Democratic self-defense and public sphere institutions

Contemporary concerns with democratic backsliding and contestation of democratic institutions, even in consolidated democracies, have reignited longstanding debates on how democratic societies should respond to perceived anti-democratic threats and what a principled “democratic self-defense” should look like (Kirshner, 2014; Müller, 2016; Malkopoulou & Norman, 2018; cf. Loewenstein, 1937). The core dilemma in these debates concerns the extent to which restrictions on anti-democratic speech, actors, and their associations can be justified in the interest of protecting the integrity of democratic institutions and strengthening democracy's guardrails.

Variations of this dilemma, traditionally concerned with the protection of democratic institutions, have increasingly come to the fore in other arenas of democratic societies. Public sphere institutions such as schools, universities, and public broadcasting organizations, as well as social media platforms have become deeply entangled in discussions on the limits of speech and political action. These institutions are expected, either by convention or legislation, to uphold and reproduce core liberal democratic values while also remaining open to a plurality of views, allowing for the free formation and expression of political ideas. Yet, the existing literature has had less to say about what values should guide decisions to restrict or call out speech deemed to challenge liberal democratic norms in the context of these public sphere institutions.

Our concern in this article is to clearly flesh out what core dilemmas of democratic self-defense in the public sphere consist of and theorize the democratic values at stake in this context. Seeing human dignity as a fundamental value for liberal democracy, we argue, helps us to more precisely identify the character of democratic threats in the public sphere, the various ways in which democratic values may be undermined, and in light of that, how public sphere institutions may respond to these challenges.

Crucially, the assumption that human dignity is a basic democratic value allows us to identify how legally protected speech can still be highly problematic from a democratic perspective. This is important, we argue, as many of the challenges to liberal democracy involve individual-level harms, instances where the human dignity of individual people is undermined. Key to this argument is theorizing the link between attacks on the equal dignity of citizens and attacks on democracy. We tie human dignity as a democratic value to the respect and status afforded to individuals as members of a political community. Paying attention to this link in the context of democracy helps highlight characteristics of speech that have not received sustained attention in current discussions on militant democracy and democratic self-defense.

Our argument emphasizes that some members of democratic society are more at risk than others by virtue of the fact that they already occupy precarious positions in terms of recognition of equal standing. Incorporating this insight should have repercussions for the strategies devised to combat anti-democratic threats at the “micro level” in democracies. This, finally, allows us to think anew regarding the strategies through which the dilemma can be approached by democratic actors. We consider both exclusionary and expressive strategies and discuss how they involve distinct trade-offs and dilemmas.

The article proceeds by first clarifying the theoretical starting point regarding anti-democratic challenges in the public sphere. We contrast our view with prevalent perspectives in democratic theory that have focused on democracy's protection. Next, we outline the significance of human dignity as a democratic value and define key ways in which dignity might be undermined in the public sphere. We illustrate our argument with discussions on dilemmas facing public broadcasting organizations, social media platforms and educational institutions. The final section outlines the menu of strategies available in pushing back anti-democratic threats in the public sphere, including both exclusionary and expressive actions. We present an argument in favor of expressive strategies while highlighting instances where both strategies may be employed, as well as instances where public sphere institutions should remain passive.

Our starting point is that contemporary threats to democracy require greater attention to the dilemmas of democratic self-defense in the public sphere. The public sphere can be defined as the “constellations of communicative spaces in society that permit the circulation of information, ideas and debates” (Dahlgren, 2005). It is composed by formal and informal institutions including schools and universities (Holmwood, 2017), public media institutions (Iosifidis, 2011), museums (Barrett, 2011) and increasingly, social media platforms (cf. Müller, 2019). These are settings of key importance for a functioning democracy where individuals should be able to participate and deliberate about public issues on an equal basis. Our argument does not assume that the public sphere is uniquely concerned with “common interests” (Taylor, 1992), or that it is necessarily “representative” and “inclusive” (Fraser, 1990). Indeed, the public sphere can also provide oxygen for anti-democratic and extremist actors, and its institutions can become vehicles of autocratization (Arato & Cohen, 2021). For these reasons, teachers, journalists, editors, their boards and directors, are forced to make often difficult decisions on where the limits should be drawn regarding which actors and which views should be given room. There are thus good reasons to theorize how public sphere institutions should approach dilemmas related to where and how to draw limits with reference to what is tolerable in a democracy. These are dilemmas that parallel those theorized in the growing literature on democratic self-defense, but they take on a different guise when we consider public sphere institutions (cf. Müller, 2016; Malkopoulou & Kirshner, 2019).

