Long-Term Effects of Applying Regulation on Agitation Against a Population Group

 

 

 

 

 

 

Long-Term Effects of Applying Regulation on Agitation Against a Population Group

An analytical research report on communication behaviour, societal well-being, and institutional trust

 

 

 

 

 

 

 

 

 

 

Research report
26 March 2026

 

Abstract

This report examines how the application of regulation concerning agitation against a population group may affect individual communication behaviour, societal well-being, and institutional trust over the long term. The analysis uses Finland as the primary legal reference point because the research prompt refers to the Finnish offence of agitation against a population group, but it evaluates the question through a broader comparative and theoretical lens. The report distinguishes between the existence of the law, prosecutorial and judicial application, administrative interpretation, and wider signal effects in public discourse. It also separates normative questions about whether such regulation is justified from empirical questions about what effects it is likely to produce. Methodologically, the report is a structured analytical review of Finnish legal sources, government and EU reports, and selected peer-reviewed research on hate speech, online harassment, self-censorship, police trust, and platform governance. The central finding is that effects are heterogeneous and conditional rather than uniform. The strongest evidence supports two propositions: first, exposure to hate speech and harassment can reduce participation, increase stress, and degrade the perceived quality of public debate; second, fair and intelligible enforcement can strengthen minorities’ sense that the state takes their security seriously. However, the evidence is much weaker for strong long-term causal claims that legal enforcement alone improves social cohesion or institutional trust at the aggregate level. The same intervention can reduce intimidation for some groups while increasing perceived censorship or selective-enforcement concerns for others. Long-run outcomes depend especially on legal clarity, procedural fairness, consistency, proportionality, communication by authorities, and interaction with platform moderation and broader political polarization. The report concludes that the application of this regulation is best understood as a trade-off shaping several partially competing goods rather than a single welfare-maximising instrument.

Keywords: hate speech regulation; agitation against a population group; self-censorship; institutional trust; social cohesion; Finland

 

Table of Contents

1. Introduction

2. Theoretical Background and Previous Research

3. Method

4. Results

5. Discussion

6. Conclusion

References

 

1. Introduction

This report analyses the following question: how does the application of regulation concerning agitation against a population group affect individual communication behaviour, societal well-being, and institutional trust over the long term? The question is empirically difficult because the relevant outcomes are multi-causal, unfold over different time horizons, and are affected by both formal legal action and broader public interpretation of that action.

The analysis therefore narrows the problem in four ways. First, it uses the Finnish offence of agitation against a population group as the legal anchor, because the phrase in the research prompt corresponds to the offence defined in Chapter 11, section 10 of the Finnish Criminal Code (Criminal Code of Finland, 1889/2022 translation). Second, it treats "application" as a layered phenomenon that includes legal existence, investigation, prosecution, court decisions, administrative interpretation, and public signal effects. Third, it distinguishes empirical effects from normative claims about whether restrictions on speech are justified. Fourth, it treats long-term effects as probabilistic and conditional rather than deterministic.

The basic empirical problem is not whether hate speech can be harmful. There is substantial evidence that hateful and hostile expression can damage targets’ mental well-being, narrow participation, and impair the quality of public debate (European Parliament, 2020; Saha et al., 2019). The harder question is whether the application of criminal regulation improves outcomes net of its own costs and, if so, under what conditions. This report argues that no single answer is defensible. The likely effects differ by group, by enforcement style, by political context, and by the interaction between state law and platform governance.

The report is organised as follows. Chapter 2 defines the main concepts and surfaces the key assumptions embedded in the research question. Chapter 3 explains the analytical method and the basis for assessing evidential strength. Chapter 4 presents the main analysis, with separate attention to communication behaviour, societal well-being, trust, time horizons, and causal uncertainty. Chapter 5 discusses alternative explanations and evidential limitations. Chapter 6 states the main conclusions and open questions.

 

2. Theoretical Background and Previous Research

2.1 Core concepts

In Finnish law, agitation against a population group refers to making available to the public, disseminating, or keeping available a message in which a group is threatened, defamed, or insulted on grounds such as race, colour, descent, national or ethnic origin, religion or belief, sexual orientation, disability, or another comparable basis. Aggravated agitation against a population group covers more serious cases involving exhortation or enticement to grave violence or international core crimes in circumstances that clearly endanger public order and security (Criminal Code of Finland, 1889/2022 translation).

