By Kavyanjali Kaushik
Understanding democratic resilience requires studying how political identities are built but it is a question the backsliding literature has left largely unasked.
Liberal democratic theory has operated on a convenient fiction that there existed a prior democratic public that is now declining. That citizens once understood constitutions, respected legal norms, valued deliberation, and are now being corrupted by populist demagogues, social media, and cultural warfare. The entire backsliding and democratic erosion literature is built on this baseline assumption. But it is largely mythological.
Mass suffrage is extraordinarily recent. Universal suffrage in most democracies, including in Western nations considered advanced in their democratic governance, is in many cases less than a century old, and in some cases like Spain, Portugal, Hungary and Poland in Europe or vast majority of nations in Latin America, Africa, and Asia, barely fifty years old. The populations now fully participating in digital public spheres, which have become the primary sites where symptoms of democratic decline are observed and studied, were never meaningfully included in the political community whose norms they are now accused of failing to uphold.
The man systematically severed from his emotions by the narrow conception of the rational citizen, the working-class woman carrying the invisible weight of care, the second-generation immigrant negotiating belonging across two worlds, the rural landholder written off as peripheral—these and many others had their relationship to liberal democracy mediated entirely through institutions that both claimed to represent them and systematically excluded them from the inner workings of political life.
Political education in schools was largely reduced to civics in its thinnest sense, transmitted as a fixed body of norms and distanced from the lived experiences through which individuals come to understand and constitute their political identities. This lack of quality civic education is widespread across the world: from the US, where just a third of the citizens have been found to possess the basic knowledge of government and constitution, and the UK, where citizenship as an antidote to declining political participation among youth was introduced only in 2002, to post-communist Eastern European states such as Romania, Poland and Hungary that are struggling to generate interest in democratic civic cultures and “good citizenship” among students, and in transitional democracies such as South Africa, Brazil, Indonesia, and India, where civic education is constrained by the intersecting legacies of colonialism and comparatively weaker institutional safeguards for its impartial delivery.
Across these contexts, political education treated learners as passive vessels into which pre-defined knowledge was to be deposited, rather than as participants in an ongoing, reflective process of political becoming. Here are the institutions, here is how a bill becomes a law, here are your representatives, here is the electoral system. It was almost never political in any real sense if it did not address political power, the inherent tensions within democratic public spheres, and the actual mechanism of identity construction through which entitlement to rights is fought for, won, and remains fragile.
In the higher education of academia, where the democratic backsliding literature is produced, the dominant polarization research measures distance between people’s emotions, issue positions, partisanship, and attitudes. This commits researchers to treating people’s identity as a given scientific input within regression models, rather than something historically and socially constructed. The whole literature is overwhelmingly political science and fundamentally institutionalist. Macro-comparative projects such as V-Dem and indices such as Freedom House, while being extraordinarily rich in data, measure democratic erosion through institutional performance proxies, without touching upon on how citizens form the political identities through which they engage with or disengage from democratic institutions.
The affective polarization literature documents growing partisan hostility and social distance but leaves the formation of those partisan identities almost entirely unexamined. The core issue here is that the surveys measure the product of identity formation (how much do you dislike outgroup partisans?) without asking how that identity was constructed.
It asks, flabbergasted, what happened to democratic norms and why are they declining? The implicit assumption is that there is a pre-political citizen whose democratic commitments erode under authoritarian pressure. But this framework treats polarized identities as the phenomenon to be explained rather than asking about the conditions of their construction.
From a more constructivist perspective, even the idea of a citizen who naturally values democratic norms is itself a historical construction, a myth, which the research frameworks of political science cannot accommodate without undermining their own basic assumptions. Relational sociology, contentious politics, social psychology and more recently, digital communication studies explicitly invert this and show that relationships, self-expression, and interactions within a community are all prior to categories of identities, but this is incompatible with the standard survey-experimental research, which treats identities as stable and already defined. For instance, scholars working in the tradition of contentious politics and sociology, including Charles Tilly, Andreas Wimmer, Sidney Tarrow, Doug McAdam, Robert Benford and David Snow, have shown that political identities emerge through collective storytelling, public claims, and shared narratives that help previously excluded people see themselves as both deserving of political voice and capable of political action.
Feminist political theorists such as Margaret Somers, Judith Butler and Kimberlé Crenshaw similarly insist that identity categories like gender are relationally produced, emerging through social interactions, narratives, and institutional arrangements, rather than residing as fixed attributes within individuals. Scholars of digital communication have extended these insights to show how social media platforms are relational architectures that actively produce political subjectivities, whether by bringing together disparate groups around shared emotions (Zizi Papacharissi), especially around anger and outrage (Manuel Castells), encouraging personalised forms of political action (W. Lance Bennett and Alexandra Segerberg) or by creating online communities built around exclusionary or hostile ideas, such as those found in parts of the “manosphere” (Sarah Banet-Weiser, Rosalind Gill, Debbie Ging and Alice Marwick).
