Social Media and Hate

Using expert interviews and focus groups, this book investigates the
theoretical and practical intersection of misinformation and social media
hate in contemporary societies.
Social Media and Hate argues that these phenomena, and the extreme
violence and discrimination they initiate against targeted groups, are
connected to the socio-political contexts, values and behaviours of users
of social media platforms such as Facebook, TikTok, ShareChat, Instagram
and WhatsApp. The argument moves from a theoretical discussion of the
practices and consequences of sectarian hatred, through a methodological
evaluation of quantitative and qualitative studies on this topic, to four
qualitative case studies of social media hate, and its effects on groups,
individuals and wider politics in India, Brazil, Myanmar and the UK.
The technical, ideological and networked similarities and connections
between social media hate against people of African and Asian descent,
indigenous communities, Muslims, Dalits, dissenters, feminists, LGBTQIA+
communities, Rohingya and immigrants across the four contexts is
highlighted, stressing the need for an equally systematic political response.
This is an insightful text for scholars and academics in the fields of
Cultural Studies, Community Psychology, Education, Journalism, Media
and Communication Studies, Political Science, Social Anthropology, Social
Psychology, and Sociology.

Shakuntala Banaji is Professor of Media, Culture and Social Change in

the Department of Media and Communications at the London School of
Economics and Political Science.

Ramnath Bhat is postdoctoral fellow at the International Centre for

Advanced Studies in New Delhi and visiting fellow in the Department
of Media and Communications at the London School of Economics and
Political Science.
Routledge Focus on Communication and Society
Series Editor: James Curran

Routledge Focus on Communication and Society offers both established

and early-career academics the flexibility to publish cutting-edge analysis
on topical issues, research on new media or in-depth case studies within the
broad field of media, communication and cultural studies. Its main ­concerns
are whether the media empower or fail to empower popular forces in
­society; media organisations and public policy; and the political and social
consequences of the media.

Bad News from Venezuela

Alan MacLeod

Reporting China on the Rise

Yuan Zeng

Alternative Right-Wing Media

Kristoffer Holt

Disinformation and Manipulation in Digital Media

Information Pathologies
Eileen Culloty and Jane Suiter

Social Media and Hate

Shakuntala Banaji and Ramnath Bhat

For more information about this series, please visit: www.routledge.com/

series/SE0130
Social Media and Hate

Shakuntala Banaji and Ramnath Bhat

First published 2022
by Routledge
4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
605 Third Avenue, New York, NY 10158
Routledge is an imprint of the Taylor & Francis Group, an informa
business
© 2022 Shakuntala Banaji and Ramnath Bhat
The right of Shakuntala Banaji and Ramnath Bhat to be identified
as authors of this work has been asserted in accordance with
sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted
or reproduced or utilised in any form or by any electronic,
mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any information
storage or retrieval system, without permission in writing from the
publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British
Library
Library of Congress Cataloging-in-Publication Data
Names: Banaji, Shakuntala, 1971– author. | Bhat, Ramnath, author.
Title: Social media and hate/Shakuntala Banaji and Ramnath Bhat.
Description: Milton Park, Abingdon, Oxon; New York, NY:
Routledge, 2022. | Includes bibliographical references and index.
Identifiers: LCCN 2021047804 (print) | LCCN 2021047805
(ebook)
Subjects: LCSH: Social media—Psychological aspects. | Social
media and society. | Online hate speech.
Classification: LCC HM742. B35 2022 (print) | LCC HM742 (ebook) |
DDC 302.23/1—dc23/eng/20211001
LC record available at https://lccn.loc.gov/2021047804
LC ebook record available at https://lccn.loc.gov/2021047805
ISBN: 978-0-367-53727-2 (hbk)
ISBN: 978-0-367-53726-5 (pbk)
ISBN: 978-1-003-08307-8 (ebk)
DOI: 10.4324/9781003083078
Typeset in Times New Roman
by Apex CoVantage, LLC
Kamala, my mother, was a proponent of Buddhism,
and a Vipassana teacher late in her life. It is she who
led by example and convinced me that it is possible to
believe in and work towards a more hopeful future.
She was sometime sceptical of my choices but always
believed in me. It is to her and her spirit that we
dedicate this book.

Ramnath Bhat
Contents

Acknowledgementsviii
List of tables and figuresix
Trigger warningx

1 Introduction 1

2 When hate-speech policies and procedures fail:

the case of the Rohingya in Myanmar 30

3 Brazil colonisation, violent ‘othering’ and

contemporary online hate 50

4 Social media, violence and hierarchies of hate in India 75

5 White male rage online: intersecting geneologies

of hate in the UK 96

6 Conclusion 119

Index126
Acknowledgements

This book has been made possible by the generous support of multiple
individuals and groups in Brazil, India, Myanmar and the UK, including
but not confined to the team of dedicated researchers who have undertaken
fieldwork with us since 2018. We can never thank our interviewees and
­participants enough for their courage and integrity, and the ways in which
they laid their experiences and wisdom open for scrutiny in order to ­enable
a way forward. We would also like to acknowledge the dedication of our
families and colleagues who supported us when the weight of the hate we
encountered became unbearably heavy, and, especially Prof. Robin Man-
sell, who read and c­ ommented on draft chapters. We are in your debt.
Tables and figures

1.1 Typology of Social Media Hate, Perpetrators and Recipients 21

3.1 Sample of hateful material received online by Djamila
Ribeiro. Credit: Djamila Ribeiro. 61
5.1 A selection of hateful content received on Twitter by
Dr Shola Mos-Shogbamimu. Image credit: Dr Shola
Mos-Shogbamimu.115
Trigger warning

Please be aware that this book discusses sensitive and traumatic subjects
from genocide to racism, misogyny and Islamophobia as well as the effects
these have had on individuals and communities; we detail these through
evidence, both explicit and recounted. Readers might find these triggering
for a number of reasons, particularly if you have also experienced discrimi-
nation, dehumanisation or online hate.
1 Introduction

On 23 February 2020, organised far right vigilante mobs targeted Muslims

in the northeast of India’s capital Delhi. Even as Delhi’s assembly elections
began on 8 February 2020, multiple BJP and other far right Hindutva leaders
held rallies and live-streamed videos inciting violence against Muslims and
Dalits. During the ensuing pogrom, vigilantes streamed and posted videos
of violence that were widely shared, celebrated and defended.1 Facebook
whistle-blower Sophie Zhang has shown that Facebook employees had red-
flagged several accounts as instances of Coordinated Inauthentic Behaviour
(CIB). However, predictably – and deliberately – Facebook failed to act.
When it did, much later, it was already too late. The mutilated and tortured
bodies of mainly Muslim victims lined the streets and were stuffed into the
drains of northeast Delhi. False narratives about the causes of the violence
circulated on WhatsApp and via other social media and were amplified by
mainstream media. Accounts that were overtly violating Facebook’s Terms
of Service by inciting violence were linked to prominent political leaders
from the ruling party – the BJP.2 Facebook as a company had decided where
its loyalties lay.
This episode helps illustrate two key approaches that we bring to our
work on social media and hate and to this book. First, we insist on the
need to locate discrimination, incitement and hate speech historically within
specific socio-political, economic and cultural contexts. For example, the
2020 Delhi pogrom cannot be fully apprehended without understanding the
preceding local resistance led primarily by Muslim women (epitomised by
the women of Shaheen Bagh3) against the discriminatory 2019 Citizenship
Amendment Act. The anti-Muslim violence was a culmination of the BJP’s
campaign to polarise Delhi by spreading propaganda and disinformation.
Approximately two months after the pogrom, Facebook invested nearly six
billion US dollars to acquire just under a 10% stake in Jio Platforms4 (a tech
subsidiary of Reliance Industries owned by Mukesh Ambani, PM Modi’s
close associate and the fifth richest man in the world). With contexts such
DOI: 10.4324/9781003083078-1
2 Introduction
as these informing our analysis of Brazil, India, Myanmar and the UK, we
seek to provide a sense of what Massey (2005) calls ‘power geometries’ –
relations of power that result in societies arranged in different kinds of
hierarchies. Studying hateful content production and ways of reducing it
without attention to power geometries is a self-defeating endeavour. The
ongoing trauma which underlies the sedimentation and rearrangement of
these geometries through discrimination and violence are key subjects in
our chapters.
In this introductory chapter, we present a concise, historicised and critical
review of research on social media and hate. The chapter also incorporates a
critical review of methods used to define, delimit and understand these phe-
nomena and possible ways to ameliorate them. We elaborate a theoretical
framework that broadens the scope of investigation into online and offline
far right activity, abuse, threat, discrimination, prejudice and dehumanisation
in peer-to-peer networks, apps, and platforms as well as on cross-platform
applications (described collectively as social media). We do so by discuss-
ing a) user practices, attitudes and experiences; b) technological and social
infrastructures within which social media operate; c) relationships between
social media and other forms of media and communication (face-to-face,
broadcast, print and so on) and d) a historicised account of socio-political
contexts that create the conditions where social media activity can legiti-
mise, contribute to, or be used to organise targeted, extreme and persistent
state discrimination, social discrimination and citizen-on-citizen violence.
When we refer to infrastructures, we mean technical and cultural systems
that create institutionalised structures, or a system of interlinked materi-
als and ideas, binding individuals and groups to specific forms of conduct
and subject positions (Bhat, 2020; Larkin, 2008, p. 6; Star, 1999, p. 330).
These infrastructures include market conditions, as well as the national
and international legal and regulatory regimes to which they are subject,
and the social milieu within which users produce, receive, share and act
on media content circulated with the aim of denigrating, discriminating
against, dehumanising, threatening and violating individuals and communi-
ties. We argue that individuals’ and groups’ experiences of themselves and
others are formed in part through historical processes and in part through
iterative engagement with communication infrastructures. Intersections
of identity (caste, race, gender and so on) strongly inflect the nature and
outcomes of these experiences. Thus, through an enhanced form of ‘lis-
tening’ which draws on Spivak’s (1988) theorisation of subaltern voice,
our theoretical framework studies the dialectical relationship between
communication infrastructure/technological affordances and the phenom-
enology of embodied, intersectional subjectivity in the context of hateful
communication. Such a phenomenology includes not only a sense of self
 Introduction 3
and other (Merleau-Ponty, 2012) but also a sense of the world produced
by the use of media (Banaji, 2017; Gray, 2020). For those at the receiving
end of discrimination, violence and hate, it includes trauma, loss and/or the
theorisation of their group experiences in ways that enable conscientised
resistance and practice (Freire, 2000). Throughout the book we argue that
analysis of social media use in the context of increasing incivility, bullying,
authoritarianism, political violence and polarisation benefits greatly from
the ­interrogation of the networked infrastructures – the ecosystems – within
which social media use occurs.
While we were researching and writing this book, several interviewees
questioned our decision to write about online hate speech when vigilan-
tism and physical violence might appear to be more urgent priorities. The
2020 Delhi pogrom, the anti-Rohingya violence and the coup in Myanmar,
the racist, homophobic and transphobic attacks in the UK and the violent
suppression of feminists, gay people, Indigenous and Afro-Brazilian com-
munities challenging Bolsonaro cropped up repeatedly during our research.
This suggests that, as Kishonna Gray has demonstrated with regard to digi-
tal games (2020) and Ruha Benjamin (2019) has demonstrated with regard
to coding, algorithms and AI, rather than constructing an artificial binary
between the offline and the online or the digital and the real, these domains
are inseparable, conceptually and materially.
Online and offline discrimination, harassment and violence are part of
the same constellation and act on each other on the local, national and inter-
national levels (subject, we argue, to different power geometries). Whether
one is critical of the disproportionate attention given to the ‘online’ or con-
vinced about the disruptive potential of the internet to change all aspects of
human life, a common tendency is to frame ‘the Internet’ as a fundamentally
ahistorical phenomenon that acts on and affects society, but not vice versa
(Morozov, 2013). Our approach does not deny the speed and specificity of
particular forms of online harassment, dehumanisation, incitement and hate
speech, but we follow Banaji and Buckingham (2013) in arguing that ‘the
online’ is itself shaped by and part of an individual and collective psychic
and politico-historical experience that is always also ‘offline’.
Whether this is termed a dialectic or not depends upon the amount of
agency (Banaji, 2017) one attributes to sociotechnical systems and infra-
structures such as the Internet. The fast-paced heartbeats and physical anxi-
ety that many people we spoke to evince on seeing rape threats aimed at
them or their children on TikTok, Messenger or Instagram, the depression
and anger they experience when their private and personal lives are targeted
by trolls or doxers because of their liberatory stances on issues of identity
or their concern for circulating factual evidence, cannot be disconnected
from the persistent discriminatory comments aimed at them, their children
4 Introduction
and their communities. Nor can hate online be dissociated from the street
harassment, stalking, physical intimidation, police brutality, legal injustice
and social exclusion that many social justice activists, women, LGBTQIA+,
disabled, Indigenous and minority ethnic or religious citizens, refugees and
asylum-seekers, recount as the quotidian backdrop to their social media use
(Awan, 2016; Elareshi, 2019; Felmlee et al., 2018). The intensity of these
experiences is often further enhanced by intersections of identities which
provide convenient targets for the politics of the far right that has swept
across the globe since 2014.
For these reasons, our research centres the experiences and views of social
media users whose communities are directly impacted by online hate. This
focus allows us to trace how harmful content including legally provable
hate speech, hate crimes, threatening content, inciting content and discrimi-
natory disinformation on social media, emerge via an interplay of social
and technological infrastructures and, equally important, via an ideological
nexus between social media and other forms of communication, including
face-to-face communication, urban and rural spatial and material practices,
and mainstream broadcast and print media.

Methodologies underpinning our study

While we draw on political economy traditions to explain corporate deci-
sions, the main focus of our book is on audiences, the distribution of infor-
mation and everyday media practices and cultures that are historically,
ideologically and contextually located (cf. Banaji, 2011; Parks & Starosiel-
ski, 2015). Between 2018 and 2021, we and our research collaborators Zico
Al-Ghabban, Marina Navarro Lins, Nihal Passanha and Letyar Tun inter-
viewed more than 100 individuals and conducted 20 focus group discus-
sions and 15 expert interviews in Brazil, India, Myanmar and the UK. This
book is based on in-depth analysis of these transcripts and a background
textual analysis of 3000 social media posts drawn from Facebook, What-
sApp, Instagram, Twitter and Tik-Tok. Some of this material was provided
by our interviewees. Other materials come from online research for this
book and for our WhatsApp Vigilantes research (2019).
Our research methods are qualitative and grounded in post-structuralist
and interpretivist epistemological traditions that pay attention to shifting
patterns in people’s expressions and understandings of their own and others’
identity, as well as to the gaps between individual and collective memories,
lived experiences and recorded events. Such approaches require considera-
ble self-reflexivity and attention to situated knowledge production (Gupta &
Ferguson, 1997; Haraway, 1988; Visveshwaran, 1996). We acknowledged
with our interviewees our own positionality, vulnerability and privilege.
Apart from the sensitive nature of the topic at hand, the second half of our
 Introduction 5
research was conducted amidst a global pandemic, where we and our inform-
ants were facing illness, bereavement and other collective or individual trau-
mas. Given these circumstances our interviews and analysis were founded
on empathy, allowing interviewees to speak at length by creating a shared
narrative space (Douglas, 1985; Zinn, 1979). Further, our interviews were
designed and carried out with a sensitivity to trauma-informed contexts in
order to ensure that our research did not reproduce trauma that had already
been experienced (Favaro et al., 1999; Dyregrov et al., 2000; Thomas et al.,
2019). In line with a desire to respect the autonomy of our participants as
experts and co-constructors of the research, we used pseudonyms or real
names where requested. They – and we – fully recognise the painful modali-
ties between these choices of partial erasure or increased visibility linked
to increased risk. As our chapters show, however, at no point did we wish
our voices to be heard above theirs. Thus, including extended excepts from
interviews and focus groups was a deliberate methodological choice.
Given the myriad possibilities of a phenomenological approach, we
examine how social media-related technological and social changes are
intertwined in each of our four case study countries – Myanmar, Brazil, UK
and India. In doing so, we do not claim the nation-state as the sole valid unit
of such analysis (Amelina et al., 2012). Arising from a post-war context,
this kind of methodological nationalism assumes the existence of universal
categories that can then be ‘tested’ in various countries. Our framework
treats every empirical investigation in a different society as constitutive of
the universal category – which by definition must remain an unfinished pro-
ject. In other words, as Kuan-Hsing Chen puts it,

to do area analysis is not simply to study the object of analysis through

a process of constant inter-referencing. . . [rather], relativizing the
understanding of the self as well as the object of the study is a precon-
dition for arriving at different understandings of the self, the Other and
world history.
(2010, p. 253)

Legal instruments and international principles

addressing hate speech
Although hateful propaganda, discriminatory disinformation and hate
speech existed long before the Internet, the amplification of particular forms
of hate on social media deserves special attention. As Alkiviadou argues:

Firstly, the sheer number of users of such networks on a global scale

results in the need to pay particular attention to this digital vehicle.
Secondly, social networks are used by individual users but also by
6 Introduction
organised and semi-organised groups to promote hateful rhetoric and
target the victims of such rhetoric. Thirdly, social networks come with
some kind of content regulation which must be assessed for purposes of
ascertaining whether or not and, if so, the extent to which this regula-
tion contributes to the effective tackling of online hate.
(2019, p. 20)

No universally accepted definition of hate speech exists in international law,

even though both hateful content and its consequences have been all too
clear, particularly during the 1930s and 40s in Europe, and with the spread
of social media and smart devices since 2000. The International Covenant
on Civil and Political Rights (ICCPR) adopted by the United Nations in
1966, specifically Article 20,5 states that ‘any advocacy of national, racial,
or religious hatred that constitutes incitement to discrimination, hostility
or violence shall be prohibited by law’. Amongst other legal instruments
and international principles related to hate speech, The International Con-
vention on the Elimination of all forms of Racial Discrimination (ICERD)
came into effect in 1969, limiting its definition of hate speech to speech
about race and ethnicity, and, in another crucial difference, expanding the
scope of liability in terms of disseminating hate speech, in contrast to the
ICCPR which limited liability to proof of intent to cause harm.
In the Genocide Convention of 19516 hate speech is limited to public
incitement of genocide, and the Convention on Elimination of all forms of
Discrimination Against Women (CEDAW) of 19817 focuses on discrimina-
tion and violence against women. In 1997, the Council of Europe’s Com-
mittee of Ministers in its Recommendation8 No. R (97) 20, defined hate
speech as

all forms of expression which spread, incite, promote or justify racial

hatred, xenophobia, anti-Semitism or other forms of hatred based on
intolerance, including: intolerance expressed by aggressive national-
ism and ethnocentrism, discrimination and hostility against minorities,
migrants and people of immigrant origin.

Amongst many definitions, this is the one that appears to encompass the
widest range of constituents and circumstances and is the one we return to
as we proceed.
Unfortunately, by and large, the law has not been able to address hate
speech against the most vulnerable. Indeed, ‘[t]he places where the law does
not go to redress harm have tended to be the places where women, children,
people of colour and poor people live’ (Matsuda, 1989, p. 2322). In Aus-
tralia, the United States and Canada, this inability of the law to bring legal
 Introduction 7
sanctions against hate speech and its attendant discriminatory or violent
material effects historically has stemmed from a commitment to preserving
an extreme and problematic version of freedom of speech. We argue that
this has resulted not only in physical harm but also in longstanding psychic
trauma to multiple communities across these countries.
The emphasis of the North American approach lies in assessing speech
in terms of the likelihood that ‘speech acts’ will result in clear danger to
life and property. However, in other parts of the world, speech is regulated
not just regarding the likelihood of harm but also in terms of whether the
intrinsic content is objectionable (Gagliardone et al., 2015; Pohjonen &
Udupa, 2017). In a review of regulatory and legal approaches to hate speech
in Latin America, Hernández (2011) argues that racist speech about Afro-
descendants is ubiquitous. Sickeningly, they are commonly likened to ani-
mals and, in particular, to monkeys. Hernandez argues that

[t]hese perspectives about Afro-descendants are so embedded in the

social fiber of Latin American societies, that Afrodescendants’ subor-
dinated status in society is viewed as natural and logical [while] . . .
the historical notion that “racism does not exist” in Latin America dis-
inclines those unaffected by hate speech to acknowledge the harms it
causes marginalized groups.
(ibid.: 820)

Since those ‘unaffected’ by racist hate in Latin America tend to be white or

white-passing and from non-Indigenous populations, the ties between strug-
gles for Indigenous rights, racial justice and against hate speech are linked.
Given the dominant (read: white) groups’ access to multiple channels
of representation via politics, media and religion – and their use of these
channels for the repeated derogatory positioning of Afro-descendants in
the social hierarchy, this racist hierarchy itself has become naturalised. The
resulting inequality is further legitimised to different degrees in different
Latin American nations by the adoption of essentialist assumptions (Hall,
1997) about the subordinate status of minoritized Indigenous and Afro-
descendant groups in education, workplaces and justice systems. Showing
existing regulatory approaches to be profoundly inadequate, Hernandez
argues for the need to bring about legislation that can specifically address
hate speech:

Because of its great symbolic power, a ban on hate speech can easily
become a symbol that is an end in and of itself rather than part and parcel
of an overarching policy against racism. It is thus centrally important to
enact hate speech legislation that focuses on its anti-discrimination role
8 Introduction
rather than viewing it as an antidefamation inspired law or simply as a
dignitary harm. Incorporating civil as well as criminal code provisions
would also enhance the anti-discrimination role of hate speech legisla-
tion. Restricting hate speech legislation to the criminal code context,
as is done in many jurisdictions, may limit its efficacy for a number
of reasons. Entrusting the enforcement of the criminal law to public
authority risks having the law undermined by the complacent inaction
of public officials who may harbor the same racial bias as the agents of
hate speech. This is a particular danger in Latin America, where police
officers are consistently found to discourage Afro-descendants from fil-
ing racial discrimination complaints, and are often the perpetrators of
discrimination and violence themselves.
(ibid., p. 829)

Here, we ask our readers to note the role and implication of the police in
racist violence and in suppressing redress for those who have experienced
it – a phenomenon which will be seen repeatedly in the chapters on Brazil,
India and the UK. Similarly, a recent report released by the International
Dalit Solidarity Network (Shanmugavelan, 2021) argues that both offline
and online caste-hate speech needs to be grounded in the historical contexts
of an Indian subcontinent shaped by caste hierarchies.
Shanmugavelan rightly notes that the dominant castes in India and the
diaspora within and outside institutions perform ‘castelessness’ that serves
to conceal the brutal history of caste oppression and provides fertile ground
for caste-pride and caste-hatred. Although caste-based discrimination and
the daily occurrence of caste-hate speech in everyday life in mainstream
media and social media has been acknowledged by various international
bodies, and there is some limited but largely symbolic support for anti-caste
activities evinced by social media companies, there is no clear set of legal
principles preventing caste-hate speech. Thus, Shanmugavelan argues that
‘it is essential that caste-hate speech is recognised as . . . a distinctive form
of hate speech – and that Dalits are included in actions to mitigate caste-hate
speech online and offline, at every level’ (ibid: 27).
In a 2015 report on hate speech and incitement to hatred against minori-
ties in the media, UN Special Rapporteur on Minority Issues, Rita Izsak,9
emphasised the need to distinguish between different types of expressions:
a) expressions that constitute an offence that can be prosecuted criminally;
b) expressions that may justify restriction and civil sanctions; and c) expres-
sions that raise concerns about tolerance, civility and respect for others. In
other words, Izsak views hate speech on a much wider spectrum than the
current narrowly defined legal category. Izsak goes on to argue, correctly in
our view, that non-legal and social responses to hate speech should be given
 Introduction 9
as much attention and discussion as legal responses. With this in mind, we
move to discussing scholarly efforts to define hateful, violent and discrimi-
natory speech.

Conceptual approaches to and empirical research on

online hate

‘Naming’ and its discontents

In an attempt to rescue free speech from the encroachment of ill-conceived
and misused hate speech legislation, Susan Benesch proposes the concept
of ‘dangerous speech’ arguing that ‘when an act of speech has a reasonable
chance of catalysing or amplifying violence by one group against another,
given the circumstances in which it was made or disseminated, it is danger-
ous speech’ (2013, p. 1). Benesch provides five variables to determine the
degree of dangerousness involved: (i) the speaker, who is much more likely
to commit successful incitement if he or she has some form of pre-existing
influence or authority over an audience; (ii) the audience, the more fearful it
is, the more vulnerable it is to incitement; (iii) the speech act itself, by way
of the use of certain rhetorical devices, such as the ‘accusation in a mirror’
strategy, persuading the audience that they are going to be attacked; (iv) the
social and historical context and (v) the mode of dissemination.
Benesch’s work brings scholarly attention to the social and historical con-
texts within which hateful and discriminatory communication takes place
as well as the distribution and infrastructural aspects which impact speech
acts. However, her focus on speech with a reasonable chance of catalys-
ing or amplifying violence excludes the amplification of discrimination and
structural inequality which is often the aim and result of what we have been
calling ‘hate’. When discussing racism in the US context, legal and critical
race scholar Patricia Williams (1987, p. 129) argues for a view of ‘ . . . rac-
ism as a crime, an offense so deeply painful and assaultive as to constitute
something I call “spirit murder” ’. Benesch’s re-labelling of ‘hate speech’ as
‘dangerous speech’ and her definition, while genuinely useful in identify-
ing speech acts geared towards lynching, pogroms and genocide, does not
account for the collective trauma and psychic harms of constant belittling,
maligning, insulting and exposure to humiliation and abuse in cases where
violence is not necessarily imminent. Nor does it address how deep-seated
prejudices attendant upon such dehumanisation influence discrimina-
tory practices in carceral systems, housing, land acquisition, employment,
schooling, higher education, the culture industries and so on.
Other authors who object to the ‘reified category’ or ‘thick concept’ of
hate speech are Pohjonen and Udupa (2017), who propose the concept of
10 Introduction
‘extreme speech’, arguing correctly that hate speech is not a binary (hate/
not hate) but lies along a spectrum. However, in their legitimate concern
to protect the free speech of actors working to critique widely accepted
social and religious practices, the authors appear to divert attention from
the plight of minoritised communities at the receiving end of discrimination
and violence whipped to a frenzy by hateful communication. Critical satire,
jokes and open critique aimed at oppressive practices and aimed to draw
attention to injustice, unfairness or imbalance and inequity (between and
within communities), are not on the spectrum of hate speech, even if they
are disingenuously mis-characterised as such by malicious politico-legal or
religious regimes.
Moving away from the concept of hate speech simply because hate
speech laws are deployed in bad faith against legitimate actors engaging
in critique or dissent has further implications for the recipients and targets
of hateful communications. In the context of structural discrimination and
psychic harms, scholarly work on disablist hate speech reminds us that fear
of impairment is projected onto the ‘other’. As Burch (2018) argues

[t]he use of ‘parasites’ as a means of identifying and marking out dis-

ability is supported by the relationship between welfare and employ-
ment, to which the first is presented as inferior to the latter. Making
this connection, one Reddit user argues that ‘you are a parasite on the
productive class’, thus confirming that the disabled figure is not only
unproductive, but burdensome to those who are productive.
(ibid., p. 401)

Benesch’s work serves as a useful framework in theorising another widely

held and common instance of hate speech known as ‘Islamophobia’ (Allen,
2010). There is scope to use the framework to interrogate the actors
involved, their influence, status and legitimacy in particular societies and
their motivations in specific social and historical contexts.
Quantitative textual analysis can yield interesting results about the viral-
ity, spread and ubiquity of Islamophobic tropes across specified online pop-
ulations. Aguilera-Carnerero and Azeez (2016) analyse more than 10,000
tweets with the hashtag #jihad to show how Islamophobia has spread glob-
ally via the misrepresentation of Muslims and Islam in the post-9/11 media-
scape. They report two important findings. First,

[W]hen used by Islamophobes the meaning of the word ‘jihad’ becomes

associated with ideas of ‘violence’ and ‘war’. From the data, we could
not even state that they are talking about a ‘holy war’ because that would
imply an ulterior religious motivation, but many of the tweets contain
 Introduction 11
information only about assorted felonies and misdemeanours. It is not a
unique phenomenon that a religious term transcends the religious lexi-
cal field and becomes part of the daily vocabulary of any language (for
example the terms ‘apocalypse’ or ‘purgatory’ from Christianity), but
most of them retain their original meaning or a part of it. In this sense,
the process the word ‘jihad’ undergoes is a different one as there has
been a lexical conversion that serves the speaker’s intention.
(Aguilera-Carnerero & Azeez, 2016, p. 30)

And second: ‘Far from countering any of the clichés previously attributed
to Muslims and Islam the corpus, on the contrary, reinforces and expands
existing negative stereotypes’ (Ibid: 30). Drawing on literature referring to
the core ideology of orientalism leading to Islamophobia at both the insti-
tutional and interpersonal levels (c.f Abu-Sway, 2005; Said, 1978), the
authors show that online Islamophobia is deeply connected to misrepre-
sentations and stereotyping of Muslims by western media since the early
2000s. For instance, in a study about representation of British Muslims in
nearly 1,000 UK newspapers from 2000–2008, most of the coverage was
found to focused on Muslims as threats in terms of terrorism, or differing
values, or both, in terms of Muslim extremism (Moore et al., 2008).
Using British legal definitions of hate speech as expressions of hatred
toward someone on account of that person’s colour, race, disability, nation-
ality (including citizenship), ethnic or national origin, religion, gender reas-
signment, or sexual orientation, a parallel study conducted on Islamophobia
using a qualitative analysis of 100 different social media pages, posts and
comments found nearly 500 instances of online hate speech directed against
Muslims (Awan, 2016). Word cloud frequency was deployed to examine
key words depicting Muslims in an overtly prejudicial way (such as ‘Paki’
or ‘Muzrats’). Awan posits a typology as a starting point for a framework
to analyse Islamophobia as expressed by users on Facebook. In line with
our findings, his typology includes ‘opportunists’ who post hate speech and
incite violence against Muslims immediately after incidents such as those
involving Daesh; ‘deceptive’ users creating fear by posting about false
events to intensify hate against Muslims; ‘fantasists’ who set up Facebook
pages to fantasise over Muslim deaths, often making direct threats to Mus-
lims; and finally, ‘systematic producers and distributors’ of Islamophobic
content.
In a similar vein, studying hate speech in Kenya, Busolo and Ngigi
(2018, p. 43)

interrogate the prevalence and development of hate speech over time,

investigate the perpetrators of hate speech and the targeted groups,
12 Introduction
critically analyse the consequences of hate speech, dissect the freedom
of speech vs. the protection from hate speech, highlight various chal-
lenges in curbing hate speech and reflect on strategies and methods of
curbing hate speech being used by various agencies.

Citing McGonagle, they argue that

[t]here are different types of hate speech perpetrators. There are offend-
ers by conviction . . . people with clear intention of engaging in hate
speech. On the other hand, incidentalists are people who may post
information without thinking about the consequences, but when legal
or social repercussions arise, they tend to be shocked.
(Ibid: 45)

Another study analyses comments on popular Slovenian news websites to

examine the unique factors that motivate different kinds of perpetrators of
hate speech (Erjavec & Kovačič, 2012). The authors argue that some users
are ‘soldiers’ who typically belong to political parties and systematically
vilify users identifying with the opposition, while yet other users serve as
‘watchdogs’ who use hate speech to draw attention to social problems. The
term ‘soldiers’ resonates with some of our own previous work.
We conducted research in four states of India to investigate the role of
WhatsApp in vigilante mob violence against minority communities in the
2018–19 period (Banaji & Bhat, 2019). Using focus group discussions and
in-depth interviews with nearly 300 users and analysing over 1000 What-
sApp messages, we found that privileged users – often Hindu and upper
caste, with sympathies for the ruling Bharatiya Janata Party (BJP) – justified
their sharing of hate speech against Christians and Muslims, Pakistanis
and others on the basis of nationalism, civic duty and the credibility of the
person who forwarded the misinformation. Importantly, these powerful
and privileged groups were found to be the ones most often involved in
systematic production and distribution of hateful disinformation and had
significant technical digital literacy skills as well as high levels of formal
education. Some members of vulnerable minorities, on the other hand, in
WhatsApp groups for professional or family reasons, became complicit in
spreading hate by forwarding the messages they received to appear compli-
ant, often without having the time or energy to engage fully with or even
read through the hundreds of banal prejudiced and derogatory memes,
GIFs, morphed images, quotes or false stats. The targets and ‘victims’ of
these messages (amongst these rural or poor urban women, Dalits, Adivasis,
Muslims, Christians and young political activists on the left, feminist or
LGBTQIA+ scenes) while heterogenous, and often lacking in social capital
 Introduction 13
were often far more alert to, critical of and attuned to discriminatory speech
and practices in everyday life, and consequently either less able or less will-
ing to pass on political misinformation received online. We return to some
of these testimonies in Chapter 4.
Maya Mirchandani (2018) in her paper on Hindu majoritarian online
hate speech targeted at Muslims in India builds on the notion of offence
and ‘hate spin’ as theorised by Cherian George (2016). She summarises his
argument:

The Islamic far right in countries such as Pakistan, Indonesia and the
Maldives, the Christian far right in the US and Western Europe, the Bud-
dhist far right in Myanmar, and the Hindu far right in India, are feeding
on people’s sentiments of being “offended” . . . . Cherian George makes
the case that political groups selectively mobilise genuine religious
devotion to manufacture both offense and a sense of being offended- or
offendedness.
(Mirchandani, 2018, p. 2)

She also summarises George’s perspective on the media ‘caught up’ in the
amplification of offendedness:

[T]he main objective of hate speech is met when the support base
is widened, a divisive narrative is created, and people are mobilised
around a political agenda. The media, meanwhile, are caught in report-
ing incidents when they happen, or else inadvertently serving as a vehi-
cle for politicians who use hate speech as a tool for identity politics.
In the process, the media often lose sight of the manufactured quality
of hate spin, especially where the line between hate speech and free
speech are blurred.
(ibid.: 3)

While George’s arguments corroborate findings in some of the contexts

studied here, they suggest far too meagre a view of the role of mainstream
media in contributing to discrimination and violence linked to hate speech.
Mainstream media outlets have political, ideological and economic links
with far-right groups which then influence the ways in which mainstream
media discourses retain an intertextual relationship with social media and
interpersonal (online and offline) discourses. For instance, mainstream
media discourses naturalise the validation of majoritarian anxieties – the
narrative of false victimhood – while hate speech on closed messaging
apps such as WhatsApp ‘picks up the baton’ by more explicitly target-
ing minoritised communities. Further, George seems to imply that it is
14 Introduction
offendedness or offence alone that is manufactured, but when amplified
by media becomes hate spin. More often than not, however, existing
hatred entwines with political subjectivities and affects about deserving
and undeserving citizens – such as feelings and discourses of disgust,
contempt, racism, misogyny, homophobia and so on – and masquerades
as offendedness which the mainstream media uncritically or complicitly,
amplifies.
Mirchandani’s own work (2018) bears out this more concerning thesis.
She posits a theoretical framework that works in two parts to explain the
emergence of majoritarian violence and hate speech in India. The first
draws on Appadurai’s (2006) Fear of Small Numbers, wherein Appadurai
outlines the notion of ‘predatory identities’, premised on the extinction
of proximate social categories that emerge especially out of pairs that
have often experienced long histories of contact, mixing and stereotyp-
ing of each other. The second part of Mirchandani’s framework deals
with when and what turns majoritarianism from affective and discrim-
inatory to actively violent. Building on scholarship that distinguishes
between radicalism (which could comprise wide-ranging hostilities to
the political status quo) and radicalisation (which includes various ‘push’
factors that turn individuals and groups towards violence for a cause and
against a group defined as ‘others’), coupled with Arendt’s (1963) notion
of the ‘banality of evil’, Mirchandani argues that these existing realities
of violent predatory identities need to be taken into account by various
actors working on counter-terrorism and prevention of violence. While
this is a fair and practical point, of course, there is also further room for
anxiety given that many of those involved in law enforcement, the justice
system and counter-terrorism often subscribe unreflexively to just such
predatory identities while attributing radicalisation only to the groups
defined as ‘other’.
Although Mirchandani’s framework is useful in thinking about hate
speech, there are at least two broader questions. First, if Hindutva in India
and the diaspora can be seen as an ideology based on emerging ‘preda-
tory identities’, how are we to reconcile this with pre-existing caste-based
divisions? From a critical caste perspective, the ‘majority’ dominant castes
(across religions) actually constitute less than 30% of the Indian popula-
tion, whereas Dalit, Bahujan and Adivasi groups constitute more than 65%
(Aloysius, 1997; Ambedkar, 1989). Second, are all individuals and groups
equally susceptible to radicalisation and in the same ways? If not, what
explains the differences? Both of these questions indicate the need for theo-
retical definitions of hate and radicalisation to be situated within specific
historical and political contexts.
 Introduction 15
A typological approach to hateful communication
There is scholarship on the potential targets of hate speech and other forms
of violent discrimination and dehumanisation. Surveying nearly 1000 young
adults in the US, Costello et al. (2017, p. 588) argue that online hate speech

differs from cyberstalking or cyberbullying in that hate materials

express hatred or degrading attitudes toward a collective instead of an
individual in isolation. Thus, hate materials express extreme attitudes
devaluing others because of their religion, race, national origin, sexual
orientation, gender, gender identity, ethnicity, or some other character-
istic that defines a group.

