitalising on web 2.0 in the social studies context. TechTrends, 54(4), 28–32. doi:10.100711528-010-0417-0 Hollis, G. (2017). Soring best-worst data in unbalanced, many-item de- signs, with applications to crowdsourcing semantic judgments. Behavior Research Methods, 50(2), 711–729. doi:10.375813428-017-0898-2 PMID:28550657 Hollis, G., & Westbury, C. (2018). When is best-worst best? A comparison of best-worst scaling, numeric estimation, and rating scales for collection of semantic norms. Behavior Research Methods, 50(1), 115–133. doi:10.375813428017-1009-0 PMID:29322399 Hope not Hate. (2017). Bots, Fake News And The Anti-Muslim Message On Social Media. Hope Not Hate. https://www. hopenothate.org.uk/bots-fake-news-anti-muslim-message-social-media/ Hopkins, D. J. (2010). Politicized places: Explaining where and when immigrants provoke local opposition. The American Political Science Review, 104(1), 40–60. doi:10.1017/S0003055409990360 Horsti, K. (2017). Digital Islamophobia: The Swedish woman as a figure of pure and dangerous whiteness. New Media & Society, 19(9), 1440–1457. doi:10.1177/1461444816642169 Howard, P., Kollanyi, B., & Woolley, S. C. (2016). Bots and automation over twitter during the second us presidential debate. Academic Press. Howard, E. (2017). Freedom of expression and religious hate speech in Europe. Routledge. doi:10.4324/9781315277257 Hughey, M. W., & Daniels, J. (2013). Racist comments at online news sites: A methodological dilemma for discourse analysis. Media Culture & Society, 35(3), 332–347. doi:10.1177/0163443712472089 Huntington, S. (1996). The Clash of Civilizations. Simon & Schuster. Huysseune, M. (2006). Modernity and secession: The social sciences and the political discourse of the Lega Nord in Italy. Berghahn Books. doi:10.2307/j.ctv287sgzd Ibarra, E. (2013). Víctimas y Seguridad ante los Crímenes de Odio. Cuadernos de Análisis. Ibri, I. A., Alves, M. A., & De Tienne, A. (2021). Semiótica e Pragmatismo: Interfaces Teóricas-Vol. II. Editora Oficina Universitária. doi:10.36311/2021.978-65-59541-28-7 Idevall, K. (2019). “She has promised never to use the N-word again”: Discourses of racism in a Swedish media debate. Discourse, Context & Media, 31, 1–8. doi:10.1016/j.dcm.2019.100322 329

Compilation of References

IFCN. (2021). The commitments of the code of principles. IFCN. https://bit.ly/3dyo5uw Iganski, P. (2008). Hate crime and the city. Policy Press. Il Fatto Quotidiano. (2020, Mar 9). Coronavirus, l’appello di De Magistris a chi è arrivato a Napoli dalle zone rosse: “Mettetevi in quarantena. Non è il momento di fare gli egoisti”. Il Fatto Quotidiano. https://www.ilfattoquotidiano. it/2020/03/09/coronavirus-lappello-di-de-magistris-a-chi-e-arrivato-a-napoli-dalle-zone-rosse-mettetevi-in-quarantenanon-e-il-momento-di-fare-gli-egoisti/5730147/ Il Gazzettino. (2020, Mar 8). Coronavirus, l’appello di Emiliano: «Fermatevi e tornate indietro, non portate in Puglia l’epidemia». Il Gazzettino. https://www.ilgazzettino.it/viaggi/news-5098641.html Il Messaggero. (2020°, Mar 25). Coronavirus, la “vendetta” del Sud: «Non si affitta ai settentrionali». Sui social è boom. Il Messaggero. https://www.ilmessaggero.it/italia/coronavirus_sud_nord_non_si_affitta_meridionali_25_febbraio_2020-5073714.html Il Messaggero. (2020b, Mar 8). Coronavirus, a Milano la fuga dalla “zona rossa”: folla alla stazione di Porta Garibaldi. Il Messaggero. https://www.ilmessaggero.it/italia/coronavirus_milano_fuga_milano_treno_romani-5097472.html Innerarity, D. (2006). El nuevo espacio público. Espasa-Calpe. Innes, M., Innes, H., Roberts, C., Harmston, D., & Grinnell, D. (2021). The normalisation and domestication of digital disinformation: On the alignment and consequences of far-right and Russian state (dis)information operations and campaigns in Europe. Journal of Cyber Policy, 6(1), 31–49. doi:10.1080/23738871.2021.1937252 Inspeção-Geral da Administração Interna. (2021). Plano de Prevenção de Manifestações de Discriminação nas Forças e Serviços de Segurança (PPMD). IGAI. https://www.igai.pt/pt/Publicacoes/PublicacoesIGAI/Documents /Plano%20de%20Preven%C3%A7%C3%A3o%20de%20Manifesta%C3%A7%C3% B5es%20de%20Discrimina%C3%A7%C3%A3o%20nas%20For%C3%A7as%20e% 20Servi%C3%A7os%20de%20Seguran%C3%A7a.pdf IPSOS/Cdec. (2017). Stereotipi e pregiudizi degli Italiani: dagli immigrati agli ebrei. CDEC. http://old.cdec.it/public/ Stereotipi_Opinioni_sintesi.pdf Ireton, C., & Posetti, J. (2018). Journalism, Fake News & Disinformation. Handbook for Journalism Education and Training. Unesco. Islamic Human Right Commission. (2016). Annual report. IHRC. https://www.ihrc.org.uk/wp-content/uploads/2019/04/ Annual-Report-2016-2017-04-FV-FB.pdf Islam, M. S., Hasan, M. M., Wang, X., Germack, H. D., & Noor-E-Alam, M. (2018). A Systematic Review on Healthcare Analytics: Application and Theoretical Perspective of Data Mining. Health Care, 6(2), 2. doi:10.3390/healthcare6020054 PMID:29882866 Istaiteh, O., Al-Omoush, R., & Tedmori, S. (October 19TH - 22ND, 2020). Racist and sexist hate speech detection: literature review. 2020 International Conference on Intelligent Data Science Technologies and Applications (IDSTA). IEEE. Ivandic, R., Kirchmaier, T., & Machin, S. J. (2019). Jihadi attacks, media and local hate crime. https://doi.org/ doi:10.2139/ ssrn.3401120 Jacobs, L., & van Spanje, J. (2021). A Time-Series Analysis of Contextual-Level Effects on Hate Crime in The Netherlands. Social Forces, 100(1), 169–193. doi:10.1093foaa102 Jacobson, M. F. (1999). Whiteness of a different color. Harvard University Press. doi:10.2307/j.ctvjk2w15 330

Compilation of References

Jakubowicz, A. (2017). Alt_Right White Lite: Trolling, hate speech and cyber racism on social media. Cosmopolitan Civil Societies: An Interdisciplinary Journal, 9(3), 41–60. doi:10.5130/ccs.v9i3.5655 Jay, T. (2023). 86 Hurtful words and hate speech. In G. Schiewer, J. Altarriba, & B. Ng (Eds.), Language and Emotion. An international handbook (pp. 1794–1815). De Gruyter Mouton. Jenkins, H. (2006). Convergence Culture: Where Old and New Media Collide. New York University Press. Jenkins, H. (2008). Convergence Culture. La cultura de la convergencia en los medios de comunicación. Paidós. Jerome, L., Elwick, A., & Kazim, R. (2019). The impact of the Prevent duty on schools: A review of the evidence. British Educational Research Journal, 45(4), 821–837. doi:10.1002/berj.3527 Jeshion, R. (2013). Slurs and Stereotypes. Analytic Philosophy, 54(3), 314–329. doi:10.1111/phib.12021 Jian, Y., Huang, D., Yan, J., Lu, K., Huang, Y., Wen, T., Zeng, T., Zhong, S., & Xie, Q. (2017). A Novel Extreme Learning Machine Classification Model for e-Nose Application Based on the Multiple Kernel Approach. Sensors (Basel), 17(6), 6. doi:10.339017061434 PMID:28629202 Jiménez Sánchez, F. (1994). Posibilidades y límites del escándalo político como una forma de control social. Reis. Revista Española de Investigaciones Sociológicas, 66(66), 7–36. doi:10.2307/40183715 Jonasson, A. K. (2022). The double-edged sword of democracy: How democracy promotion is used in the European Union’s prevention of terrorism in the southern Mediterranean. Journal of Regional Security, 17(2), 241–266. doi:10.5937/ jrs17-37878 Jones-Jang, S. M., & Chung, M. (2022). Can we blame social media for polarisation? Counter-evidence against filter bubble claims during the COVID-19 pandemic. New Media & Society, 0(0). doi:10.1177/14614448221099591 Juárez, J., & Restrepo, N. (2021). El auge de los discursos de odio y el papel de la ética y los medios de comunicación. En: Almudena Barrientos-Báez (coord.). XIII Congreso Internacional Latina de Comunicación Social 2021: Libros de Actas. Comunicación y Nuevas Tendencias. Kamada, T. & Kawai, S. (1989). An agorithm for drawing general undirected graphs. Information Pocessing Letters, 31, 7-15. Kapantai, E., Christopoulou, A., Berberidis, C., & Peristeras, V. (2021). A systematic literature review on disinformation: Toward a unified taxonomical framework. New Media & Society, 23(5), 1301–1326. doi:10.1177/1461444820959296 Kaplún, M. (1998). Una pedagogía de la comunicación. Ediciones de la Torre. Kariithi, N., Mawadza, A., & Carciotto, S. (2017). Media portrayal of immigration in the South African media 2011–2015. The Scalabrini Institute for Human Mobility in Africa-SIHMA, 1-32. Kaufman, G. (2015). Odium dicta. Libertad de expresión y protección de grupos discriminados en internet. Conapred. Kavaklı, N. (2018). Hate Speech Towards Syrian Refugees in Turkish Online Reader Comments. In Global Perspectives on Human Migration, Asylum, and Security (pp. 127-142). IGI Global. doi:10.4018/978-1-5225-2817-3.ch007 Keib, K., Wojdynski, B. W., Espina, C., Malson, J., Jefferson, B., & Lee, Y. I. (2021). Living at the speed of mobile: How users evaluate social media news posts on smartphones. Communication Research, 1–17. Keller, D., Costalonga, L., & Messina, M. (2020, November). Ubiquitous music making in COVID-19 times. In Proceedings of the 10th Workshop on Ubiquitous Music (UbiMus 2020), 3-16.

331

Compilation of References

Keller, F.-B., Schoch, D., Stier, S., & Yang, J.-H. (2019). Political Astroturfing on Twitter: How to coordinate a disinformation campaign. Political Communication, 37(2), 256–280. doi:10.1080/10584609.2019.1661888 Kepel, G. (2016). El terror entre nosotros. Una historia de la yihad en Francia. Península. Kerbrat-Orecchioni, C. (1983). La enunciación. De la subjetividad en el lenguaje, 117-135. Hachette. Khoma, N., & Oleksii Kokoriev, O. (2021). Compliance of the Baltic States with the Principle of Tolerance as Condition for the Development of the United Europe. The Age of Human Rights Journal, 16(16), 186–206. doi:10.17561/tahrj.v16.6038 KhosraviNik, M., & Esposito, E. (2018). Online hate, digital discourse and critique: Exploring digitally-mediated discursive practices of gender-based hostility. Lodz Papers in Pragmatics, 14(1), 45–68. doi:10.1515/lpp-2018-0003 Kirchgaessner, S. (2022, octubre 4). Mexico: Repor ters and activists hacked with NSO spyware despite assurances. The Guardian. https://www.theguardian.com/world/2022/oct/04/mexico-nso-spy ware-journalists-human-rights-hacked-pegasus Kiritchenko, S., & Mohammad, S. M. (2017). Best-worst scaling more reliable than rating scales: A case study on sentiment intensity annotation. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL-2017). Association for Computational Linguistics. https://doi.org/10.18653/v1/P17-2074 Kiritchenko, S., & Mohammad, S. M. (2016). Capturing reliable fine-grained sentiment associations by crowdsourcing and best-worst scaling. In Proceedings of the 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics. 10.18653/ v1/N16-1095 Kisang, A. K. (2017). The images and rights of migrants in the Kenyan media. Current Journal of Applied Science and Technology, 22(5), 1–14. doi:10.9734/CJAST/2017/32385 Koç, G. (2021). Derrida’s Concept of Hos (ti) pitality and Hate Crime in Turkey. Journal of Business Diversity, 21(2). Koufakou, A., Pamungkas, E. W., Basile, V., & Patti, V. (2020). HurtBERT: Incorporating lexical features with BERT for the detection of abusive language. In Proceedings of the fourth workshop on online abuse and harms. Association for Computational Linguistics. 10.18653/v1/2020.alw-1.5 Kreis, R. (2017). # refugeesnotwelcome: Anti-refugee discourse on Twitter. Discourse & Communication, 11(5), 498–514. doi:10.1177/1750481317714121 Kristeva, J. (1969). Le langage, cet inconnu. Seuil. Kroskrity, P. (2021). Covert Linguistic Racisms and the (Re-)Production of White Supremacy. Journal of Linguistic Anthropology, 31(2), 180–193. doi:10.1111/jola.12307 Krzyżanowski, M., Triandafyllidou, A., & Wodak, R. (2018). The mediatisation and the politicisation of the “refugee crisis” in Europe. Journal of Immigrant & Refugee Studies, 16(1-2), 1–14. doi:10.1080/15562948.2017.1353189 Kwak, H., Lee, C., Park, H., & Moon, S. (2010, April). What is Twitter, a social network or a news media? In Proceedings of the 19th international conference on World wide web (pp. 591-600). 10.1145/1772690.1772751 La Repubblica. (2020). “Torni a casa chi non è toscano”. Rossi sceglie la linea dura. La Repubblica (Firenze). https:// bit.ly/3FkiWE9 Lagos, A., Caicedo, J. E., Coria, G., Quete, A. R., Martínez, M., Suvire, G., & Riquelme, J. (2022). State-of-the-Art Using Bibliometric Analysis of Wind-Speed and -Power Forecasting Methods Applied in Power Systems. Energies, 15(18), 18. Advance online publication. doi:10.3390/en15186545 332

