Could Hate Speech Prosecution Undermine Political Support?

The European Court of Human Rights (2017) defines hate speech as ‘all forms of expression which spread, incite, promote or justify hatred based on intolerance and indignity of all human beings.’ An impulse to the prosecution of hate speech was given in 1965, when the United Nations adopted the ‘International Convention on the Elimination of All Forms of Racial Discrimination’ (ICERD). In August 2020, this convention had been ratified by 182 countries (UN OLA 2020). As mandated by this convention, most European democracies have implemented legislation that prohibits hate speech (cf. Fennema Reference Fennema, Koopmans and Statham2000). A considerable number of politicians has since been prosecuted based on these laws, including Jesper Langballe in Denmark, Flavio Tosi in Italy, and Jussi Halla-aho in Finland (for example, Askola Reference Askola2015). In the Netherlands and Belgium alone, more than 50 politicians have been prosecuted in recent decades (Van Donselaar Reference Van Donselaar1995; Vrielink Reference Vrielink2010). Specifically, the ICERD was ratified by the Netherlands in 1971. Wilders was prosecuted based on legislation that dates back to 1934 and that was adjusted following that ratification (Snijders and Shoemaker Wood Reference Snijders, Shoemaker Wood and Gitlin2018).

The concept of political support refers to citizens' evaluations of their political system and its institutions (Easton Reference Easton1975). David Easton (Reference Easton1965) distinguished three objects of support: the community, the regime, and the authorities. Easton (Reference Easton1975) furthermore conceptualized that support can range from specific (that is, support for an institution) to diffuse (that is, attachment to the system). Pippa Norris (Reference Norris1999, Norris Reference Norris2011) further specified this framework by proposing five dimensions of political support that range from diffuse to specific: (1) belonging to the nation-state, (2) agreement with core principles and normative values upon which the regime is based, (3) evaluations of the regime's overall performance, (4) confidence in regime institutions and (5) approval of incumbent office holders. Norris (Reference Norris2006, 6) furthermore proposed that measures of citizens' satisfaction can be used as indicators of ‘public evaluations of how well autocratic or democratic governments work in practice’. The current study therefore examined evaluations of the legal system and democracy as indicated by citizens' confidence and satisfaction. Evaluations of democracy fall under Norris' third dimension (that is, evaluations of the overall performance of the regime), whereas evaluations of the legal system are part of the fourth dimension (that is, confidence in regime institutions).

There are at least four reasons to expect that the prosecution of politicians for hate speech could diminish political support. Although it should be emphasized that it was not an aim of this study to determine which of these explanatory mechanisms best explains the impact of hate speech prosecution, we distinguish these four processes to theorize what groups of citizens may be affected. The first mechanism can be derived from the simple idea that citizens evaluate institutions and democracy more favorably when things go their way. Easton (Reference Easton1965) used the term ‘output failure’ to refer to instances in which political support is diminished because authorities are unable or unwilling to meet citizens' demands. Any important decision by a political institution therefore has the potential to strengthen support among those who get what they want and to weaken it among those who disagree with the outcome (Arnesen Reference Arnesen2017; Bartels and Johnston Reference Bartels and Johnston2013). A similar phenomenon can be observed after elections, when voters of winning parties generally report a greater satisfaction with democracy than supporters of the losers (for example, Dahlberg and Linde Reference Dahlberg and Linde2017). Likewise, American citizens typically lower their evaluations of the US Supreme Court when they disagree with a decision (Grosskopf and Mondak Reference Grosskopf and Mondak1998; Kritzer Reference Kritzer2001). In the case of hate speech prosecution, those who sympathize with the accused's ideas may therefore lower their evaluations of the legal system and democracy simply because they disagree with the undertaking.

Whereas the former reasoning would apply equally to any other controversial decision by an institution, the second reason why hate speech prosecution may undermine political support is both more specific and more fundamental. The decision to prosecute a politician for hate speech is not merely a decision that people can disagree with, but also something that affects the core of their democratic rights. People who agree with the idea behind the accused's statements may regard that they are no longer free to express their views or to elect someone who represents their convictions. Wilders has for example consistently framed his prosecution as a political trial against the democratic rights of his voters (Van Noorloos Reference Van Noorloos2014). Directly after his conviction on 9 December 2016, he released a video in which he put it as follows: ‘I have a message for the judges who convicted me. You have restricted the freedom of speech of millions of Dutch citizens and thereby in fact convicted everyone. Nobody trusts you anymore’ (Video Wilders 2016).