In the wake of increasing digitalization, the rise of social media platforms and the decline of traditional print and broadcast media, there is a strong tendency toward a “breakdown of gatekeeping” (Farrell & Schwartzberg, 2021: 222). This also constitutes a shift whereby nonstate institutions in the public sphere increasingly have to undertake contentious and sometimes highly consequential decisions about the inclusion and exclusion of particular actions and messages, even in cases where they fall well within what is legal (Chambers & Kopstein, 2022). Decisions by some of the leading social media platforms, most notably what was then called Twitter, to suspend the account of then US president, Donald Trump, are illustrative (Twitter, 2021a). The decision to ban Trump from Twitter, or indeed the decision to invite him back on the platform, was not taken by a public authority with reference to the breach of any laws. It was a decision taken by a private corporation with reference to its own policy against the glorification of violence (Twitter, 2021b).

The dilemma of democratic self-defense does not only apply to social media platforms; it is ubiquitous in the public sphere of contemporary democratic societies. Decisions to suspend participants, censor users, or to flag or limit the distribution of content, as well as decisions to not take actions and include democratically questionable views, are regularly made by media outlets, schools, museums, universities and other institutions that provide arenas for public debate. The dilemmas that come to the fore in such settings increasingly permeate the every-day practices of democracy. However, they are clearly distinct from the dilemmas faced by the democratic state in relation to anti-democratic political parties and organizations.

Public sphere institutions are neither judicial nor legislative institutions; they are part of what Rawls termed as the “background culture”. In a free and democratic society, the background culture is the arena for free associational life that should largely remain unregulated by law (Rawls, 1997). The public sphere thus provides ample occasions for the dilemma of democratic self-defense to materialize but where legislation does not, indeed should not, supply clear guidance. It is a dilemma that has less to do with the integrity of electoral institutions and procedures of political democracy and more with the extent to which the members of democratic societies are respected as equals. This comes with implications for how we understand how anti-democratic threats play out in these settings.

For the purposes of our argument, it is particularly important to challenge the assumption that such threats affect all citizens in a democracy equally. Rather, we assume that the burdens of anti-democratic rhetoric and actions are carried to a larger extent by some groups than by others, that they are unequally distributed. The increasing influence of political movements identified as potential threats to democracy, such as political parties with roots in the extreme right, are often theorized to pose an equally serious problem for all democratically minded citizens in a society. We agree that actual authoritarian governments do undermine the dignity of all their citizens in a fundamental sense. However, the belief that all citizens in an otherwise functioning democracy are equally affected by anti-democratic rhetoric obscures the fact that the key mechanism through which authoritarian politics gains traction in a democracy is by targeting often already marginalized groups (Tudor & Slater, 2020). Anti-democratic movements are never “just” anti-democratic in the sense of advocating for authoritarian forms of government. In general, these movements also target particular social, ethnic, linguistic groups in society, framing them in antagonistic terms, as enemies of the people, the state or the party. Anti-democratic movements often exploit pre-existing asymmetries in power and status in society.

In contemporary democracies, anti-democratic movements typically engage in continued rhetorical assaults on the status and rights of, in particular, immigrants, religious communities, women, members of the LBGTQ-community, indigenous peoples and minorities. Frustration, impatience, blame and indeed hatred directed at these groups is a fixture of the authoritarian playbook (Corrales & Kiryk, 2023; Lührmann et al., 2021; Gricius, 2022). Our argument builds on the premise that the primary target for contemporary perceived challengers of democracy, including the plethora of authoritarian populist parties in Europe, is the public status of specific groups and their individual members, rather than the integrity of democratic institutions and procedures as such. This observation does not exclude the possibility that, when in power, these parties would work to erode those institutions. Examples of this from the last decade are readily available in states like Hungary, Poland, and the United States.

However, beyond the most extreme fringe parties, such as neo-Nazis or revolutionary communists, parties marketing themselves in direct opposition to democracy as a form of government are rare. Targeting members of specific societal groups by identifying them as illegitimate or unwanted members of the political community is from this perspective a hallmark of contemporary anti-democratic politics. Anti-democratic movements in a democracy, while they may not attack democratic institutions as such, regularly work to denigrate the equal public status of some groups. This, we argue, is an affront to the human dignity of these individuals and as such in conflict with fundamental values in a democratic society. The type of anti-democratic threats that are our primary concern in this article thus start from the particular rather than the general; from targeted disdain, rather than from competing principles of government. Therefore, recognizing the unequal distribution of costs for assaults on democracy should inform attempts to theorize the dilemma of democratic self-defense. By contrast, public speech in a democracy that promotes anti-democratic forms of government, even if deemed offensive, is arguably less likely to have harmful effects in terms of undermining the dignity of particular citizens. Advocates of, for instance, military or expert rule, while clearly communicating support for anti-democratic principles, need not necessarily target the standing of particular individuals. Furthermore, the fact that we suspect that, given government power, critics of democracy would bring society in an authoritarian direction does not necessarily imply strong reasons for their blanket exclusion from public sphere arenas.

