The Speech Dilemma; Free speech vs hate speech

The Speech Dilemma

Today’s highly contentious issue is the balance between free speech and hate speech in the digital era. Since communication would eventually be online, the current digital time has emboldened both positive discourse and harmful rhetoric, engaging heated debates about the line between freedom of expression and regulation of hate speech. This debate is nothing new, but the implications have never been starker in light of the internet’s global presence and instantaneous nature.

Importance of Free Speech

The right to free speech is the very bedrock of any democratic society. It is used to voice opinions, share ideas, and criticize leaders without fear of retribution. As enshrined by international law, Article 19 of the Universal Declaration of Human Rights declares that everyone is entitled to hold an opinion and express it freely [1]. This right stands as a fundamental anchor in modern democratic principles and has been iterated in the constitutions of many countries.

On the other hand, however, the principle of free speech does not have limits. As pointed out by numerous scholars and activists, absolute freedom in any sphere leads to nothing short of chaos. The man who cries fire in a crowded theater in the case of Schenck v. United States, 1919, was famously opined by Justice Oliver Wendell Holmes to not be entitled to “the most stringent protection of free speech” [2]. This line of thinking brings to bear the fact that irresponsible speech is capable of causing damage.

Online Hate Speech

With the evolution of social media, hate speech has multiplied in ways that just a few decades ago simply would have been unimaginable. Hate speech is a form of communication that discriminates against or vilifies an individual based on aspects like race, religion, gender, or sexual orientation. In an age in which billions of people are connected through the internet, it has become alarmingly easy to send unhealthy rhetoric around, to inflame people’s passions, or to do either of the above.

According to Pew Research Center, 41% of Americans have experienced online harassment and 25% of them are harassed because of some personal characteristic [3]. Similarly, the percentage of hate speech online is up 21% from the previous year according to the 2020 report of the European Commission with much of it targeting immigrants and ethnic minorities [4].

Unchecked hate speech online has had appallingly real repercussions: anti-Muslim propaganda on Facebook helped fuel the persecution of an ethnic minority group, Myanmar’s Rohingya people [5], with what the United Nations termed ethnic cleansing; in the United States, the Anti-Defamation League reported that online hate speech had been correlated with an increase in hate crimes [6], a 17% jump in hate incidents against Jews between 2020 and 2021.

Legal Landscape

Freedom of speech is respected to different extents all over the world; however, definitions of hate speech and laws governing it are vastly different. In the United States, freedom of speech has very strong protection on the First Amendment, which even permits offensive or controversial speech, so long as it does not present a situation where imminent violence will result. The landmark Brandenburg v. Ohio case in 1969 further emphasized this position when the court ruled that inflammatory speech was protected unless it was directed at inciting “imminent lawless action.”

Laws relating to hate speech, though, are much stricter in Europe. Germany’s NetzDG, launched in 2017 [7], requires social media companies to remove illegal hate speech within 24 hours after receiving a complaint, with the threat of fines of as much as €50 million. France is also pretty severe on hate speech [8], with calls for stricter regulation to combat hate speech as online extremism increases following terrorist attacks in the country. Under French law, public incitement to hatred on grounds of race, religion, or nationality is prohibited.

These legal frameworks in both the U.S. and Europe bring forth a basic difference where nations view the crossing of free speech over hate speech. In the U.S., it is often considered a sacrosanct right, while European countries tend to look more toward the protection of individuals and communities against hate speech, as found in the rhetoric.

Role of Social Media

Companies operating in this arena include Facebook, Twitter, and YouTube. For hate speech, these are the primary platforms through which the freedom of speech is usually exercised. Faced with such a scenario, these companies developed content moderation policies to curb such harmfulness, though they are often criticized for inconsistent application.

Facebook gives one example: in its 2021 Community Standards Enforcement Report [9], it announced that it had removed 22.1 million pieces of hate speech content within the first quarter of the year. Twitter, also unable to overcome the problem, permanently suspended former U.S. President Donald Trump even after the Capitol riots of January 6, 2021[10], due to the “risk of further incitement of violence.” However, these are decisions that have raised many controversies as some think that such acts represent censorship, while others believe them necessary to avoid causing damage.

In many tech companies, algorithms are being used heavily to detect and block hate speech. These computers can flag harmful content quickly, but very far from perfect: sometimes they fail to spot nuances, leading to deletions of perfectly fine content; sometimes, they do not behead the problem with delicate content but allow dangerous content to pass through, such as the case when the Christchurch mosque shooter even live-streamed his attack on Facebook in 2019 [11].

