Author: David Blanco-Herrero
Postdoctoral researcher at the Amsterdam School of Communication Research (ASCoR) of the University of Amsterdam (UvA). [email protected]

Abstract:

We explored the convergence of misinformation and hate speech, revealing that believing and sharing false and hateful content are influenced by personal attributes. Moreover, people who did not believe a message shared it less. Media literacy and empathy are crucial in combating the spread of these toxic discourses.

Key words: disinformation; hate speech; anti-immigration discourses; media literacy

Mis- and disinformation pose, by themselves, a very concerning challenge (not only) for democratic societies. Although the effects might be similar, the particularity of disinformation (in opposition to misinformation) is its intentionality, so disinformation is produced being aware of its falsity and with an intention to harm. The risks and implications associated to this phenomenon are diverse and have been extensively studied. Among them, one of the most dangerous is the use of disinformation campaigns to promote hatred and other toxic discourses. In fact, disinformation and hate speech have been considered to follow parallel patterns and their associations have been highlighted in numerous works both inside and outside of academia.

While much attention has been devoted to understanding their similarities or each issue independently, the convergence of these two phenomena remains a relatively unexplored territory. However, comprehending this particular interconnection is relevant due to the aggravated danger that hateful discourses could present when disinformation strategies and fake news are used to further spread them. This interaction has been frequent in diverse demonization campaigns, being the most dramatic case the one of anti-Jewish libels and myths leading to the Holocaust. A more recent case is found in the attacks against Rohingya people in Myanmar. Here, hoaxes spread on Facebook led to lynching and violence, and that promotion of hatred based on disinformation was also used to support the ethnic cleansing of this group.

Just like the risks and implications of using disinformation to promote hate speech is clear, the role of social network sites (SNS) is also considered a very relevant factor in the spread of these messages. Indeed, the aforementioned anti-Jewish propaganda, -both during the 20^th Century as well as in more preterit times, such as the Middle Ages-, proves that the problem is not new or only explained by SNS. Yet, its current dimension and capacity to reach larger volumes of people in a shorter period of time are undisputed. And social media are one of the main explanations for this.

However, the specific ways this content is disseminated are less known. It is accepted that not all citizens are equally vulnerable to these messages, but understanding the process through which a person shares false and/or hateful messages is essential in order to address the challenge. Notably, the factors influencing individuals’ likelihood to believe and share such content have not been thoroughly examined, which is why we conducted an experimental research that could help us understand this issue in detail.

In a recent study involving 404 Spanish adults, participants were exposed to messages about migration, including false and true messages, and hateful and non-hateful messages. The choice of migration as the topic for our stimuli is explained by the growing salience of anti-immigration discourses in Spain. Another reason for this choice is the large record of fake news and misinformation surrounding this topic, as well as the great record of racist hate crimes in multiple Western countries, as the data from the OSCE show.

Our goal was to uncover how personal attributes such as ideology, gender, age, education, place of residence, income, and pre-existing attitudes towards immigration shape the inclination to share these messages on a social network site. And, more importantly, we wanted to examine the mediating role of believability, as it was expected that in order to share a message it was necessary first to believe it.

This role was confirmed, so we can claim that messages are more likely shared when they are perceived to be true. This also means that people usually share false and hateful information not with the intention to harm, but because they believe it’s true. Indeed, some of the false myths and claims used in disinformation campaigns, such as connection with terrorism, criminality and social burden for the welfare state, are precisely designed to be believed, thus supporting feelings of rejection towards foreign population.

This finding is also relevant because it was observed that the presence of falsehood and hatred in a message made it less believable. In turn, this means that the presence of falsehood or hatred does not reduce sharing intentions by itself, but it can indirectly lead to it via a reduction in believability.

It was also important to notice that the tendency to share false and/or hateful content diminished primarily among individuals with left-wing ideologies and among those without pre-existing negative attitudes towards immigration. For these groups, the presence of hate and/or falsehood clearly undermined the believability of the content, subsequently reducing the intention to share. This reduction did not happen (or it was less pronounced) among conservative people or people previously holding attitudes contrary to migration.

The rest of factors did not significantly impact the identification of false or hateful messages. Contrary to expectations, that also applies for educational level, which would be expected to positively impact accurate perceptions about the believability of hateful and (especially) false content. However, no significant role was observed for this variable.

Thus, one of the main implications of the study is the need for great media literacy, as this could help citizens to better identify false and hateful messages. And indeed, it should be media literacy campaigns designed specifically to increase awareness about misinformation, as higher general literacy levels or regular education do not seem to have the same effect when it comes to properly identifying false and hateful content.

The need to enhance media literacy, which is usually associated with the correct identification of false information, needs to be complemented with increased empathy and reduction of prejudice. This is needed to identify hatred, which is particularly important among people holding right-wing political ideologies and negative attitudes towards migration. These groups, are less likely to reduce believability of hateful and false anti-immigration content, thus also being less likely to reduce their sharing of these content.

Moreover, the analysis of the results hints the possibility of falsehood acting not as an independent variable, but as a moderator of the impact of hatred on believability and sharing intentions. This approach was explored in complementary analysis, showing that the presence of falsehood amplified the perceived believability and sharing intention of hateful messages. That is, hateful discourses are believed more precisely when they also include false information. Although this is only a preliminary observation and more research is needed, a possible explanation for this is the capacity of the creators of disinformation to tailor their messages to hateful narratives, whereas real information need to stay grounded to reality, thus being less ‘flexible’. This confirms the need to address disinformation and hate speech in a combined and systematic manner, particularly by considering disinformation a tool to promote hatred.

This study is part of a PhD dissertation on the use of disinformation to spread hate speech against migrants and refugees conducted at the University of Salamanca (Spain) that can be found here (in Spanish).