The Paradox of Freedom of Speech: Reconciling Democratic Integrity with Free Speech in a New World

Mario Nawar

Edited by Samantha Tonini, Jasmine Iyer, Mac Kang, and Sahith Mocharla

I. Introduction

Over two hundred years ago, the United States’ Founding Fathers enshrined our right to practice free speech. But did they do so under the belief that every person must be able to express their views without criticism or condemnation, or the more nuanced idea that criticism must be allowed by the government to uphold their own right to free speech? What happens when people use speech to express thoughts or judgments deemed as hateful or harmful? This raises the question of whether speech should remain absolutely free, even if doing so could clash with the right for others to exist without fear of persecution [1].

While the First Amendment’s protections of free speech presumably aim to safeguard individual liberty and egalitarian discourse, modern jurisprudence reveals growing contradictions and conflicts between these two, perhaps mutually exclusive, ideas. In investigating the Founding Fathers’ motivations to establish this right, one would find that they believed free speech was fundamental and essential for a functioning democracy; democratic integrity specifically refers to the conditions for democratic self‑governance and, crucially, equal participation, not just majority rule [2]. The evolution of the doctrine––from Justice Holmes’ ‘marketplace of ideas’ in Abrams v. United States to the ‘imminent lawless action’ test of Brandenburg v. Ohio to the recent tensions in 303 Creative LLC v. Elenis––exhibits the struggles courts face in balancing free speech as both a shield for dissent and, at times, a tool for exclusion. Contemporary challenges to free speech, especially social media’s extended platform for expression, call for the First Amendment to be recalibrated toward a model that better protects free expression without enabling harm or democratic erosion [3]. How does one reconcile the right to free speech with the potential dangers of hate speech and its amplification through the online world as a modern public square?

II. The Dilemma of Balancing Liberty and Harm

Any inquiry into the value the Framers assigned to ‘free speech’ is resolved by its placement in our foundational charter––positioned front and center, accorded primacy over all other enshrined rights. The First Amendment promises that our ideas, no matter how unpopular, can be spoken. The Founding Fathers envisioned liberty not just as individual autonomy but as the groundwork for collective self‑governance [4]. But the same protection that enables dissent to flourish also allows disinformation, extremism, and discrimination to diffuse through our social accord. Time after time, the courts have struggled to discern when speech nurtures democracy and when it corrodes it.

Determining when one’s speech qualifies as protected free expression is challenging, especially when one person’s right to speak or act on one’s conscience may limit another’s right to their own views [5]. Courts often treat speech motivated by harm differently from dissent, determining the legal line between ‘conduct’ and ‘speech’ that would grant the highest level of First Amendment protection [6]. While free speech may have been instituted as an instrument of civic dialogue, it increasingly functions as a claim of exemption––from public health mandates, anti-discrimination laws, employment standards, and even commonly held notions of equality and democracy. This shift toward treating free speech as a paradigm for exemption reflects a broader legal and cultural transformation in which the First Amendment is invoked to carve out zones of immunity from otherwise applicable laws, raising the tension between liberty (with speech) and harm (through disinformation and discrimination) [7]. Rather than protecting the political dissenter or the vulnerable minority speaker––as originally intended by the doctrine––contemporary litigants increasingly deploy free speech as a shield against regulation in realms far removed from the traditional speech marketplace. The evolution of the First Amendment doctrine from early defenses of radical speech to challenges of hate speech, digital communication, and religious expression in a plural democracy reveals how a fundamentally protective design holds the potential to be corrupted and misused.

To combat such threats, Karl Popper’s “paradox of tolerance” clarifies that a democracy’s commitment to protecting even offensive expression can ultimately endanger the very conditions of mutual respect that make free expression meaningful [8]. To maintain a tolerant society, it must be willing to be intolerant of intolerance, but only to the extent necessary to protect itself from destruction, emphasizing the paradoxical challenge of balancing freedom of expression with the need to protect vulnerable groups from harm and preserve a tolerant society itself. Such protections prompt doctrinal frameworks that view speech not only as an individual right but as a civic responsibility whose exercise must be weighed against the potential erosion of others’ equal liberty.