For the purposes of this report, the application of regulation does not mean only the statute’s presence on the books. It includes at least five analytically distinct layers: (1) formal legal existence, (2) police recording and investigation practices, (3) prosecution thresholds and charging decisions, (4) judicial interpretation and sentencing, and (5) signal effects produced when the public observes enforcement or non-enforcement. These layers can move in different directions. A state can have a broad statute but weak enforcement, or active enforcement but poor public understanding.

Communication behaviour refers here to changes in what people say, where they say it, how directly they say it, and whether they speak at all. This includes self-censorship, norm internalisation, strategic adaptation of wording, migration to closed channels, counterspeech, and confrontational escalation. Societal well-being is treated as a multidimensional construct rather than a single metric. Relevant dimensions include personal and group security, minority protection, freedom of expression as experienced by citizens, quality of public debate, social cohesion, and conflict risk. Institutional trust is disaggregated into trust in police, courts, legislators, and media, because the same legal development may increase trust in one institution while reducing it in another.

2.2 Assumptions embedded in the research question

The research question contains several implicit assumptions that should not be accepted without examination. First, it may suggest that applying the regulation is a unified phenomenon. It is not. The content of police decisions, prosecutorial strategy, court reasoning, and public messaging may diverge substantially. Second, it may imply that changes in communication behaviour are directly caused by legal enforcement. In practice, communication is also shaped by platform moderation, reputational sanctions, employer policies, media norms, and partisan conflict (Dubois & Reepschlager, 2024).

Third, the question might imply that societal well-being can be assessed on a single common scale. That is analytically misleading. Reducing intimidation of minorities and increasing perceived legal risk for borderline speakers can occur simultaneously. Fourth, institutional trust is unlikely to react uniformly across population groups. Members of targeted minorities, activists, journalists, politically marginal groups, and highly online subcultures may interpret the same enforcement event differently. Fifth, long-term effects are hard to isolate from broader changes such as digital platform design, shifts in elite rhetoric, geopolitical shocks, and long-run polarization.

A final assumption worth surfacing is that the main trade-off is simply between equality and freedom of expression. Empirically, the picture is more complex. Weak enforcement can also suppress speech by intimidating targets and observers. Conversely, overbroad or opaque enforcement can chill lawful speech outside the law’s intended scope, a mechanism often described as a chilling effect (Simpson, 2024). Thus, both under-enforcement and over-enforcement can reduce the practical freedom to participate in public discourse.

2.3 Previous research relevant to the problem

The literature most relevant to the present question does not usually estimate the long-term causal effect of one specific hate speech offence. Instead, it provides evidence on adjacent mechanisms. Research on exposure to hateful speech indicates harms to psychological well-being and community climate. In a large study of online U.S. college communities, exposure to hateful speech was associated with greater stress expression, with stronger effects among individuals with lower psychological endurance (Saha et al., 2019). Although the setting differs from national criminal law, it supports the broader proposition that hate expression can impose measurable social costs.

Research on public-facing professions shows that harassment can narrow participation. A Finnish study of academics, politicians, and media professionals found substantial exposure to work-related online harassment and linked harassment characteristics and weak workplace support to self-censorship (Celuch et al., 2023). Finnish government reporting also indicates that hate speech aimed at decision-makers reduces participation in public debate and can discourage seeking or retaining positions of trust (Ministry of the Interior, 2019; Ministry of the Interior, 2024). These findings are important because they show that the practical speech environment is shaped not only by state restrictions but also by intimidation from other citizens.

Research on policing and trust suggests that discriminatory or unfair treatment can reduce trust in police and courts, especially among minority groups. A recent German study found that perceived police discrimination was linked to stronger mistrust in police and, partly, courts among some ethnic minority adolescents (Kogan et al., 2024). This does not directly test hate speech enforcement, but it is highly relevant to the present problem because application of the law is mediated through police and prosecutorial institutions. If those institutions are perceived as biased or inconsistent, the legitimacy benefits of the law may be undermined.