Compilation of References

Lahnait, F. (2018). Prevenir el extremism violento para combatir el yihadismo autóctono: Aprender de la experiencia. CIDOB Report. Lambert, R. (1993). Educational Exchange and Global Competence. 46th International Conference on Educational Exchange. https://files.eric.ed.gov/fulltext/ED368275.pdf Lamerichs, N., Dennis, M. C., Puerta, R. R., & Lange-Bohmer, A. (2018). Elite Male Bodies: The Circulation of AltRight Memes and the Framing of Politicians on Social Media. Participations, 15(1), 180–206. Landowski, E. (2022). As metamorfoses da verdade, entre sentido e interação. Estudos semióticos, 18(2), pp. 1-23. doi:10.11606/issn.1980-4016.esse.2022.198273 Lankshear, C., & McClaren, P. (Eds.). (1993). Critical literacy: Radical and postmodernist. State University of New York Press. Larsson, G. (2007). Cyber-islamophobia? The case of WikiIslam. Contemporary Islam, 1(1), 53–67. doi:10.100711562007-0002-2 Law, I., Easat-Daas, A., & Sayyid. (2018). Countering Islamophobia through the Development of Best Practice in the Use of Counter-Narratives in EU Member States. https://cik.leeds.ac.uk/ Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (2021). The People’s Choice. How the Voter Makes up his Mind in a Presidential Campaign. Columbia University Press. Lebart, L., Salem, A., Bécue, M., & Peña, D. (2000). Análisis estadístico de textos. Milenio. Ledger, S., Their, M., Baile, L. & Pitts, C. (2019). OECD’s approach to measuring Global Competency: Powerful voices shaping education. Teachers College Record, 121, 1-40. doi:10.1177/016146811912100802 Lee, P., & Cheng, C. (2020). Profanity and Hate Speech Detection. International Journal of Information and Management Sciences., 31(3), 227–246. doi:10.6186/IJIMS.202009_31(3).0002 Legewie, J. (2013). Terrorist events and attitudes toward immigrants: A natural experiment. American Journal of Sociology, 118(5), 1199–1245. doi:10.1086/669605 Lekea, I. K., & Karampelas, P. (2018). Detecting hate speech within the terrorist argument: a greek case. In 2018 IEEE/ ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM). IEEE. 10.1109/ ASONAM.2018.8508270 Lenore, V. (2013). Entrevista a Diego Manrique. https://www.elconfidencial.com/cultura/2013-04-29/en-espanagustaba-la-musica-latina-hasta-que-se-instalaron-los-inmigra ntes_495090/ Leung, K.M. (2007). Naive Bayesian Classifier. Polytechnic University Department of Computer Science/Finance and Risk Engineering, 123-156. Lewis, B. (2016). Gli arabi nella storia. Editori Laterza. Ley Orgánica 3/2020, de 29 de diciembre, por la que se modifica la Ley Orgánica 2/2006, de 3 de mayo, de Educación. Boletín Oficial del Estado, 340, de 30 de diciembre de 2020, 122868-122953. Li, P., Wu, Q., & Burges, C. (2007). McRank: Learning to Rank Using Multiple Classification and Gradient Boosting. Advances in Neural Information Processing Systems, 20. https://proceedings.neurips.cc/paper_files/paper/2007/hash/b 86e8d03fe992d1b0e19656875ee557c-Abstract.html

333

Compilation of References

Liberatore, F., Camacho-Collados, M., Vitoriano, B. (2020). Police districting problem: literature review and annotated bibliography. Optimal districting and territory design, 9-29. Springer. doi:10.1007/978-3-030-34312-5_2 Liberini, F., Redoano, M., Russo, A., Cuevas, Á., & Cuevas, R. (2020). Politics in the Facebook Era - Evidence from the 2016 US Presidential Elections. CESifo Working Paper, No.8235, Center for Economic Studies and ifo Institute. Libero (2020, Mar 4). L’infezione crea l’Unità d’Italia: Virus alla conquista del Sud. Libero. https://www.vesuviolive. it/ultime-notizie/327529-coronavirus-sud-libero/ Librandi, G. (2020, Dec 11). Gianfranco Librandi e la superiorità dei migranti. La Zanzara. Rai Radio 2. https://youtu. be/bAlbIiISkmM Librandi, G. (n.d.). Personal homepage. Librandi. http://www.gianfrancolibrandi.it/ Light, D., Keller, S., & Calhoun, C. (1991). Sociología. McGraw-Hill. Lima, R., Borges, M., Salvador, J., & Freitas, M. (2022). Nothing about Us without Us: Social Worker, Harm Reduction and Anti-Racist Struggle. International Journal of Social Work Values and Ethics, 19(2), 101–123. doi:10.55521/10-019-208 Li, P., Li, W., Wang, H., & Zhang, X. (2014). Modeling of information diffusion in Twitter-like social networks under information overload. TheScientificWorldJournal, 2014, 2014. doi:10.1155/2014/914907 PMID:24795541 Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692. Llorente-Barroso, C., Kolotouchkina, O., & García-García, F. (2021). The meaning of the logo from its semiotic construction and its reliance on new formats of digital communication: The case of Apple: El caso de Apple. Revista Latina de Comunicación Social, 79, 333-356. https://www.doi.org/10.4185/RLCS-2021-1529 Lombroso, C. (1896). Il cervello del brigante Tiburzi. Nuova Antologia, 150, 591–600. Lombroso, C. (1900). L’ultimo brigante: Giuseppe Musolino. Nuova Antologia, 181, 508–516. López-Berlanga, C., & Sánchez-Romero, C. (2019). La interacción y convivencia digital de los estudiantes en las redes sociales. Revista nacional e internacional de educación inclusiva, 12(2), 114-130. López, F., & Rodríguez Rodríguez, J. M. (2020). El Fact Checking en España. Plataformas, prácticas y rasgos distintivos. Estudios sobre el Mensaje Periodístico, 26(3), 1045–1065. doi:10.5209/esmp.65246 López-Fuentes, V. (2020). La educación cosmopolita como enfoque educativo para trabajar la inclusión, la diversidad y el vínculo emocional en el aula. Dilemas contemporáneos: Educación. Política y Valores, 8(1), 1–20. López-García, G. (2016). ‘Nuevos’ y ‘viejos’ liderazgos: La campaña de las elecciones generales españolas de 2015 en Twitter. Comunicación y Sociedad, 29(3), 149–167. doi:10.15581/003.29.35829 López-Gutiérrez, J. (2022). Informe sobre la evolución de los delitos de odio en España 2021. Oficina Nacional de Lucha contra los delitos de Odio (ONDOD). https://www.interior.gob.es/opencms/es/servicios-al-ciudadan o/delitos-de- odio/estadisticas/ Louviere, J. J., Flynn, T. N., & Marley, A. A. J. (2015). Best-worst scaling: Theory, methods and applications. Cambridge University Press. doi:10.1017/CBO9781107337855 Lund, B. D., & Wang, T. (2023). Chatting about ChatGPT: How may AI and GPT impact academia and libraries? Library Hi Tech News, 40(3), 26–29. doi:10.1108/LHTN-01-2023-0009 Lyotard, J-F. ([1979] 2019). La condición postmoderna. Informe sobre el saber. Cátedra, ebook. 334

Compilation of References

Magallón, R. (2019). Verificado México 2018: Desinformación y fact-checking en campaña electoral. Revista de comunicación, 18(1), 234-258. Majmundar, A., Allem, J. P., Boley Cruz, T., & Unger, J. B. (2018). The why we retweet scale. PLoS One, 13(10), e0206076. doi:10.1371/journal.pone.0206076 PMID:30335827 Maldita.es. (2019). Por qué este proyecto. Maldita Migración. https://bit.ly/3UasWp7 Maldita.es. (2022). ¿Qué es Maldita Migración? Maldita Migración. https://maldita.es/migracion/ Maldonado, M. (2008). Aprendizaje basado en proyectos colaborativos. Una experiencia en educación superior. Laurus, 14(28), 158-180. https://www.redalyc.org/pdf/761/76111716009.pdf Marcos-Recio, J. C., & Flores-Vivar, J. M. (2023). Social Media contributes to radicalisation. How to prevent its expansion? In C. Melacarne & L. Fabbri (Eds.), Understanding Radicalization in everyday life (pp. 125-142). McGraw Hill Education. Martín Segovia, E. (2003). La campaña del euro (1996-2002): la Unión Monetaria Europea como objeto de comunicación política y opinión pública [Tesis doctoral, Universidad Complutense de Madrid, España]. Martínez de Bartolomé, I., & Rivera-Martín, B. (2022). Hate speech against the LGTBIQ+ community and its representation in Veneno series. VISUAL REVIEW International Visual Culture Review, 12(1), 1–12. doi:10.37467/revvisual.v9.3718 Martínez, L. (2023). Mensajes misóginos en lo perfiles de Tik Tok de la prensa española: Estudio y percepción por parte de los jóvenes. Ámbitos, 59, 110–127. doi:10.12795/Ambitos.2023.i59.08 Martínez-Rodríguez, J. B., & Sandoval-Esquivel, A. (2022). Educomunicación y medios digitales: Perspectivas teóricas y prácticas en la educación superior. Dykinson. Masip, P., Aran-Ramspott, S., Ruiz-Caballero, C., Suau, J., Almenar, E., & Puertas-Graell, D. (2020). Consumo informativo y cobertura mediática durante el confinamiento por el Covid-19: Sobreinformación, sesgo ideológico y sensacionalismo. El Profesional de la Información, 29(3), e290312. doi:10.3145/epi.2020.may.12 Masip, P., Suau, J., & Ruiz-Caballero, C. (2020). Percepciones sobre medios de comunicación y desinformación: Ideología y polarización en el sistema mediático español. El Profesional de la Información, 29(5), e290527. doi:10.3145/ epi.2020.sep.27 Maslova, K., & Savino, G. (2020). Pandemic all’italiana: Stress-test for the political system. Contemporary EuropeSovremennaya Evropa, 97(4), 37–49. doi:10.15211oveurope420203749 Masuzawa, T. (2005). The invention of world religions: or how European universalism was preserved in the language of pluralism. University of Chicago Press. doi:10.7208/chicago/9780226922621.001.0001 Matamoros-Fernández, A., & Farkas, J. (2021). Racism, hate speech, and social media: A systematic review and critique. Television & New Media, 22(2), 205–224. doi:10.1177/1527476420982230 Mathur, A., & Foody, G. M. (2008). Multiclass and Binary SVM Classification: Implications for Training and Classification Users. IEEE Geoscience and Remote Sensing Letters, 5(2), 241–245. doi:10.1109/LGRS.2008.915597 Mauro, M., & Martínez-Corcuera, R. (2020). Hypermasculinity and racist discourses in the Spanish media. The spectacularisation of radio sport programming. In J. O’Brien, R. Holden, & X. Ginesta (Eds.), Sport, Globalisation and Identity. New Perspectives on Regions and Nations. Routledge. doi:10.4324/9781003007104-9

335

Compilation of References

Mautone, E. (2020, Jan 20). Coronavirus, caso sospetto a Napoli: al Cotugno coppia di sposini cinesi in viaggio di nozze. De Luca convoca i manager: kit diagnostici al Cotugno. Il Mattino. https://www.ilmattino.it/napoli/cronaca/coronavirus_napoli_ricoverato_cotugno-5013206.html Mayo-Cubero, M. (2019). Uso de las redes sociales en la cobertura periodística de crisis, desastres y emergencias en España. Revista Española de Comunicación en Salud. 43-54. doi:10.20318/recs.2019.4428 Mazzoleni, G., & Bracciale, R. (2018). Socially mediated populism: The communicative strategies of political leaders on Facebook. Palgrave Communications, 4(1), 50. doi:10.105741599-018-0104-x McCombs, M. E., Shaw, D. L., & Weaver, D. H. (2014). New directions in agenda-setting theory and research. Mass Communication & Society, 17(6), 781–802. doi:10.1080/15205436.2014.964871 McDaniel, C. (2006). Critical literacy: A way of thinking, a way of life. Peter Lang. McNeil-Willson, R., Gerrand, V., Scrinzi, F. & Triandafyllidou, A. (2019). Polarisation, violent extremism and resilience in Europe today: An analytical framework. BRaVE Project. Megías, I., Amezaga, A., García, M. C., Kuric, S., Morado, R., & Orgaz, C. (2020). Romper cadenas de odio, tejer redes de apoyo: los y las jóvenes ante los discursos de odio en la red Juventud. Fad. doi:10.5281/zenodo.4288486 Meijer, A., & Wessels, M. (2019). Predictive policing: Review of benefits and drawbacks. International Journal of Public Administration, 42(12), 1031–1039. doi:10.1080/01900692.2019.1575664 Melley, M. (2021, Dec 29). In migliaia non riescono a prenotare la terza dose. Milano Today. https://www.milanotoday. it/attualita/coronavirus/terza-dose.html Merklejn, I., & Wiślicki, J. (2020). Hate Speech and the Polarization of Japanese National Newspapers. Social Science Japan Journal, 23(2), 259–279. doi:10.1093sjj/jyaa015 Meseguer, S. (2022). Incitación al odio por razón de las creencias en los medios publicitarios. In F. Pérez-Madrid. Discurso de odio y creencias (pp. 91–114). Thomson Reuters Arazandi. Messina, M. (2018c). La canzone meridionale “post-italiana” e il Mediterraneo. In Paci, D., Perri, P. & Zantedeschi, F. (eds), Paesaggi mediterranei: storie, rappresentazioni, narrazioni. Canterano (RM): Aracne Editrice, 153-170. Messina, M. (2019). CompraSud: A Collective Attempt to Overturn the National Italian Consumer Goods Market. História e Economia: Revista interdisciplinar, 23(2), 60-76. Messina, M. (2020b). Southernness, Disability and the Construction of the “Other” in Italian Cinema: Desire, Masculinities, Disfigurations and Medicalisations. Simultanea: Simultanea: A Journal of Italian Media and Pop Culture 1(1), 1-10. Messina, M. (2016). Narrativas pós-italianas: A re-imaginação da unidade nacional nas canções do sul da Itália. Muiraquitã: Revista de Letras e Humanidades, 4(1), 113–125. doi:10.29327/216341.4.1-10 Messina, M. (2018a). The Demonization of the South and the Southernification of Evil in Contemporary Italian Cinema: Belluscone and Qualunquemente. Journal of Italian Cinema & Media Studies, 6(2), 193–207. doi:10.1386/jicms.6.2.193_1 Messina, M. (2018b). Definire il post-italiano: Tentativi di superamento dell’orizzonte nazionale italiano nel Mezzogiorno. Jamaxi, 2(2), 107–140. Messina, M. (2020a). 107 days and counting.... Portal: Journal of Multidisciplinary International Studies, 17(1/2), 104–109. Messina, M., & Capogreco, S. (2019). The Silent Return of the South to the South: Verdone’s Pasquale Ametrano as a Counter-Chaplin. Tropos: Comunicação. Society and Culture, 8(1), 1–17. 336

Compilation of References

Messina, M., Costa, V. F., & Scarassatti, M. (2020). Cartridge Music in the Quarantine: Presence, Absence, Contingency Setups and (De-) territorialised Performances. INSAM Journal of Contemporary Music. Art and Technology, (5), 28–45. Messina, M., & Di Somma, T. (2017). Unified Italy, Southern Women and Sexual Violence: Situating the Sexual Assault TV “prank” Against Emma Marrone Within the Dynamics of Contemporary Italy as a Scopic Regime. Revista Tropos: Comunicação. Society and Culture, 6(1), 1–18. Metcalfe-Hough, V. (2015). The migration crisis? Facts, challenges and possible solutions. Briefing. Miller, K. (2021). Hostility Toward the Press: A Synthesis of Terms, Research, and Future Directions in Examining Harassment of Journalists. Digital Journalism (Abingdon, England), (oct), 1–20. doi:10.1080/21670811.2021.1991824 Mills, C. E., Freilich, J. D., & Chermak, S. M. (2017). Extreme hatred: Revisiting the hate crime and terrorism relationship to determine whether they are “close cousins” or “distant relatives”. Crime and Delinquency, 63(10), 1191–1223. doi:10.1177/0011128715620626 Ministerio del Interior, Gobierno de España. (2022). Portal estadístico de criminalidad. https://estadisticasdecriminalidad.ses.mir.es/publico/portalestadistico/ Minuz, A. (2020, Feb 16). Il virus da talk-show. Il Foglio. https://www.ilfoglio.it/televisione/2020/02/10/news/il-virusda-talk-show-301103/ Miró Llinares, F. (2013). La victimización por cibercriminalidad social. Un estudio a partir de la teoría de las actividades cotidianas en el ciberespacio. Revista Española De Investigación Criminológica, 11, 1–35. doi:10.46381/reic.v11i0.77 Misuraca, R., Teuscher, U., Scaffidi Abbate, C., Ceresia, F., Roccella, M., Parisi, L., Vetri, L., & Miceli, S. (2022). Can We Do Better Next Time? Italians’ Response to the COVID-19 Emergency through a Heuristics and Biases Lens. Behavioral Sciences (Basel, Switzerland), 12(2), 39. doi:10.3390/bs12020039 PMID:35200290 Mohler, G. (2014). Marked point process hotspot maps for homicide and gun crime prediction in Chicago. International Journal of Forecasting, 30(3), 491–497. doi:10.1016/j.ijforecast.2014.01.004 Molina, J. P. (2019). Unfaking News. Cómo combatir la desinformación. Estudios sobre el Mensaje Periodístico, 25(2), 1267–1269. Monárrez, J., & Cervera-Gomez, L. (2013). Spatial and temporal behavior of three paradigmatic cases of violence in Ciudad Juarez, Chihuahua México. ResearchGate, 1-23. http://dx.doi.org/ doi:10.13140/RG.2.1.4007.7289 Moon, S., Lee, G., & Chi, S. (2022). Automated system for construction specification review using natural language processing. Advanced Engineering Informatics, 51, 101495. doi:10.1016/j.aei.2021.101495 Moreno Barreneche, S. (2022). Identidades estetizadas, interacciones mediatizadas: Un abordaje de Instagram desde la sociosemiótica. Comunicación y Medios, 31(45), 129–141. doi:10.5354/0719-1529.2022.64515 Moreno, J., Quintero, S., Riascos, A., Nonato, L. G., & Sanchez, C. (2021). Homicide prediction using sequential features from graph signal processing. In Intelligent Computing (pp. 804–817). Springer. doi:10.1007/978-3-030-80129-8_54 Morrell, E. (2008). Critical literacy and urban youth: Pedagogies of access, dissent, and liberation. Taylor & Francis. Moscovici, S. (2001). Social representations: Essays in social psychology. Nyu Press. Mostert, M. (2019). Stereotypes fomenting hate. Perceptions, stigma, and real-world consequences for Africans with disabilities. In M. Sherry, T. Olsen, J. Solstad Vedeler & J. Eriksen, Disability Hate Speech. Routledge.