This question of the limits of free speech is about how to regulate online hate speech without infringing on free speech, a very defining challenge of our age. With the internet ever-expanding, so too will the boundary of acceptable speech. Ultimately, creating an online environment that is friendly to free expression and the rights of individuals requires a multifaceted approach involving governments, tech companies, but also individuals, and civil society. To walk this tightrope in a way that rewards freedom as well as respect requires collective efforts.

While governments and tech companies have much to do with addressing hate speech, the role of civil society organizations and people cannot be overlooked. Civil society groups or associations, in the form of NGOs, advocacy organizations, and academic institutions, often act as intermediaries between regulators and the public. They often advocate on behalf of vulnerable communities, conduct research on the impact of hate speech, and provide recommendations on policy. For instance, the SPLC and HRW among others have tracked hate speech incidents and have advocated for strong regulations on hate speech [12]. Furthermore, civil society is particularly effective in building awareness and education. Programs aimed at enhancing digital literacy to nurture an appropriate online culture constitute core efforts that address hate speech at its core. The Dangerous Speech Project [13], for instance, focuses on educating communities about the language and narratives that precede mass violence, helping them recognize and counter dangerous rhetoric before it escalates.

According to project founder, Susan Benesch, “Hate speech doesn’t emerge from a vacuum; it is nurtured by the culture of the society in which it occurs.” Once more, it shows the necessity for cultural change in the way we think about online discourse. Individual responsibility is equally important. Further moderation and regulations by the system may also curtail the growth of hate speech; however, individual users, in turn, must own what is produced and shared. Social movements like #NoHateSpeech, instigated by the Council of Europe, also make people voluntarily report objectionable content and be actively involved in encouraging tolerance. Studies have shown that users can collectively push back against hate speech through counter-speech or positive reinforcement, thus reducing its harmful effects. For example, a study by The Institute for Strategic Dialogue (ISD) [14] revealed that campaigns of counter-speech diminished the visibility of extremist narratives by up to 45% on Twitter. Such activities signal to everyone that regulation is essential, but community empowerment in dealing with hate speech through non-legislative activity is just as important.

The role of civil society in fighting hate speech brings about a holistic approach it whereby legal measures are balanced with grassroots activism and social mobilization. Without this multi-pronged strategy, efforts to tackle hate speech may fail to address its cultural and social underpinnings, allowing it to persist and evolve in more insidious forms. Furthermore, the unequal distribution of technology exacerbates the issue of platform accountability. Whereas Facebook and Twitter have their main bases in wealthy nations with the majority focusing the most moderation efforts within English-speaking countries, they tend to ignore content in local languages or regional dialects that would be a fertile breeding ground for hate speech elsewhere in the world.

Criticism has faced Facebook over its failure to sufficiently moderate hate speech against minority communities in India [15], mainly in regional languages such as Hindi and Bengali. In 2021, the Wall Street Journal reported that while Facebook had strict policies in place on content moderation in English, hate speech was largely overlooked in regional languages, which sometimes led to real-life violence. This gap in enforcement shows that global tech platforms must become more localized if they are going to effectively and seriously combat hate speech. As such, the unique challenge of the digital divide cannot be disregarded in efforts to regulate online hate speech, as it will remain uneven and perhaps ineffective in areas where it most often takes place. Ethics of Regulation This online regulation of hate speech raises important ethical questions. Who should decide, for instance, what hate speech is? Should governments or private companies, such as Facebook and Twitter, be the appropriate bodies to make this decision? What happens when such decisions come into conflict with the rights of freedom of expression of users?

There is also the risk that hate speech regulations might be employed to suppress dissenting or otherwise marginalized voices. In some jurisdictions, hate speech laws have been weaponized by authoritarian regimes to suppress political opposition or minority groups. For instance, in Turkey, laws designed to curb hate speech have been used to jail journalists and to silence critics of the government.

Balancing Free Speech and Hate Speech

Finding this delicate balance between protecting free speech and censuring hate speech is therefore a tricky thing. It requires thoughtful consideration of the potential harms of speech, set alongside the deeply ingrained importance of freedom of expression. Various strategies have emerged in responding to these issues that societies are wrestling with.