For the first time in democratic history, the public sphere is truly open to citizens. The eloquent political language and knowledge so cherished by political science, and on whose mantle the liberal scholarship proclaims the decline of democracies, is being reshaped by the people whose vocabularies we ought to study to understand and find the roots of democratic resilience and survival. And certainly not as the erosion of something that existed, but as the first genuine encounter of excluded populations with political life.
The radical right, particularly in its most sophisticated contemporary forms, did not accidentally stumble into this newly open public sphere and came to dominate it.
It recognized very early that the collapse of institutions and the rise of digital platforms created an opportunity to construct political identity and transmit political culture through new channels that establishment actors were too slow, too institutionally bound, and too contemptuous of vernacular political communication to use effectively.
Overwhelming research evidence, from the work of scholars such as Theda Skocpol and Vanessa Williamson on the Tea Party of the US, Bharath Ganesh and Cynthia Miller-Idriss on far-right movements across Germany, France and the UK, and Sahana Udupa on Indian nationalism, shows that early and systematic use of social media allowed activists, influencers, politicians, and supporters to find one another, build communities, and make their ideas appear more normal and widespread. Alternative digital spaces and formats such as memes, anonymous forums like 4chan, and platforms such as Reddit and X (formerly Twitter) helped these movements gain visibility and influence in the public sphere, even when they had previously existed at the edges of mainstream media and politics.
The study of how the radical right constructs identity, recruits supporters, and narrativizes grievance has generated a substantial and important body of work. From dedicated research centers such as C-REX (the Center for Research on Extremism at the University of Oslo), the Center for Right-Wing Studies at UC Berkeley, and the Institute for Strategic Dialogue, to a broader backsliding literature in which scholars like Nancy Bermeo, Steven Levitsky and Daniel Ziblatt have examined the structural conditions under which mainstream parties lose governing legitimacy. Yet what remains largely unasked across both the extremism and the backsliding literatures is the more fundamental question of how did mainstream center-left and center-right parties lose the capacity to construct political identities that people could recognize themselves in? Macro patterns of “corruption” and “poor performance”, often cited as causes of failures of mainstream parties, are outcomes, not explanations, for they do not tell us who the actors were, what identity they were offering, to whom, and through what communicative failures it ceased to resonate among masses.
Taking seriously the reasons why engagement patterns favor emotionally arousing narratives of the radical-right and how those patterns might, in some instances, function in ways that build rather than simply capture political agency, would require political science to move beyond the narrow agendas of misinformation, hate speech, or platform regulation, which have absorbed much of its recent attention, and to ask the deeper questions of identity, belonging, and democratic resilience.
If democratic backsliding is in significant part the predictable consequence of genuinely excluded populations entering political life without the conceptual and institutional resources to navigate it, then resilience building cannot be reduced to tracking institutional erosion indicators or measuring polarization.
It demands a fundamentally different starting point, which is the actual political consciousness of people, their grievances, their vocabularies and their networks through which they are learning, however unevenly, to act as political participants..
A nascent but growing body of research is beginning to move in exactly this direction. Tomáš Dumbrovský and colleagues at the ERC-funded RECONCILE project are developing a theory of identity constitutionalism that assesses which forms of identity discourse are compatible with democratic values such as tolerance, pluralism, and deliberation. The goal is precisely to ask whether certain forms of identity construction, especially those emerging through social media, can be scaled as a resource for democratic resilience. This is an empirically grounded intervention into the question the backsliding literature has left unasked: whether and how citizens can develop an emotional, identity-based attachment to democratic frameworks in the first place.
Related initiatives, including work led by Zsolt Boda at the ELTE Centre for Social Sciences and the Horizon Europe-funded MORES project, take seriously the moral-emotional foundations of political engagement as constitutive of democratic agency and how it can lead to anti- or pro-democratic politics. Similarly, the “Feeling Left Behind” project led by Silke Adam at the University of Berne seeks to better understand how unmet needs for recognition and belonging can make people more receptive to particular kinds of political content on social media platforms.
These are still minority currents within a field dominated by metrics and models that treat political identities as stable and fixed, but they suggest that the foundations of a more relational and communication-sensitive resilience research agenda are being actively laid. Because without sufficiently understanding how the attachment to democratic values is formed, the study of democratic erosion is, in a meaningful sense, studying consequences without causes.
Disclaimer: RECONCILE project is funded by the Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.
Kavyanjali Kaushik is a computational social scientist within ERC-funded research project RECONCILE, at the Institute for Interregional Study of Constitutionalism, Faculty of Law, Charles University, Prague, where she investigates how to build resilient constitutional identities within democracies. Her PhD in Social Sciences (2020-2025 with cum laude) from University Carlos III of Madrid, Spain examined how online self-expression and interactions influence support for radical-right parties. In the past, she worked as a computational data scientist at the international research project, ValCon: Value Conflicts in a Differentiated Europe and before pursuing academia, she was a journalist in India with organizations including Reuters, CNN-News18, and India Today.
This article is published under the sole responsibility of the author, with editorial oversight. The views expressed do not necessarily reflect those of the editorial team or the CEU Democracy Institute.