337

Compilation of References

Moussaïd, M., Barkoczi, D., & Helbing, D. (2017). Predicting and affecting opinion dynamics in real-world social networks. Proceedings of the National Academy of Sciences of the United States of America, 114(8), 201608473. doi:10.1073/pnas.1608473114 Mullen, B., & Leader, T. (2005). Linguistic factors: antilocutions, ethonyms, ethnophaulisms, and other varieties of hate speech. In J. Dovidio, P. Glick, & L. Rudman (Eds.), On The Nature of Prejudice: Fifty Years After Allport. Wiley Online. doi:10.1002/9780470773963.ch12 Müller, K. R., & Schwarz, C. (2020). Hate speech, social media, and the question of accountability. Media and Communication, 8(2), 443–446. doi:10.17645/mac.v8i2.3051 Müller, K. R., & Schwarz, C. (2021). Hate speech and the responsibility of the media. In H. Tumber & S. Waisbord (Eds.), The Routledge Companion to Media Misinformation and Populism (pp. 277–284). Routledge. doi:10.4324/9780367822847-25 MüllerK.SchwarzC. (2020). Fanning the Flames of Hate: Social Media and Hate Crime. https://ssrn.com/abstract=3082972 doi:10.2139/ssrn.3082972 Municio, N. (2017). Evolución del perfil del yihadista en Europa. Boletín del Instituto Español de Estudios Estratégicos, 6, 556–573. Naciones Unidas Alto Comisionado de Derechos Humanos de Naciones Unidas. [Naciones Unidas] Alto Comisionado de Derechos Humanos de Naciones Unidas (2013). Plan de Acción de Rabat. OHCHR. https://www.ohchr.org/Documents/Issues/ Najah Ahmed, A., Binti Othman, F., Abdulmohsin Afan, H., Khaleel Ibrahim, R., Ming Fai, C., Shabbir Hossain, M., Ehteram, M., & Elshafie, A. (2019). Machine learning methods for better water quality prediction. Journal of Hydrology (Amsterdam), 578, 124084. doi:10.1016/j.jhydrol.2019.124084 Narlı, N., Özaşçılar, M., & Turkan Ipek, I. Z. (2020). Turkish Daily press framing and representation of Syrian women refugees and gender-based problems: Implications for social integration. Journal of Immigrant & Refugee Studies, 18(1), 1–21. doi:10.1080/15562948.2018.1557311 Na s c e r d o S o l . ( 2 0 2 0 , M ay 2 ) . J ove m q u e a s s a s s i n o u o s t i o s e m S a n t i a g o d o C a c é m fica em prisão preventiva. Nascer do Sol. https://sol.sapo.pt/artigo/695165/jovem-que-assassinou-os-ti os-em-santiago-do-cacem-fica-em-prisao-preventiva Navarro, J. A., & Olmo, A. (2018). Political presence on Spanish Television (La1) in the previous electoral processes [in Spanish]. Studies in Journalism Message, 24, 1473–1488. doi:10.5209/ESMP.62229 Neitsch, J., Niebuhr, O., & Kleene, A. (2021). What if Hate Speech Really Was Speech? Towards Explaining Hate Speech in a Cross-Modal Approach. In S. Wachs, B. Koch-Priewe, & A. Zick (Eds.), Hate Speech - Multidisziplinäre Analysen und Handlungsoptionen. Springer VS. doi:10.1007/978-3-658-31793-5_6 Neubaum, G. (2022). ‘ It’s going to be out there for a long time’: The influence of message persistence on users’ political opinion expression in social media. Communication Research, 49(3), 426–450. doi:10.1177/0093650221995314 Newman, N., Fletcher, R., Schulz, A., Andi, S., & Kleis, R. (2020). Digital News Report 2020. University of Oxford. Newtral.es. (2022). Newtral. https://www.newtral.es/tag/inmigracion/ Nguyen, J. (2018). Politics and the Twitter revolution: A brief literature review and implications for future research. Social Networking, 7(4), 243–251. doi:10.4236n.2018.74018 Noelle-Neumann, E. (1973). Return to the Concept of the Powerful Mass Media. Studies in Broadcasting, 9, 67–112. 338

Compilation of References

Noelle-Neumann, E. (2013). The Spiral of Silence. Routledge Oxforshire. NorthWest Regional Educational Laboratory. (2002). Aprendizaje por Proyectos (trad. Eduteka). https://eduteka.icesi. edu.co/modulos/7/184/468/1 Nortio, E., Niska, M., Renvik, T., & Jasinskaja-Lahti, I. (2020). The nightmare of multiculturalism: Interpreting and deploying anti-immigration rhetoric in social media. New Media & Society. Advance online publication. doi:10.1177/1461444819899624 Notario, M. L., & Cárdenas, M. L. (2020). Xenophobia and Fake News. Analysis of the confused news flow. Maldita Migración Project. IROCAMM: International Review of Communication and Marketing Mix, 3(1), 75–91. doi:10.12795/ IROCAMM.2020.v01.i03.06 Nöth, W. (2000). Handbook of semiotics. Indiana Press University. Nöth, W. (2011). Self-referential postmodernity. Semiotica, 183(1), 199–217. 10-1515/semi.2011.010 Notícias, S. I. C. (2022, November 17). “Quando o ódio veste farda”: As vítimas do ódio de polícias nas redes sociais. Twitter. https://twitter.com/SICNoticias/status/1593371423258836992 Ntais, G., Saroukos, S., Berki, E., & Dalianis, H. (2016). Development and enhancement of a stemmer for the greek language. In Proceedings of the 20th Pan-Hellenic Conference on Informatics (PCI’ 16). Association for Computing Machinery. https://doi.org/10.1145/3003733.3003775 Nunberg, G. (2018). The social life of slurs. In D. Fogal, D. Harris, & M. Moss (Eds.), New work on speech acts. Oxford University Press. Nussbaum, M. (1997). Cultivating Humanity: A Classical Defense of Liberal Education. Harvard University Press. Odunaiya, O., Agoyi, M., & Osemeahon, O. S. (2020). Social TV Engagement for Increasing and Sustaining Social TV Viewers. Sustainability (Basel), 12(12), 4906. doi:10.3390u12124906 OECD. (2018a). Preparing our youth an inclusive and sustainable world. The OECD PISA global competence framework. Paris: OECD. https://www.oecd.org/education/Global-competency-for-an-inclusive-world.pdf OECD. (2018b). The Future of Education and Skills OECD Education 2030 Framework. Paris: OECD Publishing. https:// www.oecd.org/education/2030-project/ OECD. (2020). PISA 2018 Results (Volume VI): Are Students Ready to Thrive in an Interconnected World? OECD Publishing. doi:10.1787/d5f68679-en O f f i c i a l To u r i s m Po r t a l fo r S p a i n . ( 2 0 2 3 ) . C e l e b ra te p r i d e i n M a d r i d , I b i z a a n d S i t ges. Official Tourism Portal for Spain. https://www.spain.info/es/descubrir-espana/gay-pride-madridibiza-sitges/ Oficina de Instituciones Democráticas y Derechos Humanos de la OSCE. (2009). Hate Crime Laws. A Practical Guide. OSCE. https://www.osce.org/files/f/documents/3/e/36426.pdf Oficina de Instituciones Democráticas y Derechos Humanos de la OSCE. (2014). Prosecuting Hate Crimes: A Practical Guide. OSCE. https://tinyurl.com/nckaayk2 Oficina de Instituciones Democráticas y Derechos Humanos de la OSCE. (2015). Understanding Hate Crimes A Handbook for Ukraine. OSCE. https://tinyurl.com/ye2pama3

339

Compilation of References

Olteanu, A., Castillo, C., Boy, J., & Varshney, K. (2018). The effect of extremist violence on hateful speech online. In Proceedings of the international AAAI conference on web and social media (Vol. 12, No. 1). PKP. 10.1609/icwsm. v12i1.15040 Open (2020, Mar 8). Coronavirus, fuga dal nord: folle alle stazioni di Milano e treni presi d’assalto – Le immagini. Open. https://www.open.online/2020/03/08/coronavirus-fuga-dal-nord-folle-alle-stazioni-di-milano-e-treni-presi-dassalto-foto/ Organisation for Economic Cooperation and Development. (2022). The Protection and Promotion of Civic Space: Strengthening Alignment with International Standards and Guidance. OECD Publishing. doi:10.1787/d234e975Organisation for Security and Cooperation in Europe - OSCE. (2021). Los delitos de odio motivados por el racismo y la xenophobia. OSCE-ODIHR. Hate Crime Reporting. https://www.osce.org/es/odihr/502275 Ortega, C. (2022). Los medios de comunicación ante el discurso de odio: ¿Plataformas de difusión o “watchdogs” de los derechos fundamentales? En: A. Daunis Rodríguez. Odio y discriminación en tiempos convulsos, 69-80. Comares. Osuna, S., Frau-Meigs, D., & Lazo, C. (2018). Educación Mediática y Formación del Profesorado. Educomunicación más allá de la Alfabetización Digital. Revista Interuniversitaria de Formación del Profesorado, 91(32), 29-42. http:// hdl.handle.net/11162/179663 Ozduzen, O., Korkut, U., & Ozduzen, C. (2021). ‘Refugees are not welcome’: Digital racism, online place-making and the evolving categorisation of Syrians in Turkey. New Media & Society, 23(11), 3349-3369. Paasch-Colberg, S., Strippel, C., Trebbe, C., & Emmer, M. (2021). From Insult to Hate Speech: Mapping Offensive Language in German User Comments on Immigration. Media and Communication, 9(1), 171–180. doi:10.17645/mac. v9i1.3399 Paci, D., & Pietrancosta, F. (2010). Il separatismo siciliano (1943-1947). Diacronie: Studi di Storia Contemporanea, (3), 2. Paisana, M., Pinto-Martinho, A., & Cardoso, G. (2020). Trust and fake news: Exploratory analysis of the impact of news literacy on the relationship with news content in Portugal. Communicatio Socialis, 33(2), 105–117. Palermo Today. (2020, Apr 12). VIDEO: “Arrustute” proibite allo Sperone, sui tetti spunta la polizia e scatta il fuggi fuggi. Palermo Today. https://www.palermotoday.it/cronaca/video-grigliata-tetti-sperone-polizia-fuggi-fuggi.html Palombelli, B. (2020). Stasera Italia: 20 marzo 2020 VIDEO. Mediaset. https://bit.ly/424eqDx Palombo, L. (2010). The drawing of the sovereign line. In J. Pugliese (Ed.), Transmediterranean: diaspora, histories, geopolitical spaces (pp. 39–58). Peter Lang Publishing. Pamungkas, E.-W., Basile, V., & Patti, V. (2020). Misogyny detection in Twitter: A multilingual and cross-domain study. Information Processing & Management, 57(6), 102360. Advance online publication. doi:10.1016/j.ipm.2020.102360 Papacharissi, Z. (2010). A private sphere: Democracy in a digital age. Polity. Papacharissi, Z. (2015). Affective publics: Sentiment, technology, and politics. Oxford University Press. Pardalis, S. (2009). Terroni and Polentoni: Where Does the Truth Lie? An Anthropology of Social Networks And Ethnicity in Palermo (Sicily). Duhram University. Pariser, E. (2011). The filter bubble. Penguin. Pariyani, B., Shah, K., Shah, M., Vyas, T., & Degadwala, S. (2021). Hate Speech Detection in Twitter using Natural Language Processing. In Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV ’21). Institute of Electrical and Electronics Engineers (IEEE) 10.1109/ICICV50876.2021.9388496 340

Compilation of References

Park, S. (2017). Inventing aliens: Immigration control, ‘xenophobia’ and racism in Japan. Race & Class, 58(3), 64–80. doi:10.1177/0306396816657719 Parra, E. (2020). ¿Por qué es necesaria una historia de la semiótica?: En busca de la identidad y la memoria a reconstruir. CIC: Cuadernos de información y comunicación, 25, 47-60. Pavlides, G. (2019). The right of freedom of expression and its limits: Hate speech in Cyprus public television (CyBC). Interview given by the singer Notis Sfakianakis on the CyBC programme ‘Tête-à-Tête’. The Cyprus Review, 31(2). https://cyprusreview.org/index.php/cr/article/download/655/552/996 Pavlopoulos, J., Malakasiotis, P., & Androutsopoulos, I. (2017). Deeper Attention to Abusive User Content Moderation. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics. 10.18653/v1/D17-1117 Paz, M. A., Montero-Díaz, J., & Moreno-Delgado, A. (2020). Hate Speech: A Systematised Review. SAGE Open, 10(October-December), 1–12. doi:10.1177/2158244020973022 Paz-Rebollo, M. A., Cáceres-Zapatero, M. D., & Martín-Sánchez, I. (2021). Suscripción a la prensa digital como contención a los discursos de odio. El Profesional de la Información, 30(6), e300613. doi:10.3145/epi.2021.nov.13 Paz-Rebollo, M. A., Mayagoitia-Soria, A., & González-Aguilar, J. M. (2021). From Polarisation to Hate: Portrait of the Spanish Political Meme. Social Media + Society, 7(4). Advance online publication. doi:10.1177/20563051211062920 Peirano, M. (2019). El enemigo conoce el sistema: Manipulación de ideas, personas e influencias después de la economía de la atención. Debate. Peirce, Ch. S. (2007). La lógica considerada como semiótica: índice del pensamiento peirciano. Biblioteca Nueva. Pejchal, V. (2020). Hate Speech and Human Rights in Eastern Europe: Legislating for Divergent Values. Routledge. doi:10.4324/9781003005742 Pereira-Kohatsu, W. M., Costa-Lopes, R., & Vala, J. (2019). The role of group norms in predicting cyberbullying among adolescents. Journal of Interpersonal Violence, 36(5-6), NP2385–NP2402. doi:10.1177/0886260517735826 Pereltsvaig, A. (2020). Languages of the World. Cambridge University Press. doi:10.1017/9781108783071 Perera, S. (2014). Dead exposures: Trophy bodies and violent visibilities of the nonhuman. Borderlands-e-journal, 13(1), 1–26. Pérez, J. M., Furman, D. A., Alemany, L. A., & Luque, F. (2021b). Robertuito: a pre-trained language model for social media text in spanish. arXiv preprint arXiv:2111.09453. Pérez, J. M., Giudici, J. C., & Luque, F. (2021a). pysentimiento: A python toolkit for sentiment analysis and socialnlp tasks. arXiv preprint arXiv:2106.09462. Pérez, E. (2017). The role of the ascribee’s representations in belief ascriptions. Teorema., 36(1), 113–125. Pérez-Escolar, M., & Herrero-Diz, P. (2023). Capítulo 5. El desmentido como instrumento para mejorar la calidad del debate público en el escenario digital. Espejo de Monografías de Comunicación Social, (10), 89–108. doi:10.52495/ c5.emcs.10.p96 Pérez-Escolar, M., & Noguera-Vivo, J. M. (2022). Hate Speech and Polarisation in Participatory Society. Knowledge Unlatched. Pérez-Latre, F. (2022). Crisis de confianza (2007-2022). Eunsa. 341