This is particularly important at a time when it has become commonplace

to dismiss online aggression as an ubiquitous feature of ‘online culture’ that
is said to affect all social media users equally if posts are made about con-
tentious topics; merely being visible online can invite invasions of privacy.
Our work suggests that attempts by oppressed and minoritised communi-
ties to critique and resist their oppression are frequently miscategorised in
nations/institutions where the powerful majority racial, ethnic or political
group has cultivated a sense of victimhood against a set of ‘others’. Subal-
tern resistance is censored or sanctioned as hate speech simply because it
is aimed at a collective (albeit an oppressive one with a history of political
aggression and discrimination). To complicate matters further, the majori-
tarian community who are now oppressors might once have belonged to
a community who were themselves subjected to injustice and oppression
and the minoritised community which is collectively subjected to violence
and/or discrimination may itself subject some of its own members (women,
LGBTQIA+, non-conformists) to extensive and historically embedded
forms of discrimination and violence.
Some of this complexity can be witnessed in North American and Euro-
pean contexts in the chilling effects of the use of charges of antisemitism
levelled at those critiquing the violence of the Israeli state and settlers. In
other cases, in India, for instance, oppressor communities appeal to laws
on free speech to support their right to malign others publicly while also
levelling charges of religious hatred and offense against (secular or Mus-
lim) comedians who draw attention to the lack of care for human life which
motivates Hindu or upper caste ‘cow-protection’ lynch mobs. Both of these
forms of malign censorship are rampant across the globe and remain pow-
erful hypocrisies cultivated by supporters of North American and British
conservatism/Republicanism and the alt-right.
16 Introduction
The complexity and importance of offline environments in the genera-
tion of political hate, propaganda, disinformation and misinformation that is
knowingly and systematically targeted at particular groups and individuals
online are typically ignored by those who assume the separateness of these
domains. Costello et al.’s premier assumption that ‘[t]he extent to which
individuals’ online activities bring them into virtual contact with motivated
offenders affects their likelihood of victimization’ (2017, p. 589) is partly
banal and partly misleading. Their assertion is that:

[u]sing a modified version of RAT [Routine Activity Theory], we find

robust evidence that online habits, such as utilizing numerous SNS
platforms and visiting hostile online environments, are related to being
targeting by hate online. Indeed, SNS usage has the strongest relation-
ship with being targeted by hate; avid users being nearly 6 times as
likely to be targeted.
(2017, p. 597)

This conclusion has problematic repercussions. While the intention

might be to support interventions that could reduce the extent to which
people are targeted with hate online, its refusal or inability to consider
the socio-political factors connecting recipients and producers of online
hate means that interventions will target the wrong factors and are more
likely to fail.
Bikhu Parekh’s (2012) work on conceptualising hate speech emphasises
three key characteristics of hate speech: Directed against easily identifi-
able individual(s) based on an arbitrary and normatively irrelevant feature;
stigmatising the target group by ascribing to it qualities widely regarded
as highly undesirable; and a target group is viewed as an undesirable pres-
ence and as a legitimate object of hostility. Building on this, Gelber and
McNamara (2016) interviewed 101 individuals from indigenous and minor-
ity ethnic communities in Australia. They note the distinction between the
constitutive and consequential harms of hate speech (Maitra & McGowan,
2012) to observe a spectrum of hate speech such as verbal and symbolic
epithets, exclusion, negative stereotyping, transmission of racism, threaten-
ing and harassing behaviour and so on. Based on reported examples recol-
lected from the interviewees, the authors find a wide range of constitutive
and consequential harms such as: Feelings of being hurt and upset, a result-
ing fear, fear leading to a sense of paralysis, disempowerment, withdrawal
from spaces which offer opportunities for redress, silencing and/or being
rendered mute, silence as an avoidance tactic, feeling dehumanised and vio-
lated, feeling anger and frustration and deciding to dis-identify from their
own identities as a protective mechanism.
 Introduction 17
Scholarship that draws on the voices and lived experiences of those who
experience hate in the context of discrimination and violence (Sethi, 2018)
rather than on regulatory approaches, policies, laws, the hate speech itself
or those who propagate or amplify hate speech is an important addition to
the research literature. Gelber & McNamara explain that their methodol-
ogy favoured a bottom-up understanding of the harms of hate speech which
allowed for a much more capacious understanding than merely focusing on
the threat of immediate violence:

reflections shared by the interviewees confirm that public racism in

Australia occurs in face-to-face encounters and general circulation in
targeted communities. These two types of hate speech were not experi-
enced as qualitatively different in terms of seriousness or harmfulness.
[Indeed. . .] public hate speech is frequently experienced as an attack
on worth and dignity. As discussed by critical race scholar Delgado
(1993), harms which are non-physical and do not fall under immediate
danger are . . . enduring and not ephemeral.
(2016, p. 336)

As we move through the cases in this book, we will develop further the
notion of hate speech as an attack on worth and dignity but also on the right
to have rights, and argue that different intensities and modalities of hate
have similar seriousness and harmfulness.

Analytical framework for this book: Theory and

typology

Theoretical positioning
As research in the fields of cultural studies, media and communications,
political economy, Science and Technology Studies and infrastructure stud-
ies has shown, technological developments do not only act upon society.
Technological developments are also imagined, engineered, enacted and
acted upon in various ways, shaping communicative innovations as well as
the ways in which they operate in specific circumstances (Castells, 1996;
MacKenzie & Wajcman, 1985; Mansell, 2012; Williams, 1974; Winner,
1980). After the International Telecommunication Union (ITU) framed
telecommunications as a major ‘engine for economic growth’ in its 1984
Maitland Report (Chakravartty, 2004), countries in the Global South lib-
eralised their telecommunications sectors and attracted large investments
towards wireless infrastructures. Alongside smartphones becoming cheaper
as a result of a push in Chinese manufacturing in the early 2000s, these
18 Introduction
countries invested in increased internet penetration, hoping to see growth
in their GDPs. This political economic context is crucial in understanding
how ‘the Internet’ has emerged as an agglomeration of material practices,
infrastructures and discourses.
Our theoretical framework aims to disrupt any simplistic binarism that
assumes a separation of online hate from offline history and politics. Both
scholarly and corporate-technological attempts to discuss online hate
include a disavowal and underplaying of the social, political and historical
contexts that explain the specific circumstances under which some groups
have systematically dominated other groups thereby supplying the grounds
on which hate can be manufactured, rationalised, legitimised and normal-
ised. Because of this disavowal of historical context (except, ironically,
with regard to the historical spread of digital tools and technologies), hate
speech and the individuals and groups involved (as perpetrators or recipi-
ents) of hateful communications appear to be treated as interchangeable
actors, equal in the eyes of theory-making, law-making and policymak-
ing. Our framework, on the contrary, treats online hate as an ecosystem
consisting of various political (corporate, government and party ideologi-
cal interests), technological (media, infrastructures, algorithms, AI) and
social (identities, inequalities, histories of oppression and struggle) aspects.
Rather than focusing on the speech itself, we rely on situating users’ experi-
ences in a historical and contextual setting to emphasise the systemic ways
in which elements are interrelated to comprise what we call an ecosystem
of online hate.
Infrastructures can be theorised as technical and cultural systems that
create institutionalised structures and bind people together towards specific
subject positions. Drawing on insights from a critical political economy of
media and communications, we argue that these technical and cultural sys-
tems are themselves subject to power flows, including ones induced by eco-
nomic relations between social media companies, the state and domestic/
international large corporate entities. Our case studies illustrate this dynamic
clearly, be it the economic dominance of the Myanmar military regime con-
trolling Internet infrastructure or Meta/Facebook’s consistent prioritisation
of profits at the cost of widespread racist violence and genocide.
A clear pattern emerges from an analysis of the literature on online
(and offline) hate speech, related concepts and issues such as violence and
­disinformation. Overall, efforts to address and define hate speech seem to be
haunted by a concern to preserve the notional concept of free speech, while
actually preserving some people’s freedom to express hate at the expense of
other’s freedom to live and thrive. Whether these outcomes arise from regu-
latory approaches or the policies and terms of service of social media com-
panies, whether it is explicitly mentioned or implicitly guiding research,
 Introduction 19
hate speech and free speech are often treated as abstract objects framed in
binary opposition to each other, thus making it the apparent duty of different
actors (scholarly, political, legal or corporate) to balance the two conceptu-
ally and in practice.
Barring a few exceptions, research approaches to hate speech seldom
analyse the identities of those who face hate. The intrinsic content of com-
munication that is objectionable or harmful seems to serve as a sufficient
basis for working to prevent hate speech or for disavowing the harm it has
done and is doing in favour of an abstract notion of free speech. Listen-
ing to the voices of those who directly face hateful communication and its
attendant discrimination and violence can help acknowledge the legitimacy
and theoretical weight that should be accorded to affect and lived experi-
ence in understanding the consequences of socio-political hate as a tool
of power. Listening as both a theoretical and methodological framework
can expand our collective understanding of what ‘harm’ means in the con-
text of hate online. Such an expansion to include the ways in which local
marginalised populations theorise their own experiences of being othered,
excluded and dehumanised, and are silenced or conscientised into action,
has political, legal and psycho-therapeutic implications. In addition to
the steps being taken at present, such an expanded understanding of hate
speech – which includes its embedding in histories of othering and the crea-
tion of difference in order to gain or maintain social position, economic
profit and political power – may eventually open up new ways of counter-
ing the discrimination and violence attendant upon and surrounding hateful
communication.
Our framework draws upon a variety of literatures that highlight peo-
ple’s experiences and affects around discrimination and hate faced on and
offline. While there are phenomenological distinctions to be made between
experiences and parasocial relationships in virtual environments (always
nested within the material world) and those solely in the material world,
interconnections between these spheres bear deeper examination. Social
media and the online world has been mythologised in popular discourse
as a disruptive technology that leaves everything in its wake irreversibly
changed. Multiple studies refer to a ‘digital age’, to lives ‘lived online’ and
to ‘online worlds’. In our theoretical framing, what we perceive as ‘the
Internet’ is governed by sets of protocols controlled by specific institutions
(such as those allocating domain names and those negotiating intermediary
rights). These protocols exert power, and influence the ways in which we
use digital technologies (Galloway, 2004). Social media service provid-
ers generally prioritise and act faster when troubling incidents take place
in close proximity to their parent corporations’ geographical or imagined
communities, ignoring or mischaracterising incidents of Internet-enabled
20 Introduction
discrimination across much of the Global South and against disenfran-
chised groups in the Global North. The ways in which technology acts
upon society and the ways in which technological developments them-
selves are socially shaped give rise to complex strategies and responses,
complicated by states’ and corporations’ political and economic motiva-
tions. Given the complexity of these dynamic processes acting upon each
other, infrastructure and (the phenomena of) online hate can be difficult
to stabilise as the foundational objects of our research. We have cho-
sen instead to locate our research in the phenomenological life of those
who experience online hate as part of who they are and what they do.
Listening attentively to interviewees’ accounts of hateful experience and
embodied subjectivity enables us to deconstruct the peculiar and spe-
cific ways in which history and online ecosystems intersects with hate.
Deconstructing the foundational objects of research opens up new ways
of investigating the relationship or dialectic between online and offline
phenomena. Our framework therefore allows us to propose an inclusive
definition of online hate and a typology that is attentive to both contexts
and intersectional readings of users’ identities.

Defining social media hate and typology of hateful content

Emerging from our reading of the literature and our analysis of data pre-
sented in Chapters 2–5, we opt for an inclusive definition of what has been
called ‘online or social media hate’ as

online content which demeans, dehumanises, stereotypes, perpetuates

or legitimises discrimination against, initiates or legitimises violence
against individuals or groups based on protected characteristics such
as: Social class, caste, race, religion, ethnicity, gender, sex, sexual ori-
entation, disability, neurodiversity, age, language, body size and politi-
cal orientation.

Based on this inclusive definition and on evidence from past and current
research,10 we propose a typology of social media hate that illustrates the
complexity and diversity of the problems that need to be addressed.
This typology foregrounds three interlinked insights: First, a sense of the
spectrum of hateful content linked to collective identities that circulates on
social media; second, a spectrum of potential actors linked to collective
identities who engage in and perpetrate online hate; and third, a spectrum of
actors linked to identities who are most likely to be the targets and intended
recipients of such hate.
 Introduction 21
Table 1.1 Typology of Social Media Hate, Perpetrators and Recipients

TYPES OF Racist content including but not confined to anti-Black, anti-

HATEFUL Asian, anti-Indigenous, antisemitic, Islamophobic, anti-Dalit
CONTENT and casteist denigration, disinformation, misinformation,
stereotypes, slurs (often disguised as jokes or questions), direct
personalised denigration (sometimes disguised as intellectual
engagement or false praise), abuse, threats and still or moving
images of killings and lynchings.
Sexist and misogynist content (often aimed at a subset of
women based on an intersection of sexual, caste, racial
or religious identity) including but not confined to sexist
disinformation, misinformation, jokes, rape jokes, rape
threats, pornography, objectification, slut-shaming, victim-
blaming, personalised denigration, body shaming, indirect
group denigration (sometimes disguised as mansplaining
or apparent ‘intellectual’ challenge), patriarchal religious
edicts, private images made public, morphed images and
deep fakes.
Xenophobic and anti-immigrant content including but not
confined to denigrating and even genocidal comments about
wars, losses in wars, the superiority of particular nations, races
and ethnicities over others, slurs, jokes, morphed still and
moving images containing disinformation and misinformation,
incitement to violence against refugees and asylum seekers,
victim-blaming and images of the dying or dead.
Homophobic, transphobic and biphobic content including
but not confined to denigrating or genocidal comments
about all members of these groups, denigrating stereotypes,
slurs, misinformation and disinformation, morphed images
displaying what are assumed to be degrading sexual positions,
body-shaming, sex-shaming, dead-naming, transmisogyny,
allegations of being predators, false association with
paedophilia, as well as direct and indirect threats of violence,
rape and death.
Classist content including but not confined to denigrating
comments, classist labels, slurs, associations of particular
religious, ethnic, racial or caste characteristics with particular
class backgrounds, open or disguised snobbery and denigration
of working-class tastes.
Anti-fatness and body shaming often occurring at the
intersection of another aspect of identity such as gender, race
or sexual orientation
Ageism occasionally targeted at older people/the elderly but
primarily aimed at “teenagers” and “young people” including
intellectual denigration, slurs, demeaning stereotypes,
generalisation from incidents of public disorder, and direct,
personalised abuse.
(Continued)
22 Introduction
Table 1.1 (Continued)

Ablism including derogatory slurs around particular mental

health conditions, misinformation about long term illnesses,
physical conditions or disabilities and learning difficulties,
suspicion of claims around neurodiversity and disability,
morphed images targeting particular groups (sometimes at
the intersection of another aspect of identity), threats, abuse,
denigration and genocidal comments.
Anti-democratic and anti-justice content aimed widely
at critics of conservative or illiberal politics and at
dissidents to the state and within particular movements,
including but not confined to abuse, death-threats, rape
threats, slurs, morphed images, deep-fakes, sexualised and
racialised threats and slurs, false accusations of corruption
or nepotism, allegations of being paid supporters, visual
association with vilified public figures, images with nooses
and other weapons to indicate death threats and celebration
of pain, incarceration or torture.
TYPES of Organised state-linked groups/actors (paid and unpaid) These
HATE producers and spreaders of disinformation and hateful content
PERPE­ are usually working for the government and/or ruling party
TRATOR/ where this is rightwing and/or far right in ideology, with a
ACTOR bouquet of socially and economically authoritarian goals.
identities They operate both online and offline, with protection from the
state. In countries with weak liberal or leftist governments,
these online actors sometimes work for the main rightwing
opposition party to oppose the government.
Organised non-state groups/actors (paid and unpaid) These
groups and actors are usually working on behalf of or think
they are working on behalf of the government and/or ruling
party or on behalf of a racial or religious supremacist ideology,
with a bouquet of socially and economically authoritarian
goals. They operate both online and offline, with considerable
power and legitimacy. In countries with weak liberal or leftist
governments, these online actors sometimes work for the main
rightwing opposition party.
Unorganised non-state actors united by prejudices or by
presumed caste/religious/ethnical or racial identity, usually
digitally literate individuals acting independently (occasionally
left and/or liberal politically but usually with an affinity for
conservative, rightwing or far right ideas and systems)
Opportunist grifters who troll/spread misinformation to
increase their fame, following and/or finances – these are
usually high-profile people, or those who once held left/
liberal values and are now publicly performing their rightwing
allegiance for economic or political gain
 Introduction 23

Disruptive libertarians who regularly troll, flame or dox

with a mission to disrupt, destabilise, unsettle political
consensus on specific issues (such as masking, vaccines,
medical advice) or mock particular individuals
(who support causes that they oppose) but who want
to remain under the radar
Digital stalkers who intentionally target individual social
media handles for ideological or personal reasons (desire
to patronise, wish to get a reaction), or through a sense
of spurned affection/loyalty which may spill over into
violence
Intermittent trolls, malicious users, inexperienced users, those
piling-on through peer-pressure or out of fear of being targeted
themselves
TYPES OF Women (of specific races, castes, classes, sexualities and religions)
RECIPIENT in the public sphere. Within this category Black women with
/Context/ leftist/pro-democracy views, and Muslim women are most
Identities visibly subjected to violent forms of misogyny and hate online
Women (of specific races, castes, classes, sexualities and
religions) in the quasi-private sphere
Men (of specific races, castes, classes, sexualities and religions)
in the public sphere. Within this category trans men, Black
men with leftist/pro-democracy views, and Muslim, Dalit and
Adivasi men are most visibly subjected to violent forms of
racism, casteism and hate online
Men (of specific races, castes, classes, sexualities and religions)
in the quasi-private sphere
Trans individuals and LGBTQIA+ groups (male, female, non-
binary, agender, gender-fluid) and particularly those who are
visible at the intersection of another protected characteristic, with
the highest violence aimed at Black and Brown transwomen
Women (of all races, castes, classes, sexualities and religions) in
the public sphere
Women (of all races, castes, classes, sexualities and religions) in
the quasi-private sphere
Minoritised groups by virtue of religious, ethnic, racial, sexual,
class, caste, disability, body image – and in particular those
groups who are publicly vocal about rights
Refugees, asylum seekers and economic migrants
Young people, particularly those who are vocal on justice issues
Traveller communities, Roma
Working class activists, trade unionists and economic rights
groups
Investigative journalists, scholars, rights activists and groups
in the spheres of disability, human and civil rights, feminism,
the environment and anti-capitalism.

(Continued )
24 Introduction
Table 1.1 (Continued)
Pro-democracy fact-checkers, lawyers and judges
Medical professionals/experts who comment/campaign on
health policy
Political dissidents, community organisers and activists, with
leftist/social justice views
Celebrities, entertainers, actors, artists, writers and other high
profile professionals who express support for decolonial
causes (such as Palestine, Kashmir)
Individual men from racial or religious majority
communities who are targeted for their ally-work/
outspokenness against prejudice/critique of rightwing/
authoritarian practices and/or have been accused of public
wrong-doing.

Conclusion
This chapter has traced a pathway through a series of literatures on legal and
scholarly definitions of hate or hate speech, on methodological studies and
typologies of hate and its recipients to lay out our theoretical framework and
an original typology of social media hate. In Chapter 2 we explore the ways
in which existing ecosystems of hate online have been tackled by corporate,
national and international laws and policies. Asking who is culpable, we
explain how and when the levels of hate reach genocidal levels as they did
in the case of the Rohingya in Myanmar. The failure to act against hateful
content is linked to a refusal on the part of many in power to acknowledge
and act on the acute danger faced by target communities tied to the value (or
lack thereof) that those in power accord to the lives being ruined and lost in
comparison to diplomatic and profit-related goals. A nexus of soft and hard
prejudice in government chambers, courts, media houses and boardrooms
is shown to encourage discrimination, hate and abuse to flourish in the
streets and online, and to do its violent work unimpeded. Chapters 3, 4 and
5 analyse and illuminate the links between contemporary political power,
ideology, media infrastructures, social media hate, recipient identities and
historical trajectories of discrimination, violence and dispossession in Bra-
zil, India and the UK respectively, while Chapter 6 sums up our conclusions.
In this book we argue that in every country there is an urgent need to get a
sense of which groups have been oppressed by which other groups through
which means and serving whose interests. Alongside this it is essential to
comprehend how hegemony has been constructed in discursive and material
terms. This investigation requires an intersectional and historicised consid-
eration of cultural and social practices relating to systemic and sociocultural
 Introduction 25
discrimination, hatred and violence. For instance, the caste system as a
graded system of inequality, using endogamy juxtaposed on exogamy and
as a basis for the division of labour and labourers (Ambedkar, 1968) neces-
sarily intersects with gender and class relations across South Asia (Banaji,
2017). A critical anti-caste perspective, like one nested in Black liberation
and critical race theory, could go a long way in explaining the power geom-
etries of hate speech in India and casteist speech abroad. This explanation
works not only in terms of identifying and countering explicitly casteist
hate speech as Shanmugavelan (2021) argues, but also to explain the his-
torical reasons why chauvinist Hindus have othered Indian Muslims and
Christians in insidious ways to divert attention from caste-based violence
and discrimination. Similarly, unpacking the complicated history and rela-
tionship between colonialism, slavery, race, class and skin colour, espe-
cially in regard to contemporary anti-Blackness (cf. Telles, 2006) is crucial
to understand the matrix of power-relations which generates much contem-
porary hate speech, online harassment and violence in Brazil and the UK.
Our theoretical framework emphasising the centrality of history, political
economy, infrastructure, listening, intersectionality and conscientisation
also leads us to attend to online hate speech as systematic, intentional and
predictable – an integral facet of the contemporary creation of difference
which strengthens authoritarianism.

Notes
1 Sagar. (2021, March 1). Delhi violence unmasked. The Caravan. https://cara
vanmagazine.in/politics/part-one-how-rss-bjp-members-invoked-hindu-iden-
tity-to-mobilise-hindutva-mobs-at-maujpur
2 Julia Carrie Wong & Hannah Ellis-Petersen. (2021, April 15). Facebook
planned to remove fake accounts in India – until it realized a BJP politician
was involved. The Guardian. www.theguardian.com/technology/2021/apr/15/
facebook-india-bjp-fake-accounts
3 Arvind Chhabra. (2020, January 4). Shaheen Bagh: The women occupying
Delhi street against citizenship law. BBC. www.bbc.co.uk/news/world-asia-
india-50902909
4 David Fischer and Ajit Mohan. (2020, April 21). Facebook invests $5.7 Bil-
lion in India’s Jio platforms. Facebook. https://about.fb.com/news/2020/04/
facebook-invests-in-jio/
5 See www.ohchr.org/en/professionalinterest/pages/ccpr.aspx
6 See www.ohchr.org/en/professionalinterest/pages/crimeofgenocide.aspx
7 See www.ohchr.org/EN/ProfessionalInterest/Pages/CEDAW.aspx
8 See https://search.coe.int/cm/Pages/result_details.aspx?ObjectID=0900001680
505d5b
9 See https://documents-dds-ny.un.org/doc/UNDOC/GEN/G15/000/32/PDF/
G1500032.pdf?OpenElement
10 Particularly our report WhatsApp Vigilantes, and the typologies proposed by
Gelber and McNamara, and Awan.
26 Introduction
References
Abu-Sway, M. (2005). Islamophobia: Meaning, manifestations, causes. Palestine-
Israel Journal of Politics, Economics and Culture, 12(2), 1–15.
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.
Alkiviadou, N. (2019). Hate speech on social media networks: Towards a regulatory
framework? Information and Communications Technology Law, 28(1), 19–35.
Allen, C. (2010). Islamophobia. Ashgate.
Aloysius, G. (1997). Nationalism without a nation in India. Oxford University Press.
Ambedkar, B. R. (1968). Annihilation of caste with a reply to Mahatma Gandhi;
and castes in India: Their mechanism, genesis, and development. Bheem Patrika
Publications.
Ambedkar, B. R. (1989). From millions to fractions. In Dr. Babasaheb Ambedkar:
Writings and speeches (Vol. 5, pp. 229–246). Education Department, Government
of Maharashtra.
Amelina, A., Nergiz, D., Faist, T., & Schiller, N. G. (2012). Beyond methodological
nationalism: Research methodologies for cross-border studies. Routledge.
Appadurai, A. (2006). Fear of small numbers: An essay on the geography of anger.
Duke University Press.
Arendt, H. (1963). Eichmann in Jerusalem: A report on the banality of evil. Faber
and Faber.
Awan, I. (2016). Islamophobia on social media: A qualitative analysis of the
Facebook’s walls of hate. International Journal of Cyber Criminology, 10(1),
1–21.
Banaji, S. (Ed.). (2011). South Asian media cultures: Audiences, representations,
contexts. Anthem.
Banaji, S. and Buckingham, D. (2013). The Civic Web: Young people, the internet
and civic participation. Massachusetts, MA: MIT Press.
Banaji, S. (2017). Children and media in India: Narratives of class, agency and
social change. Routledge.
Banaji, S., & Bhat, R. (2019). WhatsApp vigilantes: An exploration of citizen recep-
tion and circulation of WhatsApp misinformation linked to mob violence in India.
The London School of Economics and Political Science.
Benesch, S. (2013). Dangerous speech: A proposal to prevent group violence.
United States Holocaust Memorial Museum.
Benjamin, R. (2019). Race after technology. Polity.
Bhat, R. (2020). The politics of internet infrastructure: Communication policy, gov-
ernmentality and subjectivation in Chhattisgarh, India [PhD thesis, The London
School of Economics and Political Science].
Burch, L. (2018). “You are a parasite on the productive classes”: Online disablist
hate speech in austere times. Disability and Society, 33(3), 392–415.
Busolo, D. N., & Ngigi, S. (2018). Understanding hate speech in Kenya. New Media
and Mass Communication, 70, 43–49.
 Introduction 27
Castells, M. (1996). The rise of the network society: The information age: Economy,
society and culture. Blackwell Publishers.
Chakravartty, P. (2004). Telecom, national development and the Indian state: A post-
colonial critique. Media, Culture & Society, 26(2), 227–249.
Chen, K. H. (2010). Asia as method: Toward deimperialization. Duke University
Press.
Cherian, G. (2016). Hate spin: The manufacture of religious offense and its threat
to democracy. MIT Press.
Costello, M., Hawdon, J., & Ratliff, T. N. (2017). Confronting online extremism:
The effect of self-help, collective efficacy, and guardianship on being a target for
hate speech. Social Science Computer Review, 35(5), 587–605.
Delgado, R. (1993). If he hollers let him go: Regulating racist speech on cam-
pus. In M. J. Matsuda, C. Lawrence, R. Delgado, & K. Crenshaw (Eds.), Words
that wound: Critical race theory: Assaultive speech, and the first amendment
(pp. 53–88). Westview Press.
Douglas, J. (1985). Creative interviewing. Sage.
Dyregrov, K., Dyregrov, A., & Raundalen, M. (2000). Refugee families’ experience
of research participation. Journal of Traumatic Stress, 13(3), 413–426.
Elareshi, M. (2019). University students’ awareness of social media use and
hate speech in Jordan. International Journal of Cyber Criminology, 13(2),
548–563.
Erjavec, K., & Kovačič, M. P. (2012). “You don’t understand, this is a new war!”
Analysis of hate speech in news web sites’ comments. Mass Communication and
Society, 15(6), 899–920.
Favaro, A., Maiorani, M., Colombo, G., & Santonostaso, P. (1999). Traumatic expe-
riences, posttraumatic stress disorder, and dissociative symptoms in a group of
refugees from former Yugoslavia. The Journal of Nervous and Mental Disease,
187(5), 306–308.
Felmlee, D., Rodis, P. I., & Francisco, S. C. (2018). What a B!tch!: Cyber aggression
toward women of color. In Gender and the media: Women’s places (pp. 105–123).
Walden University.
Freire, P. (2000[1980]). Pedagogy of the oppressed. Bloomsbury.
Gagliardone, I., Gal, D., Alves, T., & Martinez, G. (2015). Countering online hate
speech. UNESCO.
Galloway, A. R. (2004). Protocol: How control exists after decentralization. MIT
Press.
Gelber, K., & McNamara, L. (2016). Evidencing the harms of hate speech. Social
Identities, 22(3), 324–341.
Gray, K. (2020). Intersectional tech: Black users in digital gaming. LSU.
Gupta, A., & Ferguson, J. (1997). Anthropological locations: Boundaries and
grounds of a field science. University of California Press.
Haraway, D. (1988). Situated knowledges: The science question in feminism and the
privilege of partial perspective. Feminist Studies, 14(3), 575.
Hernández, T. K. (2011). Hate speech and the language of racism in Latin Amer-
ica: A lens for reconsidering global hate speech restrictions and legislation
28 Introduction
models. University of Pennsylvania Journal of International Economic Law,
32(3), 805–842.
Larkin, B. (2008). Signal and noise: Media, infrastructure, and urban culture in
Nigeria. Duke University Press.
MacKenzie, D., & Wajcman, J. (1985). The social shaping of technology. Open
University Press.
Maitra, I., & McGowan, M. (2012). Introduction and overview. In I. Maitra & M.
McGowan (Eds.), Speech and harm: Controversies over free speech (pp. 1–23).
Oxford University Press.
Mansell, R. (2012). Imagining the Internet: Communication, innovation and gov-
ernance. Oxford University Press.
Massey, D. (2005). For space. Sage.
Matsuda, M. J. (1989). Public response to racist speech: Considering the victim’s
story. Michigan Law Review, 87(8), 2320–2381.
Merleau-Ponty, M. (2012[1945]). The phenomenology of perception. Routledge.
Mirchandani, M. (2018). Digital hatred, real violence: Majoritarian radicalisation
and social media in India. ORF Occasional Paper.
Moore, K., Mason, P., & Lewis, J. (2008). Images of Islam in the UK: The repre-
sentation of British Muslims in the national print news media 2000–2008. Cardiff
University.
Morozov, E. (2013). To save everything, click here: Technology, solutionism, and the
urge to fix problems that don’t exist. Allen Lane.
Parekh, B. (2012). Is there a case for banning hate speech? In M. Herz & P. Mol-
nar (Eds.), The content and context of hate speech: Rethinking regulation and
responses (pp. 37–56). Cambridge University Press.
Parks, L., & Starosielski, N. (2015). Signal traffic: Critical studies of media infra-
structures. University of Illinois Press.
Pohjonen, M., & Udupa, S. (2017). Extreme speech online: An anthropological
critique of hate speech debates. International Journal of Communication, 11,
1173–1191.
Said, E. (1978). Orientalism. Penguin.
Sethi, A. (Ed.). (2018). American hate: Survivors speak out. The New Press.
Shanmugavelan, M. (2021). Caste-hate speech: Addressing hate-speech based on
work and descent. DSN.
Spivak, G. (1988). Can the subaltern speak? In C. Nelson & L. Grossberg (Eds.),
Marxism and the interpretation of culture (pp. 271–313). Macmillan.
Starr, S. (1999). The Ethnography of Infrastructure. American Behavioral Scientist,
43(3), 377–391.
Telles, E. (2006). Race in another America: The significance of skin color in Brazil.
Princeton University Press.
Thomas, M. S., Crosby, S., & Vanderhaar, J. (2019). Trauma-informed practices in
schools across two decades: An interdisciplinary review of research: Review of
Research in Education, 43(1), 422–452.
Visveshwaran, K. (1996). Fictions of feminist ethnography. Oxford University
Press.
 Introduction 29
Williams, P. (1987). Spirit-murdering the messenger: The discourse of fingerpoint-
ing as the law’s response to racism. University of Miami Law Review, 42(1),
127–157.
Williams, R. (1974). Television: Technology and cultural form. Routledge.
Winner, L. (1980). Do artifacts have politics? Daedalus, 109(1), 121–136.
Zinn, M. B. (1979). Field research in minority communities: Ethical, meth-
odological and political observations by an insider. Social Problems, 27(2),
209–219.
2 When hate-speech policies
and procedures fail
The case of the Rohingya in
Myanmar