However, the contention that the first casualties of anti-democratic action are more likely to be members of particular groups prompts us to reconsider dominant themes in longstanding discussions on how to defend democracy against anti-democratic challengers. The theoretical tools developed in this literature are dominated by a concern with the preservation of democratic institutions and procedures, including the integrity of elections, basic political rights, and the balance of power between the executive, the legislature, and the judiciary (Abts & Rummens, 2010). These are concerns which we share. Questions concerning to what extent democracies should formally tolerate extremist and anti-democratic actors that rally under anti-democratic political programs or if such political actors should be confronted with repressive measures to protect democratic institutions remain important (Gutmann & Voigt, 2021; Kirshner, 2014; Malkopoulou & Kirshner, 2019; Müller, 2016; Niesen, 2002; Rijpkema, 2018, 2019; Sajo, 2004; cf. Kelsen, 2006; Vinx, 2020). However, the predominant way anti-democratic threats are conceptualized in this literature limits its ability to accurately capture dilemmas and challenges of democratic self-defense in the public sphere.

The shadow of Karl Loewenstein—credited with coining the concept of “militant democracy”—looms large in its basic conceptualization of the dilemma of democratic self-defense (Loewenstein, 1937). It is a perspective ultimately concerned with protecting the integrity of core democratic institutions and its decision-making procedures. The fundamental “Löwensteinian” concern is that democratic procedures are vulnerable and easily exploited for anti-democratic ends. References to the fall of the Weimar Republic are ubiquitous in the academic literature, and increasingly in contemporary public debate, serving as the paradigmatic example of a democratic system that lacked adequate constitutional mechanisms to protect itself against anti-democratic attacks (Abts & Rummens, 2010; Müller, 2016; Kirshner, 2014; Quong, 2004; Rijpkema, 2018; Vinx, 2020).² Instead, the argument goes, anti-democrats were able to dismantle democracy “from within”.

A core idea is that the inherent openness of democratic institutions makes pre-emptive action permissible and sometimes necessary against political movements that can be suspected of subverting those institutions in the future. As Müller states, “a militant democracy does not wait until its enemies have gained majorities at the poll; it seeks to nip fundamental opposition to democracy in the bud” (Müller, 2016: 250). Although conventionally less pronounced in a US constitutional tradition, it has been prevalent in post-war democratic constitutional debates in many European countries (Bourne, 2018; Fox & Nolte, 1995; Norman, 2022; Suteu, 2021). Theorists of militant democracy are careful to distance themselves from many of the measures defended by Karl Löwenstein (Kirshner, 2014; Müller, 2016), and others have argued that the notion of a militant democracy is inherently tied to elitist and anti-participatory logics (Invernizzi Accetti & Zuckerman, 2017; Malkopoulou & Norman, 2018). It is, moreover, increasingly recognized that contemporary anti-democrats rarely voice explicitly authoritarian or totalitarian political ideals. Rather, the political platforms of authoritarian populists in Europe, the United States, and Latin America, with some clear exceptions, tend to occupy the grey area between democratic and authoritarian politics (Malkopoulou & Moffitt, 2023). The point is that contemporary anti-democratic movements may fall well outside the scope of traditional militant measures such as restrictions on assembly, political organization and party funding that have dominated discussions on militant democracy (Rovira Kaltwasser, 2019). This holds even if we observe how populist parties similar to those emerging in many consolidated democracies are instigating anti-democratic political developments elsewhere, where they may have gained government power. These observations further underpin the need to theorize the principles on which institutions in the public sphere should base their response to speech and agents perceived to challenge democratic values.

The answer, from our perspective, is not to advocate for more intrusive and repressive legal restrictions in the public sphere. Rather, we aim to highlight which values are at stake when public sphere institutions consider limits to public discourse. This leads us to theorize the nature of the threat to democracy in these settings. If, as we argued above, we are not primarily concerned with actions and forms of speech regulated by law, what is the basis for the dilemma and why is it a concern for democratic theory? How is democracy imperiled by systematic rhetorical attacks on the social and political status of particular groups? We propose that a focus on human dignity as a democratic value allows us to cut to the core of anti-democratic rhetoric and action. Indeed, these are values explicitly recognized by many militant democrats, including Loewenstein (1937, p. 658). Democratic institutions and procedures are from this perspective ultimately guarantors of human dignity, and this value should inform the protection of these institutions. Yet, we believe that the predominant focus on democratic institutions and procedures, specifically those related to electoral politics, tends to downplay both individual level and aggregated consequences of speech in the public sphere. For those reasons existing perspectives seem less well-suited to identify and address current anti-democratic challenges. By extending the analysis to public sphere institutions and by invoking the importance of human dignity to democracy, our approach offers the groundwork for a clearer discussion on the dilemmas that arise in efforts to defend democracy against such threats.