Compilation of References

Perifanos, K., & Goutsos, D. (2021). Multimodal hate speech detection in greek social media. Multimodal Technologies and Interaction, 5(7), 34. doi:10.3390/mti5070034 Perry, P. P., McInnis, B., Price, C., Smith, S., & Hollywood, J. (2013). Predictive Policing: The Role of Crime Forecasting in Law Enforcement Operations. The Role of Crime Forecasting in Law Enforcement Operations, RR233. doi:10.7249/ RR233 Pervez Akhter, M., Jiangbin, Z., Raza Naqvi, I., AbdelMajeed, M., & Zia, T. (2021). Abusive language detection from social media comments using conventional machine learning and deep learning approaches. Multimedia Systems, 28(6), 1925–1940. doi:10.100700530-021-00784-8 Piatkowska, S. J., & Lantz, B. (2021). Temporal clustering of hate crimes in the aftermath of the Brexit vote and terrorist attacks: A comparison of Scotland and England and Wales. British Journal of Criminology, 61(3), 648–669. doi:10.1093/bjc/azaa090 Piatkowska, S. J., & Stults, B. J. (2021). Brexit, Terrorist Attacks, and Hate Crime: A Longitudinal Analysis. Social Problems. doi:10.1093ocpropab005 Pierini, S. (2020, Jan 24). Coronavirus, primo caso sospetto in Italia: è una cantante barese, tornava da Wuhan. Il Messaggero. https://www.leggo.it/italia/cronache/coronavirus_primo_caso_italia_ultime_notizie_oggi_23_gennaio-5002825.html Piñeiro-Otero, T., & Martínez-Rolán, X. (2021). Say it to my face: Analysing hate speech against women on Twitter. El Profesional de la Información, 30(5), e300502. doi:10.3145/epi.2021.sep.02 Pitenis, Z., Zampieri, M., & Ranasinghe, T. (2020). Offensive language identification in Greek. In Proceedings of the 12th Language Resources and Evaluation Conference. European Language Resources Association. Pizzo, C. (2009). L’impossibile spazio del bando. Annali-Università degli Studi Suor Orsola Benincasa, 7(2). Plaza-del Arco, F. M., García-Madariaga, J., & Sánchez-Navarro, J. (2021). Influence of cyber-victimization on self-esteem, perceived social support, and life satisfaction in early adolescence. Frontiers in Psychology, 11, 608195. doi:10.3389/fpsyg.2020.608195 Plaza-del-Arco, F. M., Portillo, A. B. P., Úbeda, P. L., Gil, B., & Martín-Valdivia, M. T. (2022). Share: A lexicon of harmful expressions by Spanish speakers. In Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022). European Language Resources Association (ELRA). Plaza-Del-Arco, F. M., Molina-González, M. D., Ureña-López, L. A., & Martín-Valdivia, M. T. (2020). Detecting Misogyny and Xenophobia in Spanish Tweets Using Language Technologies. ACM Transactions on Internet Technology, 20(2), 1–19. doi:10.1145/3369869 Plutchik, R. (1980). A general psychoevolutionary theory of emotion. In: Plutchik, Robert; Kellerman, Henry. (eds.). Emotion. Theory, research, and experience: V. 1. Theories of emotion. Academic Press, (pp. 3-33). doi:10.1016/B9780-12-558701-3.50007-7 Pohjonen, M., & Udupa, S. (2017). Extreme speech online: An anthropological critique of hate speech debates. International Journal of Communication, 11, 19. Poletto, F., Basile, V., Bosco, C., Patti, V., & Stranisci, M. (2019). Annotating hate speech: Three schemes at comparison. In the 6th Italian Conference on Computational Linguistics, CLiC-it 2019, 2481, 1–8. CEUR-WS. Poletto, F., Basile, V., Sanguinetti, M., Bosco, C., & Patti, V. (2021). Resources and benchmark corpora for hate speech detection: A systematic review. Language Resources and Evaluation, 55(2), 477–523. doi:10.100710579-020-09502-8

342

Compilation of References

Poletto, M., Angelone, A., & Oreglia, E. (2021). The dark side of social media: Discrimination and prejudice in the age of digital communication. Computers in Human Behavior, 114, 106556. doi:10.1016/j.chb.2020.106556 Pomerantsev, P. (2019). This is not propaganda. Faber. Ponce, F., García, C., Cárdenas, G., Gómez, M., & Peña, A. (2022). Acercamiento al cuento “La careta” de Julio Ramón Ribeyro a partir de la semiótica de Greimas y del Grupo Entrevernes. Sincronía. Año, XXVI(82). doi:10.32870incronia. axxvi.n82 Poole, E., Giraud, E. H., & de Quincey, E. (2021). Tactical interventions in online hate speech: The case of# stopIslam. New Media & Society, 23(6), 1415-1442. Poole, E., & Richardson, J. (2006). Muslims and the News Media. I.B. Tauris. Positano News. (2020, Feb 7). Napoli. Sospetto contagio da Coronavirus per un assistente di volo, ricoverato al Cotugno per la prevista profilassi. Positano News. https://www.positanonews.it/2020/02/napoli-sospetto-contagio-coronavirusun-assistente-volo-ricoverato-al-cotugno-la-prevista-profilassi/3361402/ PowersD. M. W. (2020). Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation. CoRR, abs/2010.16061. https://arxiv.org/abs/2010.16061 Poyet, S. (2021). DOCUMENT – Charte des principes pour l’islam de France rédigée par le CFCM. Le Figaro. https:// acortar.link/Z6Jitg Prisco, F. (2020, Feb 24). Malvoluti al Sud: il Lombardo-Veneto messo in quarantena dal resto d’Italia. Il Sole 24 Ore. https://www.ilsole24ore.com/art/malvoluti-sud-veneti-e-lombardi-messi-quarantena-resto-d-italia-ACNe4aLB Priyadharshini, R. (2022). Overview of Abusive Comment Detection in Tamil. ACL. https://aclanthology.org/2022. dravidianlangtech-1.44.pdf Provost, F., & Fawcett, T. (2013). Data Science for Business: What You Need to Know about Data Mining and DataAnalytic Thinking. Puente, D. (2020, Jan 29). Due audio bufala circolano via Whatsapp: «A Lecce un cinese è stato ricoverato per il Coronavirus! State alla larga!». Open. https://www.open.online/2020/01/29/due-audio-bufala-circolano-via-whatsapp-a-lecceun-cinese-e-stato-ricoverato-per-il-coronavirus-state-alla-larga/ Pugliese, J. (2008). Whiteness and the blackening of Italy: La guerra cafona, Extracomunitari and Provisional Street Justice. Portal: Journal of Multidisciplinary International Studies, 5(2), 1–35. Pugliese, J. (2017). The Souths of “The West”: Geocorpographical Assemblages of Plants, Colonialism and Race in Sergio Leone’s Metafigural Spaghetti Westerns. Muiraquitã: Revista de Letras e Humanidades, 5(2), 11–59. doi:10.29327/216344.5.2-3 Pulido, M. P., Arroyo, C., Flores, J. A., & Ytarte, R. M. (2021). Health Care Guidelines against Hate Violence in Support and advice through health system for hate crimes victims: a socio-sanitary approach. Tirant Humanidades. Putnam, R. D. (1993). Making Democracy Work: Civic Traditions in Modern Italy. Princeton University Press. Quijano-Sanchez, L., Kohatsu, J. C. P., Liberatore, F., & Camacho-Collados, M. (2019). HaterNet a system for detecting and analyzing hate speech in Twitter (Version 1.0) [Data set]. Zenodo. Rabasa, A., & Benard, C. (2015). Eurojihad. Patterns of Islamist Radicalization and Terrorism in Europe. Cambridge University Press.

343

Compilation of References

Rajendran, L., & Thesinghraja, P. (2014). The impact of new media on traditional media. Middle East Journal of Scientific Research, 22(4), 609–616. Ramezan, C. A., Warner, T. A., Maxwell, A. E., & Price, B. S. (2021). Effects of Training Set Size on Supervised Machine-Learning Land-Cover Classification of Large-Area High-Resolution Remotely Sensed Data. Remote Sensing (Basel), 13(3), 3. doi:10.3390/rs13030368 Ramponi, A., Testa, B., Tonelli, S., & Jezek, E. (2022). Addressing religious hate online: From taxonomy creation to automated detection. PeerJ. Computer Science, 8, e1128. doi:10.7717/peerj-cs.1128 Ranieri, M. (2016). Populism, Media and Education: Challenging discrimination in contemporary digital societies. Routledge. doi:10.4324/9781315680903 Raya, C. (2018). El reguetón afecta a tus neuronas. TecnoXplora. https://www.lasexta.com/tecnologia-tecnoxplora/ciencia/divul gacion/regueton-afecta-tus-neuronas_201802065a7b860d0cf21951 69bd407b.html Regione Campania. (2020). Coronavirus, discorso del presidente De Luca. Regione Campania. https://www.regione. campania.it/regione/it/multimedia/web-tv/coronavirus-discorso-del-presidente-de-luca Regione Puglia. (2020). Il presidente Emiliano spiega in video l’ordinanza dell’8 marzo 2020. Regione Puglia. https:// press.regione.puglia.it/-/il-presidente-emiliano-spiega-in-video-l-ordinanza-dell-08.03.2020 Reinert, M. (1986). Un logiciel d’analyse lexicale: Alceste. Les Cahiers de l’Analyse des Données, 11(4), 471–484. Reinert, M. (1990). Alceste: une méthodologie d’analyse des donne textuelles et une application: Aurélia de G. de Nerval. Bulletin de Methodologie Sociologique, 26(1), 24–54. doi:10.1177/075910639002600103 Relia, K., Jakubowicz, K., & Kovács, A. M. (2019). Hate speech and hate crime: The legal and regulatory framework for the EU and its member states. European Parliament. doi:10.2861/252148 Reuters. (2020). Two first coronavirus cases confirmed in Italy: prime minister. Reuters. https://www.reuters.com/article/ us-china-health-italy-idUSKBN1ZT31H Reyes, M. A. (2006). La construcción del Estado autonómico. Cuadernos Constitucionales de la Cátedra Fadrique Furió Ceriol, (54), 75–95. Rey, F. (2015). Discurso de odio y racismo líquido. In Revenga Sánchez, M. (dir.), Libertad de expresión y discursos del odio (pp. 51–88). Universidad de Alcalá. Riccio, I. (2020, Feb 2). “Napoletani figli del colera vi mettiamo in quarantena”: lo striscione choc dei tifosi dell’Inter. Il Giornale.it. https://www.ilgiornale.it/news/napoli/napoletani-figli-colera-vi-mettiamo-quarantena-polemiche-1820626. html Richardson, R. (2009). Islamophobia or anti-Muslim racism–or what?–concepts and terms revisited. Insted. http://www. insted.co.uk/anti-muslim-racism.pdf Richardson, L., Amundsen, M., & Ruby, S. (2013). RESTful Web APIs: Services for a Changing World. O’Reilly Media, Inc. Rid, T. (2020). Active measures. The secret history of disinformation and political warfare. Profile books. Rietdijk, N. (2021). Post-truth Politics and Collective Gas-lighting. Episteme (Edinburgh), 1–17. doi:10.1017/epi.2021.24 Ring, C. E. (2013). Hate speech in social media: An exploration of the problem and its proposed solutions [Doctoral dissertation]. University of Colorado at Boulder. 344

Compilation of References

Robertson, M. (2021). Provincialising the OECD-PISA global competences project. Globalisation, Societies and Education, 19(2), 167–182. doi:10.1080/14767724.2021.1887725 Rodrigues, P. (2010). Criminalidade na imprensa: Análise do Correio da Manhã, 2000-2007. Sociologia, Problemas e Práticas, 64, 149-172. https://journals.openedition.org/spp/338#tocto2n2 Rodríguez Fernández, L. (2023). Desinformación Y Relaciones Públicas. Aproximación a Los Términos Black PR Y Dark PR, Revista ICONO 14. Revista Científica De Comunicación Y Tecnologías Emergentes, 21 (1). . doi:10.7195/ ri14.v21i1.1920 Rodríguez Fernández, L. (2021). Propaganda digital. Comunicación en Tiempos de Desinformación. Editorial UOC. Rodríguez, A. (2021). Proyecto Co-critic.ar: un trabajo colaborativo interinstitucional (coil) para promover en los estudiantes habilidades para pensar críticamente la realidad social. In Innovación e investigación docente en educación: experiencias prácticas. Dyckinson, S, L. Rodríguez, W. J. (2022). De la posverdad a la hiperrealidad: ¿Qué hay detrás de la difusión viral de “Fake News”? Quaestiones Disputatae: Temas En Debate, 15(30), 86-105. http://revistas.ustatunja.edu.co/index.php/qdisputatae/artic le/view/2599 Rodríguez, A., & Lázaro Pernias, P. (2022). Nuevas narrativas en publicidad. El storydoing como modelo de comunicación que favorece la cocreación del relato de marca entre los usuarios de Instagram. AdComunica, 24, 57–80. doi:10.6035/ adcomunica.6554 Rodríguez-Arenas, M. A., Martín-Ríos, M. D., & Gil-Borrelli, C. (2018). Intervenciones en salud pública contra la violencia de odio. Gaceta Sanitaria, 32(2), 114–116. doi:10.1016/j.gaceta.2017.10.013 PMID:29370937 Rodríguez-Fernández, L. (2019). Desinformación y comunicación organizacional: Estudio sobre el impacto de las fake news. Revista Latina de Comunicación Social, 74, 1717–1728. doi:10.4185/RLCS-2019-1406 Rojas-Torrijos, J. L., & Guerrero Caballero, M. (2021). ¿Un periodismo deportivo apolítico?: Estudio de las referencias al ‘procés’ catalán en la prensa deportiva digital en España. Estudios sobre el mensaje meriodístico, 27(3), 947-960. doi:10.5209/esmp.72554 Romero, J. A. O. (2019). Desinformación: concepto y perspectivas. Real Instituto Elcano (RIElcano), 41. Rosendo Sánchez, N. (2016). Mundos transmediales: Revisión conceptual y perspectivas teóricas del arte de crear mundos. Icono, 14(14), 49–70. doi:10.7195/ri14.v14i1.930 Rossano, M. (2011). La construcción del héroe y del antagonista a través de las imágenes: Garibaldi y los bandoleros del sur de Italia 1861-2011. Quaderns-e de l’Institut Català d’Antropologia, (16 (1-2)), 189–200. Röttger, P., Vidgen, B., Hovy, D., & Pierrehumbert, J. B. (2021). Two contrasting data annotation paradigms for subjective nlp tasks. arXiv preprint arXiv:2112.07475. https://doi.org/ doi:10.35848/1347-4065/aca256 Roviello, V., & Roviello, G. N. (2021). Less COVID-19 deaths in southern and insular Italy explained by forest bathing, Mediterranean environment, and antiviral plant volatile organic compounds. Environmental Chemistry Letters, 20(1), 7–17. doi:10.100710311-021-01309-5 PMID:34483793 Roy, O. (2017). Jihad and death. The global appeal of Islamic State. Hurst & Company. Roy, S. (2019). Role of Nonviolent Communication in our Daily Life. Pragyaan: Journal of Mass Communication, 17(1-2), 24–27.

