Free Speech: Beyond Absolutism and Paternalism

Free speech can no longer be treated as a frozen clause in our constitutions but must be reimagined as a living right fit for the twenty-first century. Against the twin temptations of absolutism and paternalism, the case is made for a centrist-liberal approach built on “active responsibility”: systemic transparency for platforms, procedural firebreaks against government overreach, anti-SLAPP protections for journalists, and proportionate judicial remedies that preserve lawful speech. Drawing on the UN Human Rights Committee’s General Comment No. 34, landmark cases such as Sanchez v. France and Counterman v. Colorado, and the work of scholars like Tarleton Gillespie, Kate Klonick, and Evelyn Douek, the piece calls for a global rethinking of how we design the institutions, data flows, and safeguards that allow diverse voices not just to speak, but to be heard.

Freedom of expression has long been treated as the secular scripture of liberal democracy, a principle etched in the marble of constitutions and invoked with ritual reverence whenever politics turns rancorous. Yet like all rights, it is not a relic to be admired behind glass but a living infrastructure that requires upkeep. To preserve the vitality of free speech in a world reshaped by demographic change, technological disruption and political fragmentation, we must begin to understand it less as a fixed negative liberty and more as a dynamic ecosystem demanding active responsibility from those who wield the greatest power over our shared discourse.

This is not to abandon the normative baselines. The International Covenant on Civil and Political Rights, ratified by 173 states, remains foundational. Its Article 19, as interpreted by the United Nations Human Rights Committee in General Comment No. 34 (2011), affirms that freedom of expression covers the right to “seek, receive and impart information and ideas of all kinds,” and makes clear that states bear positive obligations not only to refrain from censorship but to guarantee access and pluralism. The Committee’s insistence that criticism of public officials must enjoy heightened protection is not an antiquated footnote; it is a live instruction at a time when “fake news,” insult, or disinformation laws proliferate. Yet if law supplies the grammar, practice provides the narrative, and that narrative is increasingly fractured.

“What matters is whether the architecture of recommender systems, transparency dashboards, appeals, and audits produces a speech environment that is accountable and plural.”

The evidence is sobering. Freedom House’s Freedom in the World 2025 records the nineteenth consecutive year of global democratic decline, with expression singled out as the liberty most in retreat. CIVICUS’s global monitor reports that violations against expression (arrests of protesters, censorship of journalists, harassment of activists) constitute the single largest category of civic space violations worldwide. UNESCO confirms that at least sixty-eight journalists and media workers were killed in 2024, more than sixty per cent in conflict zones. And Access Now’s #KeepItOn coalition counts a record 296 government-ordered internet shutdowns in 54 countries during that same year. The trajectory is unmistakable: while the capacity to speak to millions is technically greater than ever before, the vulnerability of that speech is also historically acute.

What has changed is not merely the willingness of governments to suppress but the architecture through which speech circulates. Tarleton Gillespie’s Custodians of the Internet mapped how content moderation decisions, once hidden within technical departments, now shape the distribution of public discourse at planetary scale. Kate Klonick’s seminal article “The New Governors” described how private platforms, facing pressure from states and civil society alike, built internal systems of quasi-law, replete with due-process language and appeals, even while remaining accountable to advertising markets rather than electorates. Evelyn Douek, in her more recent work on content moderation as “systems thinking,” rightly warns that obsessing over individual takedowns misses the point; what matters is whether the architecture of recommender systems, transparency dashboards, appeals, and audits produces a speech environment that is accountable and plural. To speak of free expression today without reference to these scholars is to miss the terrain entirely.

“The most serious threats to expression today are structural, not textual. They lie in opaque recommender systems that determine virality.”

This recognition has already begun to shape regulation. In the European Union, the Digital Services Act (DSA) has shifted the focus from individual pieces of content to the systems that govern them. Article 40 grants vetted researchers access to platform data, forcing legibility where opacity was once the norm. The Commission has not hesitated to open formal proceedings: against X (formerly Twitter) for failures in risk management, against Meta for risks to children’s safety, against TikTok Lite for its “rewards” feature that was swiftly withdrawn, and against AliExpress for opaque recommender systems. These actions do not determine truth or falsehood; they demand auditable processes. In the United Kingdom, Ofcom is gradually implementing the Online Safety Act, focusing on illegal harms while resisting a general monitoring obligation. Meanwhile, universities face a separate regime under the Higher Education (Freedom of Speech) Act 2023, with the Office for Students now issuing guidance clarifying that lawful but shocking ideas must be tolerated on campus. Together, these initiatives represent a cautious but significant rebalancing: away from blunt prohibition and toward transparency, oversight, and resilience.