Introduction
Since 2005, social media companies have grown rapidly in terms of num-
bers of users and in terms of economic value. Pursuing markets, corporate
owners have paid insufficient attention to preventing discrimination and
violence targeted at vulnerable groups in society. In the introductory chap-
ter we argued that the debate around hate speech online and incitement to
violence has been framed around the preservation of an abstract principle of
free speech and a failure to acknowledge the different forms of dehumani-
sation, harassment, discrimination and violence that groups and individuals
experience daily. In this chapter, we critically evaluate some of the safe-
guarding policies agreed by social media companies – either as guidelines,
approaches or terms of service for users. Most of these documents list a
range of problematic content. Our analysis is restricted to guidelines avail-
able in the public domain. In addition, social media companies have internal
documents for content moderators, which are not accessible to the public.1
Overall, social media companies have adopted approaches to discrimina-
tory and toxic speech which combine big data, artificial intelligence and
algorithms. Most social media companies operate across a fine balance
between contradictory objectives: On the one hand, restricting the ease with
which information can be circulated (especially bulk circulation) amongst
users and, on the other hand, promoting both the number of overall users
(often linked to valuation of the company) and the ease of information cir-
culation (often linked to greater Average Revenue Per User or ARPU). This
dynamic is a result of the political economy of social media wherein sur-
plus value is generated through complex mechanisms such as commodifica-
tion of audiences, monopolisation and concentration, advertising and so on
(c.f. Arvidsson & Colleoni, 2012; Dahlberg, 2015; Fuchs, 2010; Rigi &
Prey, 2015). The primary mode of achieving this balance is through inno-
vation on affordances (Gibson, 1982, p. 403). This approach is based on
DOI: 10.4324/9781003083078-2
 When hate-speech policies fail 31
‘investigating the empirical question of embodied human practices in real
time situated interaction involving technologies’ (Faraj & Azad, 2012;
Hutchby, 2003, p. 582).
An advantage of affordance approaches is the attention they pay to the
ways in which the materiality of communication technologies and infrastruc-
tures intersects discrimination and violence. Affordance-based approaches
also maintain a relational perspective on the use of technologies, thereby
bringing attention to users and usage, rather than technologies in a vacuum.
For example, after significant evidence of WhatsApp misuse for bulk cir-
culation of disinformation in India, WhatsApp introduced features to curb
bulk distribution. These restrictions included limiting the number of mes-
sages that can be forwarded at a single time, providing identifying informa-
tion on forwards (including messages forwarded many times) and so on.
While WhatsApp has not released country-specific data on the impact of
such affordance-based interventions, large-scale disinformation, abuse, dis-
crimination and other forms of physical and symbolic violence continue in
India and across the globe.
The affordance-based approach has shortcomings too. First, it fails to
consider the complex historical contexts within which heterogenous users
have unequal access to power. Some users have power to and power over
others while others experience life as individuals or groups who may resist
but are frequently othered and acted upon. The lack of a political and his-
torical perspective obscures the complex ways in which communication
technologies and infrastructures are themselves caught up in contests over
power and dialectically related to intersecting and changing identities. Situ-
ating the emergence of social media use within a neoliberal economy and
culture, against the backdrop of race, caste, religious or gender conflict,
adds precision to sociotechnical decision-making by pointing more specifi-
cally to factors which enhance or inhibit equity and justice online.
While these approaches are most common for dealing with discrimina-
tion and violence on a global scale, even these approaches are internally
inconsistent. Market pressures and geopolitical allegiances lead to uneven
ways in which violence has been handled in different parts of the world. In
order to illustrate this, we take the case of discriminatory, violent and hate-
ful content in Myanmar. Apart from sporadic attention in 2018 to pogroms
targeted at Rohingya Muslims and more recently in 2021 with the military
coup, Myanmar has largely remained invisible to social media companies
when it comes to designing policies about online violence and hate. The
murder of and displacement of Rohingyas into neighbouring countries
has further widened the spread of disinformation, discrimination and hate
against the Rohingya in mainstream and social media. Social media com-
panies have largely failed to stem the hate in which they remain complicit.
32 When hate-speech policies fail
Social Media corporations’ responses to hateful content:
Policies, human moderation and machine learning
Facebook’s Community Standards2 (also applicable to Instagram) define
hate speech as a ‘direct attack against people based on what we call pro-
tected characteristics: race, ethnicity, national origin, disability, religious
affiliation, caste, sexual orientation, sex, gender identity and serious dis-
ease’. Attacks are defined as ‘violent or dehumanising speech, harmful
stereotypes, statements of inferiority, expressions of contempt, disgust or
dismissal, cursing and calls for exclusion or segregation’. Twitter’s pol-
icy,3 meanwhile, claims that the company will not allow users to ‘promote
violence against or directly attack or threaten other people based on race,
ethnicity, national origin, caste, sexual orientation, gender, gender identity,
religious affiliation, age, disability or serious disease’. YouTube’s policy4
on hate speech too claims that it will remove content promoting violence or
hatred against individuals or groups based on a set of attributes similar to
the ones mentioned above.
The constellation of social media also includes cross-platform apps,
peer-to-peer and peer-to-group messaging services, such as Facebook-
owned WhatsApp and Facebook Messenger, Viber, Telegram and Signal
that have gained in popularity since 2017. These messaging services and
apps enable private groups on a fully encrypted architecture that makes it
difficult to detect hate speech via content moderation unless encryption is
compromised, or until screenshots are made public on other platforms. This
dilemma over de-encryption (and the compromising of individual privacy)
epitomises how violence, hate and inciting communication are addressed in
jurisprudence in the form of a tension between preserving commitments to
freedom of speech and expression while preventing hate speech and further
hate crimes.
Telegram was founded by Pavel Durov, who also founded a Facebook-
like social media company called Vkontakte in Russia in 2006. When Durov
refused to divulge sensitive data to the Russian government in 2012, the
government acquired 48% of Vkontakte. Under pressure to preserve pri-
vacy, Durov created Telegram as a more secure service. Durov recounts how
after the Snowden revelations, especially around the complicity of technol-
ogy and social media companies in cooperating with the US government
on surveillance of citizens, he was inspired to take Telegram public.5 Tel-
egram’s Terms of Service6 document has nothing to say specifically about
incitement or hate speech, except in the prescription that users agree not to
‘promote violence on publicly viewable Telegram channels, bots’ and so on.
Two convergent events over the course of late 2020 and early 2021 have
drawn attention to links between policies on hateful content and corporate
 When hate-speech policies fail 33
actions: The Capitol riots in Washington D.C. followed by the suspension of
accounts on Facebook and Twitter, including the accounts of then American
president Donald Trump; and changes to WhatsApp’s privacy policy. These
led to a surge in users for Telegram and Signal which were seen as more
secure.7 WhatsApp’s changes were ostensibly to share data from WhatsApp
business accounts, but their failure to reassure users led to a mass switch-
ing of accounts to Signal and Telegram, including by Hindu supremacists,
white supremacists and other far right users.
WhatsApp’s Terms of Service8 claim to prevent users from using What-
sApp in ways that are ‘illegal, obscene, defamatory, threatening, intimi-
dating, harassing, hateful, racially or ethnically offensive, or instigate or
encourage conduct that would be illegal or otherwise inappropriate, includ-
ing promoting violent crimes’. Working with data that is unencrypted and
available to companies – such as phone number, Display Picture (DP), sta-
tus updates, geolocation and correlation of this data to those users’ content
on Facebook and/or Instagram that is unencrypted and easily accessible –
policy and technical teams at WhatsApp are exploring options for identify-
ing users propagating hate speech. Another avenue to curb the overflow of
antisemitic, Islamophobic, anti-Black, homophobic, misogynist and other-
wise racist expression has been to attempt to reduce the usage of unauthor-
ised ‘clone’ apps that circumvent preventative measures – such as limits on
forwarding – taken by the owners of these applications.
Given the scale of their operations, social media companies deploy Artifi-
cial Intelligence (AI) not only to respond to reports of hate speech by users
but also to identify and act on communication potentially designated hateful
before users report it. On a webpage titled ‘Safety in India’, WhatsApp, for
instance, claims9 to have developed ‘advanced learning technology to iden-
tify and ban accounts engaging in bulk or automated messaging and bans
more than two million accounts from WhatsApp per month, 75% of them
without a recent user report’. Note, however, that WhatsApp under the aegis
of parent company Facebook has restricted itself to removing accounts asso-
ciated with a specific type of coordinated activity rather than committing
to tackling casteist, Islamophobic, racist, misogynist, and/or homophobic
content per se. This deliberate choice is even more troubling if it is consid-
ered that it is perfectly possible for bulk and/or automated messaging to be
used for non-commercial, pro-social and non-threatening purposes such as
to combat common misinformation and to send out vaccination reminders.
The use of algorithms to organise and prioritise content on various social
media platforms and apps has prompted significant research. Much of this
work acknowledges the increasing prominence of algorithms in all kinds
of social relations with real-world consequences including large-scale
violence and discrimination (Beer, 2009; Edelman & Luca, 2014; Lewis,
34 When hate-speech policies fail
2018). Several scholars in the fields of Science and Technology Studies
and social media studies, as well as ethical AI teams at Google and else-
where, have explored ways in which algorithmic operations (and the biases
coded into them) increase discrimination and can be held more accountable
(Eubanks, 2018; Gebru et al., 2017; Noble, 2018). Options for ameliorating
the situation include small-scale observation, reverse engineering, ‘scraping
audits’ and so on (Bucher, 2012; Rieder et al., 2018; Sandvig et al., 2014).
The reasons and complications that reduce the possibility of using technical
means to limit hate speech and a deeper exploration of the ideologies and
predilections which shape the genesis and role of algorithms are important
but fall outside the scope of this book. It should be noted, however, that the
technicalities of algorithms notwithstanding, there have been some other
efforts, mainly in the USA, to put commercial and moral pressure on social
media companies to take down hate speech more widely and effectively.
While there is little doubt that social media companies will continue alter-
ing algorithms in order to increase their efficiency in detecting and removing
anything that could legally be proven to be hate speech on their platforms
or which is called hate speech by powerful entities such as governments,
it is worth noting that in the end these companies are heavily constrained
from acting on hate speech since these actions will increase ‘friction’ of
usage. Examples include Instagram enabling users to filter comments and
Twitter enabling users to hide replies. This friction – which decreases the
speed and ease of usability – has consequences for the growth of these com-
panies. Friction reduces unique subscribers and potentially loses users to
other platforms and apps. Their valuation is tied to the number of users and,
concomitantly, to a monetisation of user data.
Bearing in mind the contradictions between the ethical impulses of the
employees and the commercial interests of the shareholders and owners, it
is unlikely that owners will undertake any radical action to curb the vari-
ety of dehumanising and discriminatory communications on their apps and
platforms without market pressure. Indeed, researchers working at Goog-
le’s AI ethics unit, Timnit Gebru and Margaret Mitchell, were removed by
Google in late 2020 and early 2021 for planning to publish research high-
lighting flaws in Google’s AI technology. Stalled, Gebru refused to retract,
and sent internal emails to colleagues maintaining that Google was silenc-
ing marginalised voices.10 Google’s actions are indicative of less than stellar
employment practices, and of a decision to downplay and drag their feet on
evidence of digitally enhanced discrimination.
Another case, that of Facebook’s whistle-blower Sophie Zhang, who
wrote a long memo after she quit Facebook in September 2020, is also
illustrative of the corporate tech world’s response to online hate. Speak-
ing to Buzzfeed soon after her resignation11in 2020, Zhang revealed that
 When hate-speech policies fail 35
she had been a data scientist for Facebook’s Site Integrity fake engagement
team, looking into what Facebook calls ‘Coordinated Inauthentic Behav-
ior’ (CIB), in multiple countries including Brazil, India, Myanmar and the
UK. CIB encompasses fake pages, bots, troll farms and so on. In many of
these countries, much manipulation was traceable to users affiliated to the
ruling political parties during elections. In some cases, Zhang dealt with
pages that clearly propagated disinformation against specific groups based
on their ethnicity, religious identity, sexual orientation, political affiliation,
including disinformation related to Covid-19. While it is not easy to draw a
causal link between the content (including hate speech) published on these
pages and the physical consequences that followed for vulnerable groups in
these countries, Zhang said that she felt that she had ‘blood on her hands’.
It is worth noting here that the problem was not Facebook’s ability to
identify and flag CIB and associated hate speech, but rather two other issues
arising from large-scale CIB. First, for Facebook, CIB although a universal
problem, was to be dealt with through an internal evaluation process that
prioritised addressing incidents of hate speech in Europe, North America
and Australia-New Zealand rather than in ‘far away’ countries such as Azer-
baijan or Honduras. While it is possible to posit profit-based cynicism as an
explanation for the lack of action taken over countries such as Azerbaijan
or Honduras due to their relatively small user base or their marginality in
the mediated presentation of the world, it is another matter for Facebook to
ignore CIB and hate speech in vast countries such as India and Brazil. The
Wall Street Journal broke stories12 in August 2020 with evidence that senior
officials at Facebook India, such as the Public Policy director Ankhi Das,
deliberately ignored warnings from Facebook employees about Islamopho-
bic hate speech by politicians of the far-right ruling BJP. By October 2020
Ankhi Das had left Facebook to save face for the company, but the inten-
tions of Facebook were clear when she was replaced by another Hindutva
supporter, Shivnath Thukral, to lead its public policy unit. Thukral worked
for the BJP’s election campaign in 2014 when Narendra Modi became
Prime Minister. TIME magazine has since reported13 that Thukral too failed
to remove hate speech against Bangladeshi Muslims posted by Assamese
BJP leader Shiladitya Dev.
Facebook’s audience size puts India at the top with 320 million users in
2021.14 By comparison, the United States is second with only 190 million
users. Brazil too has 190 million users, and when Zhang alerted her col-
leagues to CIB in India and Brazil these reports were ignored, not because
these countries were deemed fiscally unimportant – rather, users who were
connected to the identified CIB were closely affiliated to the parties in
power in these countries. Brazil has over 120 million monthly active What-
sApp users with 98% of smartphone users surveyed reported to be users.
36 When hate-speech policies fail
India has approximately 400 million users making it by far the largest mar-
ket for WhatsApp.15 The parties in power in Brazil and India have, through
their leaders’ acts and speeches, as well as various politicians affiliated to
these parties, repeatedly and brazenly flouted social media policies on hate
speech and incitement to violence.16 Indeed, Facebook and others have
not just ‘looked the other way’ in order to continue benefiting from their
large user base in these countries, but have actively recruited high ranking
employees with sympathies for the far right regimes and their ideologies.
It is clear then, that when policies fail, they do not fail simply because the
policies are flawed or moderation teams poorly trained. Sometimes it seems
clear that the highest echelons in companies never intended those policies
to jeopardise their political and economic relationships with authoritarian
governments.
Social media companies are capable of ideological course correction
when it comes to hate speech when subjected to pressure. For example,
after Minneapolis police officer Derek Chauvin murdered George Floyd
on the 25 May 2020, groups advocating on behalf of Black communi-
ties put pressure on social media companies to improve their policies. In
June 2020, a campaign called Stop Hate for Profit lobbied large corpora-
tions to stop advertising on Facebook. Unilever’s announcement that it
would stop advertising on Facebook within the United States prompted
Mark Zuckerberg to declare that Facebook would overhaul its algorithms
detecting hate speech.17 The swiftness of this volte face following years
of equivocation indicates that the politics and ideological commitments
of corporate directors play a major role in sustaining or ameliorating the
stereotyping, dehumanisation, degradation, misrepresentation and violent
or threatening abuse that characterise the online experiences of individu-
als from ethnic, racial, religious and sexual minorities, and disabled com-
munities, worldwide.
The campaign to reduce advertising expenditure – which amounts to a call
for political boycott – is important. Advertising is the main source (97%)
of revenue for Facebook, increasing by 21% from $69.7 billion in 2019 to
$84.2 billion in 2020.18 In July 2020, Nick Clegg (Vice President of Global
Affairs and Communications) asserted disingenuously that Facebook does
not profit from hate. Faced with the threat of a boycott by advertisers, Face-
book tripled to 35,000 the employees working on safety and security, and
reported that their algorithm assessed 95% of hate speech reports in less
than 24 hours, faster than YouTube and Twitter. This suggests that when
Facebook’s profits are threatened, they have the capacity19 and the will to
improve the identification and take down of hateful communications. How
these takedowns are then circumvented by the far right and by users with
racist and/or other prejudiced opinions, and how the policies are used to
 When hate-speech policies fail 37
feed into rightwing narratives of victimhood, are beyond the scope of this
book, but provide fascinating material for further analysis.20
In July 2020 the National Association for the Advancement of Colored
People (NAACP) as part of a civil rights petition21 to Facebook demanded a
list of ten changes to bring about accountability (including civil rights infra-
structures, independent audits and refunds to advertisers); decency (includ-
ing take downs of hateful content and automatic internal flagging of hateful
content in private groups) and support (increasing personnel and expert
teams at Facebook). At a pace much slower than that seen after Unilever’s
pressure, Facebook announced in December 2020 that they would embark on
an overhaul of their algorithms to detect hate speech. Titled ‘Worst of Worst’
(WOW), The Washington Post accessed documents that showed that the pro-
ject22 would in effect prioritise hate speech against groups that have histori-
cally faced discrimination, especially by virtue of race. There is little doubt
that the crucial role of social media in mobilising Black communities after
the murder of George Floyd has resulted in the same groups now pressing
those very social media companies to undertake systemic and long-lasting
changes regarding racialised hate speech. Whether this purposeful pursuit of
a less violent and damaging online discourse will have beneficial long-term
effects or be by-passed by ever more tech-savvy supremacist and conserva-
tive systemic lobbying and infiltration remains to be seen. However, tech
corporations continue to be wary of the ever-stronger rightwing backlash.
There is much evidence that hate speech laws and other policies around
online offense tend to be used by those with power against less powerful
groups and lose their efficacy when brought into play by the communities
who fought to implement them. As we will discuss later in this book, this
twisting of corporate policies against the most vulnerable is clear in India
where Islamophobic and anti-Christian content and disinformation circulate
unimpeded while critiques of Hindutva fascism are taken down and their
users sanctioned, in Sri Lanka where Islamophobic and anti-Tamil senti-
ments have thrived online and in Myanmar where anti-Rohingya postings
stayed up long after they were reported but anti-government political actors
are censored and imprisoned. This is also clear in reports that aim to docu-
ment state crimes by Israeli authorities and settlers against Palestinians and
of posts that thoughtfully aim to discuss differences between anti-Zionism
and antisemitism. We now discuss these issues in relation to the case of
Myanmar.

Myanmar: A predictable genocide

With at least eight distinct major ethnic groups, Myanmar in South-East
Asia has a population of over 50 million, and shares land borders with
38 When hate-speech policies fail
China, India, Bangladesh, Thailand and Laos. Although the constitution
mentions freedom of religion, in practice, the majority religion is Bud-
dhism; with Christianity, Islam and Hinduism along with Indigenous faiths
practiced to a lesser extent. The region has a more or less continuous his-
tory of struggle between ‘centre’ (Bamar) and periphery (Shan, Mon and
Rakhine groups) that was suppressed with British colonial rule in 1885
(Aung-Thwin, 1985; Taylor, 1987; Than, 2005). The colonial separation
of ‘Burma proper’ from the frontier areas with non-Bamar ethnic groups
(‘Scheduled Areas’) enabled Bamar nationalists to construct an ‘imagined’
Myanmar unified by a melding of disparate ethnic-based sovereign enti-
ties (Cady, 1965; Selth, 1986). The rise of Major-General Aung San as a
leader of the Anti-Fascist People’s Freedom League (AFPFL) for national
independence and his subsequent assassination in 1947 saw the creation of
a mythic figure that still plays a role in contemporary politics (Guyot, 1989;
Taylor, 1987). With Aung-San’s assassination, the military regime produced
a routinised form of violent insurgency and counterinsurgency, ‘a knitting
together of networks of violence that constituted a tenuous but productive
form of state-building’ (Callahan, 2005, p. 115). Revolts from the other eth-
nic groups such as the Mon, Pao and Kachin prompted a military coup in
1962 (Smith, 1991). Border skirmishes and negotiations doubly benefitted
the military leaders – the Tatmadaw – by helping to justify coercion in the
face of armed insurgencies, and by allowing the junta to profit from both
licit and illicit trades in profitable jade, timber and opium, while retaining
control of sectors from banking and transport to telecommunications (Mee-
han, 2011; Selth, 1986).
Between 1988 and 2010, Myanmar’s GDP grew from $12.6 bn USD to
$45.4 bn, imports rose from $246 mn to $4.8 bn, exports rose from $167
mn to $8.7 bn and foreign investment rose from $4 mn to $8.3 bn, a size-
able growth primarily backed by China’s involvement in Myanmar (Jones,
2014). A general election was held in 2010 and the military-backed Union
Solidarity and Development Party (USDP) swept the board while the
National League for Democracy was declared illegal. However, soon after
the elections, politician Aung San Suu Kyi, daughter of assassinated Major-
General Aung San, was released and various reforms began to be introduced
such as a reform in labour laws allowing for unions and strikes, a relaxation
of press censorship and amnesties on hundreds of political prisoners.
Even while the state continued to suppress minority dissent brutally,
the military regime’s dominance seemed to end with the formation of the
Socialist Republic of the Union of Burma in 1974 under a new constitution.
The democratic movement leading up to the coup of 1988 was largely com-
prised of Bamar intellectuals along with a coalition of students, ex-military
dissidents and older political leaders (Guyot, 1989; Smith, 1991). In 1990,
 When hate-speech policies fail 39
the military junta – known as the State Law and Order Restoration Council
(SLORC) – held an election in which Aung San’s daughter Aung San Suu
Kyi leading the National League for Democracy (NLD) won 392 seats and,
along with the Shan NLD, cornered 90% of the seats. However, in 1990,
the military ensured that the victors would only be responsible for a new
constitution. Upon Suu Kyi’s release in mid-1995, the two parties remained
in deadlock, despite continued dialogue, with coercion and violence con-
tinuing (Smith, 1991). In the 2015 and 2020 elections, the NLD won with
NLD’s Htin Kyaw elected as the first non-military president since the mili-
tary coup of 1962 with Suu Kyi as state counsellor. However, in Febru-
ary 2021 the Tatmadaw declared a state of emergency, jailed senior NLD
leaders including the president and state counsellor, and imposed martial
law in many areas. As we write, mass protests have broken out, and protes-
tors have been brutally repressed.

‘Othering’ as a tool of authoritarian governance: ‘We

have been trained since we were children to be racists’
The process of ‘othering’ (Hall, 1996, pp. 4–5) links discursive practices of
identity construction to stereotyping, exclusion and violation. Some minor-
ity ethnic groups were incorporated into the Tatmadaw’s nation-building
project. Others have been excluded and systematically targeted. Although
several ethnic groups have struggled for decades against the Tatmadaw, the
Rohingyas have suffered the most (Clarke et al., 2019) by being subject to
a diverse set of strategies such as ‘lawfare’ and ‘spacio-cide’ (MacLean,
2019). These legal and governmental strategies legitimise the exclusion
of Rohingyas from territories where they have lived for generations. Such
strategies have most commonly been used, as Appadurai argues,23 by ‘para-
noid’ sovereignties and/or ‘predatory’ states against ‘biominorities’, in other
words, populations whose difference (based on ethnicity, religion, race and
so on) from the national majorities is perceived as a bodily threat to the
national ethos (c.f. Hanafi, 2009; Yiftachel, 2005).
One of the first major acts of ‘lawfare’ against the Rohingya was ‘Opera-
tion Dragon King’ in 1978, which saw foreigners screened and illegal immi-
grants expelled. Across successive ethnic cleansing operations over the next
two decades, roughly 550,000 Rohingyas fled to neighbouring countries,
especially Bangladesh, though some 250,000 gradually returned. This exo-
dus was a response to the Burmese Junta’s abuse of international principles
pertaining to citizenship. For example, the 1982 citizenship law, updating
the 1947 constitution, explicitly tied ancestry to territory to come up with
three categories of citizenship: Full citizens (descendants of residents who
lived in Myanmar before 1823); associate citizens (people who acquired
40 When hate-speech policies fail
citizenship under the 1948 union citizenship law); and naturalised citizens
(people who have been residing in Myanmar before 1948 but had failed to
apply for citizenship under the 1948 union citizenship law). All three cat-
egories of citizenship only applied to Myanmar’s ‘national ethnic races’ first
published in 1960 naming the major groups of Bamar, Kayin, Shan, Kachin,
Rakhine and so on, but not to Rohingyas. A new census was conducted in
1983 and results of 135 ethnic groups revealed only in 1990, and again
Rohingyas did not figure since Bamars constructed Rohingyas as Benga-
lis who migrated to Myanmar after 1823 and not an Indigenous popula-
tion24 (Cheesman, 2017; Cheng Guan, 2007; Ferguson, 2015; Kipgen, 2019;
Kyaw, 2015).
During the early 1990s, Rohingyas were forcibly evicted so that ‘model
homes’ could be built on their lands, with the labour of displaced Rohingyas.
On 16 October 1992, the then Special Rapporteur on Freedom of Religion
or Belief informed the Government that he had received information that25:

since late 1989, the Rohingya citizens of Myanmar . . . have been

subjected to persecution based on their religious beliefs involving
extrajudicial executions, torture, arbitrary detention, forced disappear-
ances, intimidation, gang-rape, forced labour, robbery, setting of fire to
homes, eviction, land confiscation and population resettlement as well
as the systematic destruction of towns and mosques.

In 2012, three Rohingya men received death sentences for the gang-rape of
a Rakhine woman. A week later, ten non-Rohingya Muslims were lynched,
and after a stridently effective hate speech campaign across mainstream and
alternative media, mass violence followed, targeting Muslims in Rakhine,
with full complicity from the state. More than 10,000 homes were destroyed
and 140,000 Rohingyas were displaced, most of whom live in temporary
camps set up in Rakhine. Similar to Israeli practices of settler colonialism,
the Tatmadaw battalions, the Buddhist Rakhine communities and ‘entrepre-
neurs’ seeking fresh markets combined to drive Rohingyas from their lands
and homes in Rakhine. How Facebook came to play a role in furthering the
ethnic cleansing of the Rohingya and strengthening the Tatmadaw in Myan-
mar goes to the heart of our concerns in this book.
Before 2021, incidents ranging from communal clashes based on
rumours, incendiary political speeches and announcements, misuse of old
laws and the introduction of new laws targeting Rohingyas as well as other
political dissenters were increasing. Superficial reforms and the more active
role played by politically-savvy Suu Kyi had led international organisa-
tions and the UN to ignore her Bamar chauvinism (Davis, 2021; Lee, 2014).
However, in 2015 Myanmar’s parliament approved a set of discriminatory
 When hate-speech policies fail 41
laws collectively dubbed the Race and Religion Discrimination bills. Sub-
mitted to Parliament in 2013 by Ashin Wirathu, then leader of the far-right
Buddhist and nationalist organisation Association for the Protection of
Race and Religion (also known as Ma Ba Tha within Myanmar), these bills,
now laws,26 allow regional officials to establish 36-month birth spacing for
specific communities (Rohingya), compel Buddhists and other groups to
obtain official approval to marry partners from another faith, prohibit Mus-
lim couples from having more than two children and impose monogamy to
target Muslims who are often framed as sexual deviants. Before his account
was blocked in 2018, Wirathu used Facebook unimpeded to spread disin-
formation and advocate violence against Rohingyas (Fink, 2018; Whitten-
Woodring et al., 2020). More than two thirds of Myanmar’s Rohingya
population have fled.
The Rohingyas’ suffering has commonly been framed as a ‘humanitar-
ian crisis’ (Ullah, 2011). Since the mid-1990s, Rohingya groups failed to
mobilise collectively, failed to revoke the military ban on Muslim organisa-
tions such as the Rohingya Students Union or the Rohingya Youth League
and/or split into factions such as the Rohingya Solidarity Organisation
and the Arakan Rohingya Islamic Front. Without protection, the 2016–17
ethnic cleansing led more than 600,000 Rohingyas to flee to Bangladesh.
Rohingya men are routinely demonised and killed, while women and chil-
dren are disproportionately targeted with sexual violence and torture. Sub-
sequently, denial of healthcare and other basic infrastructural facilities in
camps fuelled disease and death. Much of this has been common knowl-
edge for years. Severe travel restrictions, systematic denial of land use for
agriculture, extortion and bribery have contributed to a ‘slow genocide’
that seeks to erase Rohingya identity and culture from Myanmar (Amnesty
International, 2017; Anwary, 2021; Houtman, 1999; MacLean, 2019;
Mahmood et al., 2017; McCarthy & Menager, 2017; Ware & Laoutides,
2019; Zarni & Cowley, 2014).
Communications in Myanmar are tightly controlled by the Tatmadaw
through prohibitive pricing and centralised control over communicative
infrastructure. SIM cards and mobile phones were introduced by the Myan-
mar Post and Telephone department in 2000 at a cost of $5000. The Myan-
mar Computer Science Development Law criminalised unregistered access
to the internet with a maximum 15-year prison sentence. Pre-2014, draco-
nian legislations such as the Burma Wireless Telegram Act, Printers and
Publishers Registration Act and the Myanmar Computer Science Develop-
ment Law were in place (Sablosky, 2021). Post-2014 saw a rapid prolifera-
tion of legacy broadcast and print media, with increased internet penetration
through foreign investment, a proliferation of cheaply available SIM cards
and mobile smartphones and drastic increase in social media use (Davis,
42 When hate-speech policies fail
2021; Farrelly & Win, 2016; Renshaw, 2013). In 2016–17, Facebook tried
to penetrate emerging markets like India and Myanmar through Free Basics,
wherein users would get Facebook pre-installed and not pay for data while
using Facebook. The program failed to take off in India, but succeeded in
Myanmar. Thenceforth, Myanmar relied on Facebook as the primary news
interface (Whitten-Woodring et al., 2020, pp. 414–415). State media, mean-
while, was busy spreading hatred towards Rohingyas via misrepresenta-
tions as ‘Bengali’, ‘foreigners’, ‘terrorists’ and so on (Lee, 2019).
While Facebook provided some independent information sources for
minority groups, Facebook did nothing to take down communications incit-
ing violence even by its own definitions and standards. In 2012, President
Thein Sein’s spokesperson Zaw Htay posted the following message on his
Facebook page:

Rohingya terrorists as members of the RSO are crossing the border

into Myanmar with weapons . . . Our troops have received the news
in advance so they will completely destroy [the Rohingya]. It can be
assumed that the troops are already destroying [the Rohingya]. We
don’t want to hear any humanitarian or human rights excuses. We don’t
want to hear your moral superiority, or so-called peace and loving kind-
ness. Go and look at Buthidaung, Maungdaw areas in Rakhine State.
Our ethnic people are in constant fear in their own land. I feel very
bitter about this. This is our country. This is our land. I’m talking to
you, national parties, MPs, civil societies, who are always opposing the
President and the Government.
[Detailed findings of the independent fact-finding
mission in Myanmar, 2019, p. 167]

The Rohingya community is also well aware of the popular discourse on

Facebook (Whitten-Woodring et al., 2020, p. 418):

. . . the coverage of local media is not fair. What local media portrayed
was that the Rohingya set their own homes on fire, like the same thing
you would see on Facebook. Most people shared such news on Face-
book and the whole country believed that the Rohingya just set their
homes on fire.

Even as Suu Kyi herself tried to appease Buddhist nationalists in 2016 by

asking the UN not to use the word Rohingya (presumably preferring the dis-
information term ‘Bengali’), she and her colleagues faced abuse on social
media, were called ‘Muslim lovers’ by the Ma Ba Tha and their supporters
(Davis, 2021, p. 113). Despite this kind of inflammatory phrasing and its
 When hate-speech policies fail 43
material consequences – vicious physical atrocities, rape, arson, homeless-
ness, land-grabbing and other anti-Rohingya violence – Facebook chose
not to act. Our analysis suggests that this decision may have been based on
its estimation that Myanmar was too insignificant in terms of international
geopolitics, and too profitable as a home for Facebook’s Free Basics pro-
gramme and other potential investments.
Between 2017 and 2018 the volume and content of anti-Rohingya, anti-
Muslim, disinformation on Facebook in Myanmar rose. Repeated hostile
misrepresentations served to legitimise extreme forms of violence and
atrocity by the armed forces and Buddhist-controlled paramilitaries, which
were, in turn, represented online as efforts to ‘safe-guard’ the nation. Mili-
tary personnel stationed in mixed Buddhist and Muslim villages would
segregate Buddhist villagers from their Muslim neighbours, saying they
had come to ‘protect them’. Western governments remained determinedly
silent until the grim reports of murder and atrocity began to emerge pub-
licly through investigative journalists in Rakhine:27 Ten men shot and buried
in one village; boys and men burnt to death in torched houses in another;
women and girls gang-raped and denied medical treatment. Estimates from
the UN and Human Rights Watch in 2018–19 suggest that over 7000 were
murdered and more than 10,000 raped.
One of our key informants, HS, (anonymised for his safety) discussed the
context of the racist socialisation that intersects with hateful content online:

We have been trained since we were children to be racists. I mean,

I attended High School in Yangon. Although our teachers and head-
masters already understood the [racist] context, there were some forms
of discrimination for example . . . our school [played and still plays] a
patriotic song every day at the end of the day: ‘The Pride of Birth’, in
the lyrics there are some words like ‘The country for martyrish Bud-
dhists . . .’ We had been indoctrinated by such words since we were
young. . . . Once, I did not take notes on history in a writing book,
I just wrote a question and answer on the textbook. So some teachers
told my family that I had a Muslim friend and got spoiled. In differ-
ent ways, such as singing, routine activities and informing parents by
teachers, we were trained to hate others [especially Muslims] . . . If a
Buddhist student is poor in study, he or she is not blamed, but if stu-
dents from other religions are poor in education, it is because of their
religion.

The Myanmari diaspora are also affected by the abject failure of Facebook
to act upon years of online dehumanisation and violent incitement. Roh-
ingya refugees in India face persistent dehumanisation and threats online,
44 When hate-speech policies fail
with disinformation about them tied to anti-Muslim pogroms and changes
in laws. Camps have been burnt down in India. Even individuals living in
the West who have spoken out against the February 2021 military coup fear
for their relatives’ lives. Hate speech in Myanmar is thus contemporary as
well as historical and remains consonant with discrimination and racism in
changing circumstances.

Rohingya under constant ‘threat of genocide’: Social

media companies flailing
In the Tatmadaw’s genocidal project against the Rohingyas which has
unfolded over decades, social media has added fuel to inflammable rhetoric
and policies. In response to disquiet amongst platform owners about the
2021 coup, the Tatmadaw opted for cruder forms of infrastructural con-
trol such as Internet shutdowns, intranets, surveillance through biometric
technology, verification cards, facial recognition technologies and whitelist-
ing applications that delete the vast majority of online content except on
authorised applications.28 Meanwhile, the response of social media compa-
nies continues to be disjointed. Even as Facebook had close to seven million
users in Myanmar, a Reuters report29 revealed that Facebook in 2014 had
only one employee who spoke Burmese (based in Dublin) and a year later,
only four employees (based in Manila and Dublin). Currently, Facebook has
outsourced its content moderation in Myanmar to the global business pro-
cess outsourcing firm Accenture, based in Kuala Lumpur, through a project
called Honey Badger. Approximately 60 employees who speak Burmese
are moderating content posted on Facebook related to Myanmar. Since con-
nectivity came to Myanmar relatively late, it affected the development of a
Burmese Unicode script and online users in Myanmar had developed their
own script called Zawgyi. Oblivious to these technical issues, Facebook
created a Burmese to English interface unable to decipher Zawgyi. As a
result, their content moderators cannot even read and understand content
accurately, let alone moderate it.
For example, an inciting phrase posted on Facebook ‘Kill all the K***rs
that you see in Myanmar; none of them should be left alive’ was translated
by Facebook interface as ‘I shouldn’t have a rainbow in Myanmar’. Kalar
is a derogatory term that is commonly used to refer to Rohingya and other
minoritised groups as foreigners and outsiders. In an attempt to show the
world that it is taking firm action, Facebook banned the word, filtering out
its Burmese characters, which they later discovered literally translated as
‘from the west’. However, due to its reliance on automatic filters rather than
contextual moderation, Facebook has unwittingly been removing any word
that has kalar in it such as kalar pae (chickpeas) or kalarkaar (curtains).
 When hate-speech policies fail 45
More recently, in 2019 and 2020, Facebook removed accounts linked
to the Tatmadaw, explicitly admitting that Facebook has played a role in
the genocide against Rohingyas. However, in an attempt at supposed fair-
play and balance, Facebook also deleted30 several Facebook pages of what
are commonly known as Ethnic Armed Organisations (EAOs). While some
EAOs do discuss armed resistance and recruit on social media, many more
use social media to highlight human rights abuses by the Tatmadaw, which
otherwise have no chance of publication on mainstream media. Not all
EAOs have been affected but only those which are anti-regime, thus posi-
tioning Facebook – possibly inadvertently but more likely by choice – in
favour of the Tatmadaw, and entrenching Facebook in the entangled politics
of ethnic supremacism in Myanmar (Sablosky, 2021).
Since the coup in February 2021, minority groups including political
organisations mobilising in favour of minority rights, activists, academics,
journalists and political dissenters have to change strategy. A young activist
from Myanmar who has been speaking out against the regime since 2017
and faces death and rape threats, shares her experience31:

People can’t get real information . . . They restored the Internet but not
the television. On Facebook there is no legitimate news verification,
there’s no accountability, so now misinformation is growing danger-
ously. There is a lot of hate speech, a lot of fake accounts. I relocated,
but I don’t feel safe at all, anything can happen to us. I feel a mounting
threat. . .

Media literacy and sensitisation programmes for young social media users
are being offered by organisations such as Myanmar Institute of Theology
and Myanmar ICT for Development Organisation32 while UNICEF runs
youth media literacy and voice campaigns in Cox’s Bazaar amongst the
Rohingya refugees. However, since state and non-state actors, militias and
Buddhist vigilante mobs are bent upon eliminating the Rohingya popula-
tion, such efforts while worthy and sensible, are drops in an ocean of hate.
Given the discussion of multiple policies in the first half of this chapter
and the limited but notable achievements that international pressure and
boycott threats had in leading to acknowledgment of anti-Black racism on
Facebook, it is important to note that many platforms continue to be used
by the military and by Buddhist supremacist groups in Myanmar and there-
fore need to be scrutinised and held to account by international pressure.
The need for such pressure will become all the more evident in the case of
Brazil, which we delve into in the next chapter, and where social media dis-
information and hate is implicated in far right vigilantism and authoritarian-
populist regime change.
46 When hate-speech policies fail
Notes
1 See www.apc.org/en/pubs/statement-facebooks-internal-guidelines-content-
moderation
2 See www.facebook.com/communitystandards/hate_speech
3 See https://help.twitter.com/en/rules-and-policies/hateful-conduct-policy
4 See https://support.google.com/youtube/answer/2801939?hl=en-GB
5 See www.dazeddigital.com/artsandculture/article/24279/1/pavel-durov
6 See https://telegram.org/tos
7 See www.nytimes.com/2021/01/13/technology/telegram-signal-apps-big-tech.
html
8 See www.whatsapp.com/legal/terms-of-service-eea
9 See https://faq.whatsapp.com/general/safety-in-india/?lang=en
10 See www.bbc.co.uk/news/technology-56135817
11 See www.buzzfeednews.com/article/craigsilverman/facebook-ignore-political-
manipulation-whistleblower-memo
12 See www.wsj.com/articles/facebook-hate-speech-india-politics-muslim-hindu-
modi-zuckerberg-11597423346
13 See https://time.com/5883993/india-facebook-hate-speech-bjp/
14 See www.statista.com/statistics/268136/top-15-countries-based-on-number-of-
facebook-users/
15 See www.statista.com/statistics/289778/countries-with-the-most-facebook-users/
16 See www.theguardian.com/technology/2020/jul/02/whatsapp-groups-conspiracy-
theories-disinformation-democracy
17 See www.theguardian.com/technology/2020/jun/29/how-hate-speech-campaigners-
found-facebooks-weak-spot
18 See www.statista.com/statistics/267031/facebooks-annual-revenue-by-segment/
19 See www.vanityfair.com/news/2020/11/facebooks-election-tweaks-curb-misin­
formation
20 As Chouliaraki and Banet-Weiser (2021, p. 4) argue in their introduction to
a special issue on the logic of victimhood ‘that victimhood today operates as
a ‘master’ signifier, a dominant communicative logic that relies on auxiliary
vocabularies – of injured white masculinity, celebrated survivorship or heroic
sacrifice – to reclaim power for the powerful and retrench existing hegemonic
arrangements in liberal polities. Whether it is authoritarian populists, networked
misogynists or imperialist state actors, this logic of weaponized victimhood is,
we contend, a crucial site of political struggle and, as such, a scholarly terrain of
urgent interrogation’.
21 See www.naacp.org/latest/statement-stop-hate-profit-meeting-facebook/
22 See www.washingtonpost.com/technology/2020/12/03/facebook-hate-speech/
23 Arjun Appadurai. (2018, May 22). Across the world, genocidal states are attack-
ing Muslims: Is Islam really their target? Scroll. https://scroll.in/article/879591/
from-israel-to-myanmar-genocidal-projects-are-less-about-religion-and-more-
about-predatory-states
24 Readers familiar with recent legislation in India, especially the Citizenship
Amendment Act and the National Citizenship Register, will find similarities
between how the Rohingyas have been excluded and dehumanised in Myanmar
and how the Muslims in India are being targeted by the BJP government in
India.
25 E/CN.4/1993/62. https://documents-dds-ny.un.org/doc/UNDOC/GEN/G93/101/
09/PDF/G9310109.pdf
 When hate-speech policies fail 47
26 Michael Caster. (2015, August 26). The truth about Myanmar’s new discrimi-
natory laws. The Diplomat. https://thediplomat.com/2015/08/the-truth-about-
myanmars-new-discriminatory-laws/
27 www.reuters.com/investigates/special-report/myanmar-rakhine-events/
28 Crisis Group. (2021, May 18). Myanmar’s military struggles to control the
virtual battlefield. Crisis Group. www.crisisgroup.org/asia/south-east-asia/
myanmar/314-myanmars-military-struggles-control-virtual-battlefield
29 Steve Stecklow. (2018, August 15). Why Facebook is losing the war on hate
speech in Myanmar. Reuters. www.reuters.com/investigates/special-report/
myanmar-facebook-hate/
30 https://about.fb.com/news/2021/02/an-update-on-myanmar/
31 Peter Guest. (2021, February 2). How misinformation fueled a coup in
Myanmar. Rest of World. https://restofworld.org/2021/how-misinformation-
fueled-a-coup-in-myanmar/
32 Samantha Stanley. (2017, May 16). Misinformation and hate speech in Myan-
mar. First Draft. https://firstdraftnews.org/articles/misinformation-myanmar/