There is a dual relationship between the value of human dignity and democratic ideals. On the one hand, human dignity is the background norm—an abstract principle—that is frequently perceived as a justification for democratic rights of participation. On the other, human dignity signifies a particular social status—or social practice—that is arguably instrumental to active participation and the use of democratic rights. We discuss both these aspects of dignity as a vehicle to identify threats to democracy in the public sphere.

As a bedrock for normative claims and a social status worthy of protection, human dignity is bound up with a broader catalogue of liberal rights. Human dignity is typically incorporated as part of this catalogue and appears in written constitutions to differentiate the democratic political order from past authoritarian experiences. Human dignity emerged as the legal principle that provides the foundation for democratic and constitutional orders in post-war Europe (Dupré, 2013). A similar pattern is visible in Latin America after the fall of the juntas and in Eastern Europe after the fall of Communist regimes (Barak, 2015).³

The principle of human dignity provides a justification for democratic institutions. This is, for instance, illustrated by the fact that it figures among the founding values of the Treaty on European Union (Article 2) and that it plays a key role in the 1949 constitution of Germany. It also undergirds important decisions in the jurisprudence of the US Supreme Court that has treated “dignity as a value underlying, or giving meaning to, existing constitutional rights and guarantees” (Goodman, 2005, p. 743; cf. Kahn, 2015; Simon, 2021; Tremper, 1988). Given that human dignity is widely recognized as a key value for democracy it might appear surprising that it has not been given more room in the literature on democratic self-defense.⁴ To an extent, we believe that this has to do with the concept's pliability in terms of the different meanings to which it is associated in different literatures.

In fact, the relationship between human dignity and democracy remains contested (Baer, 2009; Rosen, 2012; Valentini, 2017; Waldron, 2012). There are conceptions of democracy where conceptions of human dignity play a limited, if any, role. For instance, justifications for democratic procedures as means for competitive but peaceful resolution of political conflict scarcely need appeal to values of equal dignity (Bobbio, 1984; Popper, 1945; Przeworski, 1999; Schumpeter [1943] 2003). This does not mean that such “minimalist” perspectives on democracy would necessarily deny that dignity is an important concern for human beings, only that it does not play into core justifications for democracy as a form of government. From a minimalist democratic perspective, even nondemocratic regimes have often done “very well in securing what most of us believe the democratic method should secure” including human dignity (Schumpeter [1943] 2003, p. 246). A related and common objection to the concept is that human dignity fails to offer any clear normative direction at all (Seglow, 2016). Others argue that human dignity may be more useful in other social and ethical contexts, for instance, as the basis for the protection of bodily integrity, privacy, or the right to life (Beyleveld & Brownsword, 1993).

We think, however, that the idea of human dignity as an important background value for democracy is useful to ground a discussion on democratic self-defense in the public sphere. Though human dignity refers to the inner human worth of individual beings, we shall here treat it as more or less coterminous with equal moral status and the moral claim that this status ought to be recognized by public institutions. Our point here is that a commitment to the principle of human dignity is fundamental not merely for liberals, but also for democrats (cf. Dworkin, 2011).

To further specify the relevance of the principle of human dignity for discussions on democratic self-defense, we highlight two aspects. First, human dignity signifies a social status that is instrumental to democratic participation (Ober, 2012). Human dignity is hence understood to require public recognition of the person as an equal member of society. A person who is publicly recognized as an equal is recognized as a person with human dignity. This overlaps with what is elsewhere termed social equality and rejects “hierarchies of social status” and “stigmatizing differences in status” (Fourie, 2012). The respect required by human dignity is in this sense more demanding than protections of political equality. Arguably, the status of “public equality” is among the great benefits conferred to all members of a democratic society by virtue of their standing as political equals (Christiano, 2009; Gonzalez-Ricoy & Queralt, 2018). The secure enjoyment of the social status associated with recognition of human dignity plays an important role in making available venues for participation. More specifically, related to the functioning of a democratic public sphere, it is a precondition for dialogue in public affairs for all members of the community.

Human dignity is, second, a basic claim about moral status that grounds a wide variety of rights, including rights to democratic participation (Rosen, 2012).⁵ The claim that all individuals share the same moral status, because of their human dignity, is intimately bound up with the justification of democracy as it allows us to regard the equal enjoyment and exercise of political rights as the full expression of our status as inviolable and dignified human beings (Baer, 2009; Dworkin, 2006). In the context of extremist threat to democracy, we thus believe the most illuminating aspect of dignity is that of a particular public status. This is consistent with the notion that it is a background value that makes democratic politics possible. The core meaning of human dignity is—we take it—that all members of the democratic community should be publicly affirmed as individuals of equal standing.