Courts, too, are recalibrating. In the United States, the Supreme Court in Counterman v. Colorado (2023) required proof of at least recklessness in prosecutions of threatening speech, narrowing the state’s criminal reach and acknowledging the difficulty of distinguishing menace from misunderstanding in online communication. In Lindke v. Freed (2024), the Court set out a test for when public officials’ social media accounts constitute state action, clarifying when blocking users crosses into constitutional violation. The decision in Murthy v. Missouri (2024), dismissing a case on government “jawboning” of platforms for lack of standing, left unresolved the line between persuasion and coercion, but made plain that future cases will turn on fact-intensive scrutiny.

The European Court of Human Rights has moved in a different direction. In Sanchez v. France (2023), it upheld the conviction of a politician for failing to remove hateful comments under his Facebook post, signalling that the duty to police online forums may fall on individuals as well as platforms. In Hurbain v. Belgium (2023), it required a newspaper to anonymise an archived article to protect the rehabilitation rights of a named individual, even at the expense of journalistic integrity of the record. Both judgments have defensible logic, yet both highlight the chilling potential of liability rules that encourage over-removal or erasure. The Court of Justice of the EU’s line, from Glawischnig-Piesczek v. Facebook (2019), allowing injunctions against “identical and equivalent” content across borders, to Google v. TU, RE (2022), affirming a right to be de-listed when content is manifestly false, similarly illustrates how the battle has shifted from the town square to the search index.

Against this shifting backdrop, the liberal centre must resist the false binary of absolutism and paternalism. American First Amendment absolutists argue that any positive obligation is the thin end of censorship; European paternalists insist that dignity, equality and privacy must override liberty. Both risk missing the central problem: that the most serious threats to expression today are structural, not textual. They lie in opaque recommender systems that determine virality, in governments’ reliance on emergency shutdowns that become permanent, in vexatious lawsuits that bankrupt journalists before truth can be adjudicated. The challenge is not whether to protect speech but how to design systems that prevent the most powerful actors from distorting it.

This requires what I call “active responsibility.” The principle is modest yet ambitious: allocate enforceable duties to the most powerful nodes in the speech ecosystem while preserving robust protection for individuals. It means data-access regimes for platforms that operate as public squares, ensuring their algorithms can be scrutinised by independent experts. It means firebreaks against executive overreach—time-limited shutdown powers, expedited judicial review, and public rationales archived for accountability. It means anti-SLAPP statutes with early dismissal procedures, shifting costs to abusive plaintiffs, so that journalists and NGOs are not silenced by attrition. It means remedies from courts that are proportionate and speech-preserving: anonymisation rather than deletion, friction rather than blanket bans, and reach reduction rather than criminalisation where the content is offensive but lawful. It also means building an economic floor under pluralism: ownership transparency, targeted funding for independent outlets, and protection against oligarchic capture, as embodied in the European Media Freedom Act of 2024. And it means universities modelling tolerance for discomfort, protecting both controversial speech and peaceful protest, thus cultivating citizens resilient to difference.

“That freedom of expression will not survive if we cling to rigidity. Its health lies in tension with dignity, with equality, with safety, and in adaptation.”

The global south offers as much to learn as Strasbourg or Washington. The Inter-American Court of Human Rights has for two decades pressed states to replace criminal defamation with civil remedies, recognising the chilling effect on democratic oversight. The African Commission’s 2019 Declaration of Principles on Freedom of Expression and Access to Information in Africa modernised regional standards by affirming internet rights and proportionality tests. India’s replacement of sedition with Section 152 of the Bharatiya Nyaya Sanhita 2023, prohibiting acts that endanger sovereignty, unity or integrity, illustrates how vague drafting can invite arbitrary use, a warning as much as a precedent. South Africa’s Constitutional Court, in cases such as Qwelane v. South African Human Rights Commission (2021), has grappled with balancing hate-speech prohibitions with constitutional guarantees, providing another model of jurisprudence in evolution. A liberal project that borrows selectively from each region’s strongest safeguards, while rejecting their worst temptations, will be more resilient than one that remains parochially transatlantic.