References
Amnesty International. (2017). Caged without a roof: Apartheid in Myanmar’s
Rakhine state. Amnesty International.
Anwary, A. (2021). Sexual violence against women as a weapon of Rohingya geno-
cide in Myanmar. The International Journal of Human Rights, 1–20.
Arvidsson, A., & Colleoni, E. (2012). Value in informational capitalism and on the
internet. The Information Society, 28(3).
Aung-Thwin, M. (1985). Pagan: The origins of modern Burma. University of
Hawaii Press.
Beer, D. (2009). Power through the algorithm? Participatory web cultures and the
technological unconscious. New Media & Society, 11(6), 985–1002.
Bucher, T. (2012). Want to be on the top? Algorithmic power and the threat of invis-
ibility on Facebook. New Media & Society, 14(7), 1164–1180.
Cady, J. (1965). A history of modern Burma. Cornell University Press.
Callahan, M. (2005). Making enemies: War and state-building in Burma. Cornell
University Press.
Cheesman, N. (2017). How in Myanmar “national races” came to surpass citizen-
ship and exclude Rohingya. Journal of Contemporary Asia, 47(3).
Cheng Guan, A. (2007). Political legitimacy in Myanmar: The ethnic minority
dimension. Asian Security, 3(2), 121–140.
Clarke, S. L., Myint, S. A. S., & Siwa, Z. Y. (2019). Re-examining ethnic identity in
Myanmar. CPCS Publication.
Dahlberg, L. (2015). Expanding digital divides research: A critical political econ-
omy of social media. The Communication Review, 18(4), 271–293.
Darusman, M., Coomaraswamy, R., & Sidoti, C. (2019). Detailed findings of the
independent fact-finding mission on Myanmar. https://www.ohchr.org/EN/
HRBodies/HRC/MyanmarFFM/Pages/Index.aspx
Davis, A. (2021). Hate speech in Myanmar: The perfect storm. In S. Jayakumar, B. Ang, &
N. D. Anwar (Eds.), Disinformation and fake news (pp. 103–116). Palgrave Macmillan.
48 When hate-speech policies fail
Edelman, B. G., & Luca, M. (2014). Digital discrimination: The case of Airbnb.
com. SSRN Electronic Journal. Elsevier BV.
Eubanks, V. (2018). Automating inequality: How high-tech tools profile, police and
punish the poor. New York: Picador.
Faraj, S., & Azad, B. (2012). The materiality of technology: An affordance perspec-
tive. In P. Leonardi, B. Nardi, & J. Kallinikos (Eds.), Materiality and organizing:
Social interaction in a technological world (pp. 238–258). Oxford University
Press.
Farrelly, N., & Win, C. (2016). Inside Myanmar’s turbulent transformation. Asia &
the Pacific Policy Studies, 3(1), 38–47.
Ferguson, J. (2015). Who’s counting? Ethnicity, belonging and the national census
in Myanmar. Journal of the Humanities and Social Sciences of Southeast Asia and
Oceania, 171(1), 1–28.
Fink, C. (2018). Dangerous speech, anti-Muslim violence, and Facebook in Myan-
mar. Journal of International Affairs, 71(15), 43–52.
Fuchs, C. (2010). Labor in informational capitalism and on the internet. The Infor-
mation Society, 26(3).
Gebru, T., Krause, J., Wang, Y., Chen, D., Deng, J., Erez Lieberman Aiden & Li Fei-
Fei. (2017). Demography with deep learning and street view Proceedings of the
National Academy of Sciences, 114(50), 13108–13113.
Gibson, J. (1982). Notes on affordances. In E. Reed & R. Jones (Eds.), Reasons for
realism: Selected essays of James J. Gibson (pp. 401–418). Lawrence Erlbaum
Associates.
Guyot, J. (1989). Burma in 1998: Perestroika with a military face. Southeast Asian
Affairs, 108–133.
Hall, S. (1996). Who needs identity? In S. Hall & P. du Gay (Eds.), Questions of
cultural identity (pp. 1–17). Sage.
Hanafi, S. (2009). Spacio-cide: Colonial politics, invisibility, and rezoning in Pales-
tinian territory. Contemporary Arab Affairs, 2(1), 106–121.
Houtman, G. (1999). Mental culture in Burmese crisis politics: Aung San Suu Kyi
and the national league for democracy. Tokyo University of Foreign Studies.
Hutchby, I. (2003). Affordances and the analysis of technologically mediated inter-
action: A response to Brian Rappert. Sociology, 37(3), 581–589.
Jones, L. (2014). The political economy of Myanmar’s transition. Journal of Con-
temporary Asia, 44(1).
Kipgen, N. (2019). The Rohingya crisis: The centrality of identity and citizenship.
Journal of Muslim Minority Affairs, 39(1).
Kyaw, N. N. (2015). Alienation, discrimination and securitization: Legal person-
hood and cultural personhood of Muslims in Myanmar. Review of Faith and
International Affairs, 13(4), 50–59.
Lee, R. (2014). A politician, not an icon: Aung San Suu Kyi’s silence on Myanmar’s
Muslim Rohingya. Islam and Christian – Muslim Relations, 25(3).
Lee, R. (2019). Extreme speech in Myanmar: The role of state media in the Rohingya
forced migration crisis. International Journal of Communication, 13, 3203–3224.
 When hate-speech policies fail 49
Lewis, R. (2018). Alternative influence: Broadcasting the reactionary right on You-
Tube. Data & Society Research Institute.
MacLean, K. (2019). The Rohingya crisis and the practices of erasure. Journal of
Genocide Research, 21(1).
Mahmood, S. S., Wroe, E., Fuller, A., & Leaning, J. (2017). The Rohingya people of
Myanmar: Health, human rights, and identity. The Lancet, 389(10081).
McCarthy, G., & Menager, J. (2017). Gendered rumours and the Muslim scapegoat
in Myanmar’s transition. Journal of Contemporary Asia, 47(3), 396–412.
Meehan, P. (2011). Drugs, insurgency and state-building in Myanmar: Why the
drugs trade is central to Myanmar’s changing political order. Journal of Southeast
Asian Studies, 42(3), 376–404.
Noble, S. (2018). Algorithms of oppression. New York University Press.
Renshaw, C. S. (2013). Democratic transformation and regional institutions: The
case of Myanmar and ASEAN. Journal of Current Southeast Asian Affairs, 32(1),
29–54.
Rieder, B., Matamoros-Fernández, A., & Coromina, Ò. (2018). From ranking algo-
rithms to “ranking cultures”. Convergence: The International Journal of Research
into New Media Technologies, 24(1), 50–68.
Rigi, J., & Prey, R. (2015). Value, rent, and the political economy of social media.
The Information Society, 31(5). https://doi.org/10.1080/01972243.2015.1069769
Sablosky, J. (2021). Dangerous organizations: Facebook’s content moderation deci-
sions and ethnic visibility in Myanmar. Media, Culture & Society, 1–26.
Sandvig, C., Hamilton, K., & Karahalios, K. (2014). Auditing algorithms: Research
methods for detecting discrimination on internet platforms. In Data and discrimina-
tion: Converting critical concerns into productive inquiry. New America Foundation.
Selth, A. (1986). Race and resistance in Burma, 1942–1945. Modern Asian Studies,
20(3), 483–507.
Smith, M. (1991). Burma: Insurgency and the politics of ethnicity. Zed Books.
Taylor, R. (1987). The State in Burma. C. Hurst.
Than, T. M. (2005). Dreams and nightmares: State building and ethnic conflict in
Myanmar (Burma). In K. Snitwongse & S. W. Thompson (Eds.), Ethnic conflicts
in Southeast Asia (pp. 65–108). ISEAS-Yusof Ishak Institute.
Ullah, A. A. (2011). Rohingya refugees to Bangladesh: Historical exclusions and
contemporary marginalization. Journal of Immigrant & Refugee Studies, 9(2).
Ware, A., & Laoutides, C. (2019). Myanmar’s “Rohingya” conflict: Misconceptions
and complexity. Asian Affairs, 50(1). https://doi.org/10.1080/03068374.2019.15
67102
Whitten-Woodring, J., Kleinberg, M. S., Thawnghmung, A., & Thitsar, M. T. (2020).
Poison if you don’t know how to use it: Facebook, democracy, and human rights
in Myanmar. The International Journal of Press/Politics, 25(3), 407–425.
Yiftachel, O. (2005). Neither two states, nor one: The disengagement and “creeping
apartheid” in Israel/Palestine. The Arab World Geographer, 8(3), 125–129.
Zarni, M., & Cowley, A. (2014). The slow-burning genocide of Myanmar’s Roh-
ingya. Pacific Rim Law and Policy, 23(3), 681–752.
3 Brazil colonisation,
violent ‘othering’ and
contemporary online hate

Introduction
This chapter presents a case study of Brazil where modest political reform
led by previous regimes – for instance, with regard to food security under
the Worker’s Party or LGBTQIA+ rights – have been followed by a virulent
rightwing backlash. This backlash is intimately connected to the rising pop-
ularity of the Brazilian far right whose authoritarian politics are most nota-
bly personified by Jair Bolsonaro. Bolsonaro’s rise to power and his current
grip on the political imagination of at least a third of the voting public, was
supported and accompanied by the systematic production and reproduction
of hateful discourse and action. This systematic hate was deployed against
specific groups – including the working classes, Indigenous populations,
Afro-Brazilians, the left, LGBTQIA+ groups – always exacerbating existing
lines of racism and ethnic difference. From the creation of difference to a
process of consolidation of political and economic power, the production of
targets for populist rage and violence has been an extended socio-political
process. This process includes but is not confined to the co-optation and
misuse of state institutions and actors specific to Brazilian history, culture
and politics, and the rapid demonisation and delegitimisation of groups and
individuals who publicly stand for peace, equity and social justice through
reference to traits of identity, character or politics which act as excuses for
violence.
Within this milieu, social media use including Facebook, YouTube, Twit-
ter, WhatsApp, Instagram and TikTok for the circulation of hate and misinfor-
mation has further intensified violence through the seeding of disinformation
that deliberately deceives specific groups about their fellow Brazilians. The
intimacy afforded by social media spaces and the illusion of privacy on some
cross-platform apps has led to the emergence of new forms of political troll-
ing where the identity traits of those championing social justice and human
rights (such as religion, gender, race, sexuality or disability) are ‘discovered’
DOI: 10.4324/9781003083078-3
 Brazil and contemporary online hate 51
by organised far right political actors and used to build narratives of deceit
and contamination, whose solution appears to be the annihilation of the
voices and bodies of the ‘other’ from the public sphere. For instance

The story that was spread via Bolsonaro’s social media channels was
that Haddad had created a “gay kit” which he planned to introduce in
primary schools so that children from the age of six “would be encour-
aged to become gay”. In fact, Haddad, as Brazil’s former minister for
education – and alongside other politicians – had promoted an edu-
cational programme for primary school students to understand sexual
diversity and combat homophobia.1

This hateful and mendacious but politically effective campaign culminated

in deep-fake videos of Fernando Haddad implicating him and his supporters
in ‘deviant’ sexual practices and, in particular, paedophilia, that were circu-
lated in a targeted manner towards sections of the Catholic and Evangelical
population, and in particular towards middle-aged women, swinging them
away from the Workers Party at the last minute.
Despite many brave, reputable and insightful journalists, the mainstream
media in Brazil, meanwhile, have played a deeply problematic role in main-
taining various forms of colonial thinking, white supremacism and hetero-
sexual privilege. Built on a model of private investment and weak regulation
since the early 1990s and in a manner startlingly similar to India, even the
few supposedly educational media outlets are run by quasi-religious or oli-
garchic interests: ‘Brazil has always had a weak public media sector which
has been composed mainly of the respected but funded-starved TV Cultura
in SP and its counter-part TVE in Rio, as well as other regional outlets con-
trolled by local politicians and by sectors of the evangelical Church’ (Matos,
2011, p. 7). While TVE has since ceased to broadcast and the primary public
channel now is TV Brasil, the broad point still holds good. In some regions
a single family with affiliation to or entanglement with conservative politi-
cians controls all of the media outlets, and therefore the political messages.
As Alfonso de Albuquerque argues powerfully with regard to Brazil,

elites and their media portray themselves as a westernized minority

endowed with a civilizing mission regarding their societies as a whole,
and manipulate the Fourth Estate discourse toward their own benefit, as
a means for securing and legitimizing their own privilege.
(2017, p. 906)

Associating the western democratic tradition with capitalism in an unregu-

lated form, these elites use mainstream news media to spin their attempts
52 Brazil and contemporary online hate
to suborn and subvert the institutions of accountability as ‘pro-democratic’
when their targets are Indigenous populations or leftist politicians. Spon-
holz and Christofoletti’s informative analysis traces the ways in which co-
optations and rightwing ownership models have led to a situation whereby
‘the media system has been enabling public figures to use hate speech to
enhance their media prominence’ (2019, p. 67). The toppling of the left gov-
ernment and the ascendence of rightwing forces who utilise hate speech is
thus, in their view, a combination of social and media capital that pre-dates
the embedding of social media as a popular tool.
As we discussed in Chapters 1 and 2, the hate, disinformation and mis-
information circulated on social media are not isolated phenomena. Nor
do they just appear out of the blue in the online world as by-products of
the affordances of new and emerging technologies. Rather, social media,
hate and disinformation spring from complex and intersecting histories and
genealogies. They are embedded in and fuelled by structural and material
violence, racism, prejudice, discrimination and inequality. Accompanying
these adverse processes and behaviours, we begin to see more recently the
entry of a techno-communicative development model, untethered from an
ethics of rights and solidarity. It is a model that thrives because of social
media’s value in advancing neoliberal capital’s agendas and in empower-
ing surveillant, carceral states. In this chapter, we ground our analysis of
the phenomena of hate-filled trolling, disinformation and misinformation in
the powerful narratives of a cross-section of Brazilian citizens whose work,
identities or politics have ‘resulted’ in their experiences of extreme trolling,
racism, misogyny, homophobia, death threats and/ or physical attacks and
violence. Before diving into these cases, however, we provide our readers
with some basic historical background.

Histories of violence: Colonialism to the present

Prior to the 16th century Brazil was inhabited and controlled by Indigenous
peoples, comprising over ten million and distributed between two thousand
tribes, some constituted into discrete nations and occupying specific parts
of the continent, others semi-nomadic, and moving between the coast and
the interior. Struggles for power were common between and amongst the
largest of these groups – for instance, Tupis and Tapuias – who tended to
settle along the coastal regions and in the interior respectively. Many of the
groups contained numerous subgroups grappling for internal dominance.
Following a 1494 treaty between Spain and Portugal which arbitrarily
allocated all territories east of a certain line and west of that line in the land-
mass of South America to Portugal or Spain respectively, from 1500 onwards
Portuguese explorers and traders led by Pedro Alvares Cabral opened a path-
way to the decimation and pillage of Indigenous territories. In the subsequent
 Brazil and contemporary online hate 53
centuries, millions of Indigenous people were subjected to colonial geno-
cide. This happened both directly – through massacres, the sacking of their
tribal homelands, and violent take-over of their land – and indirectly, through
the transmission of deadly illnesses that swept across their populations, the
hunger that ensued, and the suppression of their Indigenous spiritual prac-
tices by swathes of Jesuits and other European Christian priests.
The white Portuguese invaders and colonisers treated Indigenous South
American populations with extreme contempt, deploying mass suppres-
sion, erasure, violence and multiple forms of dehumanisation common
across Belgian, British, Dutch, French, German, Italian, Spanish, and other
colonial systems. Local populations responded by fighting back fiercely,
going underground, doing deals with opposing colonial powers in interne-
cine struggles and, on occasion, intermarrying with or joining forces with
escaped parties of African rebel slaves transported to South America by the
slave trading colonisers.
Despite early lack of interest in their new colony which seemed simply
too unyielding and distant, the Portuguese soon discovered that sugarcane –
a then almost priceless commodity – could be grown in abundance. Thus
began several centuries of slave-trading and plantation slavery to enhance
Portugal’s monopoly in the global sugar trade. We will not dwell here on the
various European and colonial wars for ascendancy that occurred between
the Portuguese, French, Dutch and British over parts of the Brazilian ter-
ritory, and over the slave, sugar, gold, diamond and coffee trades, as these
are all detailed with incredible nuance in histories such as those by Braudel
(1981) and Puntoni (2019). Highly relevant to our volume, however, are the
hidden groups of escaped African slaves, who began to form communities
of quilombos (Anderson, 1996; Ferretti, 2019), gathering strength from the
increasing numbers of slaves transported to Brazil during until the middle
of the 19th century (Brazil was one of the very last countries to officially
abolish slavery in 1888). Although there continue to be contests over the
histories of quilombo descendants in modern Brazil, up to the mid-19th
century these subaltern communities frequently banded together in insur-
rectionary movements fighting off the colonisers, sometimes with help from
surviving Indigenous populations.
The rebels were often brutally suppressed but also never surrendered and
were thus increasingly successful, particularly once the abolitionist move-
ment took hold amongst the European middle classes. Intermarriage and
sex (both coerced and consenting) between African, Indigenous and Cau-
casian European populations was common. New creole languages devel-
oped which mirrored the mixed communities living outside the purview of
the Portuguese crown. Although there are alternate histories which point to
multiple complicities with the white state, many of these mixed communi-
ties continued to operate in the 20th century as focal points for socialist
54 Brazil and contemporary online hate
mobilisation, particularly in urban areas, and for African and Indigenous
spiritual practices that contested the stranglehold of the Catholic church.
In line with Colin Snider (2018), our work takes ‘an expanded
approach . . . that explores the ways in which memories, discourses, and
policies from military regimes continue to shape politics, society, and dis-
course decades after militaries left power’ (p55). Therefore, as we highlight
historical events and processes primarily in order to elucidate the complex
contexts and theoretical underpinnings of hate in contemporary Brazil, we
skip ahead now to the 1960s, which saw the beginning of two decades of
military dictatorship. Supported by Johnson’s administration in the United
States with its virulent anti-communist propaganda, Brazil’s democratically
elected left-wing government was brought down in the coup of 1964 and
a spate of relentless state killings, disappearances and torture began crush-
ing dissent (Chirio, 2018; Schneider, 2011). Unquestionably propped up by
imperialist military support and intervention from the United States and by
the highest ranks of the Catholic church (which shed its earlier institutional
connections to socialist groups), the military dictatorship lasted until 1988.
Democracy only returned to Brazil as a result of almost unthinkable suf-
fering and courageous protest on the part of multiple civil society groups.
Unfortunately, there was no concomitant widespread change to social val-
ues and attitudes to race, gender and violence which had been shaped under
colonial and then dictatorial rule. The wounds and scars from these years
run deep, and influence much of contemporary politics and social life.
Founded in 1980, the Workers Party is the largest socialist or social demo-
cratic party in South America. During the 1980s, when unemployment and
inflation were at an all-time high and violence was rife in the urban factories
and favelas, many grassroots union organisers were threatened, beaten and
arrested for organising protests and strikes. Amongst these community organ-
isers was Luiz Inácio Lula da Silva, popularly known as Lula, who became
the leader of the Workers’ Party and the 35th President of Brazil from 2003
until his former chief of staff, economist Dilma Rousseff took over in 2011.
Lula’s and Dilma’s regimes oversaw some of the most sweeping pro-people
socio-economic reforms in Brazil’s history. These were fiercely opposed by
US corporations and the rightwing politicians allied to US interests.
Ironically, corrupt prosecutions over apparent corruption saw Dilma ousted
in 2016, at which point disagreements over the agenda of the Workers’ party
and its opponents spilled over into cyberspace. When the mainstream media
focused on the highly politicised corruption scandals — foregrounded to
undermine social democratic reforms — and rumours spread on social media,
in quick succession Lula was imprisoned followed by the impeachment of
Dilma and the electoral hobbling of the Workers party. A former member of
the military during the dictatorship, famed for his homophobic, racist and
 Brazil and contemporary online hate 55
misogynist views, Jair Bolsonaro came to power in 2016 on a tide of violent
bigotry.2 While his military service and nostalgia for the dictatorship era are
a matter of public knowledge, his connections to Steve Bannon and US far
right networks are less well known. The misogynist and white supremacist
legacy which brought Bolsonaro to power also includes the vigilante mur-
ders of multiple Indigenous land defenders, women politicians and Black
trans folk, more than half a million Covid deaths, multiple complex sets of
disinformation emanating from the dreaded “Office of Hate”3 and a flood of
homophobic deep-fakes against Bolsonaro’s opponents.
Brazil always has been and remains a ferociously racist society, with
stratification between Black and Indigenous Brazilians and their white
counterparts now at an all-time high, and overlaps between misogyny,
homophobia and racism making certain communities and people even more
vulnerable. While no modern social history of Brazil can ignore the issue of
race, the unfounded romanticist myth of “racial democracy” (the idea that
the three racial groups in Brazil – Indigenous peoples, African and Euro-
pean descendants – enjoy equal rights and that racism per se does not exist)
has dominated many accounts since the work of Gilberto Freyre in Masters
and Slaves. Now challenged both by accounts of Afro-Brazilian lived expe-
rience (Keisha‐Khan, 2004; Sheriff, 2001) and by historical and theoretical
analysis of racial hierarchies and practices (Anderson et al., 2019; de Vas-
concelos, 2019), this myth continues to serve as a salve in outward-facing
international discussions of race relations, but is no longer believed with
conviction amongst affected populations. As Keisha-Khan points out,

Resistance to urban renewal plans in Salvador demonstrates how strug-

gles for urban land rights are a crucial part of engaging in the broader
national and international politics of race. In Black communities in
Brazil and throughout the African diaspora, urban land and territorial
rights are the local idioms of Black resistance.
(2004, p. 811)

Each of these historical threads informs our analysis of the data in com-
ing sections.

Will to power as the root, hatred as the tree

In addition to being professionals who found themselves unexpectedly in
the eye of a political storm (journalists, doctors, fact-checkers) or spiritual
leaders, academics and social activists whose work for social justice made
them targets, our interviewees in Brazil often had intersecting lived experi-
ence of being members of groups against whom hate in Brazil has reached
56 Brazil and contemporary online hate
new peaks – Indigenous, LGBTQIA+, Black, women leaders, political dis-
sidents. Gradually as their stories and words flowed across our screens, the
picture which took shape became clearer and clearer: The legitimisation of
discrimination, dehumanisation and violence by powerful authorities and
government figures is linked to a steep rise in hate crimes and hate speech
online in Brazil. Beatriz Buarque, a journalist, researcher and founder of
NGO Words Heal the World consistently described how the views, val-
ues and behaviours of powerful leaders in government are linked to social
media hate amongst the populace:

[M]ost hate crimes [in Brazil today] are generated by racial prejudice.
Gender comes in second place. In third place, we find homophobic
crimes. On social media, we see a higher incidence of racist mes-
sages, homophobic messages and political-ideological messages, espe-
cially after Bolsonaro’s election. We have a president who reinforces
and legitimises this kind of narrative. Bolsonaro and his ministers are
very active on Twitter and they use this platform to post hate mes-
sages against the left, as if any person who diverged from the gov-
ernment was an enemy. This is their narrative. Besides the ideological
side, there’s also the racist one – the government is composed mostly
of white people. They don’t talk about diversity. They are extremely
homophobic. During Carnival, the president himself reinforced stereo-
types that exist in relation to homosexuals. When authorities legitimise
hate speech, we expect that some people will start reproducing this
behaviour. We are seeing far-right groups proliferating; groups that
actively hate homosexuals. And they don’t even use subtle language.
On Facebook you will find posts defending ‘killing gays and lesbians
in the name of a clean society, in the name of a Christian society’ . . .
African or African-Brazilian religions are frequently attacked. This is
closely linked to racism too. These religions have been historically del-
egitimised, they are often said to be linked to the devil . . . This narra-
tive feeds religious fanatics who are against diversity. So Brazil has an
interconnected society that uses social media to express every desire
and hate, without any fear of punishment. We have the president and
other authorities legitimising these stereotypes and discourses.

Our interviewees and other key informants were uniform in their description
of the communities and individuals most targeted by hate. Sonja Guajajara
who is finishing her second term as Executive Coordinator of a group of Bra-
zil’s Indigenous People (APIB) also confirmed that hate is directed largely at
‘Left-wing groups, Indigenous people, Black people, LGBTQIA+ and women,
with Indigenous people and Black people leading this ranking... people tell me
I’m incompetent because I’m indigenous, that I’m fat, they tell me “go back to
 Brazil and contemporary online hate 57
the jungle because that is your place”’. Although they gave different accounts
of the role of platforms and technology in facilitating this, there was agree-
ment about the root causes and perpetrators (see Table 1.1 in Chapter 1). As in
India, the situation in which discrimination, hate speech, dehumanisation and
online incitement to violence appear to carry no social stigma for perpetrators
let alone any criminal proceedings, and are sanctioned by members of the rul-
ing party, has led to ever increasing attacks.
Gilberto Scofield, a journalist who has worked for the major news organi-
sations such as Globo and now manages a well-established fact-checking
agency told us that the attacks against him were deeply personal. The attackers
use every aspect of his private life to delegitimise his public role and political
stances or interventions. The less visible he tries to be as an individual in order
to assist in the work of his organisation, the more he is made the personal
focus, with accusations of being an LGBTQIA+ activist, with details of his
private life publicised in malicious and destructive ways by organised trolls.
His attitude now is not to hide anything in his online presence. Nor does he
publicise anything about his identity online. He tries to exist and retain his
integrity, regardless of the vitriol:

I’m gay. I’ve been out of the closet my whole life. These hate posts
always say I’m an LGBT activist. I don’t consider myself an LGBT
activist, I consider myself a normal person who talks about his personal
life. People talk about their lives, their husbands, their kids and their
career. Why can’t I talk about these subjects too? We have problems
as well as virtues. We’re normal human beings. For most of my career,
I covered business/economic issues – a sector that is extremely closed to
diversity and is very uptight. Journalists who are gay are usually cover-
ing fashion, culture, music. It’s a kind of closet for gay journalists. I met
[my husband] 17 years ago. We got married and we adopted two kids.

However, while the strategy of being open but keeping a lower profile online
might maintain the careful balance of privacy for his children, he cannot
prevent them from experiencing anti-Black racism because of the colour
of their skin, their features. The accounts he gave of their experiences tally
with accounts from the United States (Bailey, 2021; Sethi, 2018) and UK (in
this book), and show the extent of everyday discrimination, microaggres-
sion and prejudice both in the association of Blackness with criminality and
in the association of Blackness with poverty and ghettos.

Racism is a major issue in Brazil. Once, my son was coming back from
school and, when he tried to enter our building, a woman was going out
and she told him he didn’t live there. My husband arrived and asked
what was happening and the lady told him ‘this boy was trying to get in
58 Brazil and contemporary online hate
the building, but he doesn’t live here’. So my husband told her that he
was our son. She felt embarrassed and tried to justify herself. It turned
out she was the one who didn’t live in the building, she was just visiting
someone. Even not living there, she assumed a Black child would never
live in that building. Things like that happen all the time. Once we were
swimming in a pool and I heard a [white] teenager saying ‘this place is
going downhill, there’s a Black kid in the pool’.

Gilberto and his fellow fact-checkers are also targeted precisely when they
are most successful at managing to get hate speech taken down or called
out online:

When you go on Facebook, there are three dots next to every post
where you can click and report that post. One option is to report it as
fake news. Every time someone chooses that option, Facebook sends
that post to us so we can fact-check it. We analyse them based on two
guidelines: How relevant is the person who is posting and how many
times the post was shared. If it’s fake, we tell this to Facebook. They
don’t delete the post, but every time a person sees that content, there
will be a pop-up saying that our agency has fact-checked it and found
out it is fake. Also, they change the algorithm so fewer people can see
the post. Hence, people who want that content to be seen and shared get
very angry; they say it is censorship, they say it’s an attack to their free-
dom of expression. If you think hate speech is freedom of expression,
I’m sorry, it’s not. There are ethical and moral limits for everything in
life.

Indeed, Gilberto’s children’s safety is often threatened simply because of

his work. He has begged them not to go on social media in order to protect
them from further exposure to the most vile abuse and also to keep their
images from being circulated and doctored by fascist trolls.

The right uses disinformation as a strategy. The groups that support the
government using bots are very active, they are professionals. Some
don’t necessarily share the ideology, they are being paid to do that.
So we avoid posting photos showing where we live, we avoid posting
family photos. We are careful not to say where our kids go to school.
Attacking our loved ones is part of their strategy. Some people send
emails saying ‘I know where your kid goes to school’ – this is more
common than you imagine, it happens all the time . . . There’s another
fact-checking agency that live fact-checked Bolsonaro’s speech at the
UN and they are still being attacked for that.
 Brazil and contemporary online hate 59
Another well-known journalist and fact-checker anonymised here (at her
request) as Antonia P, has been forced into exile after being targeted by
Bolsonaro’s regime. Asked what kinds of messages she and her female
colleagues received, Antonia went on to name others who had faced such
attacks and to associate the attacks with any attempt on the part of women to
take a role in Brazil’s public sphere either professionally or through social-
ist politics:

They [the far right] call you a hooker and worse. It is always an attack
against our bodies, our existence, our mental capability. It’s always
linked to sex, like what happened to Patrícia Campos Mello,4 Miriam
Leitão, Vera Magalhães . . . It doesn’t happen only to journalists. Maria
do Rosário, Manuela D’Ávila,5 the list of women who’ve been through
this is endless. . . . There was an account with a statue in the profile
picture that sent me an inbox message every day, saying ‘I’ll shoot you
in the face, you whore’. It’s always a statue, a comic strip, but it’s not a
bot because they react to reality, they talk.

Antonia’s verdict: ‘ “Techno-populism” is highly responsible for online

hate’ did not differ much from the narratives we were hearing in other coun-
tries, except in its insistence on the culpability of the left too in the growth
of misinformation. Another interviewee, Mariana (a pseudonym, as it is not
safe for her or her family if we use her real name), reiterated that misog-
yny becomes a blunt weapon of the far right in inciting hate and violence
towards liberal or leftist women who comment on politics. When the integ-
rity and courage of her work – which involved critical engagement with
the public sphere and evaluations of the accuracy or mendacity of political
communication – initially triggered an avalanche of hate speech and incite-
ment against her online, much of that hate took the form of misogyny:

They cursed me – the usual social media behaviour . . . things started

escalating. I received messages mentioning my son; telling me that, if
I wanted my son to be safe, I should leave the country. I reported it to
the police. A person called my mobile saying I was going to get punched
in the face; my schedule was shared in WhatsApp groups as an encour-
agement for people to confront me face-to-face. Fake news about me
was also being spread; people were sharing a doctored photo . . . People
recognised me in the streets . . . We were afraid someone would do
something physical, so [my work] hired me a security guard. . . . People
started spreading horrible memes of me, indescribable memes; they put
my face on a naked woman with her legs open, with the most disgust-
ing and pornographic subtitles; Another one was a naked woman with
60 Brazil and contemporary online hate
my face and a line of men waiting their turn. I also received messages
saying I deserved to be raped. It was so overwhelming.

Mariana’s ability to survive a routine dose of trolling which she describes

as ‘the usual social media behaviour’ suggests that we need to distin-
guish between different forms of trolling. It is the multiple forms and
sites of aggression, the threats, the formal and substantive connections
between politicians and vigilante publics (Banaji, 2018), that makes the
fascist social media sphere of hate quite distinct from some of the early
troll behaviour from individuals and groups on message boards in the
1990s and early 2000s, and connects it to older and more organised forms
of Nazi, Fascist and white supremacist action and propaganda from pre-
Internet days.
Another interviewee, philosopher and anti-racist activist, Djamila
Ribeiro has authored three books on racism and Black feminism, and writes
a weekly newspaper column. Djamila is very active on Facebook as well as
on Instagram, where she has a million followers. Since she posts frequently
about violence against women, abortion and racism, she told us that she is
used to getting hate comments and messages, and has come to expect them,
getting attacked primarily by the right but even on occasion by those osten-
sibly belonging to the left. This year, she was attacked by a leftist woman
who accused her of being against the working classes after she participated
in a sponsored post for a taxi company. She was attacked on Twitter (even
though she doesn’t have an account) and on other platforms. Someone sent
messages to her 15 year-old daughter’s phone saying: ‘Your mother is a
disgrace! Aren’t you ashamed of being her daughter? We know where your
mother lives! There is no way out!’, at which point Djamila reported the
case to the police but to no avail.
Along with other activist organisations, she also brought a lawsuit against
Twitter arguing that the platform benefits economically from racist and
misogynist attacks. While she was clear with us that the structure and pro-
tocols of spaces like Twitter and Facebook make them ‘toxic environments’
for Black women in particular, she also emphasised how the takeover of
politics and mainstream media by the right and far right were fuelling over-
lapping forms of hate and violence:

Since the coup against president Dilma, the polarisation has intensi-
fied. It was partly encouraged by the hegemonic media that has defi­
nitely contributed to the polarised discourse. But I think it can be
traced back to the criminalisation process imposed against the Labour
Party.6 Now, the current government encourages hate against other
parties’ members, against people who don’t share their ideology and
 Brazil and contemporary online hate 61

Figure 3.1 Sample of hateful material received online by Djamila Ribeiro. Credit:
Djamila Ribeiro.

against intellectuals. There is a clear anti-intellectualism. So there is

this political side of it, attacks against left-wing politicians. Manuela
D’Ávila,7 for example, has been victim of several attacks . . . These
spaces are not detached, since these are structural discriminations;
therefore, they structure all social relations. Social media platforms
are also spaces where hate speech is spread against historically dis-
criminated groups.
62 Brazil and contemporary online hate
A repeated theme in our interviews both in Brazil and elsewhere was that
historically marginalised and discriminated populations, who have faced
vigilante violence and atrocity from both state and non-state actors over the
decades, also bear the brunt of hate online. De Vasconcelos summarises the
multiple overlaps of erasure, violence, exclusion, inequality and discrimina-
tion when it comes to Brazil’s African descendant population, and connects
current circumstances to the history of slavery:

According to Brazil’s National Statistics Institute (IBGE), 53% of the

population identify as Black or mixed. According to the Atlas of Vio-
lence 2018, the Black homicide rate was more than double that of non-
Blacks (40.2% versus 16.0%). A Black person is murdered in Brazil
every 23 minutes. According to Oxfam International (2017) . . . white
Brazilians earn twice as much as Black Brazilians. Between 2003 and
2013, the number of Black women murdered rose by 54%, while the
white femicide rate fell by 10% . . . . Only 10% of Brazilian books
published between 1965 and 2014 were written by Black authors. The
Face of National Cinema . . . (UFRJ), revealed that of all Brazilian film
directors, only 2% are Black men, and none are Black women. The
racial gap also exists among writers, where only 4% are Black. Of all
the films and novels analysed, only 31% had Black actors in the cast, in
which they commonly play roles associated with poverty and crime . . .
In Brazil, slavery persisted for a longer period of time and over a larger
geographical area than anywhere else in the world. Forty percent of
enslaved Africans sold between the 16th and 19th century landed there.
The presence of Black people influenced customs, language, religion
and culture. For almost five centuries of Brazilian history, the Black
community has been marginalised.
(2019, pp. 1–2)

Whatever efforts have been made by progressive governments to redis-

tribute resources, their overall impact has not been the kind of decolonial
approach that would radically alter such statistics. And, although the Black
Lives Matter movement has seen a surge of support in Brazil, community
organisers and social justice activists are also in contention with one of the
most racist regimes in living history. The tentacles of this racism extend into
every area of life, including pedagogy and religion.

A war between narratives: Colonial dehumanisation

versus liberation
Afro-Brazilian religions face daily racism and intolerance in Brazil. Catho-
lic priests persist in trying to delegitimise them, linking them to witchcraft
 Brazil and contemporary online hate 63
and the devil. Bolsonaro-aligned evangelical churches openly preach against
them, often urging evangelical congregations to act violently against them.
Some studies show the connection between evangelical churches and drug
militias,8 which makes it even more dangerous for Afro-Brazilian religious
centres to remain open. Further emphasising the connections between vio-
lence against African descendant individuals and communities in Brazil
with religious discrimination, our interviewee MER was a religious prac-
titioner of the Afro-Brazilian religions Candomblé and Umbanda which
gained popularity amongst autonomous communities in the 19th century.
Leader of a religious house in São Paulo and holding a PhD in Theology
with a focus on Afro-Brazilian religions, MER explained to us her position
and activism as a white ally in the fight against religious racism. Evangeli-
cal Christian leaders are increasingly occupying political space; they repre-
sent one of Bolsonaro’s main support pillars. Their official sanction by the
government makes it easier for them to spread hate against Afro-Brazilian
religions and cultures. Sparked by her postings on these topics on social
media, MER has received hate messages calling her a whore and a charla-
tan. She reported only one of the posts to the police.
MER was forced to postpone her interview with us when someone painted
a swastika on the wall of her residence. Other Afro-Brazilian religious cen-
tres have been attacked and even burned to the ground. MER’s explanation
of her work sheds further light on the divides in Brazilian society:

I’m both Umbanda and Candomblé. Modern society is used to think-

ing based on written tradition; our world conception is rational and
systemic – there is a beginning and an ending. Umbanda and Candom-
blé are not based on written tradition and, therefore, they don’t fol-
low that linear mindset. They also don’t have a central power, they
are polycentric. Thus, Afro-Brazilian religions are very diverse. How-
ever, there is some common ground, there is a shared skeleton. So they
have a generational transmission and they are based on oral tradition,
which doesn’t mean that we don’t have a secondary written tradition.
We do have books, but they are not the main axis. One of their central
characteristics is religious trance. We value our ancestral memory, so
it is really important to talk about ancestry and to experience it dur-
ing our rituals. We are based on a circular time, a mythical time. It’s
not like Christianity that you can say it was invented 2,000 years ago.
We acknowledge two worlds that coexist: A human world and a super-
natural world that intervenes directly in our lives. The divine is not
outside people, we incorporate the divine, and we go into a trance with
the divine. Simplistically, this is the best way to explain Umbanda and
Candomblé . . . . I don’t post my rituals, but since 2018 I tend to post
the opening speeches I do before the rituals. These are 10, 15 minutes
64 Brazil and contemporary online hate
videos in which I talk about spirituality. Every time I talked about Exu,
I got a lot of criticism but also a lot of sympathy . . . People tend to
link Exu to the devil. Along with the colonisation process, there was
also colonisation of religious values. So Exu was linked to the devil,
this is still present in people’s imaginary. When the pandemic came
and I started posting more often, there were people calling me macum-
beira9 trying to offend me, but I would just joke and say ‘I love being a
macumbeira!’. . . . I avoid using key words because of the bots, I never
say ‘Bolsonaro’, ‘hate’. When we talk about Black people, poor peo-
ple, we get a lot of attacks. But I still talk about that. Instead of saying
‘hate’, I talk about peace, culture and otherness, empathy. When you
say ‘Afro’, ‘Umbanda’, ‘Candomblé’, ‘Exu’, you trigger these people.

Many of MER’s postings are popular and receive praise and solidarity. She
has a following across the globe. So, for her, it’s imperative not to get fright-
ened off, but to continue engaging and posting content, despite the risks.
Commenting on the vicious animosity towards Afro-Brazilian religious
beliefs and communities, MER explained:

The main reason is eugenics. Black people were objectified in order

to be enslaved. So their culture didn’t have any value because they
couldn’t be seen as human beings with values. The idea that everything
that comes from either Black or Indigenous culture is negative is still
very present in Brazil. . . . When we talk about evangelicals, we are
talking about a process of white supremacy including a high number
of Black people who have converted to evangelical religions. Studies
show that there’s a mentality that Black people feel less Black when
they are part of an evangelical church. . . . There is the idea that we have
to evolve, I can’t stand that. I tell these people that this is our choice,
I didn’t choose this religion because I didn’t have other options. This
idea is completely connected to religious racism: The idea that Black
people’s culture is primitive. Evangelicals are mostly fundamentalists
and they want to silence our voices, so they attack us. We are also
attacked because of political reasons when I talk about eugenics, homo-
phobia, gender and racial issues.