Finally, a growing literature on platform governance shows that the online speech environment has changed independently of criminal law. Policies governing harassment and hate speech on Facebook, Twitter/X, and Reddit became more complex over time, while responsibility for enforcement expanded from users alone to a mixture of users, platforms, technologies, and external actors (Dubois & Reepschlager, 2024). This is crucial for causal analysis, because a measurable change in public speech after an enforcement episode may partly reflect platform policy shifts rather than state law alone.

 

3. Method

3.1 Research design and material

This report is a structured analytical review rather than an original causal impact study. The material consists of four source types: (1) Finnish legal texts defining the offence of agitation against a population group; (2) official Finnish, EU, OECD, and Council of Europe reports and webpages on hate speech, hate crime, freedom of expression, and institutional responses; (3) selected peer-reviewed studies on online hate, self-censorship, police trust, and related behavioural mechanisms; and (4) comparative conceptual literature on chilling effects and platform governance.

The material was selected for direct relevance to the research question’s mechanisms: behavioural adaptation, well-being impacts, trust effects, and causal attribution problems. The analysis prioritised official and peer-reviewed sources over advocacy documents where possible, although some institutional reports were included because they contain policy-relevant descriptive evidence not available elsewhere.

3.2 Delimitation

The report does not attempt to determine whether the regulation is morally or constitutionally justified in the abstract. That is a normative question. Instead, the focus is empirical: what kinds of effects are plausible, which are supported by evidence, and under what conditions. The report also does not claim to identify a single average treatment effect of the Finnish offence itself. Available evidence is not strong enough for that. The long-term horizon is treated analytically, but not as a claim that the observed literature can isolate decade-long causal effects with high precision.

The analysis is limited to effects relevant to communication behaviour, societal well-being, and institutional trust. It therefore touches criminal law, public discourse, democratic participation, and social cohesion, but it does not attempt a full doctrinal analysis of Finnish constitutional or European human rights law. Legal materials are used mainly to define the regulatory object and its plausible implementation pathways.

3.3 Analytical procedure

The analytical procedure had five steps. First, the report separated the law’s existence from its application. Second, it mapped potential mechanisms linking application to outcomes. Third, it assessed evidence for each mechanism. Fourth, it identified competing explanations that could generate similar observations. Fifth, it classified claims into three categories: observation, interpretation, and speculation. In this report, observation means a claim grounded directly in the cited source material; interpretation means a reasoned inference drawn from multiple sources; speculation means a plausible but weakly evidenced possibility.

3.4 Reliability and validity

Reliability is strengthened by triangulation across legal sources, official reports, and peer-reviewed research. However, validity remains limited in three important ways. First, several studies concern online harassment or hate speech more broadly rather than the Finnish offence specifically. Second, much of the evidence is observational, descriptive, or survey-based, which restricts causal inference. Third, institutions and platforms co-regulate public discourse, making attribution difficult. Accordingly, strong claims are reserved for mechanisms that are repeatedly supported across source types, while long-term aggregate claims are treated more cautiously.

 

4. Results

4.1 What counts as application of the regulation?

Observed evidence: Finnish law criminalises agitation against a population group and aggravated forms of that offence (Criminal Code of Finland, 1889/2022 translation). Official Finnish material also treats hate speech and hate crime as linked but not identical phenomena, noting that some hate speech is criminal only when the statutory elements of an offence are satisfied (Ministry of the Interior, 2024).

Interpretation: for empirical analysis, the most important form of application is not bare legal existence but credible, intelligible, and visible enforcement. A dormant statute will likely have weaker behavioural and trust effects than consistent investigation, reasoned prosecution, and publicly understandable case law. Public signal effects also matter. Highly publicised prosecutions or acquittals may affect behaviour even among people who never directly interact with police or courts.

Speculation kept in bounds: the salience of application may sometimes matter more than the frequency of application. A few prominent cases can reshape perceived risk and social norms, but the magnitude and persistence of that effect remain uncertain.

4.2 Mechanisms affecting communication behaviour

4.2.1 Self-censorship

Observed evidence: studies of online harassment in Finland indicate substantial self-censorship among public-facing professional groups. Celuch et al. (2023) found that harassment characteristics and weaker social support were associated with self-censorship among Finnish academics, politicians, and media professionals. Government reporting in Finland similarly states that hate speech reduces participation in public debate and that many decision-makers considered leaving politics after exposure to such hostility (Ministry of the Interior, 2019; Ministry of the Interior, 2024).