III. Historical Foundations of Free Speech

A. Holmes and the Marketplace of Ideas

The contemporary free speech narrative begins with Justice Oliver Wendell Holmes Jr.’s famous dissent in Abrams v. United States (1919) , where the Supreme Court upheld convictions under the Espionage Act, a statute enacted in 1917 that criminalized the transmission of information that could harm national defense [9] [10]. In Abrams, the Supreme Court affirmed that the distribution of anti-war leaflets denounced U.S. intervention in the Russian Civil War and called for a general strike to halt munitions production––thereby constituting a potential threat to national security [11]. Holmes’ dissent in Abrams emerged partly in response to his earlier articulation of the “clear and present danger” test in Schenck v. United States (1919), where he had permitted restrictions on speech posing an immediate threat to national security [12]. By Abrams, he reconceived this standard toward greater protection, insisting that only speech producing an imminent and concrete harm beyond mere political dissent could justify government suppression. Yet Holmes’ ‘marketplace of ideas’ metaphor has since drawn sustained criticism from scholars who note that structural inequalities, concentrated media ownership, and disparities in political power distort the freedom of this marketplace, allowing the ideas amplified by wealth, institutional influence, or algorithmic visibility to unequally prevail.

Justice Holmes envisioned an, at the time, radical idea in Abrams: the truth ought not emerge from a governmental decree but rather from a ‘marketplace of ideas.’ He explained that the best test of truth is “the power of the thought to get itself accepted in the competition of the market” [13]. If such viewpoints––no matter how absurd––attain wide and quick acceptance, then the claim is more credible than even an ‘objective truth’ that is instantly rejected by the masses. Holmes’ framework envisages democratic resilience grounded in the belief that consensus counters falsehoods. However, his proposal also assumes an equal marketplace in which voices have comparable access and influence. Over the next century, this assumption has become increasingly tenuous. While Holmes prized the idea of truth being in the jurisdiction of the claims and emerging through contest as opposed to through government decree, he could not anticipate the media’s capacity to persuade masses towards ‘truths’ that may not withstand scrutiny, nor the reality that this ‘marketplace’ metaphor began to erode under fear of communism. When further considering that speech, once spoken, cannot be retracted completely due to its permeance in the 'marketplace,' as it were, it is imperative to recognize that, even if later disproven, falsehood will take root and be immensely difficult to dislodge, especially tying to libelous speech and journalists' outsized speech protections.

B. Evolution to Modern Protection

Whitney v. California (1927) paints a stark contrast to Abrams. In Whitney, the Court upheld a conviction for membership in the Communist Party. In his concurrence, Justice Louis Brandeis famously argued that “the remedy to be applied is more speech, not enforced silence,” paralleling Justice Holmes’ articulation of democracy grounded in open debate and reason [14]. However, by the time of the Cold War era, that ‘economic’ confidence gave way to fear. 

In Dennis v. United States (1951), the Court continued to uphold convictions of Communist leaders, reflecting a similar ruling that speech advocating the violent overthrow of government could be restricted if the danger was “probable” and “significant enough,” even without imminent threat [15]. The test was a refined evolution from Schenck’s “clear and present danger” to Dennis’ “grave and probable danger,” ushering in the era of “balancing tests,” in which liberty and security were weighed on uncertain scales, where, perhaps, the tense context of the Cold War prompted this approach of leaving no risk for government threat [16]. These balancing tests, however, introduced doctrinal instability, as courts were left to weigh abstract notions of ‘liberty’ against speculative threats to national security, producing outcomes that often hinged less on principled standards than on judicial intuitions shaped by fear and political pressure. Against the backdrop of the early Cold War––marked by anti-communist hysteria, loyalty investigations, and widespread suspicion of dissent––the Court became increasingly willing to restrict speech, revealing how Justice Holmes’ marketplace metaphor faltered in moments of perceived existential crisis, when the state’s fear of ideological subversion overshadowed confidence in open debate as a safeguard of democratic truth [17].