So, alongside the new populist rhetorics of the far right, the complicity of
the church and the criminalisation of dissent, it is clear that the majority of
those who face the worst violence for their social justice and pro-democracy
activism share overlapping histories of marginalisation or oppression, and
intersecting demographic characteristics being either LGBTQIA+ and/
or Black and/or Indigenous in terms of their presentation and cultural or
 Brazil and contemporary online hate 65
spiritual identifications. Like MER, Djamila had also initially gone online
to connect with communities with whose experiences she felt solidarity and
with whom she shared common praxis. She explained the trajectory of her
own online practice – from blogging to social media – and the daily barrage
of hatred that she faces as a consequence:

I started writing for a blog for Black women, in 2013. It was a very cool
website that promoted meetings with Black women and encouraged us
to write about them. In 2014, I started writing for Carta Capital maga-
zine, so I started posting my columns on social media. At first, I was
scared when I saw the kind of reactions some of my columns generated
on social media. . . . If the post is about violence against women, I get
many aggressive comments, mostly from men. At the same time, these
posts also engage people who are more aware and on our side. . . . when
I talk about sexual violence or rape, men feel more comfortable to
attack me. I think this is a curious aspect. They feel more comfortable
to say that it is a lie, that women overreact. About racial issues, there is
the myth that Brazilians are not racists, so they are more ashamed to do
that. Racist attacks happen, but they are more restrained.

In Djamila’s view the complexities of people’s overt reactions to her femi-

nist and anti-racist postings are reflective also of their beliefs about Brazil
as a society, of their socialisation into spaces of violent male dominance,
of Bolsonaro’s encouragement of anti-intellectualism and of the demonisa-
tion of Brazil’s left movements. One particularly pernicious aspect of being
harassed and violated on social media remains that those spaces for thought
and debate amongst marginalised communities which were welcomed and
often saved or still save lives, giving people the strength to resist and to
organise in ways that mainstream media has long failed to do, have now
become sites of further trauma and anxiety.
Another interviewee, Wari’u Tseremey’wa, who is Xavante from Mato
Grosso in the Central-West region of Brazil, uses emerging technology and
graphic design to challenge and rebalance misrepresentations of Indigenous
cultures. He uses his personal knowledge and experience of Indigenous pol-
itics to correct the preconceptions of those he comes into contact with. In
2018, while still a teenager, he started the YouTube channel ‘Wariu’ where
he posts videos explaining what it means to be Indigenous in today’s Brazil.
Patiently and in a calm, soothing tone of voice, he demystifies preconceived
ideas that non-Indigenous Brazilians hold about indigenous people10:

I am from the Parabubure Indigenous territory, near Campinápolis, in

the state of Mato Grosso. . . . I am Xavante mixed with Guaraní, but
66 Brazil and contemporary online hate
the Xavante culture doesn’t allow any mixture. . . [This] means that,
if the Dad is Xavante, the descendant will only be Xavante. My Mum
is Guaraní, but my name is Xavante and the cultural traditions I fol-
low are Xavante. It’s the kind of protection we have . . . to avoid cul-
tural loss. We do have relationships with other Indigenous people, even
blood relationships, but we keep our culture intact. This is not a gen-
eral Indigenous rule; the Xavantes were always ‘war people’, so these
strategies exist to protect us, our culture, our people. I belong to one
of the most traditional Xavante families, my ancestors were important
in Xavante history; even today, my father and I still make history, in a
way. My father is the president of the Federation of Indigenous People
and Organisations of Mato Grosso (Fepoimt) . . . . As an Indigenous
communicator, I currently do communications for the Federation of
Indigenous People and Organisations of Mato Grosso, where I organ-
ise events and contribute to the national Indigenous movement. We
are always in contact with Apib (Brazil’s Indigenous People Articula-
tion). That is an Indigenous social and political movement that is often
ignored.

Many of Wari’u’s postings on YouTube have drawn positive comments and

interactions, with some Brazilians ‘surprised’ to see a young Indigenous
person explaining their culture so calmly and clearly, and others joining
him in critique of Portuguese colonial mindsets. There has also been hate,
particularly on Twitter, which is a platform that he rarely uses because of
its lack of enclosed community. Wari’u’s response to hate received online is
tempered by his constant experiences of dehumanisation and racist stereo-
typing since his earliest years:

I have 25,000 followers [on YouTube] and almost 300,000 views. On

Instagram, I have 14,000 followers. . . . I don’t usually feel so bad with
online comments because I’ve already heard terrible things being said
to my face. Sometimes people send comments, but I just think ‘this
person is such a coward hiding behind a screen’ . . . . Once, a teacher
told me to “go back to the forest”. She had a joking tone of voice, she
said it was only a joke. . . . Sometimes online hate is not very clear for
me because I’ve faced hate in person. People were racist in front of
me, shamelessly. On Twitter, I’ve already received this kind of com-
ment, but I see myself in control of my content. In order to explain my
behavior, I need to tell you a bit about my culture. Xavantes are taught
to resist since they are little. Our rituals test us. In a rite of passage, you
spend one month in cold water, having water thrown on your ears. After
this, they pierce your ears with a jaguar’s bone. In another ritual, they
 Brazil and contemporary online hate 67
give us a borduna11 and we need to take care of it as if it was our baby.
So the rituals test us physically and psychologically. . . . When I think
it is hard now, I remember it was harder back when I was a child. . . .
The country’s president defends this old colonial narrative. . . . On one
of the videos I posted on Instagram, I poked the Portuguese a little bit.
Brazilians joined us Indigenous people to criticise the Portuguese. It
was like ‘post a moment when everything went wrong’ and I showed
an image of Portuguese boats arriving in Brazil. The Portuguese have
a perverted idea of Brazil’s colonisation. They learn in school that they
were the heroes, it’s shocking. So Portuguese people commented ‘if
we hadn’t colonised you, you wouldn’t be where you are today’ . . . .
This colonial narrative has always imposed itself, so when we bring
our Indigenous narrative, it bothers some people. And they attack us
to protect their narrative, they have their own interests. All informa-
tion has an underlying bias. This is the Indigenous fight now – to show
our side of the narrative. The other narrative justifies people taking our
lands, justifies a development that can’t really be called development,
it’s a destructive development.

In this fascinating account, a new form of resistance to colonial and rac-

ist dominance is outlined in the way of life of Xavante and of this young
representative of the community. The mental tenacity cultivated in order to
survive, to struggle against ongoing attempts at erasure and dehumanisation
enables Wari’u to deal placidly with hate speech or belittling and stereo-
typing online. While complex intersections of Indigenous, Afro-Brazilian
spiritual, Black, LGBTQIA+ communities and women are longstanding
targets – experiencing hate and harassment in venues both real and vir-
tual regardless of their professions – some individuals find themselves in
entirely uncharted waters.
Marcus Lacerda, a doctor and researcher specialising in infectious dis-
eases, based in Manaus, capital of the state of Amazonas, in the North of
Brazil, led a clinical research trial with Covid-19 patients. His team’s first
finding was that high dosages of chloroquine were dangerous for patients.12

When the pandemic came, I started posting videos explaining aspects

of the disease and people liked it, they thought it was important at that
moment. So I got many followers on Instagram. . . . Our study ended up
being the first randomised clinical trial with chloroquine in the world.
Its visibility increased because it was the first one. The first publica-
tion was a preliminary one; we had 81 patients at that time and it was
already clear that the highest dosage we tested had toxicity problems.
So we thought it was necessary to publish quickly and tell the world
68 Brazil and contemporary online hate
that a high dosage isn’t safe. Our idea was to block the studies that
were being conducted with high dosages; there were doctors all over
the world using these high dosages. So we published this preliminary
report and continued the study with lower dosages, which had proven
to be at least safe . . . We had to interrupt most of these studies when the
Ministry of Health started recommending the use of chloroquine. . . .
Our conclusion was that the use in high dosages had an important
toxicity.

At the outset the team had hoped to save lives. Instead, they suddenly found
themselves objects of national suspicion and hatred. The day after they pub-
lished their first findings, Eduardo Bolsonaro, the president’s son, tweeted call-
ing the researchers murderers and saying they were from the Worker’s Party
(Lula’s party). Marcus immediately received hate messages on Instagram and
had to acquire armed security. His team was threatened and as the controversy
grew many researchers became depressed and were unable to work. They con-
cluded the study sooner than planned but suffered the consequences of having
been framed by Eduardo Bolsonaro and government supporters as anti-national:

Most of the profiles that attacked us were not fake, they belong to real
people. [I received messages like]: ‘are you the doctor who killed those
people?’, ‘we are watching you, we are watching your family’, ‘we are
going to make you and your family suffer like you did the patients’ fam-
ily suffer’. It was really hard because we didn’t know if this would lead
to physical aggression. I talked to the police and to some lawyers . . . I
was escorted by a policeman hired by the Amazonas government for two
weeks. . . . my wife panicked because I have three small kids, so we never
know if something will happen, if this hate will become tangible. Up until
a month ago, when the media talked about this subject again, some peo-
ple from other states sent me messages saying I was going to burn in hell,
calling me a ‘communist son of a bitch’, saying ‘you’re going to end up
like Marielle13’ – these people have no idea how to do research, they don’t
know me, they don’t know anything. We reported it to the police depart-
ment that investigates cybercrime and to the state’s Public Prosecutor’s
Office, but nothing has been done so far. [Speaker’s emphasis]

Psychological damage from defamation, threat of violence and legal harass-

ment is not easy to quantify or overcome. Several interviewees detailed the
efforts that they go to in order to reassure friends and families, even while
suffering panic attacks and depression themselves. Other repercussions
included the destruction of the reputations of team members and repeated,
targeted harassment by law enforcement and judiciary loyal to Bolsonaro.
 Brazil and contemporary online hate 69
Pro-Bolsonaro prosecutor Bento Gonçalves14 started investigating us;
they published it on social media before I was even given the inquiry.
They did it publicly in order to harm us. They summoned the research-
ers, we had to hire a lawyer and answer many questions. . . . Bolsonaro
said that more people were dying in Manaus because we had a protocol
to use high dosages of chloroquine. He wanted to convince everyone
that the high dosages were not used only in the 40 patients we were
studying, but in every patient in Manaus, that the deaths were caused
by our study. He said it and hoped it would stick. He also said ‘accord-
ing to what I was told, the doctor is affiliated to PT, but I’m not going to
comment on that’. It wasn’t only his son’s tweet, Bolsonaro said it. . . .
The president’s supporters started to report us everywhere; it is a kind
of repression that is extremely organised. The prosecutors from Bento
Gonçalves published their inquiry online, so another prosecutor from
Goiania copied the same text and opened another investigation. Then
a Congressman from Amazonas reported us to the Federal Council of
Medicine. These investigations are endless.

The conjunction of political and legal harassment and online hate – the use
of the state apparatus to support the misinformation and hate-speech against
the teams of doctors and medical researchers attempting to win the fight
against Covid-19 – was potent both as a warning to other scientists and in
silencing critique of the government’s laissez faire policies and reliance on
conspiracy theories. While this case is somewhat unique, similar patterns of
harassment are visible in India in the case of doctors such as Kafeel Khan
and journalists or clinicians who question establishment lies or blow the
whistle in the medical field.

What is to be done?
As in India and the UK to which we turn in Chapters 4 and 5 respectively,
Brazilians’ strategies for dealing with the affective burdens placed on indi-
viduals and families targeted by hate varied significantly from trying to
respond with love to hate, trying to respond reasonably to hate, getting
one’s supporters to report or respond to hateful comments, trying to ignore
hateful comments, systematic troll blocking, complaining to platforms,
leaving platforms, complaining to the police, leaving jobs, quitting blogs,
and disengaging from platforms because they are unsafe. In some cases
our respondents who could manage to organise and afford this had been
forced into exile to protect themselves and their families. Almost everyone
we interviewed in Brazil who could afford counselling and therapy had
been forced to avail of it in the aftermath of particularly heinous incidents
70 Brazil and contemporary online hate
of online harassment and incitement against them, their families and their
communities. While the online vitriol and abuse appeared to be a trig-
ger for increasing levels of anxiety and depression, most expressed the
view that their group identities or personal identities were almost always
under threat or stress due to ongoing political circumstances regardless of
their online presence. Going offline or leaving a job were simply stop-gap
measures aimed at shielding their loved ones or selves from immediate
violence.
All of these strategies are also clearly mediated by people’s theories
about why hate happens, its linkage to histories of violent discrimination
and the clear and present danger to our interviewees and their families as
well as the responses of platforms and law enforcement to their predica-
ment. Acknowledging that in the medium and long-term there is a need for
profound social change akin to a revolution in attitudes to gender, sexuality,
race, religion, caste, class, disability and other protected characteristics, we
engaged deeply with all our interviewees about ways forward in the short-
term. Hate costs health, mental health, livelihoods and lives. In response
we heard time and again about the frustrations incumbent on being good
citizens and following policies and procedures to report hate:

I reported a tweet for using my photo three months ago. One month
ago, Twitter replied asking for my ID to prove who I was. I called
Twitter’s press office to ask if this was serious or if it was phishing. It
was serious. Yesterday, they blocked only that one tweet. The person
who posted was angry with my organisation because they fact-checked
something and he thought I was responsible for it, so he posted a doc-
tored photo of me . . . that took Twitter three months to act. It’s neces-
sary to improve the system, being very careful not to be partial, since
it’s not always easy to define what hate is. This tweet wasn’t blocked
for hate, but because they used my photo. Social media platforms have
to be more agile in acting on hate speech, they need to hire more peo-
ple to work with this goal, they need better definitions and criteria. In
second place, this subject needs to be discussed in schools. I am in my
40s, my generation is gone! My daughter is 11, we need to start talking
about hate speech in the classroom.

Education, and media education in particular, was brought up several times

as a medium-term solution. However, it was also seen to have its limits, and
to be curtailed by existing regimes of discrimination and domination which
require quite different types of action:

We should educate people, but it’s important to have a system that will
punish perpetrators. If you type ‘Hitler’ on Facebook, you will find
 Brazil and contemporary online hate 71
many pages worshiping Hitler and genocide. It’s not subtle. This is a
crime, these people need to be punished. Education won’t solve the
problem concerning those people who are already using social media
to spread hate speech. . . . We have a project to build a Stereotype
Guide: the students are writing this guide to send to journalists in order
to encourage them to stop reproducing stereotypes. We have a group
that is finishing the 2019 Hate Crime Report. In the UK, we’re going
to work with students . . . on a campaign against fascism, explaining to
people fascism’s characteristics, and also a campaign to stop spreading
hate. In Latin America, we are working on a report about femicide.

Several of our interviewees expressed deep frustration, hopelessness

and overwhelm. Mariana was suspicious of new laws to curb online hate
speech, arguing that they would be utilised in corrupt ways to curb the free
speech of human rights defenders and critics of authoritarian behaviour.
Weary as she was of being forced into hiding her identity, being threatened
with violence, having images of her sexuality and body constantly used in
demeaning and dehumanising ways, and having her child threatened, she
was sceptical of change while the ones wielding the power over courts and
laws remained unchanged.
Yet others whom we interviewed had only recently faced hate in this
form, since their lives had otherwise been protected by unexamined privi-
leges of race, gender and class.

Before, I didn’t believe in hate speech and in the existence of an “Office

of Hate”.15 I always thought it was a bit of left-wing fantasy. But it is
in fact really organised, I experienced it. The impact is huge, people
still haven’t realised it. They understand what this means when they
become the victims. Disinformation makes it even harder for people
to understand scientific research. . . . I’m very pessimistic about the
future. Hate speech is here to stay. I can’t see a way to neutralise it. . . .
On WhatsApp, there is an excess of freedom, of sharing, nobody con-
trols it. When you share content, you don’t know exactly what it is and
where it is going to go. You’re a cog in a much larger process, someone
is overlooking it. I’m going to be honest, I’m afraid hate speech will
transform our society in a bad way, make it worse.

Others told us that their solidarity overrode potential threats:

My children – blood related and of saint – are really afraid some-

thing will happen to me, especially now. They are really afraid. . . .
We reported it to Decrin,16 in Brasília. We don’t have a special police
department dedicated to these crimes where I live; one of my projects is
72 Brazil and contemporary online hate
to implement policies that protect the terreiros in this area . . . we don’t
know if our case will have any result. We’re thinking about suing this
person in a civil procedure as well. The attack was public, it was made
in a comment in one of my posts on Facebook and everybody saw it.
Yesterday, someone drew a swastika on my wall. [But. . .] I can’t stop.
I won’t be silenced, I can’t be indifferent to what is happening. My
indignation is stronger than my fear.

Conclusion
Throughout this chapter we noted the insistent relationship between the
past and the present, between seething prejudice and structural discrimi-
nation, between sanctioned disinformation and the rise in hateful trolling.
Histories of colonial repression, imperialist aggression, media manipula-
tion and religious alignment with European and US-based churches all
converge in the accounts given by our key informants and interviewees of
racist, anti-Indigenous, anti-leftist, homophobic and misogynist bullying,
harassment, trolling and aggression. On the streets and online, in grocery
stores, churches, sidewalks, newsrooms, chatrooms, message boards, mes-
saging apps and on platforms, dehumanising language and imagery, and
symbolic representations of lynching and death, run in parallel with each
other. Common denominators mentioned by everyone were the legitimacy
lent to hate by powerful politicians and religious leaders, the complicity
or co-optation of mainstream media, and the profit-seeking complicity and
inadequate response of corporate platforms. The common political theme of
this stark spectacle of hate was a push to silence and make invisible people
from Black and Indigenous, LGBTQIA+, feminist and leftist communities,
in order to continue to dominate and oppress them, and to profit from their
suppression. Their refusal to be silenced, their determination to continue to
stand for justice, and to be open about their identities in the face of struc-
tural injustice, discrimination, pain, anxiety, family pressure, job loss and
other dangers was the single most inspiring finding of our research. As will
be seen in Chapters 4 and 5 on India and the UK respectively, this courage
and determination is something that sustains not just community organisers
and activists but also entire communities in the face of hate.

Notes
1 https://theconversation.com/how-jair-bolsonaro-used-fake-news-to-win-
power-109343
2 www.theguardian.com/commentisfree/2018/oct/06/homophobic-mismogynist-
racist-brazil-jair-bolsonaro
 Brazil and contemporary online hate 73
3 https://brazilian.report/power/2020/10/24/against-vaccine-bolsonaro-son-reac-
tivates-office-of-hate/ and www.zdnet.com/article/fake-news-probe-in-brazil-
exposes-office-of-hate-within-government/
4 https://advox.globalvoices.org/2018/10/28/brazilian-journalists-face-hacking-
doxxing-and-other-threats-as-election-draws-near/
5 Left-wing politicians. Manuela D’Ávila was the vice president candidate on
Fernando Haddad’s ticket.
6 Worker’s Party – Partido dos Trabalhadores (PT).
7 She was the vice presidential candidate alongside Fernando Haddad and a mem-
ber of the Brazilian Communist Party who also ran to be Mayor of Porto Alegre.
8 https://theconversation.com/in-brazil-religious-gang-leaders-say-theyre-wag-
ing-a-holy-war-86097
9 A pejorative way of referring to Afro-Brazilian religious followers. It’s similar
to calling someone a ‘witch’.
10 https://observers.france24.com/en/20190211-brazil-indigenous-youtube-com-
bat-racism-12
11 An indigenous weapon.
12 www.sciencemag.org/news/2020/06/it-s-nightmare-how-brazilian-scientists-
became-ensnared-chloroquine-politics
13 Marielle Franco, Rio’s city councillor who was murdered in 2018.
14 A town in the state of Rio Grande do Sul, in the South of Brazil, more than 4,000
kilometres away from Manaus.
15 ‘This toxic environment has been fomented by what Brazilians call the “office
of hate,” an operation run by advisers to the president, who support a network of
pro-Bolsonaro blogs and social media accounts that spread fake news and attack
journalists, politicians, artists and media outlets that are critical of the presi-
dent. The office of hate does not have an official title or budget – but its work
is subsidized with taxpayer money.’ – www.nytimes.com/2020/08/04/opinion/
bolsonaro-office-of-hate-brazil.html
16 Special police department that investigates crimes motivated by discrimination
against race, religion, sexual orientation, and against elderly or disabled people.

References
Anderson, R. (1996). The Quilombo of Palmares: A new overview of a Maroon
state in seventeenth-century Brazil. Journal of Latin American Studies, 28(3),
545–566.
Anderson, W., Roque, R., & Ventura Santos, Ricardo. (2019). Luso-tropicalism and
its discontents: The making and unmaking of racial exceptionalism. Berghahn
Books.
Bailey, M. (2021). Misogynoir transformed: Black women’s digital resistance. New
York University Press.
Banaji, S. (2018). Vigilante Publics: Orientalism, Modernity and Hindutva Fascism
in India, Javnost - The Public, 25(4), 333–350.
Braudel, F. (1981). Civilization and capitalism, 15th–18th century. Fontana Press.
Chirio, M. (Trans.). (2018). Politics in uniform: Military officers and dictatorship in
Brazil, 1960–80. University of Pittsburgh Press.
74 Brazil and contemporary online hate
de Albuquerque, A. (2017). Protecting democracy or conspiring against it? Media
and politics in Latin America: A glimpse from Brazil. Journalism, 20(7), 906–923.
de Vasconcelos, A. V. S. (2019). The persistence of racism and its impact on the
Afro-Brazilian culture. Researchgate.
Ferretti, F. (2019). Decolonizing the Northeast: Brazilian subalterns, Non-European
heritages, and radical geography in Pernambuco. Annals of the American Associa-
tion of Geographers, 109(5), 1632–1650.
Keisha‐Khan, Y. P. (2004). The roots of black resistance: Race, gender and the
struggle for urban land rights in Salvador, Bahia, Brazil. Social Identities, 10(6),
811–831.
Matos, C. (2011). Media and democracy in Brazil. Westminster Papers in Commu-
nication and Culture, 8(1), 178–196.
Oxfam. (2017). Brazil: extreme inequality in numbers. [online] Oxfam International.
Available at: https://www.oxfam.org/en/even-it-brazil/brazil-extreme-inequality-
numbers.
Puntoni, P. (2019). “The Barbarians war”: Colonization and indigenous resistance
in Brazil (1650–1720). In N. Domingos, M. Jerónimo, & R. Roque (Eds.), Resist-
ance and colonialism: Cambridge imperial and post-colonial studies series. Pal-
grave Macmillan.
Schneider, N. (2011). Breaking the “silence” of the military regime: New politics of
memory in Brazil. Bulletin of Latin American Research, 30(2), 198–212.
Sethi, A. (Ed.). (2018). American hate: Survivors speak out. The New Press.
Sheriff, R. E. (2001). Dreaming equality: Color, race, and racism in urban Brazil.
Rutgers University Press.
Snider, C. M. (2018). The perfection of democracy cannot dispense with dealing
with the past: Dictatorship, memory, and the politics of the present in Brazil. The
Latin Americanist, 62(1), 55–79.
Sponholz, L., & Christofoletti, R. (2019). From preachers to comedians: Ideal types
of hate speakers in Brazil. Global Media and Communication, 15(1), 67–84.
4 Social media, violence and
hierarchies of hate in India

Introduction: The politics of the digital sphere

As in Brazil, many of our informants in India evinced a shared cynicism
about the chances of imminent political or policy change that would alter
the volume, types and reasons for targeted hate and violence. The subset of
Indian social media users whom we interviewed theorise online hate and
discrimination in an impassioned manner. Like us, they too link it to the
histories of postcolonial religious pogroms, poverty, caste discrimination
and gender-based violence. Addressing online hate was, for them, just one
facet of a wider need for action towards economic and social justice. Since
2014, under the aegis of the Bhratiya Janata Party (BJP), a vast swathe of
the Hindu population have weaponised their religious identity (intersect-
ing caste and gender) against overlapping minoritised groups – Muslims,
Dalits, Adivasis, Christians, queer and trans groups – with different forms
of slurs, atrocity and violence directed at men and women in these com-
munities. This weaponisation manifests offline and online in both physical
and symbolic forms.
In India, where mass digitisation has taken place through a series of forced
bureaucratic measures (Bhat, 2020; Ferguson & Gibson, 2015), these inter-
connected domains act on each other in increasingly complex ways, some
of which are addressed in this chapter. By dint of subverting democratic
processes in all but name, the BJP has managed to achieve electoral and
ideological hegemony unlike their political predecessors. The BJP’s poli-
tics is comprised of strands familiar to those studying authoritarian populist
governments across the world – vicious privatisation and neoliberal reform
with capital flowing freely between top corporate and political interests
while the populace becomes ever more destitute. This is accompanied by
the discursive championing of a faux-emasculated nation-state as the moral
horizon set up to justify and underpin all political action, cultural values,
public space and so on – a paradoxical formation of, as Tambiah (1986) puts

DOI: 10.4324/9781003083078-4
76 Hierarchies of hate in India
it, ‘a majority with a minority complex’; the concomitant diffusion of hate
against minoritised groups into the minutiae of daily life; widespread suspi-
cion and paranoia about the Other (as the cause for all real and imagined ills
of society); and deep-seated patriarchy and misogynist control of women’s
bodies and youth relationships at peril of death.
An aspect of the BJP’s dominance is its effective capture of public insti-
tutions, including universities, public health bodies, the electoral commis-
sion, banking and financial institutions, legal and regulatory institutions
and media. Captured through political appointments, co-optation, bribery,
threat, steamrolling legislation through parliament without discussion and
other strategies, these are used to discriminate against, surveille and harass
opponents and dissenters. There is also subservience to and complicity in
the BJP’s politics from the private sector – particularly from the largest
corporations. Increasingly, global social media companies have recently
acquired stakes in or are working closely with dominant domestic multi-
nationals such as the Ambani group, the Adani group, the Tata group and
so on.
A further aspect of the BJP’s dominance is the personal ‘brand’ of Nar-
endra Modi. During his previous stint as BJP chief minister of Gujarat, he
was accused of complicity in the 2002 Gujarat pogrom in which more than
2000 Muslims were killed by Hindu vigilante groups (Ayyub, 2016). Once
banned throughout Europe and North America and refused visas, with the
help of global image management firms, he has steadily consolidated his
political image through a mix of neoliberal politics (acquiring agricultural
land and providing it to large industrial investors along with heavy subsi-
dies) and a hardline Hindu nationalism that has constantly vilified Muslims
(Jaffrelot, 2007). After his entry into national politics, Modi rebranded him-
self as a champion of national security, development and technologised,
corruption-free governance (Chakravartty & Roy, 2015). Simultaneously,
his party members unabashedly used disinformation to vilify Muslims.
They promoted a culture of Hindu vigilantism that led to the public lynch-
ing of Muslim boys and men, and the rape and murder of Dalit and Muslim
women and girls, the beating and murder of Dalit men and multiple hate
crimes. Modi was an early adopter and is an active user of social media. His
supporters and his party use social media heavily for propaganda and to shut
down criticism. They employ trolling, abusive speech, doxing, hacking and
other strategies that we will expand on as we elaborate our interviewee’s
experiences.
Technological networks are embedded in socio-political, cultural and
economic contexts, but also, to a large extent, reproduce and strengthen the
abiding tendences of the contexts into which they are embedded. In the late
2000s, a series of transformative changes precipitated the widespread use of
 Hierarchies of hate in India 77
social media platforms and cross-platform applications, primarily accessed
via smartphone. Since 2015, Indians have enthusiastically taken to Google
(YouTube) and Facebook (Facebook, WhatsApp and Instagram), and, to a
lesser extent, to Twitter, Instagram and Tik-Tok (now banned in India). The
precise number of users is uncertain since financial motivations can inflate
or exaggerate the extent of social media and internet usage. While social
media usage has particularities (for instance, its limitation to at most 40%
of the population, and concentration amongst males and in urban areas)
that help us distinguish it from other aspects of daily life, we argue that this
distinction rests on a thin and blurred line between the online and the offline.
Founded in 1980, the BJP’s ideological roots are in the Rashtriya Sway-
amsevak Sangh (RSS) which was founded in 1925. In less than a hundred
years, the RSS has gone from a small cultural chauvinist initiative begin-
ning with fewer than 20 members into a mass fascist organisation with a
membership above five million. It is no coincidence that the core ideas of
the RSS draw deeply from the ideologies of fascism and Nazism that were
flourishing in 1930s Europe. The RSS has also diversified its organisational
structure to target different communities. For instance, there is a separate
wing for girls and women, a separate organisation for students, for youth
members and so on. Collectively these organisations are called the Sangh
Parivar or the Sangh Family (Andersen & Damle, 1987; Hansen, 1999). The
BJP is the political face of the Sangh Parivar. Narendra Modi was cultivated
and promoted by the RSS and most of the BJP’s senior leaders come from
the RSS. This background is important to keep in mind when we refer to
the BJP or to Modi.

A long history of inequality

For more than 2000 years, much of what is now South Asia has been sub-
ject to the caste system which, far from remaining static, has changed in
multiple ways in response to historical events and movements. The caste
system comprises four broad categories of castes, arranged in vertical hier-
archy, with Brahmins (priests) at the top, followed by Kshatriyas (warriors),
Vaishyas (traders) and Shudras (manual and menial labour). These savarna
groups imagine themselves in distinction to the fifth and outside caste ava-
rna group, commonly known as Dalits. Located at the bottom and outside
the structure, Dalits alongside Adivasis (Indigenous peoples) have suffered
continual discrimination, exclusion and violence. Online too, in what Shan-
mugavelan1 has called ‘caste-hate speech’, Dalits face extensive abuse and
discrimination.
Some aspects of the caste system invite comparison with oppressive sys-
tems in other parts of the world. The most common comparison is race in
78 Hierarchies of hate in India
the United States (c.f. Wilkerson, 2020). While we have provided an abstract
and condensed picture of the caste system, in practice it is far more complex.
Each caste category is in turn comprised of various sub-castes or jaatis, that
are also constantly competing for superiority over various other jaatis in
their caste group. Conflict in the caste system is therefore not strictly vertical
but also to some extent, horizontal (Manor, 2010). The graded inequality of
the system also distributes power in a peculiar way – since a group is incen-
tivised to ally with those above it, oppress those below it, and negotiate with
those who are on par, even if only to maintain the status quo (Gorringe and
Rafanell, 2007). The distinctive aspect of the caste system is that it, unlike
other oppressive systems, links occupations to birth and is considered perma-
nent. In other words, there is technically no control over an individual’s entry
into a caste and there is little way out of the caste you are born into – apart
from the gradual change introduced through jaati politics, inter-caste mar-
riage and religious conversion. The concept of inter-caste marriage is espe-
cially significant since the caste system is essentially a system of ‘exogamy
superimposed on endogamy’ maintained through a control over women’s
sexuality (Ambedkar, 1968[1917]). The caste system is therefore inherently
a patriarchal system and patriarchy profoundly informs how gender identi-
ties are constructed, and refracted, through caste. Much of the current online
discrimination, bullying and hate cannot be separated from misogyny and
the inability to accept any deviation from a dominant caste masculinity that
pervades Indian and wider South-Asian online culture. A caste perspective
arguably enriches the complex intersectionality that underpins these con-
temporary practices of gender struggle and identification.
This does not mean that all conflict, discrimination and violence in India
can be analysed exclusively through a caste perspective. Caste should be
seen as what Bourdieu calls a ‘structuring structure’ (1984, p. 170), a gener-
ative space within which we act but do not always or necessarily experience
as a restriction, but rather develop a ‘feel for the game’ (Bourdieu, 1990,
p. 66). For example, we do not maintain that the identities of all Adivasi
groups, or Indian Muslims are reducible to caste. Some Muslim sects and
groups – such as Sunnis, Shias, Ahmadis, Boras, Sufis and so on – impose
their own sets of divisions and rules. However, the caste perspective provides
a way of deepening our understanding of the distribution of power when it
comes to how Muslims in India are treated. Those who are lynched or face
physical attacks, such as the Delhi pogrom of late February 2020, are more
likely to be Pasmanda Muslims; whereas Muslim political dissenters who
are targeted through draconian legislation are more likely to be from domi-
nant castes. At the same time, the constant reproduction of anti-Muslim-
ness through governmental practices, academic research, media discourse,
landmark events (such as the demolition of Babri Masjid or the Shah Bano
 Hierarchies of hate in India 79
case) produces a clear sense of Muslim identity that can sometimes con-
ceal the intersection of Muslim-ness with caste (c.f. Ansari, 2009). Working
with a caste perspective necessitates an intersectional approach (c.f. Arya &
Rathore, 2020) which can help us situate contemporary problems.

Socio-political contexts and the emergence of online and

social media usage
As in Latin America, the mid-1980s to late-1990s in India were years where
neoliberal structural adjustment policies were forced on successive Indian
Governments by Bretton Woods institutions in order to mitigate global debt
crises sparked by western policies. One of the key objectives of these poli-
cies was to ‘open up’ public sectors to the ‘free market’ and western direct
investment. Governed under the colonial 1885 India Telegraph Act, media
and communications had been tightly controlled by the central government.
After the 1990s, one of the first industries opened to private investment
was telecommunications. Telecommunications was seen as a luxury service
(compared to more ‘basic needs’) by previous governments. Other socio-
political shifts took place in tandem with foreign direct investment.
After the assassination of Congress Party leader Indira Gandhi in 1984,
her son and successor Rajiv Gandhi essayed a series of symbolic actions
pandering to the sentiments of chauvinist Hindus (and in particular, domi-
nant castes) in order to consolidate the fallout from his mother’s regime.
A crucial symbolic act was the regular broadcast of the Hindu epics Rama-
yana and Mahabharatha by the sole public broadcaster in India – Doord-
arshan (Mankekar, 1999; Rajagopal, 2001). In 1990 however, the coalition
government headed by prime minister V.P. Singh announced that it intended
to implement a 27% reservation of central government and public sec-
tor jobs for Other Backward Classes (OBCs), the largest population bloc
(approximately 52–54%). This announcement about affirmative action sig-
nalled a big boost for parts of the population that had hitherto been sys-
tematically excluded from access to education and employment and other
benefits deriving from these fundamental aspects of citizenship. This move
resulted in a major recalibration of electoral politics at the national level.
The BJP, led by their then leader L.K. Advani, mobilised rightwing support
through a deployment of the so-called forward or upper castes in a vicious
and spectacularly orchestrated mosque demolition and temple building pro-
ject that left a trail of violence and death in its wake (Teltumbde, 2005).
In this journey, which utilised multiple forms of media to communicate
its message of exclusion and provocation against Muslims, we can see the
antecedents of Modi’s contemporary Hindutva fascist government (Sarkar,
1993; Banaji, 2018).
80 Hierarchies of hate in India
The BJP’s strategy for the unification of a caste-fragmented Hindu popu-
lation (wherein upper castes had no desire to give up privileges and inherited
networks of position), was through the othering of Muslims. In a stunning
feat of modern propaganda and disinformation, the BJP, aided by the RSS’s
fast-growing infrastructure of fascist training schools and programmes,
reaped the benefits of this fictional construction of Hindu identity. Capitalis-
ing on imaginary wounds inflicted against upper caste Hindu masculinity by
past (and present) Muslim ‘others’ and by affirmative action caste policies
(known as reservations), the BJP went from two seats in 1984, to 120 seats
in 1991. By 1999, the BJP led a rightwing coalition (National Democratic
Alliance). In the early 2000s, the RSS leader Narendra Modi was elected
Chief Minister of Gujarat and accused of complicity in the 2002 pogrom
where more than 2000 Muslims and several Muslim-aligned Hindus were
killed in a horrific manner by Hindu fascist mobs (Sarkar, 2002; Ghassem-
Fachandi, 2012). Since then, Modi has consolidated his political career as
a figurehead of masculinist Hindutva. His neoliberal subsidies for major
corporations and destruction of welfare schemes endeared him to elites, who
in turn helped him rebrand as a messiah of development.
Alongside Modi’s rise to power, there was a decisive shift in the telecom-
munications industry. The 1990s saw poorly regulated privatisation where
the government struggled to arbitrate disputes between private players or
between private players and the government’s own public telecommunica-
tions service providers (BSNL and MTNL). Poorly managed privatisation
led to organisational change. For example, a fixed license fee regime shifted
to a revenue sharing agreement between private telecommunications service
providers and government (Athreya, 1996; Chowdary, 1999; Roy, 2009;
Sridhar, 2012); meanwhile a rising middle class created the material condi-
tions for a booming telecommunications market. Governments too would
stand to earn substantial income from spectrum charges, tax on value added
services and so on. The 2000s saw a crowding of the market with more
than ten (domestic and international) telecommunications service providers.
Heavily skewed towards urban areas, most service providers were in long-
term debt to license spectrum at high costs – either expecting c­ onsistent
long-term growth or expecting to re-sell assets to competitors at a profit.
Voted out in 2004, the far right BJP came to power again in 2014 after
an unprecedented use of social media during the campaign with Modi at the
helm. By this point, the telecommunications market in India was crumbling
as the online ecosystem tilted towards usage related to products from compa-
nies such as Google, Amazon, Facebook and Microsoft. Supported by fascist
cadres, the Modi regime has seen a marked increase in a Jim-crow style poli-
tics involving frenzied mobs who lynched, tortured, harassed and intimidated
religious minorities (Banaji, 2018). Although several tracker websites and
 Hierarchies of hate in India 81
initiatives have been attempted in order to provide an aggregated picture of the
atrocities and violence under Modi, the government has repeatedly shut these
down.2 We estimate that there are hundreds of such instances of mob violence
connected to ‘defending’ or advocating Hindu nationalism (Hindutva).
Social media platforms and cross-platform applications have played a
role since 2014 in enabling the formation of vigilante mobs and allowing
the fascists to upload and spread their propaganda. Aside from these inci-
dents, the BJP has constructed a fictional trope of India as an ascendant
Hindu nation state vulnerable to the conspiracy of Islamic and communist-
backed terrorism. This fascist rhetoric has been amplified by a subservi-
ent mainstream media and used as a cover to target both real dissidents
and imagined opponents – including Dalit and Adivasi activists, religious
minorities, university students, human rights activists, academics, artists,
journalists and many others. Many have been jailed on trumped up charges.
In 2016, one of the richest men in the world, Mukesh Ambani owner of
Reliance Industries Group, launched Reliance Jio Infocomm Ltd. From its
inception, the group enjoyed a competitive advantage as it was allowed
by TRAI (Telecom Regulatory Authority of India) to offer free subscrip-
tions, free internet usage and free mobile phone devices for nearly a year.3
Such corrupt subsidies enabled Jio to ‘steal’ millions of customers from
competitors who could not afford equally low tariffs. In 2020, Reliance Jio
Infocomm became a subsidiary of Jio Platforms, which raised more than
20 billion US dollars by divesting a nearly 33% stake. Companies with
stakes in Jio Platforms include Facebook (9.9% stake for 6 billion US dol-
lars) and Google (7.7% stake for 4.7 billion US dollars).4 According to a
report by researchers at Azim Premji University,5 Mukesh Ambani, a key
beneficiary of the Modi government in multiple sectors from energy to tel-
ecom, and a few others such as the Adani group, have increased their wealth
many times over, while the Covid-19 pandemic has pushed 230 million
Indians into poverty. This brief account of the macrolevel structural matri-
ces within which social media companies operate in the Indian economy
sets the scene for an understanding of the role of social media in circulating
and promoting hateful content.