Adopting clearer, more transparent definitions of hate speech can help prevent the misuse of regulations for silencing and stigma-laden speech. Content moderation policies should also be more uniformly enforced, allowing for clear avenues of appeal when users believe their content has been wrongly removed. Education also has a critical role to play. Digital literacy, coupled with responsible behavior in interacting online, can set up a culture of responsibility that deprecates hate speech. “The best way to counter hate speech is more speech-better, smarter, counter-speech,” says British journalist Timothy Garton Ash.

Global Digital Divide

Another understated aspect of the hate speech debate is the global digital divide, which redefines how hate speech may be perceived, regulated, and addressed in different regions.

The digital divide is understood as the gap between people who, compared with others, have ready access to modern information and communication facilities and those who do not. Such a division can be seen as very pronounced between the developed and the developing nations, which fundamentally change how online hate speech is monitored and managed. In regions with limited access to high-speed internet and social media platforms, the spread of hate speech may appear slower or less visible, but when these regions finally gain access, they mostly lack the infrastructure, policies, or awareness to regulate such harmful content. For instance, online hate speech and misinformation are not only pertinent causes of political violence and ethnic tensions in many parts of Sub-Saharan Africa but weak governmental frameworks and low levels of digital literacy hinder efforts to address these issues effectively. The World Bank also revealed that internet usage in Sub-Saharan Africa stands at a low 29% [16] of the population: thus, most of the hate speech monitoring and intervention strategies developed in the Global North may not be applicable or scalable in these regions.

In addition to the legal and social challenges of regulating hate speech, the psychological impact on individuals and communities must also be acknowledged. Research has further established that frequent exposure to hate speech may be associated with significant emotional and mental disorders, especially in minority groups.

For instance, a report on the impact of online harassment on Youth Mental Health [17] pointed out that hate speech victims typically tend to develop more anxiety, depression, and loneliness. This is even worse among young people who have been found highly susceptible to online harassment.

In a survey carried out by the Anti-Bullying Alliance, 42% of UK young people aged 12-20 said they have experienced cyberbullying at least once, and much of that is based on hate speech. Its impact reaches far beyond individuals, also. Communities frequently exposed to hate speech online face a more general experience of social alienation, which can impair intergroup relationships and reinforce cycles of discrimination and violence. What makes online hate speech particularly damaging is its persistence and visibility. Unlike in-person encounters, hateful content on social media can be shared and spread to large audiences within minutes, magnifying its harmful effects. Victims are often reminded of the abuse each time the content resurfaces, compounding the psychological damage over time.

Internet anonymity also emboldens perpetrators, who might otherwise shut their mouths while giving hateful opinions during face-to-face interactions. As reported by Professor John Suler, a cyberpsychologist, this phenomenon, which is often termed the “online disinhibition effect,” tends to encourage people to be more aggressive and harmful online than in the real world. Addressing the psychological implications of hate speech, therefore, cannot be seen as content regulation but rather safe spaces online where the focus is on mental health and emotional well-being. AI: Algorithms and artificial intelligence are becoming ever more central to the regulation of online content, including hate speech, as online platforms become even more extensive. Social media companies such as Facebook, Twitter, and YouTube heavily depend on algorithms to scan for harmful content and remove it. These are programmed systems to scan large data sets looking for keywords, phrases, or even patterns that might include hate speech.

However, while AI has developed in leaps and bounds over the last few years, there’s a lot more work ahead of it.

One of the biggest challenges is context—algorithms often struggle to interpret the hidden meanings behind certain words or phrases. For example, a word could constitute hate speech in one context but be part of a reclaiming movement in another, where derogatory terms are repurposed for empowerment by marginalized groups. Thus, this lack of contextual understanding leads to both over-censorship or proper speech being mistakenly flagged, as well as under-censorship, where content that can cause harm escapes moderation. Furthermore, AI-based moderation systems often contain bias, reflecting the prejudices embedded in the data they have been trained on. The Alan Turing Institute conducted research [19] that showed these algorithms have a higher chance of throwing up incorrect flags on the content of minority groups, thus leading to further marginalization of such already vulnerable communities. This brings the issues of fairness and transparency in moderation processes into question. Therefore, experts argue that AI systems should not work independently but in conjunction with human moderators who can provide the necessary context and judgment.

As Professor Kate Crawford of New York University [20] points out, “AI can be an essential tool in moderating content, but without human oversight, it risks perpetuating the very problems it aims to solve.” Thus, an effective approach to online hate speech requires a hybrid system that combines AI efficiency with human discernment.