C. Culmination in Brandenburg v. Ohio

In Brandenburg v. Ohio (1969), the pendulum swung decisively back towards liberty. The Court established the modern standard for incitement: speech advocating violence is protected unless it is “directed to inciting or producing imminent lawless action and is likely to incite or produce such action” [18]. The decision fortified the First Amendment against government overreach, even in the event of vile and extremist expression. Still, the Court failed to, and could not entirely, anticipate today’s world, where a single tweet can mobilize thousands, and a fringe ideology can metastasize into mass violence months later. Brandenburg changed the standard set forward by Dennis and Whitney. The Brandenburg standard, anchored in imminence and physical proximity, shielding even hateful or extremist speech unless it triggers immediate violence, is insufficient in a digital era where incitement is global, algorithmic, and unlimited, and where online radicalization unfolds gradually, across platforms, and without the kind of temporal or spatial imminence the doctrine presupposes. The January 6th Capitol attack in 2021 starkly exposes this doctrinal gap, as months of online rhetoric that may not have satisfied Brandenburg’s requirement of imminence nonetheless culminated in coordinated lawless action [19]. Hence, hate speech is left largely outside Brandenburg’s reach and the Court is forced to manage it indirectly through other, narrower doctrinal categories.

IV. Hate Speech and the Democratic Paradox

A. Constitutional Protection of Hate Speech

The U.S. takes a distinctive stance amongst global democracies by refusing to outlaw hate speech. In R.A.V. v. City of St. Paul (1992), the Court struck down an ordinance banning cross-burning and hate symbols, holding that punishments based on the message’s viewpoint constituted unconstitutional discrimination [20]. In contrast, the Court in Virginia v. Black (2003) narrowed the circumstances under which religious discrimination could be prohibited. The Court clarified that cross burning can only face prohibition if there exists an intention to intimidate, thereby introducing the “true threat” doctrine [21]. Yet, in practice, U.S. states and localities still commonly propose hate-speech restrictions, even though First Amendment doctrine consistently pushes back on such laws as impermissible content or viewpoint-based discrimination [22]. These cases reaffirmed the principle that offense alone is not a constitutional basis for restriction. Nonetheless, they expose a troubling contradiction: the Constitution protects speech that undermines the equal dignity of those whom the First Amendment aims to empower.

B. Comparative and Theoretical Perspectives

Other democracies have taken a different approach to the question of hate speech. In Handyside v. United Kingdom (1976), the European Court of Human Rights famously declared that freedom of expression protects ideas that “offend, shock, or disturb,” though they still upheld restrictions on obscene materials [23]. Karl Popper’s principle was clear in this context: democracy requires tolerance, but not self-destruction. Similarly, the Supreme Court of Canada in R. v. Keegstra (1990) upheld hate speech laws that criminalized the willful promotion of hatred [24]. The reasoning was that such limits serve equality and social cohesion. Critics such as Owen Fiss also argue that hate speech does not merely offend, but actively distorts the democratic process by silencing marginalized voices, undermining the equal participation that theorists like Meiklejohn view as essential to a functioning democratic public sphere [25] [26].

Both courts recognize what American jurisprudence often fails to acknowledge: that unrestricted speech can undermine the very conditions necessary for meaningful discourse. This tension echoes Karl Popper’s “paradox of tolerance,” according to which unlimited tolerance may lead to the disappearance of tolerance itself [27]. Justice Holmes’ 'marketplace of ideas' may no longer suffice in a marketplace flooded with manipulation, amplification, and hate. In the U.S. setting, interpretations of freedom of expression are inadequate in a modern world, thus calling for concrete judicial reform as opposed to a more theoretical discussion of free speech. These foreign examples can be translated to the U.S. context, given First Amendment constraints, and the normative question asking if the U.S. should reconsider the Amendment in light of what other democracies are doing.

V. Free Speech in the Age of Social Media: The Brandenburg Problem

The digital age has transformed speech from a personal act into a networked phenomenon. A single post has the ability to be seen millions of times before its author even refreshes their page. The traditional safeguards of time, space, and intent––presupposed in the rationale behind the Brandenburg imminence requirement––no longer apply. In Packingham v. North Carolina (2017), the Court likened social media to a “modern public square,” underscoring the role and power of social media in contemporary discourse [28]. Yet the Court’s idealized vision of social media as an open democratic forum sits uneasily with the reality of digital spaces increasingly shaped by algorithmic amplification, harassment, and disinformation––conditions that undermine the very expressive freedom and civic equality the Packingham decision sought to protect. Although Packingham struck down a statute banning sex offenders from accessing online platforms, that very “public square” now appears to be dominated by algorithms that privilege outrage, vitriol, and polarization over substantive messages.