Modi and the media

The distinctive relationship between Modi’s regime and the media is the
absolute control that Modi appears to wield over his own image. Unlike all
previous Prime Ministers, Modi controls his media appearances by elimi-
nating any possibilities for spontaneous interaction with journalists (there
are no unscheduled interviews or press conferences). He uses the public
broadcaster to voice his rhetoric, speaks only to journalists who will report
82 Hierarchies of hate in India
favourably, and actively uses social media platforms to get his versions of
events across to millions of followers. Amit Shah, the Home Minister and
Modi’s key aide since his days as Chief Minister of Gujarat, has also made
effective use of Facebook in the 2014 and WhatsApp in the 2019 elections
to mobilise BJP voters. Multiple IT cells and hundreds of thousands of BJP-
linked social media groups circulate content to keep the BJP’s ideological
vision in supporters’ minds (Thakurta & Sam, 2019). Many (often, although
not exclusively, men and from dominant castes) have become vocal sup-
porters of Modi on the internet and brook no criticism of him or his gov-
ernment. These online Modi supporters are notorious for bullying, abusive
speech, trolling, doxing and spreading disinformation. There is sufficient
evidence to indicate that social media platforms and apps are regularly
‘gamed’ by the BJP IT cell to make topics trend or go viral, manipulating
opinion through coordinated behaviour.6 Legacy media (newspapers and
television news) then report the ‘buzz’ uncritically, selectively favouring
the BJP. Bollywood, the Hindi film industry, with a massive viewership
and disproportionate cultural and symbolic capital at its disposal, has also
promoted Modi and Hindutva7 both directly through Islamophobic and anti-
communist propaganda films and via Hindutva-supporting actors and social
media ‘influencers’.
In line with other tactics of suppression, the Modi government is trying
to control social media use and the use of other digital media and communi-
cation networks (such as streaming platforms). The government has intro-
duced new rules for intermediaries that establish procedures for addressing
complaints from users.8 However, the government has claimed for itself the
power to take final decisions on complaints that cannot be resolved through
internal mechanisms. Further, the central government has pressured com-
panies to break the encryption of peer-to-peer messaging applications such
as WhatsApp, allegedly to track the originators of disinformation, but in
actuality to surveille and incarcerate opponents of the regime. Authoritar-
ian restrictions on NGOs and digital news and current affairs initiatives
have had a ‘chilling effect’, while enabling government supporters to domi-
nate social media. In 2021, social media companies, especially Facebook
and Twitter, have been under increasing pressure from the Modi regime to
shut down even mildly critical accounts, delete posts, and stop tagging gov-
ernment supporters’ disinformation posts as manipulated and so on. As we
write, it appears likely that the courts will ask social media companies to
comply with government demands, thus further endangering social media
users who have identities targeted by Hindutva fascist mobs, are critical
of the regime or do social justice work. The situation, as our key inform-
ants explain, is already one of deep anxiety and existential threat for many
communities.
 Hierarchies of hate in India 83
Difference and discrimination
Meena, a queer activist who has been targeted by the state for their political
views, explained their experiences with social media in terms of how dif-
ference of identity and social position plays out, both with them and other
progressive people they follow on social media:

A lot of it was targeting my appearance, just saying that I’m ambigu-

ously gendered, essentially. Sort of saying that I don’t have a sense
of morals or ethics because I went out and created problems for peo-
ple. It was mostly attacking my general queerness and the fact that
I’m assigned female at birth. . . . Something recently happened to an
artist I follow on Instagram, where she parodied another Instagram
account and the person who she parodied got really upset. That person
(an Upper Caste Hindu), told her followers that she’s being bullied for
being Hindu but she wasn’t being bullied . . . Her entire follower(base)
turned into this troll army and would send the girl who posted the par-
ody so many hate messages. . . . They went after her for being fat, hav-
ing darker skin, being Dalit, being queer – every aspect of her identity,
and said absolutely horrible things. It was really painful to watch.

Preeti, an activist working in a rural part of a North Indian state, shares how
quickly an ideological argument can shift to caste-based abuse:

A few days back there was news that the Indian government is sell-
ing a part of LIC. I had posted on Facebook saying that LIC is the
backbone of many people in India and if government sells even that
it will be a loss of people. Now this is not against the government
as such. This is about ideology. There were some Modi Bhakts who
jumped on it. First a lady posted saying ‘all this is fine but Modi is
doing good work’. I didn’t respond, this is her opinion. Then a few
more people responded saying, ‘you hate Modi, what is your status
(aapki aukat kya hai?), you should remain where you have come from.
Caste system should exist, people like you should be suppressed (pair
ke neeche daba ke rakhna chahiye)’.

These experiences reveal the extent to which the online environment in

India is dictated by dominant caste, heterosexist and pro-regime users.
A particular set of techniques – trolling, hacking, abuse, doxing and so on –
ensure that those who are marked as different online by virtue of skin colour,
gender, sex, sexuality, caste, class and so on are silenced and humiliated to
the point where they do not venture online again. These experiences also
84 Hierarchies of hate in India
reveal that much like the offline world, an intersection of caste, gender and
sexuality or class and gender heightens probabilities of harassment and dis-
crimination. Indigeneity, Dalitness, Muslimness, queerness, darker skin, and
leftist and/or feminist political opinions compound each other in terms of
the consequences that individuals and groups face.
Apart from the ‘pure’ hostility and abuse towards specific identities
(LGBTQIA+, Muslims, Dalits), there is also an immediate torrent of abuse
that users – particularly from a vulnerable social identity – face for ideo-
logical dissent. In the excerpt above, critiquing the privatisation of a public
sector company was the catalyst that unleased caste-hate speech. The ease
and extent to which online ‘political’ discourse (for example, about priva-
tisation of public sector industries) transforms into casteism should chal-
lenge illusions about the Internet as a Habermasian public sphere where
rational-critical dialogue can be put to use towards deliberative democracy
(Anderson & Jaffrelot, 2018; Bürger et al., 1992; Khorana et al., 2014).
Historically formed asymmetrical hierarchies dominate the media sphere
in India while historical discrimination and violence manifest themselves
repeatedly on social media. A particularly violent maintenance of differ-
ence can be seen in the abuse targeted against Muslims. Hatred against
Muslims has political support from prominent BJP leaders who have not
hesitated to use the full power of the state apparatuses at their disposal
against Muslims.

Anti-Muslim hate online

One of the most prominent forms of social media discrimination in India is
Islamophobia – discrimination, dehumanisation and incitement to violence
against Muslims. Several interviewees spoke about this. Syed, a journalist
at a mainstream media organisation, and Shruti, who leads an independent
digital news organisation explain:

I get abuse just because of my name and my religion. . . . People try

to make it a Hindu-Muslim issue – you can check my page; very few
people write against radical Islam the way I do. But at the same time,
I write about other things as well. I won’t forget how Palestinians are
being robbed and killed and how Israel has been treating them for years
and it has the backing of the so called developed and modern world . . .
Just a few months ago I wrote something about cricket, someone started
abusing my name and just the usual abuse they used on social media
which I can’t even tell you. They say ‘go to Pakistan, you are a traitor,
you are a mullah, and you are a motherfucker’, these types of things.
Things that are so abusive that I can’t even tell you. . . [Syed]
 Hierarchies of hate in India 85
I’ve done reports on ‘love jihad’ and seen the response of people
like . . . I don’t even know how to say it . . . I’ve seen people say, ‘the
first thing you need to do when you meet a Muslim man is really make
sure he’s Muslim by taking off his pants’ . . . . And just really gross
things like ‘you should just behead them, hang them by their balls’. . . .
I really don’t know what to do about it. I think when the Rohingya
thing came up and I was reporting on it, I just saw the kind of things
people said about them, like calling them ‘terrorists’, ‘dump them in
the sea’ – who says things like that? And they said that ‘these guys are
terrorists’, but there is not one FIR against [the Rohingya] so far like
major FIR all the complaints against them have been for pickpocketing
or small theft but nothing, nothing to do with national security. . . . I
have Muslim friends who are journalists. The minute they write some-
thing they are called ‘jihadis’, all of us are called ‘jihadis’ and ‘terror-
ists’ – if I’m called that twice, that person will have been called that
100 times. I know a lot of Muslim friends who are otherwise very brave
journalists, who refrain from writing on the Internet or talking about
religion, because it is just so explosive. There was this example of a
Muslim journalist from [redacted], things got to such an extent that
there were blogs saying that he was a mujahideen member, that he’s a
terrorist – with absolutely no evidence, just because he’s a Muslim you
can write things, and there’s a percentage of people who will actually
believe it simply because of the fact that you’re a Muslim. I think the
kind of hatred Muslim journalists face is something that I cannot even
comprehend. [Shruti]

Any pretensions that social media platforms and cross-platform applica-

tions may have had to being an ‘alternative public sphere’ ring hollow given
this context. Silencing comes through Islamophobic abuse directly aimed
speakers’ identities (‘traitor’, ‘mullah’, ‘go to Pakistan’, ‘he’s a mujahideen
member’, ‘he’s a terrorist’ and so on). In most instances, it is not just the
intensity of abuse but also the magnitude that can be overwhelming (‘called
that 100 times’) making it impossible for many Muslims to post about
anything, even about ostensibly non-political domains such as cricket.
Muslimness as an internally heterogenous and complex social identity is
discursively reduced to a homogenous religious stereotype. Although these
discursive strategies against Muslims may resemble anti-Muslim discourses
in other parts of the world, noticeably in the UK and Myanmar, the hatred
against Muslims as the ‘Other’ has a distinctly Indian lineage that has been
strengthened by post-9/11 Islamophobia across the globe.
Historically, a positioning of Muslims as being on the ‘outside’ and ‘ene-
mies’ has been how the RSS and the BJP have constructed ‘Hindu’ identity.
86 Hierarchies of hate in India
This spurious religious identity overwhelms other potential ways of engag-
ing fellow citizens on or off social media, especially when these are critical
of the casteism, chauvinist nationalism and neoliberalism that the BJP has
cultivated. The ontological equation of Muslimness with threats to national
security or Hindu womanhood is now so ingrained that many caste Hindus
barely give it a second thought. While such racist imaginaries depend on
a binary of good Muslim/bad Muslim so crucial to post-9/11 imperialist
geopolitics (Mamdani, 2004), the Indian history of discrimination against
Muslims constantly references Mughal rule, a trope through which Hindu
majoritarians construct a false history of persecution for which they are
seeking redress (Anderson & Jaffrelot, 2018; Truschke, 2020). Discrimina-
tion against contemporary Muslims is especially intense when it comes to
Muslim women.

Islamophobia and misogyny

The intersection between caste and religious identity is exacerbated by the
misogyny and hate targeted at Muslim women which has risen steeply since
the early 2000s. In 2018, Amnesty International published a report about
misogyny on Twitter, detailing the ways in which women were targeted in
western countries such as the UK, USA, Denmark, Italy, New Zealand, Swe-
den, Spain and Poland. The problem is even more ubiquitous and intense in
India. A particularly pernicious instance of doxing (a process where a per-
son’s private documents are revealed without that person’s consent, in order
to silence or harass) targeting Muslim women came to light in mid-2021
when an app and website ‘Desi Sulli Deals’ was revealed. ‘Desi’ is a com-
mon term for South-Asian and ‘Sulli’ is a derogatory term used by far-right
Hindutva groups to refer to Muslim women. Developed through the open-
source software website Github, ‘Sulli Deals profiled more than 80 Muslim
women, including several outspoken, high profile and dominant caste Mus-
lim women, using images stollen from their social media and public spaces
and conducted faux auctions.9 These ‘auctions’ provided opportunities for
Hindutva trolls and sock-puppet accounts to post vicious misogynist and
anti-Muslim comments objectifying the women, reducing them to their body
parts and threatening rape. Although rights activists forced the Delhi police
cybercell to file a complaint, the issue was under-reported and political lead-
ers kept silent. It was only after several Muslim women complained vocally
that Github and various social media companies including Twitter and You-
Tube shut down the application, the website and the accounts promoting it.
Other violence goes beyond the virtual silencing of outspoken women.
Women in semi-urban and rural areas told us that their daughters’ profile
pictures are circulated accompanied by lewd and abusive comments, with
 Hierarchies of hate in India 87
‘upskirt videos’ of young teenagers and women frequently circulated with-
out their knowledge in closed all-male WhatsApp groups. In mixed gender
WhatsApp groups, women sometimes encountered hardcore pornography
shared by men, supposedly ‘by mistake’. In other places, for example in
the state of Uttar Pradesh, incidents of rape and gang rape are regularly
filmed by perpetrators and then sold for a nominal cost at local mobile
phone shops, transferred to clients’ SD cards and further shared.10 During
our research we were told of at least one public health worker who commit-
ted suicide. A sexual health activist, Preeti, who works with young women
in rural North India, and is active on social media explained:

. . . if you are online at 1am in the night, suddenly you will get a call
and obscene language is used. They say very obscene things like- ‘you
are online at this hour, are you sexually not satisfied?’ Some ask, ‘why
are you still single? If you need some, just call me, I will come there.’ If
you posted something about gender equality on Facebook, people write
so obscenely, they start commenting on your private parts like vagina,
breasts, on your weight . . . There are about 4–5 girls with me who
said they don’t want to do this work as people send obscene content on
[Facebook] messenger. Those girls went into depression, we had to put
them on medication. Some children come from small families, and they
struggle and reach here. For them these kinds of things are very new.
They are not able to handle it.
Recently, we were working on the Hathras case, there was a com-
mittee made. In this regard we had posted on Facebook – in that case
boys from Jat community were the perpetrators – we posted about the
case. The first comment that came was, ‘we will rape you, we will do
what was done to Manisha’. These comments were not even in the
inbox, they would write it directly on our Facebook post. They would
write, ‘we will rape you, will throw acid on you, you try and step out of
your house alone!’ It is a lot of mental harassment. People who make
these comments, they all have fake accounts. Nobody used their per-
sonal account to write all this. They will be in my friend list from their
personal account and will read what I have posted but will post com-
ments from a fake account. They keep themselves safe. Though from
their style of writing one can understand who it is actually.

Feminist non-profit spaces, which might have reported physical instances of

intimidation and abuse through legal protocols, struggle to deal with anony-
mously delivered abuse. Apps such as Facebook Messenger lead women to
experience this abuse as a deeply invasive phenomenon that must be suffered
in private, and internalised individually, resulting in fear, depression and loss of
88 Hierarchies of hate in India
motivation. This is not restricted to women in rural areas. A prominent Kashmiri
ex-student leader from Jawaharlal Nehru University in Delhi, Shehla Rashid,
spoke about the massive levels of abuse she has faced from men in all commu-
nities, in response to her work as an outspoken woman, Muslim, Kashmiri, and
progressive activist from university with a history of leftwing activism.

First of all these are not very innovative . . . rape threats, and you know,
‘whore’ and ‘prostitute’. . . . So you have degrees in it. You can have
rape threats, you can grade it by prostitute or a worse word. The sec-
ond kind is misogynistic content. Now that’s more difficult to identify.
Through keywords you can look at rape, whore, slut, etc., you can still
do a keyword search. How do you do a keyword search for misogyny?
Now only qualitative analysis can tell you that. It is very difficult to
pinpoint that but its very very widespread and across the political spec-
trum. So is abuse, don’t get me wrong. . . . If I were to comment on
an issue which is sensitive to Muslims in general, despite me being a
Muslim, despite me having a claim to speaking for Muslims myself, or
as a Muslim, I will get abused by Muslim men and in pretty much the
same terms.

As Rashid says, it is difficult to use quantitative methods to ascertain

instances of misogyny which go beyond expected keywords. Misogyny in
the Indian social media space, like casteist communication and Islamopho-
bia, can either be open and easily recognisable or highly contextual, cul-
turally and linguistically nuanced, presenting a challenge for social media
teams that emphasise Artificial Intelligence approaches to identify and take
down discriminatory content.
Ever more disturbing physical and symbolic violence against women cir-
culates on WhatsApp. Padmini, a postgraduate student from a well-known
academic institution in Maharashtra recounts her experience:

. . . I also talked about how the army is masculinist and assists in state
sponsored crimes. These are words that we use in academia very fre-
quently. Someone said that a WhatsApp group wasn’t an appropriate
place for this . . . But screenshots [of my posts] were circulating all
around campus. And then those were posted on Twitter by a student
who very proudly claims to be an office bearer of the ABVP [the stu-
dent wing of the BJP]. He later pulled down those posts. He blurred
everyone’s numbers but mine. This post was getting shared by a BJP
MLA and another BJP spokesperson. It was flooding all over twitter.
Then I started getting calls that I didn’t answer. Later I started getting
messages on Facebook, so I deactivated my account. But before that
 Hierarchies of hate in India 89
someone had got a photo of me and my mother and circulated that on
Twitter. My number and my email id, my Facebook, Twitter account
was made public. The institute was also getting a lot of political pres-
sure from outside to act, so they set up a committee which conducted a
hearing. They asked me to give an apology and I didn’t want to make
this into an issue, so I gave a written apology. The committee gave
me notice saying that I must vacate the hostel and that they are letting
me finish my PhD but if something like this happens again, I would
be expelled. They also said that they won’t consider me for further
employment or study in the institute.

The most startling thing about this account to those who have not followed
the rise and spread of fascism in India may well be the university admin-
istration’s abject refusal to support their victimised student and their posi-
tioning of themselves alongside her oppressors and doxers. Her comments
about the abuse meted out by the Indian army and the ‘masculinist’ nature of
the army space have ample evidence in human rights reports over the years.
However, since the BJP’s rise to power, critiquing nationalism is danger-
ous, and any critique of the armed forces is deemed unacceptable. Voicing
such opinions as a woman multiplies the danger. The excerpt above clearly
reveals the ways in which the BJP, through its various licit and illicit net-
works, create offline consequences for online utterances.

Complex psychic and social consequences

While justice and rights activists have been targeted through overt and devi-
ous means, public opinion against them is stirred precisely through the legit-
imisation and amplification of discourses that vilify them (as anti-nationals,
for instance). An activist, Father Stan Swamy, jailed under the deeply anti-
democratic UAPA law for fighting for Adivasi Indigenous rights in central
India, died in custody.11 A cruel irony is that many activists, such as Suren-
dra Gadling, Sudha Bharadwaj, Rona Wilson and Father Stan Swamy, who
have been jailed for years on false charges of being ‘Naxalites’, were fight-
ing precisely for the rights of Dalit and Adivasi political prisoners. Apart
from direct state sponsored harassment and imprisonment, public disinfor-
mation and vilification also has other effects that are internalised and have
long term health consequences.

If someone hits you, you obviously feel pain, and if someone hits you
with their words you feel pain. I don’t think there’s much you can do
about it. You have to become like some cold-hearted monster where
nothing affects you anymore; but that’s not a desirable thing to be.
90 Hierarchies of hate in India
So yeah, it does affect you very badly, it lowers your self-esteem . . .
I mean the government and ministers have called us “anti-nationals”.
But social media never lets you forget that. . . . Even today they are
using the word in official communications. [Shehla]
I was already in a pretty bad headspace, and seeing the media cov-
erage of me was really difficult. Part of it is that they were constantly
misgendering me and referring to me as a girl, but I think the part
that really hurt is that I felt like my whole character was being put on
trial. . . . And then I read the comments, and it made me feel scared for
myself and for my own physical safety. . . [Meena]
. . . the first thing I did is make my Instagram private and for a
while I was still posting things on my Instagram story and I had a close
friends list, so it would only go to them, but now I don’t use social
media at all . . . I only read things on Twitter to keep up with the news.
[Meena]
I used to get badly affected, I used to have nightmares and I couldn’t
sleep. I used to think that some of them are going to come to my house.
I remember one election where I was really scared of going out because
someone did recognise me and said ‘you were the one who tweeted
against. . .’ Initially I used to be very closed about this feeling of fear,
I wouldn’t even tell my husband, who actually works with me. I never
used to share with anyone that I’m feeling very disturbed. I used to act
very brave and say it doesn’t affect me but inside it was eating me up, it
was impacting my work and sleep. [Now. . .] I don’t give my opinions
on social media anymore, I am very cagey about saying stuff – I self-
censor a lot. [Shruti]

The twin constructions of nationalism and the figure of the ‘anti-national’

or ‘urban Naxal’ has been a longstanding authoritarian political project,
with the previous Congress and coalition regimes complicit in previous
decades, particularly targeting Adivasis working for labour and land rights,
and critics in Kashmir and the Northeast. Under the BJP, targeting critics
and dissenters across the country has become a vast government-sponsored
strategy, with systematic discursive support from mainstream media and
social media trolls. The ‘drip-drip’ of hate that our interviewees receive
seeps into and acts on how they construct their identities and live their lives.
Much of the policy debate around taking down problematic content hinges
on the extent to which content may incite violence or poses a clear and pre-
sent danger conceived as physical danger. However, as explained in Chap-
ter 1, we argue, ‘violence’ and ‘danger’ are defined far too narrowly in an
attempt to protect an abstract and often disingenuous idea of free speech. The
actual experiences of our subjects demonstrate that it is precisely a destruc-
tion of their free speech which recurs in response to subtler forms of threat
 Hierarchies of hate in India 91
and symbolic violence. This may not always fall under the ambit of the law,
but must be addressed by social media companies to protect multitudes of
socially and politically at-risk users. It is imperative that affective forms
of violence (which cause pain, fear, lack of sleep, and silencing) should be
considered hate crimes. Since such violence has historically been targeted
towards excluded and discriminated groups, there is a greater probability that
those groups face different and tailored modes of violence online as well. Our
analysis has found this pattern repeated across all the cases and countries.

What is to be done?
As with users from other countries, our research subjects in India spoke
about a wide range of strategies and recommendations which we believe
should be taken up seriously by social media companies and policymak-
ers, as well as by international human rights organisations that work on
discrimination and violence.
In India, there are (very few) institutional resources available for recipients
of hate who need support, and most of these are created by feminist, Dalit
and anti-Fascist activists themselves, in a personal or collective capacity:

But I do report, I tell my groups of friends to report . . . For me, what

has helped me to cope is common support groups, understanding what
my rights are and recognising that it is a problem. [Shruti]
I’ve become very reclusive and if you knew me before any of this
happened that’s very much not how I am. It made me quiet down a lot.
[Meena]
One shouldn’t get scared. When one moves ahead without fear – that
kills their power. Their aim is to scare you. But when you continue to
do your work while scared then they get scared that we are doing so
many things but she is not getting scared. So we should keep our focus
on work, you should not ignore other things but should deal with it. . . .
All the girls in our circle who are usually trolled, we all have made a
group on WhatsApp and Facebook. So if a girl is trolled then we all
get down to trolling them back. We get number of that boy then we
take his case. We record the call and out it on Facebook. . . . 3–4 boys
deleted their Facebook account. We are about 68 members in the group.
At times there should ‘behengiri’ [aggressive sisterhood] also. [Preeti]

Inspiring as these accounts are, they demonstrate the immense labour that
individual social media users in India have to undertake to grapple with
discrimination and violence levelled at them, often demonstrating remark-
able patience and resilience in the face of consistent threat, hate and hostil-
ity. There are signs of more experienced social media users self-organising
92 Hierarchies of hate in India
and fighting back, not just around specific incidents, but also in a more
systematic way.

There’s an organisation in Karnataka called hate speech beda [we

don’t want hate speech]. They brought out a report last year on how
Karnataka media is spreading hatred. It’s documented for posterity
but what is the action that we are aspiring for on the ground? The
government is not reacting, nor is the police. Our documentation is
all that we’re doing now, are we creating any impact? Are we going
to judge only by election results whether this kind of polarisation
works? I’m hoping that people will see that this is not right, bigotry
is not right. Once that realisation happens, or the other way is, if you
have ten people who are behaving in this way, like what’s happening
in America, the voices of these people are getting drowned and these
people are getting stronger [depicts a model with hands], it does not
mean that these people don’t exist, their racism and bigotry are still
there. The other voices have become more powerful. That’s the only
way out for India. The so-called fence-sitters have to choose their
sides. [Shruti]

Concluding on a slightly more optimistic note, the Hate Speech Beda

group’s efforts in Karnataka finally paid off. Following their campaign, on
23 June 2021, the self-regulatory broadcasting body National Broadcasting
Standards Authority (NBSA) ordered a TV news channel, News18 – part of
the powerful Network 18 group – to air an apology and pay a fine.12 They
also censured Suvarna News and Times Now for provocative and hateful
coverage of Tablighi Jamat Muslims in 2020.

Conclusion
Several of our key informants, interviewees and focus group participants
expressed disappointment and resignation with regard to social media com-
panies taking prompt action against the hateful and discriminatory commu-
nication to which they find themselves subjected. Social media companies
must recognise the current conjuncture in India as a crisis for democracy
and for mental health that needs to be comprehensively and courageously
addressed, beyond profit, algorithms and AI. Rather, if the social media
companies are serious about democratic mandates and free speech, they
must listen carefully to human and civil rights activist and users from histor-
ically marginalised groups who have been consistently targeted both online
and offline, and act to protect them.
 Hierarchies of hate in India 93
Groups that have been physically harassed and targeted historically based
on their caste, gender, sexual orientation, class and religious identities, and/
or intersections of these, continue to be targeted on social media. Although
the latter mode (of social media abuse, incitement and threat) uses differ-
ent techniques, it has no less serious consequences particularly in the area
of free speech and mental health, leading in extreme cases to suicide, rape,
lynching and/or targeted pogroms in which entire communities are expro-
priated and their homes burnt to the ground or razed. Given the enabling
environment for such discrimination and hate against these groups at the
highest levels of the ruling party and Modi government and from most state
institutions including the police and local magistrates, it is disingenuous to
tell affected individuals and groups to seek legal protection, report online
abuse to the police and so on. Many of them as we have seen, have indi-
vidually and collectively come up with strategies to fortify their own well-
being. At the risk of imprisonment or other serious forms of censure and
detriment, they continue speaking against authoritarian and fascist develop-
ments in their own local and national contexts. That our collective global
democratic well-being rests on the shoulders of some of the most vulnerable
and discriminated persons should be a matter of deep shame to technolo-
gists, scholars and policymakers alike.

Notes
1 Dr. Murali Shanmugavelan. (2021, March). Caste-hate speech: Addressing
hate speech based on work and descent. International Dalit Solidarity Net-
work. https://idsn.org/wp-content/uploads/2021/03/Caste-hate-speech-report-
IDSN-2021.pdf
2 Scroll Staff. (2019, September 12). FactChecker pulls down hate crime data-
base, IndiaSpend editor Samar Halarnkar resigns. Scroll. https://scroll.in/lat-
est/937076/factchecker-pulls-down-hate-crime-watch-database-sister-web
sites-editor-resigns
3 Daniel Block. (2019, February 1). Data plans: How government decisions are help-
ing Reliance Jio monopolise the government sector. The Caravan. https://caravan
magazine.in/reportage/government-helping-reliance-jio-monopolise-telecom
4 Scroll Staff. (2020, July 16). Why Google and Facebook are investing in Mukesh
Ambani’s Reliance Jio. Scroll. https://scroll.in/article/967618/explainer-why-
google-and-facebook-are-investing-in-mukesh-ambanis-reliance-jio
5 Azim Premji University. (2021). State of working India 2021: One year of
Covid-19. Centre for Sustainable Employment, Azim Premji University. https://
cse.azimpremjiuniversity.edu.in/wp-content/uploads/2021/05/State_of_Work
ing_India_2021-One_year_of_Covid-19.pdf
6 Buddhadeb Halder. (2020, December 17). How the BJP tried to manipulate pub-
lic opinion on social media in favour of the CAA. The Wire. https://thewire.in/
politics/how-bjp-tried-manipulate-public-opinion-social-media-favour-caa
94 Hierarchies of hate in India
7 Kamayani Sharma. (2019, April 1). Supporting role: How Bollywood acted
under the Modi government. The Caravan. https://caravanmagazine.in/
perspective/how-bollywood-acted-under-modi-government
8 Internet Freedom Foundation. How the intermediary rules are anti-democratic
and unconstitutional. https://internetfreedom.in/intermediaries-rules-2021/
9 Geeta Pandey. (2021, July 10). Sulli deals: The Indian Muslim women “up for
sale” on an app. BBC. www.bbc.com/news/world-asia-india-57764271
10 Asad Ashraf. (2016, October 31). A dark trade: Rape videos for sale in India.
Al-Jazeera. www.aljazeera.com/features/2016/10/31/a-dark-trade-rape-videos-
for-sale-in-india
11 Mrityunjay Bose. (2021, July 5). Tribal activist Fr Stan Swamy passes away
hours before the hearing of his bail plea. Deccan Herald. www.deccanherald.
com/national/tribal-rights-activist-fr-stan-swamy-passes-away-hours-before-
the-hearing-of-his-bail-plea-1005065.html
12 Campaign Against Hate Speech. (2021, June 29). News 18 Kannada: Why
fighting hate speech in news is crucial. The Quint. www.thequint.com/voices/
opinion/network-18-apology-why-fighting-hate-speech-in-news-is-crucial-for-
democracy

References
Ambedkar, B. R. (1968). Annihilation of caste with a reply to Mahatma Gandhi;
and castes in India: Their mechanism, genesis, and development. Bheem Patrika
Publications.
Andersen, W. K., & Damle, S. D. (1987). The brotherhood in saffron: The Rashtriya
Swayamsevak Sangh and Hindu revivalism. Westview.
Anderson, E., & Jaffrelot, C. (2018). Contemporary South Asia Hindu nationalism
and the “saffronisation of the public sphere”: An interview with Christophe Jaf-
frelot. Contemporary South Asia, 26(4), 468–482.
Ansari, K. (2009). Rethinking the Pasmanda movement. Economic and Political
Weekly, 44(13), 8–10.
Arya, S., & Rathore, A. S. (2020). Dalit feminist theory: A reader (S. Arya & A. S.
Rather, Eds.). Routledge.
Athreya, M. (1996). India’s telecommunications policy: A paradigm shift. Telecom-
munications Policy, 20(1), 11–22.
Ayyub, R. (2016). Gujarat files: Anatomy of a cover up. Rana Ayyub.
Banaji, S. (2018). Vigilante publics: Orientalism, modernity and Hindutva fascism
in India. Javnost – the Public, 333–350.
Bhat, R. (2020). The politics of internet infrastructure: Communication policy, gov-
ernmentality and subjectivation in Chhattisgarh, India. PhD thesis, London School
of Economics and Political Science. Etheses. http://etheses.lse.ac.uk/4175/
Bourdieu, P. (1984). Distinction: A social critique of the judgment of taste. Harvard
Universtiy Press.
Bourdieu, P. (1990). The logic of practice. Polity.
Bürger, T., Lawrence, F., & Habermas, J. (1992). The structural transformation of
the public sphere: An inquiry into a category of bourgeois society. Polity.
 Hierarchies of hate in India 95
Chakravartty, P., & Roy, S. (2015). Mr. Modi goes to Delhi: Mediated populism and
the 2014 Indian elections. Television & New Media, 16(4), 311–322.
Chowdary, T. H. (1999). Telecom demonopolization: Why did India get it so wrong?
Info, 1(3), 218–224.
Ferguson, J., & Gibson, T. (2015). Give a man a fish: reflections on the new politics
of distribution. Duke University Press.
Ghassem-Fachandi, P. (2012). Pogrom in Gujarat: Hindu nationalism and anti-
Muslim violence in India. Princeton University Press.
Gorringe, H., & Rafanell, I. (2007). The embodiment of caste: Oppression, protest
and change. Sociology, 41(1), 97–114.
Hansen, T. B. (1999). The saffron wave: Democracy and hindu nationalism in mod-
ern India. Princeton University Press.
Jaffrelot, C. (2007). Hindu nationalism: A reader. Princeton University Press.
Khorana, S., Parthasarathi, V., & Thomas, P. N. (2014). Public spheres and the
media in India. Media International Australia, 152(1), 75–76.
Mamdani, M. (2004). Good Muslim, bad Muslim: America, the cold war, and the
roots of terror. Pantheon Books.
Mankekar, P. (1999). Screening culture, viewing politics: An ethnography of televi-
sion, womanhood, and nation in postcolonial India. Duke University Press.
Manor, J. (2010). Prologue: Caste and politics in recent times. In R. Kothari (Ed.),
Caste in Indian politics (pp. xi–lxi). Orient Blackswan.
Rajagopal, A. (2001). Politics after television: Religious nationalism and the
reshaping of the Indian public. Cambridge University Press.
Roy, H. (2009). Telecom growth trajectory in India. SSRN.
Sarkar, S. (1993). The fascism of the Sangh Parivar. Economic and Politicaln
Weekly, 28(5).
Sarkar, T. (2002). Semiotics of terror: Muslim children and women in Hindu Rash-
tra. Economic and Political Weekly, 37(28), 2872–2876.
Sridhar, V. (2012). The telecom revolution in India: Technology, regulation, and
policy. Oxford University Press.
Tambiah, S. (1986). Sri Lanka: Ethnic fratricide and the dismantling of democracy.
University of Chicago Press.
Teltumbde, A. (2005). Hindutva and Dalits: Perspectives for understanding com-
munal praxis. Samya.
Thakurta, P., & Sam, C. (2019). The real face of Facebook in India: How social
media have become a propaganda weapon and disseminator of disinformation
and falsehood. Oxford University Press.
Truschke, A. (2020). Hindutva’s dangerous rewriting of history. South Asia Multi-
disciplinary Academic Journal, 24–25.
Wilkerson, I. (2020). Caste: The origins of our discontents. Random House.
5 White male rage online
Intersecting geneologies of
hate in the UK

This chapter is based on the testimonies of multiple informants ranging in

age from their early twenties to their seventies. They have experienced troll-
ing, doxing, hacking, stalking, targeted harassment, death-threats and vio-
lence on and offline in the UK. Contrary to the scholarly and media narrative
which sees this kind of hate originating in ‘Russian troll farms’ or ‘abroad’,
most of this hate is homegrown. We find ourselves at a frightening histori-
cal juncture which is repeatedly misunderstood and downplayed by being
called ‘post-truth’ or ‘populist’. This neglects the elaborate infrastructures
of the political right and far right that now permeate everyday conscious-
ness. Ruptures have reached breaking point between those who actually
(rather than rhetorically) uphold freedom of speech, equality before the
law, transparency, democracy or human rights, and those white and white-
adjacent supremacist ideologues who mobilise around retrograde ideologies
masquerading as pride in nation, rights of the individual and freedom of
speech. Although we know the UK intimately, the painful and grotesque
testimonies of our UK interviewees in the context of ongoing public strug-
gles over definitions of antisemitism, anti-Zionism, feminism, sexual and
gender identity, race, racism and resistance make this country strange in
multiple ways.
As elsewhere, we insist on the overwhelming importance of history in
understanding different forms of violent incitement, dehumanisation, dis-
crimination and hateful content. We want to eschew presentist claims, even
while the present envelops us in ever more bizarre circumstances. Writing in
2021, degraded public services are groaning under the onslaught of govern-
ment cuts, mismanagement and graft. Children are going hungry and being
abused while also being exhorted to learn online, to fulfil their potential
and to achieve. While maintaining its own very British fiction of freedom
and fairness, a largely unregulated media-sphere has moved significantly
towards the expression of authoritarian and majoritarian values (now with
the launch of a new alt-right news channel to mimic Fox news).1 To these
DOI: 10.4324/9781003083078-5
 White male rage online 97
processes, one might add a series of intertwined media representations and
Government actions that are fuelling vicious collective prejudices. Curbs
on migration and a suspicion of refugees and asylum seekers have led to de
facto murder, with the Home Office criminalising the saviours of drown-
ing refugees. Definitions of antisemitism, such as that by the IHRA, which
encompasses those who uphold the humanity of Palestinians, have made it
ever more challenging to point out the horrific humanitarian repercussions
of settler colonialism and European exceptionalism in Palestine. The rise of
celebrity Islamophobes and anti-Black racists whose visibility is linked to
the anti-democratic opinions they spew in tandem with a deliberate fuelling
of suspicion against anti-racism and anti-racists has had direct repercus-
sions. This was clearly demonstrated when three young Black players for
the England football team missed penalties in Euro2020 in July 2021 and
found themselves on the receiving end of racist hate and disapproval.
Debilitating divides between those who voted for and against Brexit
remain, while the worst economic repercussions of leaving the European
Union are only beginning to be felt. Resurgent transphobia amongst parts of
the British intelligentsia has mobilised diverse publics against an already dan-
gerously marginalised group and exacerbated homophobia at the same time.
All of this has been and continues to take place against the backdrop of the
second year of the global Covid-19 pandemic which has caused a hundred and
forty four thousand officially counted deaths in the UK as we go to print, left
several million in poverty and jobless and provided an opportunity to an end-
lessly corrupt Conservative party and Government to line the pockets of their
friends and donors. In the pandemic’s wake, we have barely had time to count,
name and mourn the dead, here and in diasporic ‘home’ countries. Mean-
while millions of people find themselves diverted by anti-vaccine conspiracy
theories or caught up in the incapacity and unwillingness of capitalism and its
favoured private systems to adjust to the shared burden and responsibility for
material and psychic survival. In the UK these events and processes are fil-
tered through relative and respective class privileges or burdens as educators,
cleaners, bus drivers, retail, care and healthcare workers attempt to bear the
unbearable risks of maintaining – and changing – this failing system.
Meanwhile, migration from former colonies in Asia, Africa and the Car-
ibbean in line with government policies of the 1950s, 60s and 70s has been
caught up in the post 9/11 demonisation of Muslims in the UK. British
participation in wars against Afghanistan, Iraq and Libya, as well as the
supply of weapons to Yemen and Syria has provoked an influx of newer,
educated migrants and refugees from these countries. The drastic destruc-
tion of social welfare by a decade of ideologically motivated privatisations,
cuts and closures covered up and excused by a media steeped in neoliberal
dissimulation around economic mismanagement has ensured a constant
98 White male rage online
sense of resentment on the part of the mainly white citizens who view peo-
ple of colour and migrant communities as competing with them for scarce
resources which they think of as their birth-right.
As we analyse overlapping accounts of political abuse, racism, hom-
ophobia, transphobia, misogyny, trolling and physical violence in this
chapter, we find ourselves returning time and again to these events and
processes to explain, and counter, the simplistic and vicious imaginaries
revealed in racist tweets and direct messages, dehumanising Facebook and
Instagram groups, posts, racist viral videos on TikTok and YouTube and
newspaper comments’ sections. We also find ourselves repeatedly grateful
to scholarship which theorises intersectionality (hooks, 2014; Crenshaw,
1989) and interrogates race and caste critically with attention to sociolegal
frameworks and atrocity (Bell, 1995; Delgado & Stefancic, 2013; Davis,
1981; Teltumbde, 2010), since almost all accounts detail how overlapping
identities and positionality within constructions of race, class, gender and
sexuality exacerbate the harmful, vitriolic and violent content of social
media hate.