In her book Atlas of AI, Professor Kate further notes the limitations and potential biases of AI systems, she writes, “AI systems are not autonomous, objective, or neutral. They are embedded in social, political, and economic worlds, and as such, they produce and reinforce power structures.”

Freedom of Speech in Pakistan

Discourse on freedom of speech finds a parallel in Pakistan, which has enshrined this principle in the Constitution under Article 19. This provision ensures the right to freedom of speech and expression in the interest of national security, public order, and morality. These restrictions embody the unique interplay of democracy and Islamic values in Pakistan, reflecting the socio-political fabric of the state as the Islamic Republic. The framing of Article 19 allows for significant flexibility in curbing speech deemed detrimental to societal cohesion, a feature that differentiates Pakistan from Western democracies with broader guarantees for free speech.

The judiciary in Pakistan has defined the contours of this right in many ways. Perhaps most significantly, the court rendered martial law invalid in Asma Jilani v. Government of the Punjab [21], establishing a crucial juncture for democratic rights, like free expression. This move set the stage for more aggressive challenges to authoritarian speech constraints emanating from the judiciary. The judgment, though based on political rights, reflected the judiciary’s more general duty to protect individual freedoms even when state security is pleaded as a reason for limitation.

A similar conflict arose in Begum Nusrat Bhutto v. Chief of Army Staff and Federation of Pakistan [22], where the court attempted to address the impact of martial law on civil liberties. Here, freedom of expression was analyzed in the context of state stability and security, reflecting the precarious balance Pakistan has often maintained between individual rights and collective harmony. The judgment highlighted that while the judiciary might validate certain state actions for the sake of public order, it also delineates the limits of such justifications, emphasizing that freedoms cannot be suppressed indefinitely.

Miss Benazir Bhutto v. Federation of Pakistan [23] has given some important lessons in the arena of political speech. It was a case challenging political censorship under the military regime. The court found that speech could be restrained on grounds of public order; however, the restriction must not exceed the limit and should not result in the defeat of democratic processes. This judgment is a typification of the evolving jurisprudence in Pakistan, where courts are sometimes forced to reconcile state power with a right to dissent.

More contemporary, Shafqat Ali v. State [24] dealt with some of the challenges thrown in by digital platforms and the problem of hate speech. Courts invoked the Prevention of Electronic Crimes Act 2016 relating to online misinformation and offensive rhetoric. The case illustrated how Pakistan’s courts are navigating the complexities of digital speech, seeking to curb harm without overstepping to censorship. However, the broad language of PECA and its implementation has been criticized as being open to misuse, and there is a clear need for greater legal clarity.

Finally, in Al-Jehad Trust v. Federation of Pakistan [25], the emphasis put on judicial independence indirectly speaks to the importance of freedom of expression. An untouchable judiciary is something necessary for providing security to civil liberties, most of all speech. A case like this is a real-life example of how even structural elements of governance make all the difference in their actualization of constitutional rights in practice.

Pakistan’s approach to free speech embodies a complex interplay between its foundational Islamic values and the aspirations of a modern democratic state, forging a path that is both uniquely principled and deeply pragmatic. Unlike the United States, where the First Amendment prioritizes speech except in cases of imminent lawless action, or Europe, where hate speech laws reflect a rights-based approach, Pakistan’s framework incorporates moral and religious considerations into its legal reasoning. This has come out in a jurisprudence that seeks to balance individual freedoms against collective societal values. However, the practical application of those principles remains very challenging to apply, especially in these days of digital communication when the judiciary and legislature is continually called upon to adapt constitutional protections to changing modes of communication.

Conclusion

Achieving an equilibrium between free speech and hate speech is not only a legal regulatory challenge but also reflects how we define our shared values in the age of the internet. The internet is no longer a neutral space; it shapes societies, influences elections, and fosters global movements. Yet to protect free speech without enabling harm, we have to recognize this paradox: freedom is not the absence of rules but the presence of principles.

As the philosopher John Stuart Mill opined, “The worth of a state in the long run is the worth of the individuals composing it.” In that regard, regulation isn’t censorship – it’s a social contract. Companies must not adopt blanket, feel-good bans but rather have open-ended policies based on accountability and context. Similarly, users need to be responsible for what they say online: their speech will have weight and consequences.