Eight years apart, the cases of Elonis v. United States (2015) and Counterman v. Colorado (2023) both grapple with the criterion of what constitutes a ‘true threat’ online [29] [30]. Both decisions acknowledge the blurred line between expression and intimidation in digital spaces, yet the Court only offers minimal guidance on the structural role of message amplification. The Brandenburg focus on imminence collapses when harm is delayed, collective, or data-driven. Today, incitement rarely occurs on street corners; it happens in online comment sections, encrypted chats, and online public forums. Thus, the law must evolve accordingly to reflect such shifts. A ‘foreseeable amplification’ standard––recognizing how speech can predictably metastasize through digital ecosystems––could preserve liberty while addressing twenty-first-century challenges.

Such a ‘foreseeable amplification’ standard would fill the doctrinal gap left by Brandenburg’s insistence on imminence, which is increasingly unwieldy in an environment where a post made today may mobilize violence days or months later through algorithmic ranking and networked diffusion. Drawing on algorithmic radicalization, networked harassment, and platform governance, this proposed standard would recognize that the predictable spread of harmful speech differs fundamentally from spontaneous incitement. Applied only to narrow categories, such as extremist advocacy or targeted hate campaigns, the standard would avoid sweeping First Amendment overreach. Nevertheless, critics may still argue that it risks chilling lawful expression, especially since the Court has never explicitly integrated digital amplification into its incitement jurisprudence, underscoring the need for judicial evolution attuned to technologically mediated harm.

VI. Reconciling Free Speech and Free Exercise in a Plural Democracy

Religious expression occupies a uniquely fraught space in First Amendment law. What starts as speech grounded in one’s personal conscience can become hateful conduct that denies others the ability to live (and practice) freely in the public sphere. In 303 Creative v. Elenis (2023), the Court sided with a designer who refused to create websites for same-sex weddings, ruling that compelling her to do so would violate her freedom of expression [31]. The decision blurred the boundary between free exercise and free speech, transforming religious belief into a right to freedom of speech that allows expressive conduct exempt from anti-discrimination laws and from being classified as hate speech. To reconcile these rights, courts must distinguish between speech as participation (expression that enriches democratic dialogue) and speech as exemption (expression that withdraws from shared obligations of equality). Where ‘speech as participation’ assumes an outward-facing contribution to public discourse, ‘speech as exemption’ functions inwardly as a claim to opt out of generally applicable civic duties. A balancing test based on the potential for harm could help restore coherence by asking the following questions:

I. Is the speech central to public discourse?

II. Does it materially impair others’ participation in that discourse?

III. Does it function as a shield for belief, or a sword against equality?

Applying this framework to 303 Creative, Brandenburg, and Counterman shows that, even though the First Amendment protects dissent, it does not protect the right to weaponize expression against others’ democratic belonging. This ‘democratic belonging’ refers to the baseline condition in which all individuals can enter public life as equals, able to both speak and be heard without exclusion or stigma. As for the balancing test’s risk of discretionary inconsistency, this framework must incorporate clear, predictable criteria, including the ‘foreseeable amplification’ inquiry, to recognize that even religiously motivated expressive conduct online can metastasize through digital networks in ways that threaten others’ equal participation. This amplification could thus fracture the very conditions of belonging the First Amendment is meant to secure.

VII. Conclusion

Ultimately, Justice Holmes’ 'marketplace of ideas' remains a powerful metaphor––but a marketplace requires fair access, good faith, and safeguards against domination. Today, the loudest voices can often be the most toxic, and algorithms commonly reward division over dialogue. The First Amendment’s promise to protect individual liberty and egalitarian discourse as fundamental for a functioning democracy cannot survive on formal neutrality alone. Courts and legislatures can take actionable steps to reconcile the right to free speech with the dangers of hate speech and uniquely modern potential for amplification. These include the adoption of a foreseeable‑amplification standard, and a clearer balancing framework for religious‑speech or exemption cases. Renewed jurisprudence should embrace pluralism, responsibility, and equality as democratic imperatives, not as threats to liberty. Protecting speech must also mean protecting the conditions under which speech remains meaningful and free. In an age of algorithmic amplification and ideological polarization increasing threats of hate speech, the true test of the First Amendment is not how much it tolerates, but how well it sustains democracy itself.