Doing God’s work: Faith-based support for LGBTQIA+

rights around the world
The intersection between organised religion and sexuality has historically
been a troubled one, not only for the Church of England and Catholicism
but across other religions too. LGBTQIA+ people of faith have experi-
enced suppression, exclusion and active persecution both in the global
north and in the global south. Worse still, much violent homophobic and
transphobic persecution has been enshrined in law and is often first expe-
rienced in childhood. From the trauma of conversion therapy (Adamson
et al., 2019), homelessness and playground bullying to that of being spat at
and assaulted on public streets (Tyler & Schmitz, 2018), many LGBTQIA+
teenagers in the UK remain closeted for survival, with religious rhetoric
often providing a convenient disguise for parental and community lack of
support. Being a migrant or person of colour almost always adds a further
layer to the hate levelled at LGBTQIA+ people of faith. While discourse
in the UK has changed somewhat, largely due to the constant struggles of
activist groups and individuals, much of the above still holds true in British
faith communities.
We interviewed British-Nigerian Reverend Jide Macaulay only weeks
after he had become a priest in the Church of England. We discussed the
14 years since he had founded House of Rainbow, a faith-based campaign
group and space for ‘sexual minorities, lesbian, gay, bisexual, transgender,
 White male rage online 99
intersex, and queer people, and more particularly, Black Africans, Black
Caribbeans and others’. House of Rainbow has extensive collaborations –
from work in over 22 countries including Jamaica, Guyana, Lesotho,
Malawi, the United States and Colombia. These collaborations mean that
they have influence beyond Nigeria and the UK where much of the day-to-
day work takes place. They hold workshops both for support and education,
connect Buddhists, Muslims, Sikhs and other individuals of faith to Chris-
tians, have prayer, music, sermons, and food-based events to build com-
munity solidarity. They maintain safe spaces for improving mental health
and well-being, with regular training events and webinars on combatting
misinformation about what the Bible teaches about homosexuality. This can
be seen collectively as a form of ‘sanctuary’.
With a long history in African Indigenous and independent churches,
including Pentecostalism and other charismatic ministries, Reverend Jide
is a critical insider. His stance and that of the organisation is both personal
and political:

[House of Rainbow] started 14 years ago out of the need that also
affected me as a gay man who had come out about his sexuality and
was looking for a safe space where I can express my Christian beliefs
and my understanding of God . . . the early missionaries and even some
modern day missionaries, largely within the evangelical or extreme
right-wing conservative Christians are making this forceful claim that
God hates gay people and that homosexuality is an abomination. And
of course, for the millions of sexual minorities around the world, it
is inconceivable to have them brush on this canopy of hate and mis-
interpretation of the religious texts, regardless of the faith, regardless
of the religion, whether it’s Christianity or Islam or Buddhism or any
other religion. I think that the essence of queer people is very impor-
tant . . . House of Rainbow is the work of creating spaces for queer
people of faith, and also bringing the message of affirmation to the peo-
ple contrary to many conservative beliefs that ‘God condemns homo-
sexuality’. We believe God did not, nor did Christ, condemn same sex
relationships. . . . Homosexuality is a taboo. HIV is a taboo. . . . So that
is why we use social media, so that we can connect with people and
people can connect with us. The traffic of people connecting with us
is higher than what we can handle as an organisation. . . . We are faith
based because we do not discriminate against people of other religions
outside of Christianity. . . . We have colleagues who are gay imams or
lesbian imams who take on the responsibility of teaching and nurturing
and providing pastoral care for queer Muslims.
100 White male rage online
Reverend Jide and the other organisers of House of Rainbow face multiple
forms of hate – from discrimination and legal threat to abuse and violence.

We are considered an abomination. We are considered against nature.

They believe that we came from the bottomless pit of hell. And so here
we are, saying that god is loving, god is inclusive, god is liberating,
god is a freedom and god is queer because this is all of who we are. . . .
House of Rainbow started in the atmosphere where the Nigerian gov-
ernment had introduced the anti-gay bill to parliament . . . 14 years
imprisonment with hard labour for anyone convicted of homosexual-
ity. There was, about 5 to 10 years imprisonment for organising a gay
group or assembly of gay people [including a religious one] . . . . From
day one I received hate messages on social media. . . . [Under the blog
we wrote] the hatred was just unbelievable. And just two days ago,
I came across an article and I was reading it and then I read the com-
ments about me and my ministry. A few days ago, people were asking
for me to be killed and executed . . . . I report them to the police as
soon as I get them. People have taken out a petition page on change.org
against me as a “fake pastor”. People have set up an alternative Twitter
account in order to create a massive following of people that will hate
me and cuss me. Yeah, so I have a designated police officer in London
who I will just send all these things to. [Emphasis added]

One thing that struck us about the way in which Reverend Jide has been
targeted and his account of the intense anxiety, grief, anger and fear that he
has felt was the intersectional nature of the abuse, discrimination and vio-
lence he faces. From being stalked and threatened on the streets of Nigeria
for being gay and openly Christian to having his profession mocked and
ridiculed and being treated with contempt in the UK because of the colour
of his skin on the street or because of his accented English when he went
to an acting trial for a Shakespeare play in Glasgow, all the different facets
of his life come under pressure and scrutiny from people’s prejudices. This
intersectional experience of hate was repeated time and again during our
interviews.

The intersections of race and online abuse: ‘If you’re

Black or Brown your life is political’
Grace Blenkinsop, a young BLM activist, told us that her passion for the
movement was about a concern for the Black lives that often are left out of
debates: ‘Queer black lives, working class black lives, women of colour’.
Grace uses Twitter, Instagram and TikTok to comment on current events
 White male rage online 101
such as the death and ‘memeification’ of Breonna Taylor, with a high level
of engagement, but also high levels of targeted hate messages:

I’ve seen a lot of hate towards Black women, women of colour pre-
dominantly . . . Tik-ToK is such a horrible, horrible site for breeding
hate, it’s really nasty. . . . I have a lot of people comment on my videos,
who don’t follow me, and say that I’m always on their ‘for you’ page.
I guess if the wrong person can see your videos, there’s just a lot of
racist hate, a lot of sexist, misogynous comments, comments on girls
weight, appearance, especially with people that have links to the BLM
movement, just racist, derogatory terms surrounding that. [Speaker’s
emphasis]

The overlap between different forms of identity such as race and gender or
race and transness were key issues drawing the largest numbers of dehu-
manising and abusive comments online. Neither trying to ignore the com-
ments knowing that there are a lot of sock-puppets (fake identities) and
bots involved, nor putting on filters, has reduced the shock and distress of
encountering hate in the intimate spaces of inboxes and comment sections:

[D]uring the start of the year, when the whole BLM movement resur-
faced, I was active making videos about protests and calling systems
out, like the police and the government. . . . But people took it very
personally. It was actually really troubling. I had a lot of comments on
my videos – people saying that I should die or get hit by something,
or really graphically explaining how they want me to die or my family
to die. That was on Tik-Tok. Then, because my Instagram was linked
to my Tik-Tok, after I’d block those people or delete the comments, it
came on Instagram. I was just so confused as to why I was trying to
help educate people and then suddenly I was getting told to get hit by a
bus. It wasn’t just comments directly about me, it was comments about
the movements. People would DM me on Instagram with jokes about
George Floyd’s death. . . . I made a video about police officers being
racist and I had loads of people message me like ‘my dad’s a police
officer, f*** you’. Some people were really angry to the point where
I would read their message and not reply because I wouldn’t want to
spur it on and they’d keep messaging me like, ‘f*** you, N-word’, and
all of these racial slurs. [Speaker’s emphasis]

Reading through the transcripts of our interviews for this book in the context
of the torrent of racial abuse endured by Black British footballers Raheem
Sterling, Marcus Rashford, Bukayo Saka and Jadon Sancho in July 2021
102 White male rage online
after the Euro2020 cup final penalty defeat, that reaction by a large sec-
tion of the British public to these successful young stars (which shocked
many white British commentators and engaged a response of disavowal)
was utterly predictable. It reveals only the tip of the iceberg when it comes
to everyday life for Black British citizens and many other people of colour
in the UK.
Comedian Sandy A. (a pseudonym, at her request) explained forcefully
that intersections of gender and race provoke much hateful content she
receives and underpin experiences of discrimination. Given her occupa-
tional visibility and the fact that she’s British Caribbean, she cannot hide.
She told us passionately, ‘If you’re Black or Brown your life is political,
isn’t it? Unless you’re stupid. It has to be, we’re politicised as kids’ and later
‘Being an opinionated Black or Brown person in this country is going to
bring you grief from everywhere’. In Sandy’s experience, racism is a driver
of much social media hate:

Racism would be the biggest one. Sexism. And in racism, I include

Islamophobia. That’s another one that gets people going. Transgender
issues get people going – that generates a lot of hate speech. Black
people in adverts seems to upset people! Anything to do with identity,
I think. Anything outside of white identity, and it used to be white male,
but I think it’s white female as well now because I think white women
are just as bad . . . I think white supremacy adapts, doesn’t it, and
changes itself to survive. . . . So, the kinds of hateful stereotypes I see
online are ‘transgender women are really rapists in disguise – they’re
just trying to get at women’; ‘Muslim people are extremely violent and
they just want to Islamify Britain and steal Christmas’; ‘Black people
are stabbing everybody and lazy’; ‘Jewish people are greedy’, but that
one’s not so much anymore. . . . The really nasty ones will be talking
about the death of my daughter . . . .
I wrote an article about Islamophobia and then I had to write a fol-
low up article because the response was so bad. It ranged from, ‘go and
live in Iran, or why don’t you go and live in Iran’ – I had a dress on in
my profile picture at the time – to ‘go and live in Iran in that dress’.
I was like, ‘why would I? It’s a winter dress?’ So, that would be the
mild. Then it’s like, ‘take Muslims into your house, I hope they rape
you’. Then, ‘get out the country’, then accusations of being a secret
Muslim. . . . I had EDL [English Defence League] members saying
they were going to come to my [workplace], and they were posting
pictures of a noose and saying they were going to lynch me, and ‘Hope
you die like your daughter, you fucking Golliwog’. You’ll report it and
 White male rage online 103
[the Platform] goes, ‘It hasn’t breached our community standards’.
[Speaker’s emphasis]

Failed by the inadequate and biased implementation of rules on hate speech

and harassment to apps and platforms, Reverend Jide, Grace and Sandy, like
many other key informants, took issue with being told to just ‘ignore’ the
racist, misogynist or homophobic threats and slurs, ‘don’t feed the trolls’;
or to just ‘have fun with the trolls’ by trolling them back. As Sandy summed
up: ‘We can’t have fun with them, because they literally want us dead’.
The refusal and/or inability to recognise racism in language or action,
denial, identification with racists and systemic racism are all named as
issues preventing moderation teams and the police from taking adequate
action against those posting hateful content and misinformation. Reverend
Jide and Sandy in particular have faced such serious threats that they now
report regularly to cybercrime units and pursue platform moderation teams.
They are sceptical about the ability and commitment of these institutions
and intermediaries to change the overall context in which hate is manufac-
tured, circulated and does its divisive political work.
Clayton Wildwoode, a working-class, white queer student who co-
founded ‘All Black Lives, Bristol’2 at the age of 19, insists that racism,
classism, misogyny and transphobia intersect, underly rising social tensions
in the UK, and have been fuelled strategically by the ‘Tory’ elites with Gov-
ernment connections. As someone who counts himself an anti-racist ‘ally’,
Clayton was clear that he was still learning about the profound effects of
racism, misogyny and transmisogyny on Black communities and Black
women, in particular. He explained how All Black Lives, Bristol often gets
trolled on Facebook and Twitter by white people ‘who write essays under
our posts explaining that we’re wrong because racism doesn’t exist, or that
we’re being racist to white people’. In his experience dehumanisation and
misrepresentation are incredibly varied, covering body shaming within the
gay community, comments on appearance and gender, as well as suspicion
of bisexuality and Biphobia. All of these, he noted, are exacerbated at the
intersection of race and class. As such it is worth ending this section with
Clayton’s reflexive commentary, which emphasised that there is often an
assumption that all minoritised groups support each other. There is, how-
ever, sexism amongst Black cis and gay men and racism and misogyny
within the white gay community that comes out in online spaces as well as
in discriminatory actions, attitudes and violence:

Everyone that’s not a straight white cis man thinks that, oh, ‘if I’m
queer or if I’m a different race, it automatically means I’m woke’ and
104 White male rage online
that you are accepting of everyone, but it’s not true. There’s so much
misogyny in the gay community and then there might be different types
of racism from different races. So just because you’re not the stereo-
typical racist or sexist person, I think you kind of shrug it off as, ‘oh I’m
gay and there’s prejudice against me so I’m obviously not prejudiced
against other people’, but that’s not true.

We were deeply moved by and learnt a lot from the ways in which Clayton
and our other interviewees used the space of the interviews to reflect on
their relationships to different facets of identity in their chosen and birth
communities.

‘We’re not allowed to be Jews because we’re not

the sort of Jews that they want us to be’: Political
censorship and the destruction of democracy
Clayton’s and Sandy’s commentaries were pointed with regard to UK insti-
tutional politics. In their opinion, stereotypes about Black people, Muslims
and Jews could be found amongst commentators on the right and the left, and
amongst majority and minoritised groups. Sandy was, however, very clear
that this was most evident amongst members of the Conservative Party, and
in factions on the right of the Labour Party who ousted Jeremy Corbyn (leader
of the Labour Party from 2015–2019) after an insidious campaign of disinfor-
mation. To get a closer insight into the disingenuous and destructive tactics
utilised by the Labour Right against their own members who support the Pal-
estinian people’s struggle or call out Zionist racism, we interviewed two long-
time left-wing activists. Their experience is in trade unions and the Labour
Party. Mike Cushman, a founder member and Membership Secretary of Jew-
ish Voice for Labour (JVL) who frequently represents the group, explained
how complex the tensions have become around trolling, anti-Zionism and
antisemitism in the UK. He explained that although there are some ‘lone rang-
ers’ who engage in trolling or sock-puppetry, much of it is coordinated by
groups of hard-line Zionists with access to funding, media and social media:

The fact that we are dissident Jews offends them greatly because they
really don’t know what to do with us. So, a lot of it is trying to deny that
we’re Jewish. We’re not allowed to be Jews because we’re not the sort
of Jews that they want us to be. . . . [The trolls have] a vastly inflated
idea of our impact and reach. They complain about conspiracy theories
about Jews. Not incorrectly, because these can be very dangerous theo-
ries. But they then seem to engage in the same conspiracy theory about
[members of JVL], ascribing to us all sorts of malevolent power and
 White male rage online 105
influence. . . . So things like we are ‘pulling Jeremy Corbyn’s strings’,
‘we had all this privileged access to Jeremy’ (In fact, Jeremy kept us at
a very great distance.) That we are ‘a nest of anti-Semites’, ‘holocaust
deniers’. The phrases go on and on and reverberate and reverberate.
Their favourite label is ‘cranks’. As soon as you say you’re not a crank
you’re proving your crankishness. It’s a double bind. . . . There was a
whole Twitter feed – it was recently taken off – called “JVL Watch”
which was people who just continually trolled us and harassed us . . .
We’re Jewish, and it’s very hurtful. I know of at least one person who
attempted suicide [because of this trolling].

In a similar vein, Labour councillor Jo Bird is on the left of the Labour party,
and recounted the impact on her life when online and offline harassment
against her intensified, including an unjustified suspension from the party
in February 2019 over her support for Palestine. She has found herself with
doors slammed in her face on doorsteps because she was canvassing for
the Labour Party (which at the time had endless disinformation published
against it in the mainstream media). She also had non-Jewish people telling
her that she doesn’t understand antisemitism. She’s experienced bullying
and harassment from those on the rightwing of the Labour party3 amplified
by the media because, as an anti-Zionist Jewish socialist, she did not ‘fit’
the Labour right and Zionist narrative about institutional antisemitism in
Jeremy Corbyn’s Labour Party.

From my lived experience as a Jewish person in the Labour Party it’s

been a very welcoming place. I very quickly became a councillor. I’ve
not, in day to day, face to face interactions seen antisemitism and I would
notice [if it were there], because I do know what it looks like. I come
from a generation of a family that know what anti-Jewish racism looks
like. That’s why my family is in the UK, not in Prussia, in Poland where
it started off. So, me saying that doesn’t fit the narrative that is being
pursued by other organisations and the [Zionists inside and outside the
Labour party] don’t like that I continue to speak. . . . In October, maybe
November 2019, [the hate campaign] effectively stopped me becoming
a candidate to be an MP. So, the Jewish Leadership Council and Mer-
seyside Jewish representative committee put out a joint Tweet statement
saying how awful it would be [if] I was selected as an MP candidate . . .
and how it would exemplify the antisemitism in the Labour Party – they
didn’t mention that I was Jewish. The Jewish Chronicle ran a story on
it amplifying those voices – didn’t mention I was Jewish again, erased
that protected characteristic and the key part of who I am and what I do.
And I wasn’t shortlisted, wasn’t invited for interview. . . . and back to
106 White male rage online
my [suspension] in March, they tried to stop me from speaking. They
were basically saying that the allegations that had been made against
me were enough that I shouldn’t be heard even to defend myself against
those allegations. Three independent councillors walked out when I,
a Jewish councillor, was talking about the Holocaust [in the council
chamber]. And that is hatred. That is anti-Jewish hatred.

Similar to the Brazilian activists discussed in Chapter 3, Jo has endured an

excruciating cocktail of hateful misinformation, with ‘dossiers’ of false or
sensationalist stories denigrating her work and character spread by right-
wing commentators and amplified by mainstream media. While much of the
hate was aimed at the left of the Labour party, the toll on Jo’s life has been
severe, and in a recent statement on her unfair expulsion from the Labour
Party, she evinces relief rather than disappointment. The political implica-
tions in terms of democracy and voice for those who challenge political
authoritarianism and settler colonialism are equally acute. As Jo notes, the
ultimate goal is to silence all dissent.

The attacks on people like me are constant; it’s just like these never-
ending picking up on what you’ve said, what you’ve written, trolling,
going through Facebook pages. Hundreds of people have been sus-
pended or investigated for things that are just not at all a breach of
the rules, for political disagreements. And the combined effect of all
of that is, is silencing. So, people, including myself, we use Facebook
less. We use Twitter less. We don’t talk about Israel and Palestine as
much. And we don’t talk about antisemitism as much, and people don’t
respond as much either. It’s no longer seen as a legitimate topic for
political debate. It’s seen as something that if you say anything about
it, you could be suspended. You could have your reputation trashed in
the press. You could lose your job. You can lose your position, elected
position or your candidacy. There’s fairly severe penalties in this soci-
ety for that. . . . there was factions within the Labour Party that wanted
the party to lose the last two general elections. . . . And the evidence is
there in the public domain. And antisemitism, that constant smear of
‘the leadership is racist’, was an integral part of that strategy.

Jo’s and Mike’s accounts of political manipulation of public discourse in

order to further particular political causes, prevent the ascension to power
of pro-Palestinian activists, and ensure that any criticism of Israel’s politi-
cal violence or discrimination, and of Zionism is classified as a form of
antisemitic hate speech, have been confirmed multiple times by scholars
of this period and this politics from the late David Graeber to the poet
Michael Rosen and the actress and activist Miriam Margolyes. These
 White male rage online 107
Jewish critics and commentators have neither denied the existence of anti-
semitism within British society and in the Labour party more broadly, nor
suggested that being Jewish means that someone is incapable of antisem-
itism; their positions are subtle, well thought-out and rights-based. They
have all supported mechanisms for ensuring that racism of all kinds is
called out and that Jewish individuals and communities in the UK are pro-
tected. Their well-evidenced stance, however, is that the exceptionalist
narrative which conflates the actions of a state (in this case Israel) with
a protected group across the globe should be challenged. In challenging
antisemitism, anti-Palestinian and Islamophobic racism at the same time,
like Jo and Mike in this chapter, all have followed a long history of Jewish
socialism and internationalism. Losing this voice from the history taught to
young people and from the media that circulates to make sense of contem-
porary events increases rather than lowers the visceral dangers of racism
faced by both Jewish and Muslim communities in the UK.
The narrowing of the school curriculum4 and focus to a Zionist ideology
that pushes to exclude Palestinian voices within the British history curricu-
lum is, to these commentators, another example of the selective and hypo-
critical championing of freedom of speech by the now mainstream alt-right.
Fuelled by rightwing politicians in the UK with links to alt-right think tanks
in the US, funded by corporate and political interests, and circulated on both
social and mainstream media, the idea that anyone who stands up for Black
or Palestinian rights is a ‘racist’ or is ‘antisemitic’, that those who challenge
capitalism are vicious authoritarians harking back to Stalinism, has taken a
strong hold of the imaginations of a significant section of the British public.
Untethered from ethics, and targeted at particular demographics through
Facebook, Twitter and Instagram, these ideas have gained currency in ways
which the more complex counter-narratives have been unable to. Like the
duplicitous slogans of the ‘Vote Leave’ campaign around the Brexit Refer-
endum, ideas about ‘mad Marxists’ bankrupting the treasury and destroying
British culture by pandering to minorities have a long history in far right
propaganda campaigns from Nazi Germany and Fascist Italy to Hindutva
India (Koonz, 2003; Caprotti, 2005; Banaji, 2018).
Mike turned to history to illuminate why we find ourselves in this situ-
ation of exacerbated public acceptance of mistrust and hatred against anti-
racists and socialists:

Hate speech has always existed, but when you say it to five people, in
a pub at 11 o’clock at night, it isn’t the same as saying it on Twitter or
Facebook . . . so the technologies are significant. . . . Add to that, the
collapse of a socialist ideal has all sorts of effects. . . . We get this bizarre
statement from the Department of Education that schoolteachers are
108 White male rage online
not allowed to use anti-capitalist texts. You defend freedom by abolish-
ing it! [Speaker’s emphasis]

Jean Clemens (a pseudonym, at the request of our informant) is an academic

at a well-respected British university who researches far right populism,
racism, and the mainstreaming of the far right through an examination of
elite discourse. Jean, who has a significant following on Twitter and also
receives a range of vicious trolling there, points to the ways in which both
mainstream and social media giving platform to the views of extreme rac-
ists has made fascist ideas increasingly acceptable: Through ‘the hyping of
the far right, whereby far right ideas have been given exaggerated platforms
and ways to portray themselves as the voice of the people or as bigger than
they really are in many ways’:

The trolling has been mostly on Twitter . . . I’ve been trolled by the
typical trolls, the people who are anonymous, who just latch on. Inter-
estingly, some of the more violent insults that I got were addressed to
me as a woman because my name, Jean, for some people sounds like
it’s a woman’s name. . . . it’s not surprising that people who are racist
are usually also sexist. For them, it’s a way to justify them not reading
my article at all. Just saying, oh look at that woman saying something
rubbish. Quite interestingly, as well, I’ve been attacked sometimes by
French people for commenting on French politics as they thought that
I was American or British.

In our discussion with Jean it became evident that random posters’ hateful
commentaries on people’s online posts are often based on xenophobic,
racist and gendered assumptions about the recipient, so ostensible politi-
cal disagreement is tethered to other, more visceral phobias and preju-
dices. Jean has managed to ignore much of the trolling, assuming that he
isn’t a big enough threat to the powerful for his data to end up on organ-
ised troll farm lists or machines. However, he has been targeted through
his work and his job, which has caused him extreme stress and anxiety. He
became a target of violent incitement as a consequence of signing a letter
against hate:

The final trolling that I got was death threats in the comments of an
article that was written by other academics on a right-wing website and
the article only mentioned my name, even though it was talking about
a letter that had been signed by more than 200 people. And in the com-
ments, there were some very clear death threats. Like some guy was
 White male rage online 109
talking about his gun in great detail and about how that gun could be
the solution to political correctness. I challenged them, the office of the
article, online on Twitter and they couldn’t care less. . . . I think there’s
a bit of a difference between saying ‘your research is a bit shoddy’ and
‘here is a gun’.

Jean was candid that putting himself and his research out there was a choice
that he as a white male was making and that this distinguished him from
others who receive hate due to protected characteristics. Yet he did not mini-
mise the ways in which the trolling affected him. The emotional cost was
anxiety and exhaustion. From discussions in Brazil, India and the elsewhere
in the UK, one of the key distinctions we draw in regard to online hate
and trolling is between the systematic, strategic and politically manipulated
groups of trolls and the ‘lone warriors’ either seeking thrills, performing for
their own followers, or airing their resentments and passionate beliefs in
rightwing ideological positions. Discussing the kinds of people who engage
in violent online trolling, Jean added another group – those who deliberately
stoke controversy and prejudice for profit:

The death threats make me feel scared. But then again, [I’m far safer]
than many other people who are at the sharp end of racism, sexism. I’ve
chosen to be a public figure, I’ve chosen to fight racism and the far right
publicly . . . it’s not like me suffering because of my gender, my race,
my sexuality, something that is just who I am. But it’s extremely tiring.
And I’m not even talking about the death threats . . . just the basic troll-
ing gets to you because when you dedicate so much time backing up
your work, creating arguments, but also arguments that are really the
basic kind of tenets of my work are: we live in societies that are deeply
unequal, wouldn’t it be a lot better for everyone if they were more
equal, right? It’s pretty basic. If we were more democratic, if we were
more equal, if people had equal say. And then you get people who will
fight against that idea not because of ignorance, but because they have
something to gain out of it; and they will be disingenuous. They will be
well-funded, they will have massive platforms and it’s just massively
disheartening. . . . For this group, I think the more dangerous ones, it’s
grift. They know that they can make money. They know that they can
find followers using hate and hatred towards other people. Right. They
know that racism sells. They know that dodgy understandings of free
speech these days sell massively. . . . And they use the anger of certain
sections of a population to sell books, to get money from right-wing
think tanks. . . . [Speaker’s emphasis]
110 White male rage online
Religious groups and individuals can engage in both individual and organ-
ised hate online and, while some is motivated by inegalitarian or conserva-
tive belief and fervour, there is also some that falls into Jean’s category of
‘grift’ and ‘profit’. As Reverend Jide recalled, this kind of hate can be a
multimedia affair starting in a traditional medium such as radio or televi-
sion and spilling over onto blogs, websites and social media platforms. The
continuum of offline and online hate and harm can be even more damaging
for those surrounded by communities and families who share their perse-
cutors’ hateful beliefs and values. The end result in Reverend Jide’s view
is a strong possibility of mental health breakdown and suicide attempts or
suicidal ideation:

There was one year where an entire radio station in Nigeria dedi-
cated their entire day program on ‘what should the Nigerians do to
the homosexual pastor Jide Macaulay?’ An entire day. Now, if you
are a listener to that radio channel . . . let’s not undermine the power
of communication online. I mean, we talk about terrorism. We talk
about home grown terrorists. Many of them have been impacted by
the words that they share online within those communities. So, the
homophobia that is driven online is also fueling homophobia in many
places. Let’s not underplay or downplay how it would then affect
people like myself, psychologically and mentally. . . . If you have
a Facebook handle with about a million people or Instagram with a
million, you have a responsibility when you post something on your
page because it will reach . . . 10 to 20% of those people will be
picked up. . . . If your message is extremely racist and homophobic,
it will ruin lives, there is no doubt about it. . . [W]hat we don’t know
is what happens to people in various spaces who do not have the
power to deal with it. I can only imagine somebody who is in a house
where they are homophobic and then their social media is filled with
homophobic messages.

In Reverend Jide’s view, entities or individuals with large social media

followings, like mainstream media, have ethical and legal responsibility
to stop the spread of misinformation. Yet time and again, we were told
of the misuse of power and lack of action on the part of platforms, when
those posting dehumanising and hateful content are savvy enough to dis-
guise their language or couch stereotypes and disinformation in ways
that appear to be valid debate. In this penultimate section, we turn to the
fraught arena of LGBTQIA+ rights, homophobia and transphobia, where
despite decades of systematic campaigning and greater public acceptance,
 White male rage online 111
legally protected characteristics are not guarantors of safety at home, on
the streets, or online.

LGBTQIA+ rights, homophobia and transphobia:

‘A seed that grew into a hideous tree of hate speech’
In order to understand why it is both pointless and patronising to tell peo-
ple who face daily abuse and threat online to ‘just get off social media’,
it is important to examine the experiences of people who must perforce
use social media regularly as part of their jobs. Ben (a pseudonym, at his
request), an LGBTQIA+ charity communications officer, told us that he
couldn’t carry out his role at work without social media. His work requires
the use of different social media for different audiences, and for different
purposes:

My whole life revolves around social media. So, I find that before
I even start working, I’m already browsing Twitter and then when
I start working, I immediately browse Twitter more. My job is incred-
ibly varied. My job will be everything from running and planning our
social media, so everything from writing a tweet to planning a com-
munications strategy [and. . .] events as well. . . . in a work capacity,
we use Instagram, but that’s a very different kind of platform for us
because our following is mainly young people . . . We have LinkedIn
for advertising jobs. Facebook is, again, a slightly different audience
of general supporters. I’m writing for all these different platforms with
slightly different voices. Then, in my own life, I have no interest in
Facebook, myself. So, Twitter and Instagram for myself. . . . I’ve gone
through occasional periods of deliberately not looking at Twitter out-
side of workdays for mental health reasons, purely because there’s been
moments through the past few years where it’s become really emotion-
ally involving to my detriment.

To expand on why social media becomes such a drain on his mental health,
Ben explained to us that transphobia and homophobia have grown in the
UK, despite some improvements since 2000.

I started to notice back in 2017 that there was a bit of a push back
against the work that we were doing around supporting trans young
people. When the government started to look at reforming the Gender
Recognition Act at Westminster and in Scottish parliament, there was a
112 White male rage online
lot of misinformation flying around which essentially led to the paint-
ing of trans people as a threat to women. I feel like that was the seed
that has then grown into this really hideous tree of quite unveiled hate
speech. . . . Back in 2017, I posted something that I thought was vaguely
uplifting, something vaguely cheesy about recognising the humanity of
everyone involved and being compassionate. And then it received over
a hundred replies from people that wholeheartedly disagreed. I think
that was the moment when I thought, oh, we haven’t come as far as
we thought we had . . . there’s people out there searching for the word
‘trans’. And then, there is a whole bubble of them that will then reply
to each other’s tweets and to each other’s replies. . . [For example]
Mumsnet was a site which, essentially, is like an old school message
board site, but they also now have big advertising partnerships with
companies. It was set up to be a forum for mums to share insights and
build a kind of community. But there is a particular channel on there,
called “Mumsnet women’s rights” . . . Five minutes on there and you’ll
see that every single post is about trans people. I feel like some of the
heat on social media, some of the debate is sparked by Mumsnet users
screenshotting other people’s tweets, posting them in these forums and
that then generating another social media storm. I feel like that particu-
lar bubble is almost a bit of a radicalization community at the moment.
It’s pretty scary. . . . I have older trans friends who are literally strug-
gling to leave their houses now for fear of how people react to them.
Personally, as a gay man, I think there’s a lot of parallels between what
trans people are facing now and what gay men experienced decades
ago. It’s just like history repeating itself. [Ben ext. 2, emphasis added]

Ben recounted how comments targeted at him would say things like ‘How
disgusting, your mother would be ashamed of you [for supporting trans
rights]’. His mother was upset by the comments, rather than by his princi-
pled position. Equally worrying, he felt, was the reaction to the educational
work that his charity does. When they released a carefully evidenced and
compassionate information pack for schools on LGBTQIA+ rights and sup-
port, they were deluged with hateful messages from people accusing them
of promoting child abuse and trying to get the pack banned. Referring to the
Thatcher-era law, Section 28, which saw the teaching of any material which
‘promoted’ homosexuality banned in UK schools, Ben explained that his
own adolescence had been profoundly affected by the lack of information
other than the homophobic content in the tabloids at the time.
Like Ben, Raymond Howell (a pseudonym, at his request) is a young,
Black gay man who works for ‘an LGBT international human rights
 White male rage online 113
charity’. He too described how a large volume of hate encountered during
his job and rights advocacy work is against trans people.

I see a lot of misinformation, and willfully misrepresented information

on trans identities: a lot of cherry picking of particular facts around
trans people; a lot of trying to paint trans people as inherently antag-
onistic for existing; trying to paint trans people as potential abusers,
when actually the vast majority of trans people that I know and work
with are just people who are trying to get on with their lives.
[RH. Ext. 1]

One of the most insightful aspects of our discussion with Raymond centred
on his understanding of the intersection of racism and dating/sex – what he
called ‘sexual racism’. His attempt at pedagogic intervention online, how-
ever, proved to be a major trigger for all kinds of racism and violent threats
against him.

I’ve made infographics about sexual racism and racism in dating and
in romance and in sex, because I think that’s a topic which is really
misunderstood and there’s really interesting research on it that peo-
ple don’t know about. Another one was talking about racism and
queerness and how there’s a very strong relationship between homo-
phobia, transphobia and racism. When I posted the infographic on
sexual racism, it’s a very controversial topic in the sense that peo-
ple . . . have lots of preconceptions around sexual preferences and
who they date, thinking that’s personal and you can’t touch that. And
some of that is true, but there are ways that inherent biases interact
with who you do or don’t find attractive and it’s important to unpack
those, but it was something which went a little bit viral. A lot of
people who agreed with me and found it interesting shared it and
that led to people seeing it and disagreeing with it. Other people
who disagreed with it shared as well . . . That was by far the worst
instance of online abuse because it was very targeted. So I have some
YouTube videos and there were people commenting and putting a
thumbs down on every single video. People found my email address
through a well-being workshop for queer people of colour and they
started like sending me threats through email. Someone contacted
my mother on Facebook . . . people sent me death threats, slurs and
words that are related to parts of my identity, taking apart my appear-
ance . . . It was crazy. It was really crazy. It had a massive impact on
my well-being.
114 White male rage online
Raymond was accosted on the street by a man who filmed him on his phone
and started telling him how much he hates Black people, using the N-word,
and threatening him. After that incident and the barrage of homophobic and
racist abuse and death threats, Raymond was exhausted, frightened, anxious
and dejected and took several months away from social media. In discus-
sion after discussion, we were told of the toll that attempting to engage with
a wider audience online around racism, homophobia and transphobia takes
on people’s confidence, mental health and social interactions. And, while
in this instance, much of the abuse was coming at Raymond from whites
who could not bear to have their own racism analysed or pointed out, abuse,
threat and violence is also directed at Black and Brown LGBTQIA+ people
from within Black and Brown communities.
Ferhan Khan, another young LGBTQIA+ rights advocate, was keen to
discuss the issue of identity, and to emphasise that the kind of pigeon-holing
and typecasting of identities which takes place in online environments by
those doling out dehumanisation and hate is far from the fluidity of experi-
enced identity:

I’m a non-binary male . . . but Muslim as well. At the same time as

being Muslim, I’m also an atheist. So I’m a Muslim atheist. I’m a Brit-
ish Pakistani, I’m a Scottish Pakistani. I’m all of these things and that’s
okay, I don’t have to be one thing. I’m also a transportation planner, a
project manager, and a political activist, and an LGBTQIA+ activist,
and trans right activist . . . I’m a Stonewall BME leader.