345

Compilation of References

Rubio, D. (2022). Protocolo y semiótica como estrategia de comunicación política. SIGNUM: Revista Internacional De Investigación En Eventos, Protocolo Y Relaciones Institucionales, 1(1), 1–10. https://publicaciones.protocoloimep. com/signum/article/view/5 Ruiz-Díaz, L. (2017). La prevención de la radicalización en la estrategia contra el terrorismo de la Unión Europea: Entre soft law e impulso de medidas de apoyo. Revista Española de Derecho Internacional, 69(2), 257–280. doi:10.17103/ redi.69.2.2017.1.10 Rullo, L. (2021). The COVID-19 pandemic crisis and the personalisation of the Government in Italy. International Journal of Public Leadership, 17(2), 196–207. doi:10.1108/IJPL-08-2020-0083 Said, E. (1978). Orientalism. Pantheon Book. Said-Hung, E. M., Merino-Arribas, M. A., & Martínez-Torres, J. (2021). Evolución del debate académico en la Web of Science y Scopus sobre unfaking news (2014-2019). Estudios sobre el Mensaje Periodístico, 27(3), 961–971. doi:10.5209/ esmp.71031 Said-Hung, E., & Ocarranza-Prado, I. (2022). Actores políticos y promoción de contenidos desinformativos en Twitter. Caso España. Revista Sociedade e Estado, 37(2), 575–598. doi:10.15900102-6992-202237020009 Sakki, I., & Hakoköngäs, E. (2022). A Critical Discursive Psychological Study of Dialogical Constructions of HateSpeech in Established Media and Online Discussions. In K. Pettersson & E. Nortio (Eds.), The Far-Right Discourse of Multiculturalism in Intergroup Interactions. Palgrave Studies in Discursive Psychology (pp. 85–111). Palgrave Macmillan. doi:10.1007/978-3-030-89066-7_4 Salaverría, R., Buslón, N., López-Pan, F., León, B., López-Goñi, I., & Erviti, M.-C. (2020). Desinformación en tiempos de pandemia: Tipología de los bulos sobre la covid-19. El Profesional de la Información, 29(3). doi:10.3145/epi.2020.may.15 Salminen, J., Veronesi, F., & Almerekhi, H. (2018). Online Hate Interpretation Varies by Country, But More by Individual: A Statistical Analysis Using Crowdsourced Ratings. Fifth International Conference on Social Networks Analysis. Management and Security, 88-94. 10.1109/SNAMS.2018.8554954 Salvetti, P. (2017). Rope and soap: Lynchings of Italians in the United States. Bordighera Press. Sälzer, C., & Roczen, N. (2018). Assessing global competence in PISA 2018: Challenges and approaches to capturing a complex construct. International Journal of Development Education and Global Learning, 10(1), 5–20. doi:10.18546/ IJDEGL.10.1.02 Sánchez-Carballido, J.-R. (2008). Perspectivas de la información en Internet. Zer, 13(25), 61-81. https://ojs.ehu.eus/ index.php/Zer/article/view/3574 Sánchez-Holgado, P., Amores, J. J., & Blanco-Herrero, D. (2022). Online Hate Speech and Immigration Acceptance: A Study of Spanish Provinces. Social Sciences (Basel, Switzerland), 11(11), 515. doi:10.3390ocsci11110515 Sánchez-Junquera, J., Chulvi, B., Rosso, P., & Ponzetto, S. P. (2021b). How Do You Speak about Immigrants? Taxonomy and StereoImmigrants Dataset for Identifying Stereotypes about Immigrants. Applied Sciences (Basel, Switzerland), 11(8), 3610. doi:10.3390/app11083610 Sánchez-Junquera, J., Rosso, P., Montes y Gómez, M., & Chulvi, B. (2021a). Masking and BERT-based Models for Stereotype Identification. Procesamiento del Lenguaje Natural, 67, 83–94. Santana-Vega, L. E., González-Morales, O., & Feliciano-García, L. (2021). Are we Europeans? Secondary education students’ beliefs and sense of belonging to the European Union. Journal of Youth Studies, 24(10), 1358–1372. doi:10. 1080/13676261.2020.1839634 346

Compilation of References

Santos,C.D.,Correia,G.,&Figueiredo,I.A.(2021,Maio24).André Ventura condenado por ter chamado “bandidos” a moradores do bairro da Jamaica. Líder do Chega vai recorrer. Observador. https://observador.pt/2021/05/24/andre-ventura-condenado-por -ter-chamado-bandidos-a-moradores-do-bairro-da-jamaica/ Sanz-Leal, M., Orozco, M., & Toma, R. (2022). Construcción conceptual de la competencia global en educación, Teoría de la Educación. Teoría de la Educación, 34(1), 83–103. doi:10.14201/teri.25394 Sayad, A. (1999). La double absence. Seuil. Sayce, D. (2020). The number of tweets per day in 2020. https://bit.ly/3CZbUB2 Sayimer, İ., & Derman, M. R. (2017). Syrian refugees as victims of fear and danger discourse in social media: A YouTube analysis. Yeni Medya’da Korku ve Tehlike Söylemlerinin Kurbanları Suriyeli Mülteciler: Bir Youtube Analizi, 8(15), 384–403. Sazzed, S. (2021). A lexicon for profane and obscene text identification in Bengali. In Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021). INCOMA Ltd. 10.26615/978-954452-072-4_145 Scharwächter, E., & Müller, E. (2020). Does terrorism trigger online hate speech? on the association of events and time series. The Annals of Applied Statistics, 14(3), 1285–1303. doi:10.1214/20-AOAS1338 Scheufele, D. (1999). Framing as a theory of media effects. Journal of Communication, 49(1), 103–122. doi:10.1111/j.1460-2466.1999.tb02784.x Schmitt, J. B., Rieger, D., Rutkowski, O., & Ernst, J. (2018). Counter-messages as Prevention or Promotion of Extremism?! The Potential Role of YouTube: Recommendation Algorithms. Journal of Communication, 68(4), 780–808. doi:10.1093/joc/jqy029 Scognamiglio, R. (2020). “Chiudono tutta la Lombardia”. Assalto alla stazione di Milano. Il Giornale. https://www. ilgiornale.it/news/milano/pasticcio-governo-sul-decreto-scatta-fuga-milano-1837402.html Scolari, C. (2012). Narrativas transmedia: cuando todos los medios cuentan. Deusto. Searle, J. (1969). Speech Acts. An essay in the philosophy of language. Cambridge at the University Press. doi:10.1017/ CBO9781139173438 Searle, J. (1995). The Construction of Social Reality. The Free Press. Searle, J. (2010). Making the Social World. The Structure of Human Civilization. Oxford University Press. doi:10.1093/ acprof:osobl/9780195396171.001.0001 Sebastiani, G., Massa, M., & Riboli, E. (2020). Covid-19 epidemic in Italy: Evolution, projections and impact of government measures. European Journal of Epidemiology, 35(4), 341–345. doi:10.100710654-020-00631-6 PMID:32306149 Seeta, J. (2019). Social media, race, and gender: A critical review. Council on Contemporary Families. https://contemporaryfamilies.org/social-media-race-and-gender-a-critical-review/ Seoane, J. (2019). Opinión pública. Eunomía. Revista en Cultura de la Legalidad, 17, 235–248. doi:10.20318/eunomia.2019.5028 Serra, L., Vall-Llosera, L., Varga, D., Saurina, C., Saez, M., & Renart, G. (2022). Analysis of the geographic pattern of the police reports for domestic violence in Girona (Spain). BMC Public Health, 22(1), 1–11. doi:10.118612889-02212916-4 PMID:35313832

347

Compilation of References

Shah, D. V., Cho, J., Eveland, W. P. Jr, & Kwak, N. (2005). Information and expression in a digital age: Modeling Internet effects on civic participation. Communication Research, 32(5), 531–565. doi:10.1177/0093650205279209 Shane, T., Willaert, T., & Tuters, M. (2022). The rise of ‘gas-lighting’: Debates about desinformation on Twitter and 4chan, and the possibility of a ‘good echo chamber’. Popular Communication, 20(5), 178–192. doi:10.1080/15405702 .2022.2044042 Shigeo, A. (2005). Support Vector Machines for Pattern Classification. Springer. doi:10.1007/1-84628-219-5 Siapera, E., Boudourides, M., Lenis, S., & Suiter, J. (2018). Refugees and network publics on Twitter: Networked framing, affect, and capture. Social Media + Society, 4(1). doi:10.1177/2056305118764437 Siegel, A. A. (2020). Online hate speech. Social media and democracy: The state of the field, prospects for reform, 56-88. Silva, S., Castro Ferreira, T., Silva Ramos, R. M., & Paraboni, I. (2020). Data Driven and Psycholinguistics Motivated Approaches to Hate Speech Detection. Computación y Sistemas, 24(3). doi:10.13053/cys-24-3-3478 Silverstone, R. (2002). Complicity and collusion in the mediation of everyday life. New Literary History, 33(4), 761–780. doi:10.1353/nlh.2002.0045 Silverstone, R. (2004). Regulation, media literacy and media civics. Media Culture & Society, 26(3), 440–449. doi:10.1177/0163443704042557 Simmel, G. (1971). On Individuality and Social Forms. Ed by Janowitz, M., The Heritage of Sociology. University of Chicago Press. Simpson, A., & Dervin, F. (2019). Global and intercultural competence for whom? By whom? For what purpose? An example from the Asia Society and the OECD. Compare: A Journal of Comparative Education, 49(4), 672–677. doi:1 0.1080/03057925.2019.1586194 Simpson, G. E., & Yinger, J. M. (2013). Racial and cultural minorities: An analysis of prejudice and discrimination. Springer Science & Business Media. Sirvent, M., & Rigal, L. (2012). Investigación acción participativa. Un desafío de nuestros tiempos para la construcción de una sociedad democrática. Páramos Andinos. Slavíčková, T., & Zvagulis, P. (2014). Monitoring anti-minority rhetoric in the Czech print media. Journal of Language and Politics, 13(1), 152–170. doi:10.1075/jlp.13.1.07sla Smith, P., Phillips, T. L., & King, R. D. (2010). Incivility: The rude stranger in everyday life. Cambridge University Press. doi:10.1017/CBO9780511781803 Snoussi, T., & Korbi, W. (2021). Cyber anti-Hate speech during the Covid-19 pandemic: Semiotic Analysis. The Asian ESP Journal, 17(4.2), pp. 169-189 Soldatov, A., & Borogan, I. (2015). The Red Web. Public Affairs Books. Soscia, M. (2014). Tra storia e letteratura. Il colera in Italia e a Napoli. Sinestesieonline, 10, 7–28. Spagnolo, P. (2020). Contro il calcio moderno. Odoya. Spena, A. (2020). Coronavirus, perché nasce la psicosi. Vita. https://www.vita.it/it/article/2020/01/31/coronavirusperche-nasce-la-psicosi/153939/ Spies, A. (2020). The portrayal of victims of intimate femicide in the South African media. Journal of African Media Studies, 12(1), 41–59. doi:10.1386/jams_00010_1 348

Compilation of References

Sreelakshmi, P. R., Binu, V. S., Kumar, R., & Ramakrishnan, J. (2020). The impact of cyberbullying on psychological well-being: A cross-sectional study among college students in Kerala. Indian Journal of Community Medicine, 45(2), 208–212. doi:10.4103/ijcm.IJCM_166_19 Stamou, V., Alexiou, I., Klimi, A., Molou, E., Saivanidou, A., & Markantonatou, S. (2022). Cleansing & expanding the HURTLEX (el) with a multidimensional categorization of offensive words. In Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH). Association for Computational Linguistics. 10.18653/v1/2022.woah-1.10 Staples, B. (2019). How Italians Became ‘White’. The New York Times. https://www.nytimes.com/interactive/2019/10/12/ opinion/columbus-day-italian-american-racism.html Statista (2020c). Ranking de las principales marcas de medios de comunicación online según el porcentaje de población que las usaba de forma semanal en España en 2020. Statista. https://es.statista.com/estadisticas/670743/principalesmedios-online-porcentaje-de-usuariossemanales-espana/r Statista. (2020a). Ranking de los 10 medios de comunicación con más seguidores en Twitter en España en abril de 2020. Statista. https://es.statista.com/estadisticas/518802/Twitter-perfiles-de-medios-decomunicacion-con-mas-seguidos-enespana/ Statista. (2020b). Páginas oficiales de medios de comunicación más populares en Facebook en España en abril de 2020, por número de seguidores. Statista. https://es.statista.com/estadisticas/518580/paginas-demedios-de-comunicacion-deFacebook-con-mas-seguidores-en-espana/#statisticContainer Stevenson, N. (2018). Education and the alterity of democracy. Cadernos de Pesquisa, 48(167), 150–171. doi:10.1590/198053144668 Strossen, N. (2020). Hate: Why We Should Resist It With Free Speech, Not Censorship. Oxford University Press. Sulis, G., & Gheno, V. (2022). The Debate on Language and Gender in Italy, from the Visibility of Women to Inclusive Language (1980s–2020s). The Italianist, 42(1), 153–183. doi:10.1080/02614340.2022.2125707 Sullivan, A. (2022). Semantic Dimensions of Slurs. Philosophia, 50(3), 1479–1493. doi:10.100711406-021-00427-2 Sumalla, J. M. (2018). Los delitos de odio en las redes sociales. IDP Revista de Internet Derecho y Política, 27(27). doi:10.7238/idp.v0i27.3151 Suzor, N., Dragiewicz, M., Harris, B., Gillett, R., Burgess, J., & Van Geelen, T. (2019). Human rights by design: The responsibilities of social media platforms to address gender‐based violence online. Policy and Internet, 11(1), 84–103. doi:10.1002/poi3.185 Syahputra, I. (2019). Expressions of hatred and the formation of spiral of anxiety on social media in Indonesia. SEARCH Journal of Media and Communication Research, 11(1), 95–112. Tagg, P., & Clarida, B. (2003). Ten Little Title Tunes. The Mass Media Music Scholars’ Press, Inc. Taher, K. I., Abdulazeez, A. M., & Zebari, D. A. (2021). Data Mining Classification Algorithms for Analyzing Soil Data. Asian Journal of Research in Computer Science, 17-28. doi:10.9734/ajrcos/2021/v8i230196 Tchakhotine, S. (1992). Le Viol des foules par la propagande politique. Gallimard. Te l e s , F. , & C o e l h o , P. ( 2 0 2 2 , Nove m b e r 1 6 ) . Po l í c i a s s e m l e i : o ó d i o d e 5 9 1 a ge ntes de autoridade. Setenta e Quatro. https://setentaequatro.pt/investigacao-74/policias-sem-lei-o -odio-de-591-agentes-de-autoridade Teti, V. (1993). La razza maledetta. Origini del pregiudizio antimeridionale. Manifestolibri. 349