This quote by Wilders also illustrates the third reason why hate speech prosecution could damage political support, namely that politicians provide important cues to their supporters about what political views to adopt (for example, Bakker, Lelkes and Malka Reference Bakker, Lelkes and Malka2019; Lau and Redlawsk Reference Lau and Redlawsk2001). Most people may not have direct experience with the legal system and hence rely on messages from opinion leaders to determine their trust in the institution (Caldeira and Gibson Reference Caldeira and Gibson1992). When the prosecuted politician repeatedly states that the legal system is corrupted, this could therefore persuade many sympathizers to also distrust it. Accordingly, research shows that anti-establishment parties can fuel political distrust (Rooduijn, Van der Brug and De Lange Reference Rooduijn, Van der Brug and De Lange2016; Van der Brug Reference Van der Brug2003). This may apply in particular to citizens who psychologically identify with the accused's party, since the extensive literature on party identification demonstrates that this group is most likely to echo a party's message (Campbell et al. Reference Campbell1960; Carsey and Layman Reference Carsey and Layman2006).

The fourth reason why hate speech prosecution could erode political support is that it may politicize the legal system. Institutions usually enjoy more support when they are seen as independent authorities and less when they are perceived as ‘political’ (for example, Elchardus and Smits Reference Elchardus and Smits2002). It is likely for this reason that the legal system is generally among the most trusted institutions whereas for example political parties are often distrusted (for example, Dekker and Den Ridder Reference Dekker, Den Ridder, Van der Meer, Van der Kolk and Rekker2018). Gibson, Caldeira and Spence (Reference Gibson, Caldeira and Spence2003; Gibson Reference Gibson2007) for example argued that the legitimacy of the US Supreme Court has largely been unaffected by partisan and ideological polarization because citizens distinguish courts from other political institutions due to legitimizing judicial symbols. As Gibson (Reference Gibson2007, 516) put it: ‘The message of these powerful symbols is that courts are different, and owing to these differences, courts are worthy of more respect, deference, and obedience – in short, legitimacy.’ The prosecution of politicians may however detract from the legal system's apolitical reputation by drawing it closer to the political arena in the eyes of the public. In other words, citizens may intuitively approach an institution with a more critical attitude when it is subject to conflict over divisive societal issues.

Only four empirical studies have so far investigated the impact of the prosecution of politicians for hate speech on public opinion and only one of these focused on political support. This three-wave panel study by Van Spanje and De Vreese (Reference Van Spanje and De Vreese2014) followed Dutch citizens in the weeks before and after the 2009 court decision to prosecute Wilders for hate speech. It revealed that citizens who reject the multicultural society lowered their evaluation of democracy during this period. This decline was stronger among those who were aware of the court's decision compared to others who were unaware, which tentatively suggests a causal effect of the ruling. In the absence of an experimental design, this causal inference however remains uncertain and this study only examined satisfaction with democracy as outcome variable. Indirectly related to the issue of political support, a second study used a similar design to demonstrate that Wilders' electoral support surged substantially in the polls after the court's decision to prosecute him (Van Spanje and De Vreese Reference Van Spanje and De Vreese2015), while a third study examined the explanatory mechanisms of this effect (Jacobs and Van Spanje Reference Jacobs and Van Spanje2019) and a fourth study investigated the role of news media in this electoral impact (Jacobs and Van Spanje Reference Jacobs and Van Spanje2020). Based on our theoretical reasoning and these few earlier findings, we postulated our first hypothesis as follows:

For Whom Would Hate Speech Prosecution Undermine Political Support?

If hate speech prosecution indeed damages political support, this raises the question for whom this is the case. The four aforementioned explanatory mechanisms lead us to different expectations about how many citizens may be affected. Based on the first (simple disagreement) and second reasoning (restriction own political freedoms), we may expect that hate speech prosecution reduces political support among everyone who supports the idea behind the accused's statements. In the case of Wilders, this would imply that political support diminishes among everyone who opposes the multicultural society, whom we refer to as assimilationists. Regardless of whether or not they voted for Wilders, assimilationists may perceive that their freedom of speech is being restricted. Even assimilationists who disapprove of Wilders' choice of words may therefore oppose his prosecution.