The future of online regulation will depend on how we, as a society, value empathy alongside expression, rather than by what governments and tech companies say it should be. Freedom of speech and hate speech are not two opposing elements but an experiment to see whether we can exercise freedom responsibly in an interconnected world.

References:

1. Universal Declaration of Human Rights 1948, Article 19.

2. Schenck v. United States (1919) 249 U.S. 47. US Supreme Court
3. Pew Research Center, “The State of Online Harassment in America.” (2021) <https://www.pewresearch.org/internet/2021/01/13/the-state-of-online-harassment/> (last accessed: 2^nd December, 2024)
4. European Commission, “Countering Illegal Hate Speech Online – 5th Monitoring Round” (2020) <https://ec.europa.eu/commission/presscorner/detail/en/ip_20_1134> (last accessed: 2^nd December, 2024)
5. United Nations, “Report of the Independent International Fact-Finding Mission on Myanmar” (2018) <https://www.ohchr.org/sites/default/files/Documents/HRBodies/HRCouncil/FFM-Myanmar/A_HRC_39_64.pdf >(last accessed: 2^nd December, 2024)
6. Anti-Defamation League, “Audit of Anti-Semitic Incidents: 2020” (2020) <https://www.adl.org/audit2020> (last accessed: 2^nd December, 2024)
7. German Federal Ministry of Justice, “Network Enforcement Act (NetzDG)”. (2017) <https://sciencespo.hal.science/hal-03586791/document> (last accessed: 2^nd December,2024)
8. The Guardian, “Calls for Crackdown on Online Hate Speech after France Attacks.” (2021) <https://www.theguardian.com/world/2021/aug/16/hate-speech-inquiries-launched-in-france-over-antisemitic-protest-banners> (last accessed: 2^nd December,2024)
9. Meta, Community Standards Enforcement Report – Q1 2021 (2021) <https://transparency.fb.com/data/community-standards-enforcement/ > (last accessed: 2^nd December, 2024)
10. Twitter Safety, “Permanent Suspension of @realDonaldTrump.” (2021) <https://blog.twitter.com/en_us/topics/company/2021/permanent-suspension-of-realDonaldTrump.html> (last accessed: 2^nd December, 2024)
11. BBC, “Christchurch Mosque Shootings: Social Media Scrambles to Remove Video.” (2019) <https://www.bbc.com/news/technology-47583393> (last accessed: 2^nd December, 2024)
12. Southern Poverty Law Center, Hate Map and Extremism Tracker <https://www.splcenter.org/hate-map > (last accessed: 2^nd December, 2024)
13. Dangerous Speech Project. (2020), Countering Dangerous Speech. Available at: <https://www.dangerousspeech.org/> (last accessed: 2^nd December,2024)
14. Institute for Strategic Dialogue, “Impact of Counter-Speech on Online Extremism.” (2020) <https://www.isdglobal.org/research/> (last accessed: 2^nd December, 2024)
15. The Wall Street Journal, “Facebook’s Language Blind Spots Weaken Moderation.” (2021) <https://www.wsj.com/articles/facebooks-language-blind-spots-weaken-hate-speech-moderation-11634362745> (last accessed: 2^nd December, 2024)
16. World Bank. (2021) Internet Access in Sub-Saharan Africa <https://data.worldbank.org/indicator/IT.NET.USER.ZS?locations=ZG> (last accessed: 2^nd December, 2024)
17. Impact of Online Harassment <https://ojs.aaai.org/index.php/ICWSM/article/view/31355 > (last accessed: 2^nd December,2024)
18. Anti-Bullying Alliance Survey. <https://anti-bullyingalliance.org.uk/sites/default/files/uploads/attachments/Focus_on_Bullying_2021_0_0.pdf> (last accessed: 2^nd December,2024)
19. Alan Turing Institute, “Bias in AI Moderation Tools” (2021) <https://www.turing.ac.uk/courses/assessing-and-mitigating-bias-and-discrimination-ai> (last accessed: 2nd December,2024)
20. Kate Crawford, Atlas of AI (2021)
21. Asma Jilani v. Government of the Punjab (PLD 1972 SC 139)
22. Begum Nusrat Bhutto v. Chief of Army Staff and Federation of Pakistan (PLD 1977 SC 657)
23. Miss Benazir Bhutto v. Federation of Pakistan (PLD 1988 SC 416)
24. Shafqat Ali v. State (PLD 2020 SC 420)
25. Al-Jehad Trust v. Federation of Pakistan (PLD 1996 SC 324)