Human dignity as equal public status in a community is similar to the recognition of “the right to have rights” (Arendt, 1951; Hann, 2016; Isaac, 1996). It avoids theoretical discussions on the contents of specific rights such as the right to freedom or the right to life and instead seeks to establish the fundamental condition for such rights, however they are defined, to be respected. Respect for dignity in this perspective serves a meta-function related to the recognition of full membership in a political community that is necessary for more specific rights to be respected, including democratic ones. Our thesis here is that dignity is a social status that depends on—and is therefore also vulnerable to—practices in the wider public sphere. To protect that status, theories of democratic self-defense need to ground themselves in a perspective that fleshes out what constitutes harms to equal dignity.

To concretize how the erosion of human dignity in the public sphere of democratic societies can manifest itself, we invoke a distinction from social-psychology and bioethics between two main ways in which speech may be contrary to the equal public status of others: delegitimization and dehumanization. These two concepts depict degrees to which the public standing of a person can be questioned or undermined by speech. They are related to the well-known category of “hate speech” that predominates in legal discourse. We believe, however, that the concepts proposed here open up for a more graded and nuanced discussion on which types of speech might work to undermine dignity, even when such speech falls within what is legally permitted.

Delegitimization is defined as efforts to fundamentally undermine an individual's status as legitimate member of a political community by reference to their association with a particular societal group and attributing deeply negative characteristics as intrinsic traits to that group (cf. Bar-Tal, 1989). Delegitimization thus takes aim at individuals as members of groups rather than as representatives of particular political positions and it is characterized by implicit or explicit attempts to undermine the status of others as equal rights-holders and as equal contenders for political influence. Delegitimizing speech can be understood as intimately tied to exclusionary notions of “The People” and associated to understandings of the people's interests in fixed and holistic terms (Ochoa Espejo, 2017; Urbinati, 2019; Wolkenstein, 2019). It includes statements that negate the equal public standing of individuals in that particular community. Yet, not every claim that seeks to delegitimize the equal public status of another person is necessarily an affront to the democratic value of human dignity.⁶ Moreover, as outlined above, human dignity can also be undermined by measures that do not attack the equal public status of individuals as citizens.

Claims that some citizens be excluded from the moral, political or legal community may nevertheless constitute an attack on their equal public status and as such pave the way for infringements of their democratic rights as well as of other fundamental rights. The democratic significance of this point trades on an insight in the previous literature, which is that recognized membership in a political community is a precondition for secure enjoyment of basic political rights. As assaults on equal public status are instrumental in denying particular citizens their status as members of the political community, assaults on dignity also work to undermine basic political rights generally. This is what makes speech that attacks the public status of particular groups so insidious, and why it is inherently a concern for democracy.

Dehumanization is the category of speech with the most obvious and far-reaching implications in terms its potential to undermine the equal public status that is necessary for respecting the human dignity of all citizens. Delegitimization, deployed as a political strategy, may come with serious consequences, especially in terms of paving the way for the infringement of other fundamental rights of individuals. Dehumanization, however, cuts to the core of such infringements. The common denominator of speech that dehumanizes others is that it seeks to exclude them from membership in humanity by portraying them as “animal-like” or otherwise devoid of the traits that are uniquely human (Haslam, 2006). It is thus more far-reaching in that it does not limit itself to questions of individual's legitimate right to participate in a particular community but works to undermine the status of individuals as human beings altogether. Historically, dehumanizing speech has been deployed to establish the inferiority of members of ethnic, linguistic, religious, and other social groups and has been part and parcel of violent persecution, slavery and genocide. In a US context, the historical legacy of slavery is illustrative for our argument, where recognition of human dignity was withheld on the basis of radical exclusion of certain groups from the category of humanity. The far-reaching denial of human dignity in this context is precisely evidenced not merely by the denial of slaves and their descendants as members of the political “people” but by the denial of them as members of the category of people tout court (Canovan, 2005). Here, again, the problem was not that those in power were unable to conceptualize what particular rights entailed but that those rights emerged in a political context which rendered them null and void when considered in relation to parts of the population.

The underlying logic of speech that seeks to undermine the human dignity of others is instructive to the aim of theorizing how democratic values are at risk in the public sphere of democratic societies. Recent experimental studies in psychology indicate how dehumanization is associated with a permissive attitude toward harm with respect to targeted individuals and groups (Dalsklev & Ronningsdalen Kunst, 2015; Markowitz & Slovic, 2020; cf. Over, 2021). The same attitudes are reinforced by, for instance, the depiction of undervalued minority groups as unhygienic, which has been shown to trigger feelings of disgust (Dalsklev & Ronningsdalen Kunst, 2015). Dehumanization comes close to what would in many cases count as hate speech, which is recognized as an offense by criminal law in many countries. However, as the examples above illustrate, speech that is not necessarily hate speech can still provide subtle cues instrumental for undermining human dignity. Like explicit anti-democratic positions, assaults on the dignity of other members are a form of extremism that prompts a response from democratic institutions also in the public sphere.