---

[1] Jeffrey W. Howard, Free Speech and Hate Speech | Annual Reviews (2019), https://www.annualreviews.org/content/journals/10.1146/annurev-polisci-051517-012343.

[2] Sunstein, Cass R., Free Speech Now Symposium: The Bill of Rights, University of Chicago Law Review 59 U. Chi. L. Rev. (1992), https://heinonline.org/HOL/Page?collection=journals&handle=hein.journals/uclr59&id=263&men_tab=srchresults 

[3] Evelyn Douek, Governing Online Speech: From “Posts-as-Trumps” to Proportionality and Probability, 121 Colum. L. Rev. 759 (2021).

[4] see [2]

[5] Maleiha Malik, Religious Freedom, Free Speech and Equality: Conflict or Cohesion?, 17 Res Publica 21 (2011).

[6] Haun, William J., A Standard for Salvation: Evaluating “Hybrid-Rights” Free-Exercise Claims, Catholic University Law Review (2011), https://scholarship.law.edu/cgi/viewcontent.cgi?article=1008&context=lawreview

[7] see [6]

[8] Gregory Emeka Chinweuba & Jerry C. Nwobodo, Tolerance and the Challenge of Insecurity in Nigerian Society: Perspective from Karl Popper’s Paradox of Tolerance, 5 Int’l J. Mgmt., Soc. Scis., Peace & Conflict Stud. (IJMSSPCS) 33 (Dec. 2022).

[9] Abrams v. United States, 250 U.S. 616 (1919).

[10] The Espionage Act Is Being Used Against Journalists. Here’s What You Can Do., National Press Club Journalism Inst. (Jul. 7, 2025), https://www.pressclubinstitute.org/2025/07/07/the-espionage-act-is-being-used-against-journalists-heres-what-you-can-do/.

[11] see [8]

[12] Schenck v. United States, 249 U.S. 47 (1919).

[13] see [8]

[14] Whitney v. California, 274 U.S. 357 (1927).

[15] Dennis v. United States, 341 U.S. 494 (1951).

[16] Hayden Thorne, Cold War Supreme Court: Analysing the Change in Constitutional Interpretation between Dennis v. U.S. (1951) and Yates v. U.S. (1957), Te Herenga Waka—Victoria University of Wellington (2018), https://doi.org/10.26686/wgtn.17072069

[17] see [16]

[18] Brandenburg v. Ohio, 395 U.S. 444 (1969).

[19] Juliet Dee, Charlottesville, January 6 and Incitement: Can Civil Conspiracy Laws Permit an End-Run around Brandenburg v. Ohio?, University of Delaware (2022), https://udspace.udel.edu/bitstreams/c1db77c3-d98d-4b73-81b0-75c83910631f/download 

[20] R.A.V. v. City of St. Paul, 505 U.S. 377 (1992).

[21] Virginia v. Black, 538 U.S. 343 (2003).

[22] Hate Speech and Hate Crime, AM. LIBR. ASS’N (Dec. 12, 2017), https://www.ala.org/advocacy/intfreedom/hate

[23] Handyside v. United Kingdom (1976).

[24] R. v. Keegstra, [1990] 3 S.C.R. 697 (Can.).

[25] Owen Fiss, The Irony of Free Speech, Harvard University Press (2009), https://books.google.com/books?hl=en&lr=&id=Q9aqkB4zZRMC&oi=fnd&pg=PA1&dq=Owen+Fiss+hate+speech&ots=0gDIfkH8Gb&sig=OPGdlDc6HIZyfVxnm4SMnWXSWvw

[26] Alexander Meiklejohn, Free Speech and Its Relation to Self-Government, Harper & Brothers (1948)

[27] see [7]

[28] Packingham v. North Carolina, 137 S. Ct. 1730, 1735 (2017).

[29] Elonis v. United States, 575 U.S. 723 (2015).

[30] Counterman v. Colorado, 600 U.S. ___, 143 S. Ct. 2106 (2023).

[31] 303 Creative v. Elenis, 600 U.S. ___, 143 S. Ct. 2298 (2023).