Based on the fourth account (politicization of the legal system) the affected group could be even broader. The legal system could lose some of its status as an apolitical authority for every citizen due to the prosecution of a politician, even for those who reject the accused's views and applaud the prosecution. Following the third account (opinion leadership), the affected group could however also be narrower. Only citizens who support the accused's party are likely to take cues from it about what institutions to trust. Those who merely agree with the accused's statements, but vote for a different party, will probably not follow him or her as an opinion leader. Particularly in a multiparty context, the group that actually votes for a party can be considerably smaller than the group that agrees with it on a particular issue. For example, about one-third of Dutch citizens reject immigration (for example, Van der Brug and Van Spanje Reference Van der Brug and Van Spanje2009), but only 10–15 per cent votes for the PVV.

The only earlier study on hate speech prosecution and political support examined attitudes about the multicultural society, but not vote choice, as a moderator (Van Spanje and De Vreese Reference Van Spanje and De Vreese2014). Although this study revealed a decline in political support among assimilationists after a court decision to prosecute Wilders, it consequently remains uncertain if this effect could have been driven entirely by PVV voters. The current study aimed to provide clarity on this issue by examining if hate speech prosecution diminishes political support only among the accused's electorate, or more generally among everyone who supports his or her views. Based on this earlier finding, our tentative expectation was that the latter would be most likely. We therefore formulated our second hypothesis as a stricter version of our first hypothesis:

Could Hate Speech Prosecution Strengthen Political Support?

Although hate speech prosecution may damage political support for some, it could potentially raise it among others. An obvious candidate to endorse the prosecution is the community that the alleged hate speech was aimed against. In the case of Wilders' second trial, people in the Netherlands of Moroccan descent may for example perceive that the legal system is effectively protecting their rights as an ethnic minority. A much larger group that may applaud the prosecution however consists of everyone who rejects the accused's views. Proponents of multiculturalism, whom we refer to as multiculturalists, will likely welcome Wilders' prosecution as an instrument to defend multicultural values such as tolerance of diversity. As such, their evaluations of the legal system and democracy may be strengthened by the endeavor. Theoretically, this can be viewed as an effect of citizens simply agreeing with the court's decision. Indeed, research on the US Supreme Court reveals that Americans commonly improve their evaluations of this institution when they agree with a decision, even though this favorable response is not as strong as the loss of support among citizens who disagree (Grosskops and Mondak Reference Grosskopf and Mondak1998). Table 1 provides an overview of how different groups of citizens may respond based on each of the four hypothesized mechanisms.

Table 1.

Overview of explanatory mechanisms and their possible implications for political support

The only previous study on this topic (Van Spanje and De Vreese Reference Van Spanje and De Vreese2014) failed to provide a conclusive answer to the question of whether those who oppose the accused's views increase their political support in response to hate speech prosecution. It showed that multiculturalists' satisfaction with democracy decreased, rather than increased, in the weeks after the court's decision to prosecute Wilders. This decline was however weaker among multiculturalists who were aware of the court's decision compared to multiculturalists who were unaware, which suggests that the court's decision may nonetheless have had a positive impact among this group. The current study therefore aimed to provide a clearer picture of multiculturalists' response to hate speech prosecution by examining the following hypothesis:

Finally, some citizens may reject or endorse hate speech prosecution based on universal principles. For example, some multiculturalists may nonetheless reject Wilders' prosecution out of a fundamental belief that free speech should never be restricted. Conversely, some assimilationists may support his prosecution because they believe minorities ought to be protected. Because such universal principles can work in both directions, they are not a reason to expect a structural increase or decrease in support for a particular group of citizens.