Once dignity is identified as a value to be preserved in the face of anti-democratic challenges, the sphere of concern expands to include institutions and practices in the public sphere generally, and not just the electoral arena. Here, the dilemma consists in how to safeguard democratic and inclusive arenas of opinion-formation while at the same time protect human dignity from the corrosive effects of delegitimizing or dehumanizing speech.

For the purpose of illustration, consider the cases of public media institutions, public schools and social media platforms. Public service broadcasting organizations are expected to adhere to ideals of inclusive representation in their coverage of politics. Truthful, impartial and objective reporting on political actors, public policy and law is a key role to be played by the public service in a liberal democracy (Goidel et al, 2017; Newton, 2016; Norris, 2000). This is particularly the case in Europe where state funded public broadcasting organizations still hold important positions as public sphere institutions. There are strong and compelling reasons for these organizations to be as inclusive and pluralistic as possible. To that extent, public service media institutions should include political voices of any kind, even, it could be argued, those of actors engaging in delegitimizing speech. However, in order to protect the dignity of all citizens, there are good reasons for why public media institutions should take measures to protect both participants and audiences from attacks on their status as equal members of society. Thus, the best policy, all things considered, remains uncertain. Public broadcasting organizations often struggle to translate a principled commitment to democratic principles into a defensible response to speech deemed democratically problematic. Heated debates, in many European countries not least, have revolved around perceived biases of public broadcasting organizations on the one hand and, on the other, problems related to providing room for, and thus legitimizing, parties deemed by some to threaten democracy (Hien & Norman, 2023). The dilemma of democratic self-preservation facing public media institutions is that democratic values are potentially threatened whether speech that undermines the dignity of some is suppressed or not.

Similar conditions apply to public schools. Public schools in many countries organize mock elections as part of their civic education, often in election years, to foster citizens’ engagement in politics (Ohrvall & Oskarsson, 2020). By being inclusive with respect to political ideologies, schools can serve as a training ground for critical thinking and public deliberation. However, when radical political parties are invited, there is a risk that they will undermine the sense of equal worth and dignity among groups of students, again considering that such potential harms can be expected to be distributed unequally among the members of such groups. In order for public schools to take responsibility for democratic values they may need to take measures that protect students for particular egregious forms of political speech. The example of schools is instructive precisely because it clearly highlights the responsibility of the institution, especially as the audience in this setting are not fully formed members of the political community. Again, here emerges the dilemma of democratic self-preservation: there are democratic reasons for both including representatives of all political parties and for excluding at least some of them. Recently, in some contexts, for instance, in Scandinavian countries, schools have addressed the problem by discontinuing the practice of inviting political representatives. Although this might be a reasonable course of action it is also a solution that clearly limits the functioning of that part of the public sphere. A strategy of retreat replaces dealing head on with the difficult dilemma relating to the limits of permissible speech in political debate.

We mentioned social media platforms above and they exemplify clearly that this dilemma is not confined to institutions formally tied to the state. Although they are private corporations, dominant platforms have gained a prominent position as arenas for public debate. Exclusion from access to social media platforms may in some cases have more far-reaching consequences than exclusion from more conventional public sphere institutions, such as public service media outlets. In the United States, this came to a head with the post-election exclusion from these platforms by former president Donald Trump. The fact that platforms like Facebook, X, and YouTube have gained such prominent positions as vehicles of contemporary political debate has come with expectations that these companies should be “unbiased” in terms of the contents they allow. Due to these expectations and similar to public sphere institutions, dilemmas arise due to conflicting demands. On the one hand, there is the need to safeguard openness, inclusion, and pluralism, while on the other, there is a requirement to ensure safety, authenticity of published material, and in the case of Facebook's oversight board charter, to protect dignity of individuals (Facebook, 2021).⁷

To acknowledge the existence of dilemmas of democratic self-defense in the broader public sphere is to appreciate the need for a more graded and nuanced perspective than has often been the case in debates about democratic self-defense. This is particularly true when threats to equal dignity take the form of actions or speech that seek to either delegitimize or dehumanize specific members in society.