Interpretation: application of the regulation can reduce some forms of self-censorship by signalling that intimidation has legal consequences. This is the anti-chilling mechanism for targets and bystanders. At the same time, it can create legal-risk-based self-censorship among speakers who are uncertain about the boundary between robust criticism and punishable agitation. Whether the net effect is more or less public participation depends on which chilling process is stronger in the relevant speech environment.

What would strengthen the claim? Panel data showing that groups previously deterred by harassment became more willing to participate after credible enforcement. What would weaken it? Evidence that participation declined broadly across lawful political speech after ambiguous enforcement events. What remains open is the comparative magnitude of these two effects in Finland over a long period.

4.2.2 Norm internalisation

Observed evidence: institutional and European policy documents repeatedly frame hate speech as damaging to dignity, equality, social peace, and democratic debate (Council of Europe, 2023; European Parliament, 2020).

Interpretation: repeated, procedurally legitimate application of the law may gradually shift norms by signalling that certain forms of dehumanising or threatening expression are outside acceptable democratic contestation. This is not merely deterrence. It is a social learning mechanism whereby law helps stabilise a norm against publicly degrading targeted groups.

Uncertainty: norm internalisation is plausible but difficult to isolate empirically. A decline in overtly hateful speech may reflect true norm change, strategic concealment, or migration to other venues. Therefore, observed decreases in public expressions should not be interpreted automatically as deeper moral internalisation.

4.2.3 Strategic adaptation and coded speech

Interpretation grounded in comparative research: when direct hostile expression becomes more legally risky, actors may shift toward euphemism, insinuation, humour, symbolism, or coded language. This does not require strong evidence unique to Finland; it follows from basic strategic behaviour under sanctions. The mechanism is especially likely in digital environments where users rapidly learn platform and legal boundaries. However, the present source base supports this mainly as a theoretical inference rather than a well-identified Finnish long-run effect.

A related possibility is selective moderation arbitrage: actors adapt content to avoid both criminal liability and platform enforcement while preserving out-group hostility. This interpretation is made more plausible by research showing that platform hate speech policies changed over time and became more complex (Dubois & Reepschlager, 2024).

4.2.4 Polarization and movement to closed channels

Interpretation: enforcement can have polarising effects when it is framed by some actors as evidence of elite repression or double standards. In such cases, sanctioning may strengthen group identity among committed offenders and push communication into encrypted, pseudonymous, or fringe channels. The evidence base for this mechanism in the present review is modest, but the causal logic is strong enough to take seriously.

Observed evidence relevant to plausibility: platform governance and online hostility demonstrably shape where and how people engage. Research on social media hostility indicates that exposure can prompt avoidance, curation of one’s discursive environment, blocking, and selective engagement rather than simple silence (Matthes et al., 2025). This suggests that behavioural adaptation to hostile or regulated environments often takes the form of relocation and filtering, not only reduced speech.

4.3 Effects on societal well-being

4.3.1 Security and minority protection

Observed evidence: official European and Finnish materials consistently describe hate speech and hate crime as threats to dignity, equality, public peace, and the security of targeted communities (European Parliament, 2020; Ministry of the Interior, 2024). The Ministry of the Interior reports that suspected hate crimes in Finland continued to grow in 2023 and that only a small share of offences are reported to police, indicating both continuing exposure and measurement limits (Ministry of the Interior, 2024).

Interpretation: where enforcement is credible and procedurally fair, application of the regulation can increase minority protection by reducing impunity, signalling public recognition of harm, and supporting reporting behaviour. The logic here is strengthened by the broader finding that law-enforcement responsiveness can encourage hate crime reporting (European Union Agency for Fundamental Rights [FRA], 2021).

Important limitation: official growth in recorded incidents does not by itself show worsening underlying prevalence. It may also reflect improved reporting, changes in classification, or shifting police practices. Therefore, safety and victimisation should not be inferred directly from recorded offence counts.