Ferhan recounts their early involvement in politics and LGBTQIA+ rights as

a participant in a film for the Naz and Matt Foundation,5 started by Matt after
his beloved partner of 12 years, Naz, killed himself when he came out to his
family and they rejected him. Ferhan’s activism involves making videos with
complex and diverse representations so that the media narrative does not pit
Muslims (such as the conservatives in England who protest against LGBT-
QIA+ sexual and relationship education in schools) against the gay commu-
nity (assumed to be white). However, while there are moments of connection
and solidarity, and it brings them closer to other Black and Brown queer
people, Ferhan also receives an avalanche of abuse for their social media
postings on race and sexuality. Visibility and abuse are directly related.

I tend to post a lot of stuff on social media, Facebook, Twitter, maybe

even Instagram, about mental health, racism, LGBTQIA + topics. I put
them out there, and then I just see all the racists come and find me and
then scrap at me and say, ‘you should just be grateful that you’ve not
been thrown off the top of a building by ISIS’. And I’m like: ‘okay
 White male rage online 115
I don’t feel like I should be grateful for anything like that, respect
me’ . . . . I have been getting a lot of hatred in the form of racism,
in the form of homophobia from Muslim South-Asians, and racism
from white people and white gay people, especially. I get a little bit
of transmisogyny as well for my feminine side. . . . I’ll get direct mes-
sages saying, ‘you ought to go to hell. Being gay is wrong. It’s against
Islam’, ‘You’re going to go to hell’ or ‘I hope you go to hell, hope you
die’, ‘Hope you get bombed’. And then a racist person will say, ‘I hope
you got thrown off the top of a building by ISIS’ and things like that.
Then, there’ll be those that do it publicly but framed as a debate, but
actually it pretty much feels like racism, feels like homophobia. But,
because it’s framed in such an almost innocent childlike way, it just

Figure 5.1 A selection of hateful content received on Twitter by Dr Shola Mos-

Shogbamimu. Image credit: Dr Shola Mos-Shogbamimu.
116 White male rage online
passes under the radar. So there’s that kind. I got recognised on the
street because I took part in a TV show. I was chased down the street
before – that happened three times. That was in [London]. That was a
group of South-Asian men.

When Ferhan appears on television, especially on the BBC, there is a

spike in hate and threats, some of which spills over into the physical world
before slowly lessening as time passes. We’ve heard this called the para-
dox of visibility, implying that the more visible one is, the more hate one is
likely to attract. However, with one or two notable exceptions (for instance
Jeremy Corbyn) our research in the UK confirms that the stereotyping,
dehumanisation, abuse, incitement to violence and actual physical attacks
attached to politically outspoken Black and Brown people of whatever
religion, sexuality or gender, are generally far worse than those attached
to the same sort of visible politics for white counterparts. This intersection
of racism, misogyny, transphobia, Islamophobia and politics equates to a
steady and deliberate effort to keep certain voices out of the public sphere.
Dr Shola Mos-Shogbamimu, an outspoken critic of racism and corrupt
governance, has given us permission to reproduce some of the hundreds of
abusive Twitter messages which she receives on a daily basis (see Figure 5.1).
Shocking as these messages are, they are the tip of the iceberg when it
comes to the daily vitriol levelled at outspoken Black, Brown, Indigenous
and Muslim women in Brazil, in India and the UK – and as we know, in
Pakistan, Colombia, the United States and many other countries too. So, as
we conclude this chapter, it is important to dwell on what we might call the
‘hierarchy of hate’ and its antecedents.

Conclusion
Some of the historical processes and events discussed in this chapter are
traceable within the histories of colonial and nationalist fervour invoked
by rightwing politicians and latterly by a host of other febrile nationalists:
‘Take Back Control’ was, after all, the slogan of Brexiteers. Interlinked his-
torical events and processes stand out as the backdrop to the ways in which
discrimination, abuse and violence, both on and offline, impinge on and
injure the lives and life chances of contemporary communities in the UK.
In the language that was reported to us as standard fare of trolling for Black
and Brown commentators, activists, students, journalists and academics, we
noted constant stark reminders of the British empire and its violent colo-
nial legacy. Chattel slavery and the building of British institutions based on
capital accumulated through genocide are frequently whitewashed or disa-
vowed even in the institutions to which we are affiliated and within which
we teach. Myths about the second world war — and the misleading rhetoric
 White male rage online 117
British exceptionalism which denies the links between the British far right
and Nazism — have led to a cosy relationship between supposedly centrist
or conservative politicians, media houses and far right ideas. This has led to
a mainstreaming of far right populism even by the supposedly left-leaning
press (Brown & Mondon, 2021).
Other processes which we deem relevant to understanding the testimo-
nies of the massive number of people who are at the receiving end of dis-
crimination, prejudice and hate on a daily basis can be traced to deep-rooted
social conservatism and the lack of a systematic social revolution in the UK.
The ease with which transphobia has swept through the country and the sud-
den resurgence of biological arguments about gender amongst women who
were supposedly committed feminists is deeply troubling. Further, when
all is said and done, and despite the Chartists, the Fabians, the Levellers,
the Suffragettes and many generations of abolitionists – much white work-
ing-class solidarity in overlapping struggles of Black and Brown members
of the working class has only ever been fragile, while white middle class
solidarity has often atrophied in the face of an assumed competition from
Black and Brown fellow citizens. Farcically, but also with fascist under-
tones, recent Government-sponsored reports have loudly denied the exist-
ence of structural racism6 and have tried to lay the blame for a lack of white
working-class progression at the door of anti-racist teaching.7 In tandem,
the concept and foundations of feminism have been derided by an insidious
backlash in popular culture and politics, and trade unions have been weak-
ened by decades of rightwing curbs and legislation as well as anti-union
propaganda in the mainstream media.
In the UK, which some like to claim is one of the ‘oldest and most
mature democracies’, it is frighteningly easy to recognise parallels with
the hate politics and media propaganda of the USA, India and Brazil.
Recent eulogies of empire parallel the apparently popular scepticism of
social justice work (Social Justice Warrior or ‘SJW’, ‘woke’ and ‘can-
cel culture’ now being terms of abuse used on message boards of the
alt-right but also in the parlance of centrist journalists, politicians and
academics). We are alert to the destructive effects of increasingly swift
and bitter public shaming and factionalism amongst members of pro-
gressive groups online, and aware of instances of misinformation and
sensationalism by left-leaning politicians or media. However, those who
engage in struggles against racism, classism and sexism find themselves
vilified by a majority of the UK’s press and public service media which
leads to a deep legitimisation of the kinds of vitriolic sentiments that are
expressed against them on Twitter, Facebook, Instagram and other social
media platforms. We return to this point, made in passing above, because
it is the British mainstream media and their servile allegiance to a bru-
tal Conservative regime alongside the vindictive machinations of a now
118 White male rage online
largely ineffective opposition party, which have forced dissenting voices
into marginal spaces or into small but thriving alternative media enclaves
and tied them ever more firmly to the use of social media.

Notes
1 www.theguardian.com/media/2020/sep/25/andrew-neil-launches-24-hour-
new-channel-to-rival-bbc-and-sky
2 www.vogue.co.uk/arts-and-lifestyle/article/organisers-bristol-black-lives-
matter-protest
3 www.independent.co.uk/news/uk/politics/labour-leak-report-corbyn-election-
whatsapp-antisemitism-tories-yougov-poll-a9462456.html
4 www.theguardian.com/education/2021/jun/08/uk-history-education-row-israel-
palestine-textbooks-pulled
5 www.nazandmattfoundation.org
6 www.runnymedetrust.org/sewell
7 www.forbes.com/sites/nickmorrison/2021/06/22/its-not-white-privilege-thats-to-
blame-for-failure-of-poorer-white-students/?sh=58412035f470

References
Adamson, T., Wallach, S., Garner, A., Hanley, M., & Howell, S. (2019). The global
state of conversion therapy – a preliminary report and current evidence brief.
https://osf.io/preprints/socarxiv/9ew78/
Bell, D. (1995). Who’s afraid of critical race theory. University of Illinois Law
Review, 893. https://heinonline.org/HOL/LandingPage?handle=hein.journals/
unilllr1995&div=40&id=&page=
Brown, K., & Mondon, A. (2021). Populism, the media, and the mainstreaming of
the far right: The Guardian’s coverage of populism as a case study. Politics, 41(3),
279–295.
Caprotti, F. (2005). Information management and fascist identity: Newsreels in fas-
cist Italy. Media History, 11(3), 177–191.
Crenshaw, K. (1989). Demarginalizing the intersection of race and sex: A black
feminist critique of antidiscrimination doctrine, feminist theory and antiracist
politics. University of Chicago Legal Forum.
Davis, A. (1981). Women, race and class. Penguin.
Delgado, R., & Stefancic, J. (Eds.). (2013). Critical race theory: The cutting edge
(3rd ed.). Temple University Press.
hooks, b. (2014[1984]). Feminist theory: From margin to center (3rd ed.). Routledge.
Koonz, C. (2003). The Nazi conscience. Belknap Press.
Teltumbde, A. (2010). The persistence of caste: The Khairlanji murders and India’s
hidden apartheid. Zed Books.
Tyler, K. A., & Schmitz, R. M. (2018). A comparison of risk factors for various
forms of trauma in the lives of lesbian, gay, bisexual and heterosexual homeless
youth. Journal of Trauma & Dissociation, 19(4), 431–443.
6 Conclusion

Social media usage and the solidaristic, mutual, collegial and parasocial
relationships it enables are an immensely important feature of contemporary
society. This holds regardless of the varying numbers of social media users
in each country, in our study, and across the globe. Politicians across the
ideological spectrum routinely generate and attempt to influence political
events via social media. This political discourse is amplified or critiqued
and embroidered by broadcast and print media and woven into everyday
discourse. Some of this discourse is treated as freedom of speech and expres-
sion. Some is not. Many ordinary users as well as activists or well-known
journalists we observed and/or interviewed have faced systematic punitive
action because of their social media posts related to social and economic
injustice. They have been the targets of hateful communication, with dehu-
manising, discriminatory, threatening and/or abusive messages posted to and
about them in public and private messages online. Of the hundreds of mes-
sages we evaluated to build our typology (see Table 1.1 in Chapter 1) and to
give context to our interviews, and the thousands we collected since begin-
ning our work on social media and disinformation in 2018, most constitute
hate speech, some are implicated in hate crimes, and many remained online
for extended periods, despite being reported to platforms and/or to the police.
Social media has become an important, sometimes the dominant, way
for hundreds of millions of individuals to articulate their sense of self, con-
struct their (racial, caste, gender, sexual, religious) identities, and address
demands to politicians and corporations. For some social media is a de facto
‘marketplace of ideas’, where opinions are expressed and shaped, affects
explored, solidarities built, and information or misinformation exchanged
and contested. Although online usage is restricted to a relatively small
part of the population in Myanmar and India, and mostly skewed towards
dominant castes, males, urban areas, dominant languages and those who
can afford the data or coverage, the perception of its influence (amongst

DOI: 10.4324/9781003083078-6
120 Conclusion
different actors) also produces real effects that inform daily practices which
spill outside the domain of social media.
The clear-cut distinction often drawn between the online and the offline
in terms of hate has been shown to be a distraction. Caste, class, race, dis-
ability, gender, sexual and religious inequalities and prejudices from real
life are rife on social media, while social media use also informs the ways
in which we construct and shore up our racialised, caste-informed, gender,
sexual and class identities. Existing power geometries are enhanced through
and inform relationships online, pointing to the online and offline, the mate-
rial and the discursive as being inseparable.1 Infrastructural and political
economic approaches which informed our analyses in Chapters 2 and 4 of
this book are also helpful in illustrating the unstable distinctions between
online and offline, media, state and business, public and private, national
and international, local and global.

The advantages of our theoretical framework

Our theoretical framework in this book combined critical race theory, cul-
tural studies, cultural materialist, historical, infrastructural, phenomeno-
logical and political-economic approaches to social media hate ecosystems.
Market pressures and geopolitical allegiances lead to uneven ways in
which hate and violence have been handled in different parts of the world
by governments and conglomerates. These different institutional responses
are, in turn, implicated in the rise or fall of extreme authoritarian regimes,
the undermining or shoring up of democratic practices and the push towards
or disavowal of social justice and equity.
On the one hand, drawing on the concepts of intersectionality and othering
(and concomitantly invoking the notion of intersectional othering), our frame-
work directed attention to the implication of histories and infrastructures of colo-
nialism, casteism, racism and capitalism in hierarchies of hate. From the creation
of difference through mediated propaganda and disinformation to the consolida-
tion of political and economic power in the hands of elites and majorities, the
production of targets for populist rage and violence has been an extended socio-
political process that now encompasses mainstream and social media.
On the other hand, paying close attention to individual and community
memories and discourses of struggle, solidarity, exclusion, othering, dehu-
manisation, threat and violence, we developed a particular form of deep lis-
tening as the core of our methodological and theoretical approach. Applied
to original accounts from recipients and survivors of hate, such phenomeno-
logical ‘listening’ foregrounds the contribution of memories and discourses
not only to hateful communication and conscientised resistance but also to
legal and social media policies on hateful communications, and to the way
 Conclusion 121
in which the balance between freedom of speech and the right to life and
dignity are imagined and enacted.
Based on this, and emerging from our literature review and analysis of
the data, we offered a definition and typology of social media hate, and its
producers, circulators and recipients which will have far-reaching conse-
quences if adopted and understood as a backdrop to social media hate poli-
cies, media literacy programmes and in the development of ethical AI to
limit hate online and moderate or remove it systematically.

Typologies of hate
As we discussed in Chapter 1, there have been numerous attempts to cat-
egorise and classify different forms of hateful communication in different
national or community-based contexts. The typology of hate that we deline-
ated in the Introduction (see Table 1.1) emerged from our analysis. Amongst
perpetrators, as much as amongst those they target, overlapping aspects
of identity are crucial in determining their social and political status and
interests in hateful content. There were certain groups – particularly white
people, evangelicals and/or upper-caste Hindu males from digitally literate
backgrounds – who were most likely to be highlighted as perpetrators of hate
on social media in Brazil, India, and the UK. The kinds of hate we exam-
ined are targeted both at individuals and more broadly at othered groups.
The literature, meanwhile, shows that aggressor groups are some denomina-
tion of Muslim in many Muslim-majority countries (Bangladesh, Indone-
sia, Malaysia, Pakistan, Saudi Arabia), Jewish in Jewish-majority countries
(Israel) and Buddhist in some Buddhist majority countries (Myanmar, Sri
Lanka). To be precise, it is not just outspokenness, not just frequency of use,
and not just the intersection of race or caste with gender/sexuality/disability/
age that makes some individuals and groups more likely to be targets: Even
within racial and sexual groups there are subgroups who are most frequently
and virulently targeted ‘just for existing’. So Black women – especially
working class, Muslim, gay, and trans Black women, non-binary individu-
als and trans men face some of the highest levels of violence and abuse on
and offline while Dalit women and transpeople, and visibly Muslim women
(especially those who veil or those who openly reject the veil) are targets of
multiple forms of discriminatory and abusive communicative behaviours.
East-Asian women too face a growing amount of racist misogyny online.
This is the case even when individuals do not speak out publicly about issues
of rights and justice, but simply comment publicly on mundane issues.
These discriminated groups have been shown to be at the sharp end of
online hate, closely followed by Black men, Dalit men and Muslim men
in all these intersecting categories in all our case study countries (and in
122 Conclusion
particular, any men in these categories who present as femme and/or advo-
cate for the poor and/or minoritised and oppressed groups and are broadly
on the left). Notably, hate speech is also aimed from within oppressed groups
at other minoritised and oppressed groups. In all of our case study countries
and more generally, this is most clearly the case in regard to the disabled and
to sexual and gender minorities within racial, ethnic and religious minorities.
We also observed that there are other types of uncivil speech online, often
aimed within quasi-progressive groups at those who deviate or dissent from
their views or from the accepted or agreed position and line, or at those
who have fallen foul of a dominant block within their own communities
of choice or practice. Based on our analysis of the literature in Chapter 1,
most of these communications, while potentially unpleasant or distress-
ing, and which may cause serious mental health issues and forms of shame
to individuals in these circumstances, do not fall into the category of hate
speech. Nevertheless, there are times when such discussions make use of
the same tactics and prejudiced tropes and/or stereotypes as dehumanising
and discriminatory communication does, reinforcing marginalisation, and
thus spilling over into toxic or hate speech.

Moving forward: Suggestions for policy and politics

Our analysis of social media and hate incorporates the dialectical relationship
between social media, users, production and reproduction of hate (in its mate-
rial and symbolic forms) within specific national and historical contexts. Pre-
cisely because of its dialectical nature, this proves to be a space of flux, trauma
and opportunity depending on the context. When groups have been histori-
cally vulnerable, social media gives them a public platform but also bombards
them with new forms of violence and violation which have far-reaching con-
sequences for their minds and bodies, work, home lives and communities of
birth, choice and practice. Our analysis provides the grounds for arguing that
international political bodies such as the African Union, the UN and the Coun-
cil of Europe, and social media corporations such as Alphabet, Facebook (or
whatever other name this entity chooses to call itself), ByteDance and Twit-
ter, have a duty to act now, before matters reach extreme stages as they have
in several of our case study countries, in taking decisive ethical action. This
action should be planned and monitored transparently by well-trained (his-
torically and politically literate) AI, moderation and policy teams, and expert,
cross-national, cross-stakeholder groups. Alongside this, both political move-
ments and corporations need to bring pressure to bear on national govern-
ments to abide by humanitarian standards and national legal frameworks with
regard to the right to life and dignity of those with protected characteristics.
Social media companies must invest greater financial and human
resources in countries where there are significant number of users as well
 Conclusion 123
as historical political and religious patterns of harm, such as Brazil and
India, and in countries where their platforms and/or applications have been
used to exacerbate historical discrimination and violence, such as the UK,
Indonesia, Myanmar and so on. This holds true for other social media giants
with Chinese and Russian stakeholders operating in China and Russia too.
While it may not always be possible to have all personnel located within
the country for various reasons, social media companies should have clear
and transparent policies to employ staff (moderators, content managers or
other kind of editorial resources) proportionate to the number of subscrib-
ers in that country, the languages used within the country, and the scale of
violence that social media usage enables.
Further, individuals from vulnerable communities need to be employed
by mainstream media houses and social media companies as fact-checkers
and to train and sensitise staff appropriately so that content moderators and
moderation guidelines are compatible with the lingusitic, socio-political
and historical contexts that inform much of the discrimination, threat and
violence on social media. Our analysis for this book and for other projects
over the past five years shows consistently that hate and disinformation cir-
culates transmedially and in intertextual ways: Namely, it is not solely a
social media problem – it crosses genres and often appears in fiction and
non-fiction contemporaneously. Disinformation that targets minoritised
groups and propagates hateful content online feeds off, is linked to and
inseparable from content circulating in political speeches, and on main-
stream and hyperlocal media systems, as well as within entrenched com-
munity imaginaries.
Engaging children, young citizens, the middle-aged and elderly in sepa-
rate, carefully fact-checked and humane community education with a digi-
tal media literacy component may be the bedrock of imparting civic values
and human rights as things currently stand. Even these kinds of programmes
often don’t result in widespread change since they are few and far between,
often preach to the converted, exclude the most disadvantaged and allow
systemic trolls and ideologically motivated racist and misogynist actors to
avoid them altogether. To be even moderately effective, such critical digi-
tal literacy programmes would need be ubiquitous, generated and run by
combinations of marginalised and oppressed groups and communities, and
reflexive about change to demonstrate and illuminate the intimate connec-
tions between current and historical conflicts, representations, meaning-
making and flows of power.
Meanwhile, technological affordance-based approaches such as AI and
algorithms to take down problematic content need to be significantly and
fundamentally overhauled after detailed consultations with members of
communities who have faced such violence, scholars who have studied it,
and ethical technologists. Facebook’s actions in engaging with the Black
124 Conclusion
Lives Matter movement after George Floyd’s racist murder, in order to
overhaul its algorithms, needs to be applied urgently and widely to other
parts of the world too and by other social media platforms and apps includ-
ing, but not confined to Google, Twitter, YouTube, Instagram, WhatsApp,
Telegram, and TikTok owners, ByteDance.
Given individuals’ and groups’ deeply unsatisfying experiences of report-
ing hate-speech and overt incitement to moderators on platforms such as
Facebook, Twitter, TikTok and YouTube, and their equally depressing
experiences of inaction, apathy, contempt and harassment from police and
legal entities when they attempt to pursue their attackers in order to gain
a modicum of safety, it is clear that following regulations and policies on
discrimination, hate speech and incitement to violence does not always pro-
tect against it or even prevent ensuing violence. Efforts to reduce social
media hate to a technological phenomenon in which algorithms are gamed
for popularity or prestige, one that is engaged in by only fringe extremists,
one that is populist and used both by far left and far right or one that is
about ‘fake news’ spread by digital illiterates all play into the hands of fas-
cists, rightwing groups and the biggest, most organised spreaders of hateful
disinformation as well as unaffiliated individuals who harbour racist and
misogynist views.
Likewise, the attribution of all malign online influence to deranged loners,
Russian troll farms or western imperialism obfuscates rather than addresses
the multiple and powerful sources and beneficiaries of systemic violence,
dehumanisation, abuse and prejudice. In this context, a mixture of short and
medium-term strategies mentioned by our interviewees – ethical AI, media
education and fact-checking, need to be deployed urgently to prevent even
worse excesses and crimes. At an individual level, blocking, filtering, tak-
ing frequent breaks from social media, leaving platforms entirely, getting
therapy when affordable and speaking out about harassment and threat are
also important medium-term protections. In the long-term, however, social
media hate can only be defeated by international struggles over rights and
justice, strikes, boycotts and social movements pushing for profound and
far-reaching social and economic change that encompasses all current axes
of inequality and injustice.

Discouraged but still hopeful

By keeping the voices of marginalised communities and targeted users at
the heart of our book, we have highlighted the long-term individual and
collective effects and affective consequences associated with the dramatic
increase in the circulation of hate catalysed by social media. While their
experiences speak to the immense varieties of hate – including an overlap
 Conclusion 125
of direct and indirect violence, discrimination, threat, incitement, dehuman-
isation, disinformation and abuse – our key informants and interviewees
have also convinced us of the possibilities for a better future. Astoundingly,
despite the bleakness of the situation, despite the lack of structural support
and the multiple realms of structural violence and discrimination they face,
we also noted within our interviewees’ narratives something like Camus’
wrenchingly beautiful lines:

In the midst of hate, I found there was, within me, an invincible love. In
the midst of tears, I found there was, within me, an invincible smile. In the
midst of chaos, I found there was, within me, an invincible calm. . . . In
the midst of winter, I found there was, within me, an invincible summer.
(From ‘Return to Tipasa’)

We hope that this was also your experience of our book. The only thing
that we would add is that this resilient and powerful sense of hope, and the
courage to act which transcended the fear induced by being the targets of
historic and contemporary hate, were not individual phenomena, as in the
case of the cited existentialist theorist. Rather, these feelings and the actions
they gave rise to were expressed both through personal courage and through
solidarity that can best be theorised as a commitment to conscientised and
conscientising praxis.

Note
1 This intersection of discursive and material worlds has been acknowledged in
the critical media and cultural studies tradition for decades. See work by scholars
such as Stuart Hall (1997) and Roger Silverstone (1999) for context.

Reference
Hall, S. (1997). The work of representation. In S. Hall (Ed.), Representation: Cul-
tural representations and signifying practices (pp. 13–64). Sage Publications.
Index

abolition/abolitionist 53, 117 Black Lives Matter (BLM)/All Black

abuse 2, 9, 21, 36, 39, 45, 58, 70, 77, Lives 62, 100, 101, 103, 123, 124
83, 84, 85, 87, 88, 89, 93, 98, 100, Blackness (in white supremacist
101, 112, 113, 114, 116, 117, 121, imagination) 57
124, 125 Bollywood 82
Accenture 44 Brahmins 77
Adani group 76, 81 Bretton Woods 79
Adivasis 12, 14, 23, 81, 89, 90 Brexit 97, 107, 116
affect 19, 23
affordance-based approaches 31 Capitol riots 33
Afro-Brazilian 3, 50, 55, 62, 63, 64, 67, carceral system (and racism) 9
73, 74 caste 8, 14, 15, 20, 21, 22, 23, 25, 32,
Afro-descendants 7 33, 70, 75, 77, 78, 79, 81, 83, 84, 86,
agency/agentic 3 88, 93, 119, 120, 121
algorithms 33, 34, 36, 37, 123, 124 Catholic (church) 51, 54, 98
Ambani, Mukesh 1, 76, 81 censorship 15, 38, 58, 104
Ambedkar, B.R. 14, 25, 78 Chauvin, Derek 36
Anderson, Robert Nelson 53 Christians 12, 13, 25, 36, 38, 53, 56,
Ankhi Das 35 63, 99, 100
anti-Muslim 1, 43, 44, 78, 84 – 87 Church of England 98
antisemitism 37, 104, 105, 106, 107 Clegg, Nick 36
anxiety 14, 65, 70, 72, 82, 100, 108, colonialism 25, 40, 51, 52, 53, 54, 62,
109, 114 66, 67, 72, 75, 79, 97, 106, 116, 120
Artificial Intelligence (AI) 34, 121, commodification 30
123, 124 Congress (Party, India) 79, 90
Aung San (general) 39 conscientise/conscientisation 19,
Average Revenue Per User (ARPU) 30 120, 125
Ayyub, Rana 76 consciousness 96
conservative (and Conservative party,
Bahujan 14 also Tory) 22, 51, 97, 99, 117
bell hooks 98 conspiracy 69, 81, 97, 104
Benesch, Susan 9, 10 Convention on Elimination of all forms
biometric 44 of Discrimination Against Women
BJP 1, 35, 75, 76, 77, 79, 80 (CEDAW) 6
 Index 127
Coordinated Inauthentic Behaviour experience (and lived experience) 2, 3,
(CIB) 1, 25, 35 4, 8, 14, 17, 19, 20, 23, 30, 31, 36,
Costello, Matthew 15, 16 45, 52, 57, 63, 65, 66, 71, 76, 78,
Covid-19 35, 55, 67, 69, 81, 97 83, 87, 88, 90, 91, 96, 98, 100, 102,
Crenshaw, Kimberly 98 103, 104, 105, 111, 112, 124, 125

Dalits 1, 8, 12, 77 Facebook 1, 4, 11, 18, 32, 33, 34, 35,

dangerous speech 9 36, 37, 40, 41, 42, 43, 44, 45, 50, 58,
deep fakes 21, 22, 51 60, 72, 80, 82, 83, 87, 88, 89, 91, 98,
defamation 8, 68 103, 106, 107, 110, 111, 113, 117,
dehumanisation 2, 3, 15, 36, 43, 53, 56, 123, 124
57, 62, 66, 67, 84, 103, 116, 124, 125 facial recognition 44
Delhi 1, 3, 78, 88 fact-checker 24, 58, 70, 123, 124
democracy 23, 24, 54, 64, 92, 94, 96, far right 1, 2, 4, 13, 22, 33, 35, 36, 41,
104, 106 45, 50, 51, 55, 56, 59, 60, 64, 80, 86,
dialectic 2, 3, 20, 31, 122 96, 107, 108, 109, 117, 124
diaspora 8, 14, 55 fascist 58, 60, 77, 80, 81, 82, 93, 107,
disability 11, 22, 23, 32, 70, 121 108, 117; see also far right; Hindutva
discourse 13, 14, 18, 19, 42, 50, 54, 56, feminist/feminism 3, 23, 60, 72, 84, 117
60, 85, 98, 106, 119, 120 Floyd, George 37, 124
discrimination 1, 2, 3, 6, 7, 8, 10, 13, focus group discussions 5, 83
15, 19, 20, 24, 25, 30, 31, 34, 37, 41, Free Basics 42
43, 44, 52, 56, 61, 63, 70, 72, 73, 75,
77, 78, 84, 91, 93, 100, 102, 123, Gandhi, Indira 79
124, 125 gay 3, 51, 57, 98, 99, 100, 103, 104,
disinformation 1, 4, 5, 16, 21, 22, 31, 112, 115, 118, 121
35, 37, 43, 44, 52, 55, 58, 71, 72, 76, Gebru, Timnit 34
80, 82, 110, 119, 120, 123, 124, 125 genealogy 52
Display Picture (DP) 33 genocide 6, 9, 37, 41, 44, 45, 53, 71, 116
Doordarshan 79 geopolitics 31, 120
doxing (and doxers) 3, 76, 83, 89, 96 Global South 20, 98
Durov, Pavel 32 Google 34, 77, 80
graded inequality 25, 78
East Asian 121
ecosystems 3, 18, 24, 80, 120 hacking 76, 83, 96
egalitarian 110 Haddad, Fernando (deep-fakes of) 51
embodied subjectivity 2, 20 Hall, Stuart 7, 39
emergency 39 hegemony 24, 75
encryption/de-encryption 82 Hindu 76, 79, 80, 81, 83, 84, 85,
endogamy 25, 78 86, 121
epistemology 4 Hindutva 1, 14, 35, 37, 79, 80, 81,
ethnic cleansing 39, 40 82, 86, 107; see also far right;
evangelical 51, 63, 64, 99, 121 fascist
exceptionalism 97, 117 homophobia 14, 51, 52, 55, 97, 110,
exclusion 4, 16, 32, 39, 62, 77, 79, 98, 111, 114, 115
120 Honey Badger 44
existentialist 125
existential threat 82 identity/identification 2, 3, 4, 13, 15,
exogamy 25 21, 22, 32, 35, 39, 41, 50, 57, 71, 75,
128 Index
79, 80, 83, 84, 85, 86, 96, 101, 102, legitimisation 50, 56, 117
104, 113, 114, 121 LGBTQIA + 4, 12, 15, 23, 50, 56, 57,
ideology 11, 14, 22, 24, 58, 60, 83 64, 67, 72, 98, 110, 111, 112, 114
imaginaries 86, 98, 123 lived experience 4, 17, 19, 105; see
imperial/imperialist 54, 72, 86, 124 also experience
incitement 1, 3, 6, 8, 9, 21, 30, 32, 36, Luiz Inácio Lula da Silva/Lula 54, 55, 68
43, 57, 70, 84, 93, 96, 108, 116, 124, lynching 9, 21, 72, 93
125
India Telegraph Act 79 MacLean, Ken 39
indigenous 3, 4, 7, 21, 38, 40, 50, 52, majoritarian 13, 14, 86, 96; see also far
53, 54, 55, 56, 64, 65, 66, 67, 72, 77, right; fascism; Hindutva
89, 99, 116 marginalised 19, 34, 62, 65, 92, 97,
infrastructures 2, 3, 4, 17, 18, 20, 24, 123, 124
25, 31, 41, 80, 96, 120 masculinity 46, 78, 80
Instagram 3, 4, 32, 33, 34, 50, 60, 66, media literacy 45, 121, 123
67, 77, 83, 90, 98, 100, 101, 107, mental health 22, 70, 92, 93, 99, 110,
110, 111, 114, 124 111, 114
insurgency/counterinsurgency 38 Messenger 3, 32, 87, 87
International Convention on the microaggression 57
Elimination of all forms of Racial Microsoft 81
Discrimination (ICERD) 6 Mirchandani, Maya 13, 14
International Covenant on Civil and misinformation 12, 13, 16, 21, 22, 33,
Political Rights (ICCPR) 6 45, 50, 52, 59, 69, 99, 103, 106, 110,
International Dalit Solidarity Network 112, 113, 119
(IDSN) 8 misogyny 14, 21, 23, 52, 55, 59, 78, 86,
intersectionality 2, 20, 25, 78, 79, 98, 88, 98, 103, 104, 115, 116, 121
100, 120 Mitchell, Margaret 35
intertextual 13, 123 mob violence 12, 81; see also violence
interviews/interviewee/interviewer 3, 4, moderation 32, 36, 44, 103, 123
5, 12, 16, 17, 20, 56, 59, 60, 62, 63, Modi, Narendra 1, 35, 76, 77, 79, 80,
65, 68, 69, 70, 71, 72, 75, 76, 81, 84, 81, 82, 83, 93
90, 92, 96, 98, 100, 101, 104, 105, monopolisation 30
119, 124, 125 mujahideen 85
Islamophobia/Islamophobic 10, 11, Muslims 1, 10, 11, 12, 13, 25, 31, 35,
84, 85, 86, 102, 116; see also anti- 40, 41, 43, 75, 76, 78, 79, 80, 84,
Muslim 85, 86, 88, 92, 97, 99, 102, 104, 114;
IT cells 82 see also anti-Muslim; Islamophobia/
Islamophobic
jaatis 78; see also caste mystification (demystification) 65
Jesuits 53
Jio Platforms 1, 81 narrative 1, 5, 13, 37, 51, 52, 56, 59,
Jones, Lee 38 62, 67, 96, 105, 107, 114, 125
National Association for the
kalar 44 Advancement of Coloured People
Kashmir(i) 24, 88, 90 (NAACP) 37
Keisha-Khan, Y. Perry 55 nationalist (nationalism) 5, 12, 38, 41,
Kshatriyas 77 42, 76, 81, 86, 89, 90, 116
Nazi 60, 77, 107, 117
Labour Party 60, 104, 105, 106, 107 neoliberal 31, 52, 75, 76, 80, 86, 97
lawfare 39 Noble, Sofia 34
 Index 129
Operation Dragon King 39 self-reflexivity 4
oppressed 15, 24, 122, 123 sexuality (including bisexuality,
othering 19, 39, 50, 80, 120 homosexuality) 50, 70, 71, 78, 84,
98, 99, 103, 109, 112, 114, 116, 121
Palestine (Palestinians) 24, 37, 84, 97, Shah, Amit 82
104, 105, 106, 107 Shudras 77
Parekh, Bikhu 16 Signal 32, 33
Pasmanda 78 slaves/slavery 25, 53, 55, 62, 64, 116
patriarchy/patriarchal 21, 76, 78 Snider, Colin 54
pedagogy 62, 113 Social Justice Warrior (SJW) 117
Pentecostalism 99 sociolegal 98
phenomenology 2, 5, 19, 20 sociotechnical 3, 31
pogrom 1, 3, 9, 31, 44, 75, 76, 78, 80, South Asian 25, 77
93 stalking 4, 15, 96
policy/policies (against hateful stereotyping 11, 14, 16, 20, 21, 32, 36,
communication, toxic speech, hate 39, 56, 66, 71, 85, 102, 104,
speech; also debates about) 7, 18, 32, 110, 122
33 – 35, 75, 90, 91, 93, 122 – 123 Stop Hate for Profit 36
poststructuralist 4 subaltern 2, 53
prejudice 2, 9, 12, 22, 24, 36, 52, 56, subjectivity, embodied 2, 20
57, 72, 97, 100, 104, 109, 117, 120, Sulli 86
122, 124 surveillance 32, 44
propaganda 1, 5, 16, 54, 60, 76, 80, 81, Suu Kyi 38, 39, 40, 42
82, 107, 117, 120
Tatmadaw 38, 39, 40, 41, 44, 45
qualitative 4, 11, 17, 88 techno-populism 59
quantitative 10, 88 telecommunications 17, 38, 79, 80
queer 75, 83, 84, 99, 100, 103, 113, 114 Telegram 32, 33, 41, 124
quilombos 53 Teltumbde, Anand 79, 98
TikTok 3, 4, 77, 98, 100 – 101, 124
race/racism 7, 9, 14, 16, 17, 23, 44, 45, trans (transness, transphobia,
50, 52, 55, 56, 60, 62, 63, 64, 92, 96, transmisogyny) 3, 21, 23, 55, 75,
97, 98, 102, 103, 104, 105, 107, 108, 97, 98, 101, 102, 103, 110, 111, 112,
109, 113, 114, 115, 116, 117, 120 113, 114, 115, 116, 117, 121
racial hierarchy 55 trauma 2, 3, 5, 7, 9, 65, 98, 122
Rashtriya Swayamsevak Sangh trolling 50, 52, 60, 72, 76, 82, 83, 91,
(RSS) 77, 80, 85; see also fascism; 98, 103, 104, 105, 106, 108, 109, 116
Hindutva Twitter 4, 32, 33, 34, 36, 50, 56, 60, 66,
Reliance Industries Group 1, 81 70, 77, 82, 86, 88, 89, 90, 100, 103,
representation (mis-representation) 7, 105, 106, 107, 108, 109, 111, 114,
10, 11, 36, 43, 65, 72, 97, 103, 114, 116, 117, 122, 124
123 typology 11, 17, 20, 21, 24, 119, 121
Rohingya 3, 24, 30, 31, 37, 39, 40, 41,
42, 43, 44, 45, 85 UAPA 89
Rousseff, Dilma 54, 55, 60 Unicode 44
urban Naxal 90
Sangh Parivar 77; see also Hindutva;
Rashtriya Swayamsevak Sangh Vaishyas 77
(RSS) victim(ised)(hood) 1, 6, 12, 13, 15, 16,
Sein, Thein 42 21, 37, 61, 71, 89
130 Index
vigilante 1, 4, 12, 45, 55, 60, 62, 76, 81 WhatsApp 1, 4, 12, 13, 31, 32, 33, 36,
violence (including: anti-Indigenous, 50, 59, 71, 77, 82, 87, 88, 91
Islamophobic, gender based, Wirathu, Ashin 41
misogynist, political, racist, Workers Party 51, 54
religious, sexual and transphobic) 1, Worst of Worst (WOW) 37
2, 3, 6, 8, 9, 10, 11, 13, 14, 15, 17,
19, 20, 21, 22, 24, 25, 30, 31, 32, YouTube 32, 36, 50, 65, 66, 77, 86, 98,
36, 38, 39, 40, 41, 43, 50, 52, 53, 113, 124
54, 56, 57, 59, 60, 61, 62, 63, 64,
65, 68, 70, 71, 75, 77, 81, 84, 86, Zawgyi 44
88, 90, 91, 98, 100, 103, 114, 116, Zhang, Sophie 1, 34
120, 121, 122, 123, 124, 125 Zionism (anti-Zionism) 37, 96, 104, 106
Vkontakte 32 Zuckerberg, Mark 36