We consider two types of strategies for protecting the human dignity of participants in public sphere institutions from delegitimization and dehumanization: exclusion and expressive action. These strategies build on long-standing discussions on democratic self-defense, which have mainly focused on when it is legitimate for the state to respond with repressive measures. When transposing such considerations to public sphere institutions we think it is necessary to also discuss instances when these institutions should deviate from their role as impartial facilitators of public debate and use their expressive capabilities to push back at certain statements. Exclusion includes various degrees of denying individuals access to institutions as well as limiting the distribution of content and various forms of speech. Expressive action refers to how an institution may communicate how content and speech run contrary to specific democratic values. Of the two, exclusion from public sphere institutions tends to be more problematic. Excluding perceived anti-democratic agents risks undermining the respect required by democratic values for their equal participation in the public sphere. On a systemic level, it further risks introducing anti-participatory and elitist practices into the public sphere, similar to those highlighted by critics of the militant democratic tradition (Invernizzi Accetti & Zuckerman, 2017; Malkopoulou & Norman, 2018; Norman, 2021; Stahl & Popp-Madsen, 2022). However, expressive action, whether in the form of public endorsement or public condemnation of particular views, is not uncomplicated. It can have highly significant consequences for individuals and potentially stifle a climate of open debate.

In this context, various strategies analogous to those implied by classical dilemmas of democratic self-defense resurface, albeit in a new form. The dilemma revolves around the requirements for public sphere institutions to meet criteria of pluralism and promote the free exchange of ideas conducive to the formation of opinions, all while protecting the dignity of all individuals. Furthermore, the dilemma arises due to the potential costs incurred by individuals whose views are publicly denounced or excluded by an institution. When and to what extent is there a democratic justification for public sphere institutions to exclude political representatives, private citizens or specific content? When and to what extent should institutions in the public sphere engage in expressive action, by actively challenging or flagging speech that is perceived to undermine or attack the dignity of other citizens?

An expressive approach is analogous to Brettschneider's (2016) claim that the state should rely on its expressive powers to condemn “hateful viewpoints” by articulating the reasons and values that undergird democratic rights. However, as is regularly the case in the media, including social media, populist and far-right rhetoric does not necessarily meet the criteria for hate speech, even when it directly targets the public status of specific groups.

An alternative approach, therefore, is for public sphere institutions—schools, the public media, universities, museums, and social media platforms—to articulate the deeper justifications and reasons for their commitment to democratic values. In other words, they should publicly affirm the human dignity of all without denying anyone the democratic right to participate, speak, and be included in the public sphere. The basic idea is that respect for human dignity provides strong reasons for institutions in the public sphere to challenge content that seeks to delegitimize or dehumanize members of society. This includes a variety of possibilities, from discursive engagement to distancing, as acknowledged in the growing literature on counter-speech (e.g., Fumagalli, 2020). The expressive strategy can also be pursued by means of “flagging” content that is considered suspect, unreliable or contrary to democratic values. Flagging is a well-established practice by social media providers, executed either manually, by scripts or self-learning algorithms. Though flagging is clearly a soft sanction, evidence suggests that it is effective in slowing the spread of problematic content (Mena, 2020). Flagging is a measure that has been introduced, to a great degree, to cope with the sheer scale of content on social media apps and often contains little more than a reference to the platform's user agreement. Other institutions, such as universities or public broadcasting organizations, should be expected to provide elaborate justifications for flagging and other responses.

A benefit of the expressive approach is that it remains consistent with permissive norms of freedom of speech and freedom of association that also extend to anti-democratic rhetoric and actors. Expressive action is a form of reason-giving and is premised on the value of open debate. At the same time, it may also, to a degree, offset the often-unequal burden carried by those who are the explicit targets of efforts to delegitimize and dehumanize. The expressive strategy thus strikes a balance by offering a measure of protection while also trusting individuals, even those who might be targeted by efforts to undermine human dignity, to possess both the resilience and the agency to creatively engage in counterstrategies.

A commitment to a public sphere characterized by wide-ranging pluralism calls for tolerance and, therefore, for participants to be prepared to accept considerable discomfort or even offence. At a fundamental level it relies on a view of democracy that places trust in individuals to withstand offensive speech and to recognize the value of a permissive attitude toward speech as fundamental for democracy. We believe that norms against delegitimization and dehumanization can help distinguish speech that would necessitate an institutional response from speech that might be seen as offensive, uncivil or disrespectful but should not trigger such a response.

Far-reaching threats to the dignity of the members of society sometimes require exclusionary measures. The problems with exclusionary strategies are still numerous and mirror many of the concerns that might be raised against repressive responses, such as party bans in conventional notions of militant democracy. However, here it is also important to differentiate between various forms of exclusionary practices, each of which is associated to different problems that need to be mitigated in various ways and weighed against their possible gains.