4.3.2 Quality of public debate and experienced freedom of expression

Observed evidence: Finnish official sources and studies on online harassment indicate that hostile speech can reduce willingness to participate in public debate and can distort who remains visible in that debate (Celuch et al., 2023; Ministry of the Interior, 2019; Ministry of the Interior, 2024). European Parliament (2020) similarly argues that hate speech erodes social cohesion and blocks rational public debate.

Interpretation: if the law is applied narrowly and predictably against serious group-directed threats, defamation, and degrading incitement, it may improve deliberative quality by raising the cost of intimidation. However, if its boundaries are communicated poorly or applied inconsistently, some citizens may experience a reduction in expressive freedom even where their intended speech would be lawful. Thus, "freedom of expression" is not one variable moving in one direction. Targets may feel freer while borderline speakers feel less secure.

4.3.3 Social cohesion and conflict risk

Interpretation: long-run social cohesion can improve when enforcement communicates equal membership in the political community and reduces public degradation of protected groups. Yet cohesion can deteriorate if large segments of the population view enforcement as partisan, selective, or symbolic rather than principled. In that scenario, the law may still protect vulnerable groups while simultaneously increasing resentment and perceived distance from institutions among others.

Evidential status: the proposition that hate speech harms cohesion is strongly supported at a general level by policy and comparative literature (European Parliament, 2020; Wardle, 2024). The proposition that the application of one specific criminal offence improves cohesion in the long run is only moderately supported and remains highly context dependent.

4.4 Effects on institutional trust

Institutional trust should be disaggregated. Different institutions encounter different legitimacy tests, and different social groups interpret the same enforcement actions through different prior experiences.

4.4.1 Trust in police and courts

Observed evidence: research in Germany shows that perceived police discrimination is associated with stronger distrust in police and, partly, courts among some ethnic minority adolescents (Kogan et al., 2024). FRA (2021) emphasises that victims’ willingness to report hate incidents depends heavily on how law enforcement and other authorities respond.

Interpretation: application of the regulation can increase trust in police and courts when minority communities perceive that authorities are willing and able to protect them consistently and respectfully. The same application can decrease trust if victims perceive indifference, under-classification, or dismissive treatment, or if other groups perceive politicised or selective enforcement. Thus, trust depends less on punitive severity alone than on procedural justice, transparency, and consistency.

4.4.2 Trust in legislators and the media

Interpretation: trust in legislators is likely to increase among citizens who view the law as a credible defence of equal citizenship, and decrease among those who see it as vague, overbroad, or politically instrumental. Trust in media may also move in opposite directions. Coverage that explains court reasoning carefully may increase perceived legitimacy; coverage framed as moral spectacle may intensify distrust and polarisation.

Evidential status: these propositions are plausible and consistent with broader research on public trust and political communication, but the present evidence base does not support precise long-run effect sizes for Finland.

4.5 Conditions under which application is likely to help or harm

Application is more likely to increase well-being and trust when at least five conditions are met: legal boundaries are relatively clear; enforcement is consistent across ideological camps; police and prosecutors communicate decisions intelligibly; procedures are perceived as fair and proportionate; and enforcement is embedded in a wider strategy that includes victim support, reporting channels, and non-criminal responses where appropriate. Under these conditions, the law is more likely to deter intimidation without producing large spillovers into lawful speech.

Application is more likely to reduce well-being or trust when the opposite conditions obtain: vague standards, highly visible inconsistency, politicised rhetoric around prosecutions, weak procedural fairness, and poor institutional communication. Under those conditions, enforcement may still punish some harmful conduct, but its legitimacy effects deteriorate. The result can be a divided trust pattern: more trust among some targeted groups, less among groups that perceive the state as censorial or selective.

4.6 Short-term versus long-term effects

Short-term effects are likely to be easier to observe. A prosecution, police statement, or major judgment can immediately change perceived legal risk, attract media attention, and alter speech in visible forums. These changes may include rapid deletion, euphemistic reformulation, counterspeech, strategic silence, and migration to smaller communities.

Long-term effects are harder to identify because they depend on repeated interaction between law, institutions, norms, and technology. Over time, three trajectories are possible. First, beneficial consolidation: a stable norm emerges against dehumanising public attacks on groups, participation among targeted groups improves, and trust grows because enforcement is seen as fair. Second, superficial suppression: overt speech declines in mainstream venues, but hostility persists in coded or closed settings without broader trust gains. Third, backlash: enforcement becomes a symbol of partisan domination, feeding grievance and lowering trust among significant audiences. Current evidence does not justify assuming one trajectory in general.