None of this will persuade absolutists who insist that any expansion of obligations is censorship, nor those who argue that the harms of hate speech and harassment are so corrosive that only aggressive prohibition will do. But there are answers to both. To absolutists, one can point out that the status quo is not freedom but private bureaucracies setting policy in secret; procedural oversight is not censorship but discipline. To paternalists, one can concede the real harms of algorithmically amplified hate yet insist that the most effective remedies are systemic, not doctrinal: redesigning recommendation engines, disclosing enforcement error rates, and providing friction in moments of civic vulnerability. The lesson from Douek’s systems thinking is that efficacy, not moral satisfaction, should be our lodestar.

What, then, is the provocation for our time? That freedom of expression will not survive if we cling to rigidity. Its health lies in tension with dignity, with equality, with safety, and in adaptation. The liberal centre must reimagine free speech not as the absence of restraint but as the presence of conditions that allow diverse voices to be heard, contested, and remembered. That means humility: no single court or parliament will settle the balance once and for all. It also means ambition: designing institutions, funding mechanisms, and oversight structures that can evolve with empirical evidence and demographic change. Rights endure not by freezing in time but by adapting to it.

The marble tablets gave us the words. Our century demands that we give them life.

Sources

Access Now. (2024). #KeepItOn 2024: Internet shutdowns report. Retrieved from https://www.accessnow.org

African Commission on Human and Peoples’ Rights. (2019). Declaration of principles on freedom of expression and access to information in Africa. Banjul: ACHPR.

CIVICUS. (2024). People power under attack 2024: A global analysis of civic space. Johannesburg: CIVICUS Monitor.

Court of Justice of the European Union (CJEU). (2019). Glawischnig-Piesczek v. Facebook Ireland Limited (Case C-18/18). Luxembourg: CJEU.

Court of Justice of the European Union (CJEU). (2022). TU and RE v. Google (Case C-460/20). Luxembourg: CJEU.

Douek, E. (2020). Content moderation as systems thinking. Harvard Law Review Forum, 134, 61–75.

European Commission. (2024). Digital Services Act: Formal proceedings against X, Meta, TikTok Lite and AliExpress. Brussels: European Commission.

European Parliament & Council of the European Union. (2024). European Media Freedom Act (Regulation (EU) 2024/… of 11 April 2024). Official Journal of the European Union.

European Parliament & Council of the European Union. (2024). Directive (EU) 2024/1069 on protecting persons engaging in public participation from manifestly unfounded or abusive court proceedings (Anti-SLAPP Directive). Official Journal of the European Union.

Freedom House. (2025). Freedom in the world 2025: Democracy under threat. Washington, DC: Freedom House.

Gillespie, T. (2018). Custodians of the Internet: Platforms, content moderation, and the hidden decisions that shape social media. New Haven, CT: Yale University Press.

Inter-American Court of Human Rights. (2008). Kimel v. Argentina (Judgment of May 2, 2008, Series C No. 177). San José, Costa Rica: IACtHR.

Klonick, K. (2018). The new governors: The people, rules, and processes governing online speech. Harvard Law Review, 131(6), 1598–1670.

Office for Students (OfS). (2025). Guidance on freedom of speech duties under the Higher Education (Freedom of Speech) Act 2023. Bristol: OfS.

Ofcom. (2024). Online Safety Act: Illegal harms codes of practice and enforcement guidance. London: Ofcom.

Reporters Without Borders (RSF). (2025). World Press Freedom Index 2025. Paris: RSF.

Supreme Court of the United States. (2023). Counterman v. Colorado, 600 U.S. (2023). Washington, DC: SCOTUS.

Supreme Court of the United States. (2024). Lindke v. Freed, 601 U.S. (2024). Washington, DC: SCOTUS.

Supreme Court of the United States. (2024). Murthy v. Missouri, 602 U.S. (2024). Washington, DC: SCOTUS.

UNESCO. (2025). Safety of journalists observatory: 2024 killings and attacks data. Paris: UNESCO.

United Nations Human Rights Committee. (2011). General Comment No. 34: Article 19, Freedoms of opinion and expression (CCPR/C/GC/34). Geneva: United Nations.