Legal bans, such as those directed at anti-democratic parties in some democracies, are obviously beyond the reach for institutions in the public sphere. Yet, these institutions have the power to design their own formal or informal norms that exclude either actors or content in so far as they are consistent with the law. When implementing exclusionary measures aimed at content, private individuals, or organizations, there are well-known risks associated with infringements on the rights of those actors. If public sphere institutions are expected to reproduce norms of inclusiveness, impartiality, and openness and to safeguard individual rights such as freedom of speech, exclusionary practices become problematic. Apart from the democratic harms imposed on individuals there are systemic risks associated with the implementation of exclusionary measures. Chief among these might be that they short-circuit an important component of a vital public sphere, namely a principled level of trust extended to all members of society to process and respond to any viewpoints expressed in that sphere. A public sphere built on the notion of individuals’ inability to do so not only risks stifling public debate. It also comes with problematic implications in terms of infusing the public sphere with a paternalistic logic that goes against the grain of the democratic principle of self-government. Coercive measures, such as banning actors from participation and removing certain kinds of content should then be considered with great care before they are implemented.

To assess if exclusionary measures are justifiable from a democratic perspective, it is useful to think about exclusion in more differentiated terms than is conventionally the case. Here, considering the type of actor targeted by exclusionary measures, such as private individuals or organizations, might lead to different conclusions. To exclude representatives of groups, organizations or parties if they attack or undermine the dignity of other citizens, from spreading information in schools, public places, and social media platforms is not unproblematic. However, targeting actors in their capacity as private citizens, not only as representatives for political organizations, is more serious in terms of the risk of undermining the respect for human dignity of such individuals. As Kirshner (2014) has argued, the political interests and identities of anti-democrats, racists, and extremists are not solely determined by their hostility to democratic institutions and values. As private citizens, they may also have distinct political interests that deserve protection and inclusion in processes of collective will-formation. Blanket exclusions of private citizens from institutions in the public sphere thus represent a greater infringement of democratic norms than exclusions targeting representatives of organizations. Damages incurred by excluding representatives from particular settings also have broader implications as it excludes from discussion the views of a broader set of actors. Yet another aspect that brings nuance to the discussion on exclusionary measures is to consider time-limitations on sanctions rather than open-ended or indefinite ones. Such considerations were part of the stated reasons for the recent reinstatement of Donald Trump's Facebook and Instagram accounts by Meta (Meta, 2023).⁸

Exclusionary strategies can also be directed toward specific content or forms of speech. Again, militant democrats typically endorse legal bans on speech that attacks democratic institutions or elected representatives. Nazi ideology and propaganda have repeatedly been struck down by the European Court of Human Rights and many European constitutions either permit or require restrictions on “abuses” of freedom of speech that threaten the democratic system (Tsoumidis, 2021). In the context of social media platforms this corresponds to the removal of content rather than “deplatforming”. One problem with the removal of content, often referred to as content moderation, is that the myriad of such decisions that are taken every day by the main social media platforms tend to be nontransparent and provide few means for appeal. Content moderation is thus facing specific challenges of democratic legitimacy. These challenges can be mitigated by expressive actions. The responsibility on the part of public sphere institutions to provide reasons for exclusion and to provide individuals with possibilities for appeal is important. Expressive action is thus a necessary aspect of any effort to implement exclusionary measures. Moreover, exclusionary measures have in themselves an expressive aspect that can be used to signal the limits of publicly acceptable political action to actors beyond those directly targeted.

Like all general principles, decisions on the application of the strategies outlined here need to take into account case-specific circumstances. Of key importance for such decisions is the historical legacy of power asymmetries and domination between different groups in particular societies. Historical legacies shape and give substance to what delegitimization and dehumanization is, and how the dignity of individuals is harmed by them, and must therefore inform the strategy that best serves to protect democratic values. Understanding such legacies will provide insights into the uneven distribution of threats to human dignity. Although the specific resolution of the dilemma we have discussed depends on the context in which it arises, we believe that the general conceptual tools developed here are essential for fully recognizing the stakes involved and for modeling a response that is appropriate for a democratic society.

In this article, we have offered the theoretical underpinnings for democracy's protection in the public sphere by elaborating on the democratic significance of human dignity. This shift in perspective was partially motivated by an empirical assessment of the types of anti-democratic threats facing contemporary democratic societies. Overt anti-democrats and outright extremist political movements still exist and should not be overlooked. However, we perceive a more acute problem among political actors who often remain broadly committed to a minimalist notion of democracy while trading on ideas, increasingly prevalent in the public sphere, which undermine the human dignity of particular societal groups. The analytical shift toward public sphere institutions, we argued, helps us reconsider the theoretical tools that we use to make sense of dilemmas of democracy's protection.

Focusing on human dignity as a fundamental democratic value highlights the unequal costs borne by already marginalized groups due to anti-democratic assaults. We illustrated this argument by discussing ways in which speech and other acts may undermine dignity by delegitimizing and dehumanizing members of particular social groups, undermining their claims to have their rights recognized, or, in the most egregious forms, denying them equal status as members of humanity. The choice between expressive actions or exclusions will always depend, in part, on factors specific to particular cases. However, we believe that the clear specification of a key value that underlie these democratic dilemmas, which we have attempted in this article, provides firmer theoretical grounding for such decisions.