Compilation of References

Tewsbury, D., & Scheufele, D. (2019). News Framing Theory and Research. In M. B. Oliver, A. Raney, & J. Bryant (Eds.), News Framing Theory and Research (pp. 1–18). Routledge. Thorp, H. H. (2023). ChatGPT is fun, but not an author. Science, 379(6630), 313–313. doi:10.1126cience.adg7879 PMID:36701446 Tibi, B. (2010). Ethnicity of fear? Islamic migration and the ethnicization of Islam in Europe. Studies in Ethnicity and Nationalism, 10(1), 126–157. doi:10.1111/j.1754-9469.2010.01038.x Tontodimamma, A., Nissi, E., Sarra, A., & Fontanela, A. (2021). Thirty years of research into hate speech: Topics of interest and their evolution. Scientometrics, 126(1), 157–179. doi:10.100711192-020-03737-6 Törnberg, A., & Törnberg, P. (2016). Muslims in social media discourse: Combining topic modeling and critical discourse analysis. Discourse, Context & Media, 13, 132–142. doi:10.1016/j.dcm.2016.04.003 Torres-Lugo, C., Pote, M., Nwala, A. C., & Menczer, F. (2022). Manipulating Twitter Through Deletions. In Proceedings of the International AAAI Conference on Web and Social Media (Vol. 16, pp. 1029-1039). PKP. 10.1609/icwsm. v16i1.19355 Tort, P., & Chiesura, G. (2000). L’antropologia di Darwin: la laicizzazione del discorso sull’uomo. Manifestolibri. Tosar, B. (2021). Literacidad crítica. Enseñar a leer el mundo para transformarlo a través del encuentro. Universitat Autònomade Barcelona. Tribalat, M. (2015). Une estimation des populations d’origine étrangère en France en 2011. Espace, Populations, Sociétés, (1-2), 1–2. doi:10.4000/eps.6073 Troian, J., Baidada, O., Arciszewski, T., Apostolidis, T., Celebi, E., & Yurtbakan, T. (2019). Evidence for indirect loss of significance effects on violent extremism: The potential mediating role of anomia. Aggressive Behavior, 45(6), 691–703. doi:10.1002/ab.21863 PMID:31452219 Trujillo, H. M., & Moyano, M. (2008). El sistema educativo español en la prevención de la radicalización yihadista. Athena Intelligence Journal, 3(2), 75–83. Trust, R. (1997), Islamophobia: A Challenge for Us All. Runny Med Trust. https://www.runnymedetrust.org/publications/islamophobia-a-challenge-for-us-all Tufekci, Z. (2014). Big Questions for social media big data: Representativeness, validity and other methodological pitfalls. In Proceedings of the 8th International Conference on Weblogs and Social Media, ICWSM 2014 (pp. 505–514). The AAAI Press. https://doi.org/10.1609/icwsm.v8i1.14517 Túñez-López, M., Vaz-Álvarez, M., & Fieiras-Ceide, C. (2020). COVID-19 and public service media; the impact of the pandemic on public television in Europe. El Profesional de la Información, 29(5), e290518. doi:10.3145/epi.2020.sep.18 Twitter. (2023). Hateful Conduct. Twitter Help Center. https://help.twitter.com/en/rules-and-policies/hateful-condu ct-policy Ullmann, S., & Tomalin, M. (2020). Quarantining online hate speech: Technical and ethical perspectives. Ethics and Information Technology, 22(1), 69–80. doi:10.100710676-019-09516-z UNESCO & Oxford. (2021). Hacer frente al discurso de odio en las redes sociales: desafíos contemporáneos. Unesco. UNESCO. (2011). Media and information literacy curriculum for teachers. https://unesdoc.unesco.org/ark:/48223/ pf0000192971

350

Compilation of References

UNESCO. (2015): Countering online Hate Speech. UNESCO Digital Lybrary. https://unesdoc.unesco.org/ark:/48223/ pf0000233231 UNESCO. (2017). Migrant/migration. United Nations Educational Scientific and Cultural Organization. http://www. unesco.org/new/en/social-and-human-sciences/themes/international-migration/glossary/migrant UNESCO. (2018). Media and information literacy. https://unesdoc.unesco.org/ark:/48223/pf0000265509 U N E S CO . ( 2 0 2 0 ) . COV I D - 1 9 — Fi g h t i n g ‘ i n fo d e m i c ’ a n d s o c i a l st i g m a th ro u g h c o m m u n it y m e d i a i n I n d i a . U N E S CO . h t t p s : / / w w w. u n e s c o . o rg / e n / a r t i c l e s / c ov i d - 1 9 - f i g h t i n g - i n fo d e m ic-and-social-stigma-through-community-media-india UNHCR. (1951). Convention and Protocol Relating to the Status of Refugees. The UN Refugee Agency. https://www. unhcr.org/protection/basic/3b66c2aa10/convention-protocolrelating-status-refugees.html UNHCR. (2016). ‘Refugee’ or ‘migrant’ – Which is right? The UN Refugee Agency. https://www.unhcr.org/news/stories/ unhcr-viewpoint-refugee-or-migrant-which-right UNHCR. (2021a). Global Trends Forced Displacement In 2021. The UN Refugee Agency. https://www.unhcr. org/62a9d1494/global-trends-report-2021 UNHCR. (2021b). Refugees and Asylum Seekers in Turkey. The UN Refugee Agency. https://www.unhcr.org/tr/en/ refugees-and-asylum-seekers-in-turkey UNHCR. (2022). Refugee Data Finder. The UN Refugee Agency. https://www.unhcr.org/refugee-statistics/ UNHCR. (2023). Understanding Hate Speech. The UN Refugee Agency. https://www.un.org/en/hate-speech/understandinghate-speech/what-is-hate-speech Un i t e d Na t i o n s . ( 2 0 1 9 ) . P ro m o t i o n a n d p ro te c t i o n o f t h e r i g h t to f re e d o m o f o p i n i o n and expression. United Nations. https://documents-dds-ny.un.org/doc/UNDOC/GEN/N19/308/13/PDF /N1930813.pdf?OpenElement United Nations. (2022). United Nations Office on Genocide Prevention and the Responsibility to Protect. Office on genocide prevention and the responsability to protect. https://www.un.org/en/genocideprevention/hate-speech-strateg y.shtml Uyheng, J., Bellutta, D., & Carley, K. M. (2022). Bots Amplify and Redirect Hate Speech in Online Discourse About Racism During the COVID-19 Pandemic. Social Media + Society, 8(3). doi:10.1177/20563051221104749 Vala, J. (2021). Racismo, hoje: Portugal em Contexto Europeu. Fundação Francisco Manuel dos Santos. Valdez-Apolo, M. B., Arcila-Calderón, C., & Jiménez Amores, J. (2019). El discurso del odio hacia migrantes y refugiados a través del tono y los marcos de los mensajes en Twitter. RAEIC. Revista de la Asociación Española de Investigación de la Comunicación, 6(12), 361–384. doi:10.24137/raeic.6.12.2 Valente, R. (2019). Spatial and temporal patterns of violent crime in a Brazilian state capital: A quantitative analysis focusing on micro places and small units of time. Applied Geography (Sevenoaks, England), 103, 90–97. doi:10.1016/j. apgeog.2019.01.006 Vallinskosky, K., Benjamin, S., & Elliot, K. (2022). By correcting disinformation and encouraging them towards tolerance: Finnish educators’ considerations of PVE-E. Journal for Deradicalization, 23(33), 181–220. Vampa, D. (2020). Competing forms of populism and territorial politics: The cases of Vox and Podemos in Spain. Journal of Contemporary European Studies, 28(3), 304–321. doi:10.1080/14782804.2020.1727866 351

Compilation of References

Van der Haak, B., Parks, M., & Castells, M. (2012). The future of journalism: Networked journalism. International Journal of Communication, 6, 16. Van der Nagel, E. (2018). ‘Networks that work too well’: Intervening in algorithmic connections. Media International Australia, Incorporating Culture & Policy, 168(1), 81–92. doi:10.1177/1329878X18783002 Van Dijk, T. A. (1990). Social cognition and discourse. Handbook of language and social psychology, 163, 183. APA. van Dijk, T. A. (2000). New(s) Racism: A discourse analytical approach. Ethnic minorities and the media, 33-49. Van Dijk, T. (1993). Elite discourse and racism. SAGE Publications, Inc. doi:10.4135/9781483326184 Van Dijk, T. (1997). Racismo y análisis crítico de los medios. Paidós. Van Dijk, T. (1999). El análisis crítico del discurso. Anthropos, 186(1), 23–36. Van Dijk, T. A. (1991). Media contents: The interdisciplinary study of news as discourse. In A handbook of qualitative methodologies for mass communication research (pp. 108–120). Routledge. VanDijk,T.A.(1992).Discourseandthedenialofracism.Discourse&Society,3(1),87–118.doi:10.1177/0957926592003001005 Van Dijk, T. A. (2017). Discourse and power. Bloomsbury Publishing. Van Dijk, T. A. (2021). Discurso y poder. Gedisa. Van Patten, I. T., McKeldin-Coner, J., & Cox, D. (2009). A microspatial analysis of robbery: Prospective hot spotting in a small city. Crime Mapping, 1(1), 7–32. http://ivanpatt.asp.radford.edu/Research/Microspatial%20Anal ysis%20of%20Robbery.pdf Vardal, Z. B. (2015). Nefret söylemi ve yeni medya (Hate speech and new media). Maltepe Üniversitesi İletişim Fakültesi Dergisi, 2(1), 132–156. VareseNoi. (2020, Mar 9). FOTO. «Un grazie non indifferente a chi salva la nostra gente»: striscioni della curva al Circolo e in città. VareseNoi. https://www.varesenoi.it/2020/03/09/leggi-notizia/argomenti/varese/articolo/foto-un-grazie-nonindifferente-a-chi-salva-la-nostra-gente-striscioni-della-curva-al-circolo.html Vasallo, A. (2020). ¿Se discrimina a los inmigrantes latinoamericanos en España? La Vanguardia. https://www.lavanguardia.com/participacion/cartas/20200310/474046212445/discriminacion-maltrato-inmigrantes-latinos-latinoamericanosespana-debate.html Vasso, G. (2020, Mar 4). La pizza al coronavirus fa infuriare il Comune di Napoli. Il Giornale. https://www.ilgiornale. it/news/napoli/pizza-coronavirus-fa-infuriare-comune-napoli-1835258.html Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30. Vega-Gutiérrez, A.M. (2023). La educación como herramienta de prevención del radicalismo violento: aportaciones del curriculum de religión islámica desde un enfoque competencial. Anuario de Derecho eclesiástico del Estado. Vela, C. (2020). A semiotic approach to the study of emojis. Círculo de Lingüística Aplicada a la Comunicación, (84), 153–165. VeraLeaks. (2020, Apr 20). Il coronavirus non ferma le attività illecite nel mercato ittico di Taranto davanti all’ufficio dei Vigili Urbani. VeraLeaks. https://veraleaks.org/2020/04/19/il-coronavirus-non-ferma-le-attivita-illecite-nel-mercatoittico-di-taranto-davanti-allufficio-dei-vigili-urbani/

352

Compilation of References

Verdú, D. (2020, Apr 21). El virus invierte los papeles históricos del norte y el sur en Italia. El País. https://elpais.com/ internacional/2020-04-21/el-virus-invierte-los-papeles-historicos-del-norte-y-el-sur-en-italia.html Verkuyten, M. (2018). Religious fundamentalism and radicalisation among Muslim minority youth in Europe. European Psychologist, 23(1), 21–31. doi:10.1027/1016-9040/a000314 Vicente, Á., & García-Calvo, C. (2020). Evaluación de proyectos a pequeña escala en materia de prevención de la radicalización violenta: Una visión aplicada. Análisis del Real Instituto Elcano, 121(1), 1–14. Vidales, C., Parra, E., & Lanigan, R. L. (2021). Presentación: Comunicación y Semiótica. Comunicación y Sociedad (Guadalajara), 1-3(0), 1–3. Advance online publication. doi:10.32870/cys.v2021.8112 Vidgen, B. (2019). Tweeting islamophobia [Doctoral thesis, University of Oxford]. British Library Ethos. https://ethos. bl.uk/OrderDetails.do?uin=uk.bl.ethos.786187 Vinuesa Tejero, M. L. (1997). Opinión pública y cultura política en la España democrática: un estudio empírico de las elecciones de 1993 (Tesis doctoral, Universidad Complutense de Madrid, España]. VirgilioNotizie. (2020, Apr 8). Coronavirus, a Napoli folla in strada nei Quartieri Spagnoli. VirgilioNotizie https://notizie. virgilio.it/coronavirus-napoli-folla-strada-quartieri-spagnoli-1358953 Vitullo, A. (2021). The Online Intersection among Islamophobia, Populism, and Hate Speech: An Italian Perspective. Journal of Religion. Media and Digital Culture, 10(1), 95–114. doi:10.1163/21659214-bja10028 Viúdez, J. (2019). Récord de Pateras y de Muertes de Migrantes en 2018. El País, 16. Vorobyeva, T., Mouratidis, K., Diamantopoulos, F. N., Giannopoulos, P., Tavlaridou, K., Timamopoulos, C., Peristeras, V., Magnisalis, I., & Shah, S. I. H. (2020). A Fake News Classification Framework: Application On Immigration Cases. Communication Today, 11(2), 118-131. https://tinyurl.com/y6zasn49 Waisbord, S. (2018). Truth is what happens to news: On journalism, fake news, and post-truth. Journalism Studies, 19(13), 1866–1878. doi:10.1080/1461670X.2018.1492881 Waldron, J. (2012). The Harm in Hate Speech. Harvard University Press. doi:10.4159/harvard.9780674065086 Wall, M. (2015). Citizen Journalism: A retrospective on what we know, an agenda for what we don’t. Digital Journalism (Abingdon, England), 3(6), 797–813. doi:10.1080/21670811.2014.1002513 Wardle, C. (2017). Fake news. It’s complicated. First Draft Foot Notes. Medium. https://medium.com/1st-draft/fakenews-its-complicatedd0f773766c79 Wardle, C., & Derakhshan, H. (2018). Thinking about ‘information disorder’: formats of misinformation, disinformation, and mal-information. In C. Ireton & J. Posetti (Eds.), Journalism, ‘fake news’ & disinformation (pp. 43–54). Unesco. Warren-Gordon, K., & Rhineberger, G. (2021). The “Trump effect” on hate crime reporting: Media coverage before and after the 2016 presidential election. Journal of Ethnicity in Criminal Justice, 19(1), 25–45. doi:10.1080/15377938 .2021.1895944 Waseem, Z., Davidson, T., Warmsley, D., & Weber, I. (2017). Understanding abuse: A typology of abusive language detection subtasks. In Proceedings of the First Workshop on Abusive Language Online. Association for Computational Linguistics. 10.18653/v1/W17-3012 Weine, S., & Osman, A. (2012). Building Resilience to Violent Extremism Amongst Somali-Americans in Minneapolis-St. Paul. Final Report. National Consortium for the Study of Terrorism and Responses to Terrorism (START). Department of Homeland Security Science and Technology Center of Excellence. University of Maryland. 353