The Case Under Investigation

The present study examined the case of Geert Wilders (see, for example, Van Noorloos Reference Van Noorloos2014), who is the founder of the anti-immigration Freedom Party (PVV). The PVV was the first Dutch anti-immigration party to enjoy durable electoral success with 5.9 per cent of the votes in 2006, 15.5 per cent in 2010, 10.1 per cent in 2012, and 13.1 per cent in 2017. Alongside his more general critique of multiculturalism and European integration, Wilders is best known for his rejection of Islam. In 2007, he referred to Islam as ‘a fascist ideology’ and proposed to ban the Quran. Several people pressed charges and, although the public prosecutor initially decided not to instigate legal proceedings, an Amsterdam court ordered in 2009 that Wilders was to be prosecuted for these statements. The proceedings commenced in 2010 and resulted in a full acquittal in 2011. A second prosecution started after a political rally for the 2014 municipal elections, where Wilders asked his supporters if they wanted ‘more or fewer Moroccans’. The mob answered by chanting ‘fewer’, after which Wilders responded, ‘We will arrange it.’ The public prosecutor decided to prosecute Wilders for these statements in December of 2014 and the proceedings started in March of 2016. On 9 December 2016, the court ruled that Wilders was guilty of hate speech. A sentence was however not imposed.

Although this examination was a case study of one politician, the same mechanisms and effects may characterize other prosecutions for hate speech. The extent to which our findings generalize to other cases depends on the relevant characteristics of both the Dutch context and the specific case of Wilders. With regard to context, the immigration debate in the Netherlands is very typical of the divide over this issue that exists throughout Western Europe. For example, the Netherlands is about average compared to other West European countries in terms of the size of its immigrant population (Eurostat, 2019), the vote share of anti-immigration parties (Mudde Reference Mudde2013), and public support for multiculturalism (Bohman and Hjerm Reference Bohman and Hjerm2016). Moreover, the number of previous hate speech prosecutions in the Netherlands is not exceptionally small or large compared to other countries (Fennema Reference Fennema, Koopmans and Statham2000; Jacobs and Van Spanje Reference Jacobs and Van Spanje2018). Consequently, we see no obvious reasons why the prosecution of politicians for hate speech would have a different impact on political support in the Netherlands than in most other West European countries.

With regard to the specific case, three core characteristics may delimit generalizability. First, Wilders was prosecuted for statements that related to immigration and the multicultural society. The reasoning and analyses in this study therefore focussed specifically on citizens' attitudes about multiculturalism. Other attitudes may matter when politicians are prosecuted for other hate speech. For example, attitudes towards homosexuality may have played a similar role regarding the trial of Dutch MP Leen van Dijke for defamation of homosexuals in 1998–1999. A second core characteristic of the Wilders case is that he was the leader of a major political party. As a result, his prosecution was covered extensively by Dutch media and Wilders may have had a substantial ability to influence public opinion. The dynamics might have been very different for a leader of a minor party, such as Hans Janmaat in the Netherlands, Udo Voigt in Germany and Daniel Féret in Belgium. A third relevant characteristic of the Wilders case is that his statements seem sufficiently ambiguous to allow for different interpretations. The incident in which Wilders asked his supporters if they wanted ‘more or fewer Moroccans’ may have been perceived as extremely offensive and dangerous by his opponents, whereas Wilders' supporters may have viewed it as ‘simply a question’ about immigration policy. Combined with the fact that public opinion is highly divided on the underlying immigration issue, this ambiguity of Wilders' statements may therefore have contributed to a more polarized public opinion about his prosecution and a stronger potential to damage political support. Some caution is therefore warranted in generalizing our findings to cases in which the prosecuted statements were more universally rejected. An example is the prosecution of Jean-Marie Le Pen, whose antisemitic references to the Holocaust were widely condemned across the French political spectrum, as indicated by Le Pen's suspension from his own party. Other examples include the cases of Nick Griffin in Britain and Günter Deckert in Germany, who were also prosecuted for Holocaust revisionism.

In sum, we see little reason why hate speech prosecution would affect political support differently in most other West European countries or in other cases that resemble the Wilders case on relevant characteristics. Our findings may generalize in particular to other cases in which the accused's statements relate to immigration, the defendant is the leader of a major political party, and the prosecuted statements are ambiguous enough to divide public opinion on their acceptability. An example that resembles the Wilders case on all three characteristics is the prosecution of Marine Le Pen. As the leader of Front National, Le Pen was prosecuted in 2015 for a comparison that she made in 2010 between Muslim prayer and Nazi occupation. Indeed, the Wilders case seems to be indicative of a growing phenomenon that the leaders of anti-immigration parties explore the boundaries of what is legally permissible by making ambiguous statements about immigrants that their sympathizers would generally still find acceptable.