4.7 Which claims are empirical, theoretical, or uncertain?

Relatively well supported empirically are the following claims: hate speech and harassment can impose psychological and participatory costs; public-facing professionals may self-censor in response to harassment; hate crime reporting depends in part on institutional responsiveness; and discriminatory policing is associated with mistrust in police and courts (Celuch et al., 2023; FRA, 2021; Kogan et al., 2024; Saha et al., 2019).

Supported mainly at the level of reasoned interpretation are these claims: credible enforcement can reassure minorities and improve the effective conditions for speech; legal regulation can help internalise norms against dehumanising expression; and platform governance changes can confound measured effects of criminal law. These are well grounded, but not identified as clean long-run causal estimates for Finland.

More uncertain are claims that the application of the offence has a durable aggregate effect on national social cohesion, that it unambiguously increases trust across the population, or that it predictably reduces extremist mobilisation in the long run. Those claims may be true in some contexts, but the available evidence here is insufficient for confident generalisation.

 

5. Discussion

5.1 Alternative explanations and opposing viewpoints

Changes attributed to the application of the regulation may in fact arise from other processes. One major alternative explanation is platform moderation. Social media companies have repeatedly revised their harassment and hate speech policies, changing both what is removed and what remains visible (Dubois & Reepschlager, 2024). A second alternative explanation is the wider political climate. Finnish official sources themselves connect hate speech with broader political polarisation and adversarial politics (Ministry of the Interior, 2019; Ministry of the Interior, 2024). A third is media culture: sensationalist coverage may amplify extreme cases and shape risk perceptions independently of actual legal practice. A fourth is changes in reporting and classification. Rising recorded incidents may reflect better recognition by authorities or greater victim willingness to report, not necessarily more underlying hostility.

An important opposing viewpoint holds that criminal regulation of hate speech is especially likely to generate chilling effects, strategic vagueness, and selective enforcement concerns. This concern is not baseless. The conceptual literature on chilling effects shows that legal restrictions can deter lawful speech outside their intended scope when people lack confidence about where the boundary lies (Simpson, 2024). In addition, OSCE material warns that loosely formulated or selectively applied hate speech provisions may chill freedom of expression and freedom of religion or belief (OSCE, 2024).

However, the opposing viewpoint also risks a one-sided baseline. It can treat the absence of enforcement as neutral, when in practice under-enforcement may leave targets and observers feeling unsafe, unsupported, or excluded from meaningful participation. The practical comparison is therefore rarely between "speech restriction" and "free speech" in the abstract. It is more often between different distributions of silence, fear, and social cost across groups.

5.2 Strength of evidence

Strong evidence: legal definitions of the offence; descriptive official statistics and reporting patterns; repeated findings that hate speech and online harassment can impose psychological, participatory, and deliberative harms; and evidence linking discriminatory policing experiences to mistrust in police and courts. These claims rest on legal texts, official reports, and peer-reviewed studies that converge on the same mechanisms.

Moderate evidence: claims that fair and visible enforcement can support reporting, minority protection, and institutional legitimacy; claims that the law can shape social norms over time; and claims that platform governance materially affects the speech environment. These are plausible and supported, but often through indirect evidence or adjacent literatures rather than direct long-term causal identification of the Finnish offence.

Weak evidence: broad claims about aggregate national welfare effects over the long term; claims that enforcement has a stable directional effect on polarisation; and claims that one can neatly separate legal effects from parallel technological and cultural changes. These remain open because the causal environment is too entangled and the longitudinal evidence too limited.

5.3 Main limitations

The main limitation is causal inference. The sources support mechanism-based reasoning better than they support precise causal estimation. Another limitation is transferability. Some evidence comes from Finland, some from broader European policy work, and some from other countries or occupational settings. This broadens analytical relevance but reduces jurisdiction-specific precision. A third limitation is measurement. Self-censorship, norm internalisation, and migration to closed channels are difficult to observe directly. Recorded offences and prosecutions capture only a subset of the underlying phenomenon.