Compilation of References

Weine, S. (2017). Resilience and Countering Violent Extremism. In U. Kumar (Ed.), International Handbook of Psychosocial Resilience. Routledge. Weisburd, D., & Telep, C. W. (2011). The efficiency of place-based policing. Evidence Based Policing, 17, 247–262. Wernet, A. (2014). Hermeneutics and Objective Hermeneutics. Ed by Flick, U., The SAGE Handbook of Qualitative Data Analysis. Los Angeles: SAGE Publications, 235–246. doi:10.4135/9781446282243.n16 White, A. (Ed.). (2015). Moving stories: International review of how media cover migration. Ethical Journalism Network. Whitney, P. (2018). The Oxygen of Amplification: Better Living through Anonymity, Attention, and Scale. Data & Society. https://apo.org.au/node/172901 Williams, M. (2021). The Science of Hate. Faber&Faber. Williams, M. L., & Burnap, P. (2016). Cyberhate on social media in the aftermath of 44 Woolwich: A case study in computational criminology and big data. British Journal of Criminology, 56(2), 211–238. doi:10.1093/bjc/azv059 Williams, M. L., Burnap, P., Javed, A., Liu, H., & Ozalp, S. (2020). Hate in the machine: Anti-Black and anti-Muslim social media posts as predictors of offline racially and religiously aggravated crime. British Journal of Criminology, 60(1), 93–117. doi:10.1093/bjc/azz064 Wittgenstein, L. (2009). Investigaciones filosóficas. In I. Reguera (Ed.), Obra completa. Vol. 1: Tractatus logico-philosophicus. Investigaciones filosóficas. Sobre la certeza. Gredos. Witzenberger, K. (2018). The hyperdodge: How users resist algorithmic objects in everyday life. Media Theory, 2(2), 29-51. https://hal.archives-ouvertes.fr/hal-02047585 Wodak, R. (2015). The politics of fear: What right-wing populist discourses mean. The Politics of Fear, 1-256. Woodzicka, J. A., Mallett, R. K., Hendricks, S., & Pruitt, A. V. (2015). It’s just a (sexist) joke: Comparing reactions to sexist versus racist communications. Humor: International Journal of Humor Research, 28(2), 289–309. doi:10.1515/ humor-2015-0025 Wylie, C. (2020). Mindf*ck: Cambridge Analytica. la trama para desestabilizar el mundo. Roca Editorial. Xu, Y., & Wen, J. (2013). Detecting robbery and violent scenarios. In 2013 Second International Conference on Robot, Vision and Signal Processing, (pp. 25-30). IEEE. https://doi.org/10.1109/RVSP.2013.14 Xue, L., Constant, N., Roberts, A., Kale, M., Al-Rfou, R., Siddhant, A., Barua, A., & Raffel, C. (2020). mT5: A massively multilingual pre-trained text-to-text transformer. arXiv preprint arXiv:2010.11934. Yamaguchi, T. (2013). Xenophobia in Action. Radical History Review, 2013(117), 98–118. doi:10.1215/01636545-2210617 Yao, X. (2021). Face masks, materiality and exclusion in the COVID-19 semiotic landscape. Social Semiotics, 1–21. do i:10.1080/10350330.2021.2016032 Yar, P., & Nasir, J. (2016). GIS Based Spatial and Temporal Analysis of Crimes, a Case Study of Mardan City, Pakistan. International Journal of Geosciences, 7(03), 325–334. doi:10.4236/ijg.2016.73025 Yenigün. (2016, July 10). Konya’da Tehlikeli Gerginlik [Dangerous Tension in Konya]. Yenigün. https://www.gazeteyenigun.com.tr/konyada-tehlikeli-gerginlik/146928/ Youtube. (2017). Hate speech policy. Youtube Help Center Policies. https://support.google.com/youtube/answer/2801939

354

Compilation of References

Ypma, P., Drevon, C., Fulcher, C., Gascon, O., & Brown, K. (2021). Study to support the preparation of the European Commission’s initiative to extend the list of EU crimes in Article 83 of the Treaty on the Functioning of the EU to hate speech and hate crime. European commission. https://op.europa.eu/en/publication-detail/-/publication/f86 6de4e-57de-11ec-91ac-01aa75ed71a1/language-en Zabr iskym, Z. (2020). Big Lies & Rotten Her r ings 17 Kremlin Disinfor mation Techniques You Need to Know Now. Byline Times. https://bylinetimes.com/2020/03/04/big-lies-and-rotten-herri ngs-17-kremlin-disinformation-techniques-you-need-to-know-no w/ Zamri, N. A. K. (2021). Coronavirus Exacerbates Xenophobia: Deconstructing Otherness. In The Twitter. European Proceedings of Social and Behavioural Sciences., doi:10.15405/epsbs.2021.06.02.43 Zattar, M. (2017). Competência em informação e desinformação: Critérios de avaliação do conteúdo das fontes de informação. Liinc Em Revista, 13(2). Advance online publication. doi:10.18617/liinc.v13i2.4075 Zaya, G. M., Vázquez, Á. F. J., & Morgunova, E. (2015). Metodología para la enseñanza del análisis semiótico del texto poético. Didasc@ lia. Didáctica y Educación, 6(1), 167–190. Zhang, Z., Robinson, D., & Tepper, J. (2018). Hate Speech Detection Using a Convolution-LSTM Based Deep Neural Network. In Proceedings of ACM The Web conference (WWW’2018). ACM. https://doi.org/10.475/123_4 Zhou, P., Shi, W., Tian, J., Qi, Z., Li, B., Hao, H., & Xu, B. (2016). Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (vol. 2, pp. 207-212). ACL. 10.18653/v1/P16-2034 Zitara, N. (1971). L’Unità d’Italia: Nascita di una colonia. Jaca Book. Zitara, N. (1973). Il proletariato esterno: Mezzogiorno d’Italia e le sue classi. Jaca Book. Zuiderveen, B. F., Trilling, D., Möller, J., Bodó, B., De Vreese, C. H., & Helberger, N. (2016). Should we worry about filter bubbles? Internet Policy Review. Journal on Internet Regulation, 5(1). doi:10.14763/2016.1.401

355

356

About the Contributors

Elias Said-Hung is a tenured lecturer at the Faculty of Education, member of the SIMI research group and Director of the University Master’s Degree in Inclusive and Intercultural Education at the Universidad Internacional de La Rioja (UNIR), and President of the Science, Technology and Society Association (CITESOC). He received his Bachelor of Arts in Sociology from the Central University of Venezuela and his Ph.D. in Communication Sciences from the Complutense University of Madrid, Spain. He has published more than 100 academic publications (articles, chapters and books, among others). He has participated in more than 10 R & D & I projects financed in competitive calls for Administrations or public and private entities. Julio Montero Diaz is a Professor at the Universidad Internacional de La Rioja (since October 2014) and on leave of absence as a University Professor in Journalism (since 2007) at the Complutense University of Madrid. He is President of the Commission of degrees and Master of Social and Legal Sciences 1 of the VERIFICA Program of ANECA between 2013 and 2016. Since 2016, he is President of the same commission at Fundación Madri+d (bachelor’s, master’s, and doctoral degrees). He has worked in institutional evaluation programs since 2003 in the Valencian, Andalusian, and Galician Community agencies. He is also an evaluator of research projects in various state agencies (ANEP, corresponding ministries, and State Research Agency). *** Iakovi Alexiou is a Graduate Student in Department of Philology, Sector of Linguistics, NKUA. David Ramalho Alves holds a degree in Asian Studies and is a MSc student in International Studies, currently working in the Language department at Soft Skills Lab - Iscte - Instituto Universitário de Lisboa (LCT-Iscte). Dimosthenis Antypas is a PhD Student at Cardiff University with his main research revolving around NLP and social media. By utilising some NLP techniques he aims to develop tools and methodologies that can help us understand and explain human behaviour in social media, as well as identify common issues that are present in them such as misinformation. He also works part-time as a Research Assistant at Cardiff University and is a member of the Cardiff NLP team

 

About the Contributors

Óscar Javier Arango Arboleda is a composer, writer and editor. PhD in Contemporary Thought from the University of Barcelona, Master’s in Contemporary Thought from the University of Barcelona, undergraduate studies in Literary Studies from the Pontificia Universidad Javeriana and studies in Music from the National University of Colombia. Sergio Arce-Garcia is a lecturer and researcher at the Universidad Internacional de La Rioja (UNIR) in Spain. PhD in Humanities and Communication from the University of Burgos, with an extraordinary doctorate award. Degree in Chemistry from the University of Valladolid. He teaches at the UNIR’s School of Engineering and Technology, where he also researches on communication (study of media and social networks using machine learning techniques and network theory). Accredited by Aneca as a Associate Professor and Professor of Private University. Recognised for a six-year research period. César Arroyo López has a Degree in Humanities from the University of Castilla - La Mancha Doctoral student in Research in Humanities, Arts and Education from the University of Castilla - La Mancha. Arantxa Azqueta holds a PhD. in Educational Sciences from the Universdad de Navarra, with a thesis on education for qualified entrepreneurship with outstanding and mention cum laude. Currently, she is Professor in the Department of Theory and History of Education, Faculty of Education of the Universidad Internacional de La Rioja (UNIR). His research topics are preventing radicatisation, civic and intercultural education and entrepreneurship education. Fabienne Baider is a PhD in Linguistics and Gender studies from the University of Toronto. Full professor at the University of Cyprus where she is the director of the DISCONSO lab (Discourse, Context, Society). Her work focuses on the discursive strategies used in online conversations used to discriminate against minorities, with a specific attention to hate speech. She has published in top journals of her field including Journal of Pragmatics, Pragmatics and Society, Politics and Governance, Journal of Language Aggression and Conflict as well as with top publishers such as Cambridge University Press and John Benjamins. Editorially, she has coordinated a dozen special issues and books indexed in Scopus. She has led as PI or participated in more than 12 research projects funded in public calls advertised in Cyprus, the Middle East and Europe. She has led one of the four European union projects awarded in 2014 by the Social Justice and Consumer Rights Department; she is currently a partner in a Horizon 2020 project. She is also currently the PI of a national pro-ject She is a reviewer for the major journals in her field of research indexed in WoS and Scopus. Maximiliano Bron is a professor in Multimedia Communication at National University of La Rioja (Argentina). Teacher Assistant in the Graphic Production Workshop at National University of Córdoba (Argentina). Supervisor of the Editorial Project “Libro – E” (National University of La Rioja). Phd of Social Sciences (Rey Juan Carlos University), Master’s Degree in Educational Processes Mediated by Technologies (National University of Córdoba), Bachelor of Social Communication. Editorial Supervisor of the Journals In Iure and Oikonomos. Director of the Master’s Degree in Digital Journalism at National University of Córdoba. Director of the Degree in Social Communication at National University of La Rioja. He has participated in several research projects as a researcher and as an executive director. He works as a Digital Communication consultant in both public and private institutions.

357

About the Contributors

Jose Camacho-Collados is a Senior Lecturer and UKRI Future Leaders Fellow at Cardiff University, leading the Cardiff NLP group. Before joining Cardiff University, he completed his PhD in Sapienza University of Rome and was a Google AI PhD Fellow. Until recently, his research has focused on various semantics aspects in NLP with a distributional perspective. He wrote the “Embeddings in Natural Language Processing” book and is the current program co-chair of *SEM 2023. More related to the topic of the publication, in the last few years Jose has been working in social media analysis and applications, developing NLP tools specifically targeted to this domain. Inês Casquilho-Martins is a social worker and holds a PhD in Social Work. She is Assistant Professor at the Instituto Superior de Serviço Social de Lisboa (Universidade Lusíada) and Invited Assistant Professor at Soft Skills Lab (LCT-Iscte). She is also a researcher at Centro Lusíada de Investigação em Serviço Social (CLISSIS) and Centro de Investigação e Estudos em Sociologia (CIES-Iscte). Óscar De Gregorio Vicente is a professor and researcher at the Universidad Internacional de la Rioja (UNIR) since 2021 at the School of Engineering and Technology, and at the Vice Chancellor for Research since 2022. PhD in Mathematical Engineering, Statistics and Operations Research (2021). Adjunct Professor of the Department of Statistics and Operations Research at Universidad Complutense de Madrid (UCM) (2018-2022). Member of: Bayesian Methods research group; MATGEN math work group at CEMAT (Government of Spain); Institute of Interdisciplinary Mathematics (IMI); Spanish Association for the History of Statistics and Probability (AHEPE). His research focuses on Artificial Intelligence (machine and deep learning), Modelling, Simulation, and detection of math patterns in large volumes of data. Its leading publications in the WoS and Scopus indexes focus on different mathematics and engineering application areas. Since 2022, collaborated on a project to measure and analyse hate speech in Spanish digital media (project PID2020-114584GB-I00), funded by the Spanish Ministry of Science and Innovation (HATEMEDIA). Semra Demirdis is an assistant professor at Cankiri Karatekin University, Turkey. Her research interests are in the areas of new media and society. She has published her work in research publications including social media engagement during conflicts and crises, digital activism and new media. She received her PhD from the University of Sheffield and MA in Digital Global Cultures from SOAS, University of London and her BA in Journalism from Selcuk University. Beatriz Esteban Ramiro, PhD in Social Sciences, graduated in Social Work. Equality Specialist: Social Intervention from a gender perspective. Research Master in Applied Psychology. Research career focused on gender and feminism perspectives as well as social policy analysis. Professional career linked to social action in situations of social vulnerability in different public-private entities. Tomás Fernández-Villazala is a Captain of the Civil Guard and Head of Service of the Spanish National Office Against Hate Crimes (ONDOD) of the General Directorate of Coordination and Studies, in the Secretariat of State for Security (Ministry of the Interior). He holds a PhD in Law from the National University of Distance Education (UNED), a Degree in Law from the University of Valladolid (UVA), a Degree in Psychology from the National University of Distance Education (UNED), a Degree in Criminology from the Catholic University San Antonio of Murcia (UCAM) and he has been assigned to ONDOD since its formal creation in 2018. 358

About the Contributors

Jesús Gómez did a PhD in the Spanish National Research Council and he is actually working at the Spanish National Office Against Hate Crimes as a data scientist. He collaborates carrying out criminological studies and projects. He has more than 15 scientific articles and he has worked at the University. Marcell Lörincz is former vice-chair of European Network Against Racism (ENAR) and CEO of Subjective Values Foundation. Carlos J. Máñez is Head of Section of the Spanish National Office Against Hate Crimes (ONDOD) since 2021. Researcher and co-author of other publications related to hate crimes and hate speech. Stella Markantonatou is Research Director at the Institute for Language and Speech Processing/ ATHENA RC. Ángela Martín-Gutiérrez is an International Doctor in Education and Master’s Degree in “Management, Evaluation and Quality of Training Institutions” from the University of Seville (US). Lecturer at Faculty of Education, International University of La Rioja (UNIR) and University of Seville (US) in the Department of Theory and History of Education and Social Pedagogy and Counsellor in educational centres. Member of the Research Group “Socio-educational and Intercultural Inclusion, Society and Media” (SIMI). Member of the Editorial Committee of the Journal “Cuestiones Pedagógicas”. Her main research areas are focused on vocational training, entrepreneurship, teacher training (initial and continuing), inclusive and intercultural education, gender and educational collaboration. https://orcid. org/0000-0001-9847-245X.4aee8edd-941f-4abc-9b44-b04779d24bf6 Sandra Martínez Costa has an Advertising and Public Relations degree and Ph.D. Professor at the University of Coruña since 2004. She has taught about filmmaking, cinematography, and montage, among other subjects. Her research works focus on the use of product placement in audiovisual fiction, and also studies on advertising in new media, augmented reality, and video games. Alicia Méndez-Sanchís is a psychology student at the Autonomous University of Madrid; in her fourth year. She is collaborating in the department, and she did an internship at ONDOD. Marcello Messina holds a PhD from the University of Leeds. He is currently working as chief researcher in History and International Relations at Southern Federal University, and is also affiliated to the Federal University of Acre and the Federal University of Paraíba. He writes extensively about Amazonia, Sicily and the Italian South, and his academic interests include cultural studies, critical musicology, decolonial theory, critical race and whiteness studies and collective memory. Roberto Moreno López is a PhD in Humanities and Education. Author and coordinator of 6 books, 53 book chapters, and more than 15 articles published in refereed scientific journals. Professor-researcher at the University of Castilla-La Man-cha and Vice Dean of Research and Postgraduate Studies at the Faculty of Social Sciences. Member of the board of directors of the Ibero-American Society of Social Pedagogy (SIPS). Member of the Research Group on Education and Society of the University of CastillaLa Mancha (GIES) and the Research Group on Socio-educational and intercultural Inclusion, Society and Media (SIMI) of the International University of La Rioja. 359

About the Contributors

Teresa Nozal Cantarero has a Degree and Ph.D. in Journalism, Spain. Professor at the Faculty of Communication Sciences of the University of Coruña since 2003. She currently teaches Script and her research work focuses on the study and analysis of audiovisual content through television, new media and mobile devices, among others. Helena Belchior-Rocha has a PhD in Social Work, is an Assistant professor at ISCTE-University Institute of Lisbon in the Department of Political Science and Public Policies and deputy director of the Transversal Skills Laboratory. Integrated researcher at CIES, Centre for Research and Studies in Sociology, linked to national and international research projects, namely 2 from Marie Curie Actions. Author of papers and communications at national and international congresses, in the areas of social work theory and methodology, environment, sustainability, community Intervention, ethics, human rights, social policies and well-being, education and soft skills. Member of the Editorial Board of national/ international journals. C. Vladimir Rodríguez Caballero is an assistant professor and researcher in the Department of Statistics at Instituto Tecnológico Autónomo de México (ITAM) from 2018. He holds a PhD in Economics from the Center for Research in Econometric Analysis of Time Series (CREATES) at Aarhus University in Denmark working under the supervision of Niels Haldrup. He was postdoctoral researcher and assistant professor at Universidad Carlos III de Madrid from 2017 to 2018. His research line relies on econometric analysis of time series. Particulalry, he has worked on topics related with dynamic factor models, panel data, and forecasting. Ángel Rodríguez-Torres is a Ph.D. in Teaching and University Management. Teacher-Researcher at the Faculty of Physical Culture, Central University of Ecuador Registered Researcher at SENESCYT. Author of several publications in scientific journals. Reviewer of Scientific Journals. Max Römer-Pieretti, in addition to being a writer of opinion articles, is a professor and researcher in the area of communication and semiotics. His activity in research on social networks, digital, media and critical literacy, analysis of photographs related to political communication stands out. In addition, he is a political and corporate communication consultant. He has held important academic management positions, among which the direction of the School of Social Communication of the Andrés Bello Catholic University (1997-2007, Caracas-Venezuela) stands out. Today he is the Research Coordinator of the Faculty of Communication and Humanities of the Camilo José Cela University (Madrid-Spain). Antonio Sanjuán Pérez has a Degree and Ph.D. in Information Sciences from the Complutense University of Madrid. Master Business Administration. He worked as journalist and audiovisual producer for twenty years. Since 2003 he is a tenured professor of Audiovisual Communication at the University of Coruña. He currently teaches audiovisual analysis and audiovisual journalism. He has published many books and articles, with topics ranging from crisis communication to television communication and the press in new media and supports. Vivian Stamou is an Associate Researcher at ILSP, Athena RC and PhD student at the University of Athens.

360

About the Contributors

José Juan Videla Rodríguez has a Degree and Ph.D. in Information Sciences from the Complutense University of Madrid. He is currently a professor of Audiovisual Communication at the University of A Coruña (UDC). He worked for twenty years as a journalist in radio and press. He has directed news programs on Radio Galega for sixteen years, and has been chief editor at El Ideal Gallego for four. He has published research articles on radio and television journalism. Alessandra Vitullo is researcher at Sapienza University of Rome, where she teaches Sociology of Communication and Sociology of Migration. She is associated director of the Network for New Media Religion and Digital Culture Studies. She was research fellow in several universities and research centers, such as Texas A&M University, KU Leuven, Uppsala University, “Bicocca” University of Milan, and the Bruno Kessler Foundation.

361

362

Index

#VatanimdaMülteci 237, 239, 244-253

A Accuracy 34-35, 45, 69-71, 74, 76-77, 79-80, 85-86, 294 Algorithms 51-53, 55, 59, 61, 65, 68-72, 76-77, 79-81, 85, 99-100, 103, 105, 109, 112, 221, 223, 232 Artificial Intelligence (AI) 57, 65, 79-80, 105, 179, 183, 185 Automatic Detection 65, 256

84, 102, 107, 187, 199, 204, 208, 210, 212, 215, 217-218, 220-221, 223-227, 229-233, 240-242, 262, 281, 287-289 Discursive Violence 283, 289 Disinformation 50, 52-64, 114, 136-137, 139, 145, 151, 157-161, 188, 220-222, 235, 241, 246, 281, 285, 293, 298

E

Bias 2-3, 7, 13, 18-20, 27, 57, 101, 163, 165

Educommunication 262-263, 272, 274-275, 281 Emotions 50-52, 54-56, 104, 145, 158, 245 Experimental Research 186, 190 Extremist Violence 31, 106, 258, 281

C

F

Confusion Matrix 73-75, 80 Connotative 111, 113-114, 118, 120, 123-125, 128 Content Analysis 186, 188-191, 193, 196, 198, 232, 237, 247, 264 Counter-narratives 59, 210-211, 215-216, 297 COVID-19 63-64, 66, 79, 112, 114, 130, 134, 148, 160-163, 165-166, 168-172, 174-177, 179-183, 191, 194-196, 202, 211, 215, 221-222, 225-226 Critical Literacy 283, 285-286, 291-293, 297, 299300, 303

F1-Score 79-80 Fact-checking 136-143, 145, 151, 153, 156-158, 160 Fictional Content 186, 188, 190, 194-196 First Wave 162-163, 175 FN 80 FP 80 Free-To-Air Television 186-187, 190, 192

B

D Denotative 111, 113-114, 118, 120, 123, 128 Digital Crime 65 Digital Culture 216 Digital media 53, 67, 76, 111-114, 116, 136-137, 139, 145, 147, 151, 156, 220-222, 225, 227, 229, 232233, 241-242, 281 Digital Violence 65, 67 Disarm 57, 61, 63 Discrimination 3, 6, 8, 10, 13, 16, 28, 30, 33, 54, 68,  

G Global Competence 260, 262-264, 269-270, 275, 278-281 Global Competence Educommunication 260 Greece 16-17, 19, 32, 39, 46, 136-141, 144-147, 157158, 204, 242, 264

H Hate 1-3, 5-8, 10-22, 27-31, 33-34, 37, 46-48, 50, 53-56, 61-62, 64-65, 67-71, 73-78, 81-93, 95109, 111-114, 116-119, 122, 125-126, 128-135, 138-139, 141-142, 146, 157-158, 161, 186-187,

Index

189-190, 192-206, 208-216, 218-225, 227-233, 235, 237-240, 242, 244-248, 252-260, 262-263, 269, 273, 281, 283-305 Hate Crimes 1-18, 20-22, 27-29, 54-55, 81-84, 98-99, 102-104, 107, 109, 186-187, 193, 195, 197, 221, 223, 232, 255, 283, 288-290 Hate Speech 1-5, 7-8, 11, 13-22, 28-31, 33-34, 37, 4648, 50, 54, 67-69, 76-78, 81-88, 90-93, 95-107, 109, 111-114, 117, 119, 126, 128-135, 138-139, 141, 146, 157-158, 161, 186-190, 192-193, 195-206, 209-216, 219-225, 227-230, 232-233, 235, 237-248, 252-260, 262-263, 269, 273, 281, 283-301, 303-305 Health Care System 1 Hoaxes 136-139, 141-146, 148-149, 151, 153, 156-159 Hospitality 237, 249, 253

I Immigrants 20, 83, 85, 103-105, 113, 117, 129-130, 136-142, 145-148, 151-158, 184, 201, 205, 211, 220, 238, 242-244, 255, 261, 270-272, 283, 288291, 294-295, 297 Infection 162-166, 168-170, 175-176, 211 Influencing 50-51, 61, 297 Insults 67, 124, 138, 190-197, 217, 228, 251, 284, 288 Integration 4, 73, 253, 258, 261, 270-271, 274, 277, 281, 299 Intercultural Education 260, 262 Internet 10, 15-16, 21, 53, 61-66, 82, 84, 102, 112, 133134, 138, 159, 187-189, 196, 198, 203, 206, 208, 213-214, 216, 220, 222-226, 232-234, 240-241, 254-256, 258, 263, 267-268, 293, 303 Intertextuality 111, 113, 124-126 Intolerant Speech 84, 95, 98-101, 109, 287 Islam 68, 78, 103, 130, 203, 205, 207, 209-213, 215216, 218, 242-243, 246, 261, 270-271, 276, 279 Italy 17, 19, 32, 53, 136-141, 144, 151-153, 156-158, 162-167, 169, 171-172, 175-178, 180-184, 203206, 208-211, 214-215, 218, 242, 264

L Lexical Resource 32-33, 35

M Mass Media 63, 65, 69, 73-74, 188-189, 225, 238, 304 Media Education 263, 281 Misinformation 61, 69, 106, 112, 137, 139, 141, 145, 156, 159, 161, 220-223, 233, 285, 298

Misogynistic 111-114, 117, 230 Modern Greek 32-34, 38, 40-42, 45-48

N Narratives 57, 118, 157, 162-163, 169-171, 199, 203208, 210, 216, 220-221, 232, 271, 274, 283-285, 292-293, 296, 298-301 Natural Language Processing 37, 46, 48, 52, 70, 77, 80-81, 84, 106, 109, 201 NIS Directive 67, 77, 80 NLP 37, 39, 48, 52, 70, 77, 80, 84-85, 98 Non-Governmental Organizations (NGO) 2, 8, 55, 59, 206, 218 North vs. South 162

O Offensive Language 30, 32-33, 41-42, 45-48, 82, 191, 201, 217, 242, 255, 283-284, 286, 288-289, 293, 296, 303 Offensive Speech 15, 45, 81, 84, 99-101, 221, 287 Offensive Thematic Categories 32, 35 Online Media 15, 20-21, 220, 222, 227, 232, 241, 243, 253 Orientalism 204, 207, 215-216

P Perceptual Experiment 14, 18, 21, 27-28 PISA 166, 177, 260, 262, 264, 277, 279-281 Populism 106, 160-161, 215-217, 234 Precision 71-72, 79-80, 190 Predictive Policing 83, 102, 106-107, 109 Preventing Radicalization 260 Preventing Violent Extremist (PVE) 282 Prevention 1, 28, 67-68, 79, 102, 166, 227, 231, 233, 260-263, 265-270, 272-275, 278, 282 Protocol 10, 13, 102, 259 Public Opinion 50-51, 53, 61-62, 113, 137-138, 142, 145, 156, 213-214, 255, 260, 262

Q Quantitative Research 29, 204, 209

R Racism 2, 4-5, 12, 16, 18, 20, 29-31, 33, 76, 84, 103, 112, 117, 122, 130-131, 133, 138, 162, 164, 168, 183, 187, 198-200, 212-213, 215-216, 218, 221, 363

Index

223-224, 227-228, 232-235, 241-242, 245, 247, 252, 255, 257-259, 286, 288, 304 Radicalisation 2, 54, 260-275, 277-279, 281-282 Recall 71, 79-80

S Semiotics 111-114, 118-119, 128, 131, 133-134 Shitstorm 217 Social Events 81-87, 97-99, 101-102 Social Important Events 81, 109 Social Media 15-16, 19, 29-31, 34, 37, 46-48, 51-63, 66-67, 76-77, 82, 84, 99, 101-102, 104-108, 111114, 116, 118, 129-130, 132, 137-138, 142-143, 145-151, 153, 156-157, 159, 161-164, 168, 170, 173-174, 184, 198-201, 203-204, 208, 214-218, 221-222, 226-227, 230-232, 234, 237, 239-241, 243-244, 247, 253-259, 273, 279, 281, 285-286, 291 Social Network 52, 65, 84, 97, 99, 101, 109, 145, 187188, 229, 232, 241, 257 Social Networks 10, 15, 46-47, 50-51, 65-67, 73, 75-76, 81, 106, 114, 170, 180, 187-189, 196-198, 206, 253, 255, 257, 263, 269, 272, 284-285, 293-294, 299, 301 Southern Italy 162-164, 171, 175, 184 Southernification 162, 164, 176, 179 Spain 1-2, 10, 14, 16-17, 19, 30, 50, 52, 62, 65, 81-82, 84, 90-91, 95, 98-100, 102, 106-107, 111, 114-116, 119-120, 128-129, 135-142, 144, 157-158, 161, 186-187, 191-192, 222, 224, 234, 242-243, 254, 260, 264-265, 268-271, 274-275, 283, 288-292, 294, 297, 300-301 Spanish 11, 48, 53, 65-70, 73-77, 81-82, 84-88, 98, 107, 109, 115-116, 119-121, 126, 129, 133-135, 141-142, 144, 158, 167, 186, 190-192, 196-198, 200, 242, 258, 269-271, 283, 289, 291-292, 297 Spanish National Office Against Hate Crimes (ONDOD) 98, 109 Speech 1-5, 7-8, 11, 13-22, 28-34, 37, 39, 41, 45-48, 50, 54, 65, 67-69, 71, 76-88, 90-93, 95-107, 109, 111-114, 117, 119, 121, 124, 126, 128-135, 138139, 141, 146, 157-158, 161, 168, 170, 186-190,

364

192-206, 209-216, 219-225, 227-230, 232-233, 235, 237-248, 252-260, 262-263, 269, 273, 281, 283-301, 303-305 Split-Half Reliability 32, 38 Stereotypes 2, 15, 20, 28, 35, 51, 56, 112-113, 129-130, 138, 141, 163, 175, 189, 196, 200-201, 203-213, 217-218, 225-226, 242, 269, 273, 287, 303 Symbolic Structures 111 Syrian Refugee 237

T Television 31, 142, 163, 186-194, 196-197, 199-202, 207, 220-221, 226-227, 229-230, 258, 263 TN 80 TP 80 Turkey 67, 237-239, 242-244, 246-253, 255-259, 270 Twitter 16, 30, 34, 62-64, 67, 69, 73, 81, 84, 86-87, 98-99, 101-109, 111-112, 114-116, 120, 123, 128-131, 133-134, 136-139, 141, 143-145, 157-158, 161, 194, 197-198, 201-202, 208-209, 222, 230-231, 234-237, 239-241, 244, 246-248, 250-257, 259

V Victims 1-12, 17, 19, 21-22, 28-29, 54, 67, 82-83, 99, 109, 138, 147, 168, 170, 201, 210, 212, 233, 238, 258, 260, 274, 288-291, 296 Violent Actions 157, 196, 229, 283, 286, 291 Violent Extremism 102, 261-262, 269, 271, 273, 275, 278-282

X Xenophobic 111-114, 117, 126, 128, 138, 141, 184, 222, 224, 226-230, 288-290, 293

Y Youngsters 1, 6, 14