 

6. Conclusion

The most defensible conclusion is conditional. The application of regulation concerning agitation against a population group can, under some conditions, improve the effective speech environment for targeted groups, strengthen minority protection, and support trust in police and courts by signalling that severe group-directed intimidation is taken seriously. Yet these benefits are not automatic and should not be assumed merely from the law’s existence. They depend on clarity, consistency, proportionality, procedural fairness, and credible communication by authorities.

The same regulatory application can also generate costs. It can produce legal-risk-based self-censorship, contribute to perceptions of selective enforcement, and push some communication into coded or closed channels. These costs are especially likely when legal boundaries are opaque or enforcement is experienced as politicised. Consequently, long-term effects on societal well-being and institutional trust are unlikely to be uniform. They may be beneficial for some groups and harmful for others at the same time.

Empirically, the strongest claims concern mechanisms rather than total net outcomes. Harassment and hate speech can narrow participation and degrade public debate; responsive and fair institutions can matter for trust and reporting; discriminatory institutional practice can undermine legitimacy. What remains insufficiently established is the size and durability of the specific long-run contribution made by applying this offence, as distinct from platform moderation, political climate, media culture, and wider polarisation.

Further research should therefore prioritise longitudinal designs that combine legal events, survey data on perceived speech freedom and trust, platform-level changes, and differentiated group responses. Especially valuable would be panel studies examining whether visible enforcement increases participation among targeted minorities while also testing whether lawful speakers perceive a broader chilling effect. Without such evidence, strong one-directional claims about the law’s long-term aggregate effects remain unjustified.

 

References

Celuch, M., Oksa, R., Ellonen, N., & Oksanen, A. (2023). Self-censorship among online harassment targets: The role of support at work, harassment characteristics, and the target’s public visibility. Information, Communication & Society. https://doi.org/10.1080/1369118X.2023.2289978

Council of Europe, European Court of Human Rights. (2023). Hate speech (Fact sheet).

Criminal Code of Finland. (1889/2022 translation). Chapter 11, sections 10 and 10a. Ministry of Justice, Finland, unofficial English translation updated to amendments up to 433/2021.

Dubois, E., & Reepschlager, A. (2024). How harassment and hate speech policies have changed over time: Comparing Facebook, Twitter and Reddit (2005–2020). Policy & Internet, 16(3), 523–542. https://doi.org/10.1002/poi3.387

European Parliament, Policy Department for Citizens’ Rights and Constitutional Affairs. (2020). Hate speech and hate crime in the EU and the evaluation of online content regulation approaches.

European Union Agency for Fundamental Rights. (2021). Encouraging hate crime reporting: The role of law enforcement and other authorities.

Kogan, I., Weißmann, M., & Dollmann, J. (2024). Police discrimination and police distrust among ethnic minority adolescents in Germany. Frontiers in Sociology, 9, 1231774. https://doi.org/10.3389/fsoc.2024.1231774

Matthes, J., Atamimi, N., Khoirunnisa, I. A., & Arif, A. (2025). Discursive consequences of social media hostility: Chilling effects, avoidance, and intervention behaviors of emerging adults in the U.S. and Indonesia. Computers in Human Behavior.

Ministry of Finance. (2021). Citizens’ panel on the freedom of expression: Recommendations for measures to be taken in Finland to protect people in public professions from hate speech and to safeguard free expression of opinion.

Ministry of the Interior. (2019). Study: Hate speech deters decision-makers from participating in public debate.

Ministry of the Interior. (2024). Hate crime.

Organisation for Security and Co-operation in Europe. (2024). Hate crime prosecution at the intersection of freedom of expression and freedom of religion or belief.

Organisation for Economic Co-operation and Development. (2021). Civic space scan of Finland.

Saha, K., Chandrasekharan, E., & De Choudhury, M. (2019). Prevalence and psychological effects of hateful speech in online college communities. In Proceedings of the ACM Web Science Conference 2019 (pp. 255–264). https://doi.org/10.1145/3292522.3326032

Wardle, C. (2024). A conceptual analysis of the overlaps and differences between hate speech, misinformation and disinformation. United Nations Department of Peace Operations and Department of Political and Peacebuilding Affairs.