Online Repertoires of Discursive Delegitimation Through Critical Online Comments—The Case of the Pandemic Crisis in Romania

Abstract

The COVID-19 pandemic represented a critical stress test for institutional trust and legitimacy, particularly in societies characterized by pre-existing deficits of confidence in public authorities. In Romania, the health crisis unfolded against a background of low institutional credibility and widespread skepticism toward political and administrative actors. This study examines how institutional authority was discursively evaluated, contested, and delegitimized through public online comments during the most severe phase of the pandemic. The analysis is based on a corpus of 457 comments collected from major Romanian news websites and YouTube channels between 20 and 26 October 2021, corresponding to the peak week in terms of infections and mortality. Drawing on Discourse Analysis and Discursive Psychology, the study combines quantitative coding with qualitative analysis to identify recurrent forms of moral and epistemic criticism, inflammatory discourse, and sarcasm, organized into interpretative repertoires. The findings show that online criticism is structured primarily around accusations of hypocrisy, incompetence, corruption, insensitivity, and authoritarianism. These discourses function not merely as expressions of dissatisfaction but as practices through which commenters articulate moral order, contest institutional legitimacy, and position themselves as morally vigilant and epistemically competent actors entitled to judge public decision-making. Online comment spaces thus emerge as arenas of discursive delegitimation in times of crisis, with important implications for democratic resilience and crisis governance.

1. Introduction

In democratic societies, institutional legitimacy is not a stable property of governing structures but an ongoing social accomplishment, continuously negotiated through the discursive practices of citizens, media actors, and political elites (Beetham 1991). Crises represent privileged moments in this negotiation: they compress into a short temporal window the evaluations that citizens make of institutional competence, moral credibility, and responsiveness, rendering visible the trust dynamics that ordinarily remain latent (Devine et al. 2021; Easton 1975). When those evaluations are predominantly negative—as tends to occur in contexts where institutional trust was already fragile before the crisis began (Esaiasson et al. 2021; Radu 2022)—crises do not merely test the administrative capacity of governing institutions; they reveal the normative structure of the society itself. Through their public reactions, citizens make visible the expectations, values, and moral standards against which they measure institutional decisions and performance, articulating a counter-discourse that sets what authorities should have done against what they actually did. Online comment sections attached to news media constitute particularly transparent sites for this process: spaces of mass participation where citizens spontaneously and often unfiltered express evaluations of crisis management, producing critical positionings that simultaneously reflect and reconstitute the normative frameworks through which institutional authority is judged (Tileagă 2018; Márquez-Reiter and Haugh 2019). This article examines how Romanian citizens constructed such critical positionings during the most acute phase of the COVID-19 pandemic—a moment that concentrated, in an unusually visible and consequential form, the pre-existing deficit of institutional trust that characterizes the Romanian political landscape.

1.1. The Importance of Institutional Trust and Legitimacy: The Case of Romania

Understanding why institutional trust matters for crisis governance requires attention to both its structural determinants and its dynamic character. The study of crises can reveal essential aspects of the functioning of democratic institutions through three interconnected mechanisms. First, crises activate latent trust: citizens are suddenly required to act on the basis of their institutional evaluations, rendering previously implicit attitudes publicly observable and analytically accessible (Devine et al. 2021). Second, pre-existing trust levels function as a structural resource that either enables or forecloses effective crisis governance: Kritzinger et al. (2021) show that rally-round-the-flag effects emerged only where baseline institutional trust was sufficiently high, while Radu (2022) documents that Romania registered only a modest and short-lived increase during the same period. Third, post-crisis trust is jointly determined by these pre-existing baselines and by actual institutional performance (Devine et al. 2024), meaning that crises do not merely provide a snapshot of trust dynamics at a given moment but also reflect pre-existing trajectories and actively shape their long-term evolution. The COVID-19 pandemic represented one of the most intense and dramatic recent social crises precisely in this sense, strongly testing institutional trust across multiple national contexts (Lazarus et al. 2021; Moucheraud et al. 2021).

Transnational surveys, such as the Wellcome Global Monitor included in the Gallup World Poll, further document the link between vaccine hesitancy and distrust in public administrations perceived as distant, unfair, or unrepresentative (Moynihan 2022; Yang 2021). Trust in institutional capacity, once lost, can persist in social memory over the medium and long term, significantly affecting public support for and compliance with subsequent policy decisions—a dynamic of particular relevance in the Romanian context, where the pandemic crisis unfolded against a background of chronically low institutional credibility.

In Romania, the vaccination strategy and the associated communication campaign constituted major areas of vulnerability, reflected in the low levels of trust expressed by the population. Numerous studies conducted at the time documented how citizens related to public authorities. Sandu (2021) shows that trust in institutions and medical professionals is positively correlated with pro-vaccination attitudes, while trust in social media—particularly among younger people—is more strongly associated with hesitant or anti-vaccination attitudes. Unfortunately, according to 2021 data, nearly 80% of Romanians did not consider the national government a trustworthy source of information regarding COVID-19 vaccines (European Commission 2021). In addition, 45% of the population declared dissatisfaction with the way the government managed the vaccination campaign (European Commission 2021). Public communication was perceived as lacking transparency by 71% of respondents (European Commission 2021).

According to an INSCOP survey conducted between 17 and 25 October 2021 on a sample of 2830 respondents, vaccinated individuals expressed the highest level of trust in doctors, whereas unvaccinated individuals reported the highest level of trust in people who had previously contracted COVID-19. Trust in information sources varies significantly: among vaccinated individuals, only 7.5% reported trusting politicians for information about vaccines, while among unvaccinated individuals the percentage drops to 3.2% (INSCOP Research 2021). In this context, doctors and medical staff could have played a key role in transmitting pro-vaccination messages. However, by the end of 2022, 40% of Romanians declared that they did not trust doctors, and 33% expressed distrust in nurses (IRES 2022). Distrust is even more pronounced in the case of politicians, with 91% of Romanians stating that they do not trust them (IRES 2022).

In conclusion, in Romania, trust in vaccines and in the actors involved in their promotion is fragmented and shaped by perceptions of institutional competence, transparency, and integrity. Relevant statistical indicators reveal not only generalized uncertainty, but also a vulnerability in the public health system’s capacity to mobilize the population in the face of future threats.

Another important dimension that must be taken into account is the parallel production and circulation of delegitimizing narratives, sometimes with a conspiratorial tone, which come to replace, within the public discourse, the communication and explanations provided by authorities and certified experts (Toth 2020). Of particular interest are Romanians’ beliefs regarding the emergence and effects of the SARS-CoV-2 virus. Thus, a survey of unvaccinated respondents showed that 25.1% fully agreed with the statement “The virus does not exist; it is just propaganda,” while 21.8% fully disagreed with the statement that “COVID-19 can be fatal.” (INSCOP Research 2021).

Similarly, approximately two thirds of Romanians (65.7%) believed that the COVID-19 pandemic was caused by global elites in order to impose control over the world’s population (INSCOP Research Report 2022). Comparable beliefs were also identified in an IPSOS Research Report (2021) online survey conducted between 3 and 4 February 2021 on a sample of 503 respondents, representative of the urban population aged 16 and over in Romania. According to this survey, 17% believed that the vaccine alters human DNA, 22% believed that the vaccine is more dangerous than the disease, 31% considered the vaccine unsafe because people had allegedly died after vaccination, 35% believed that the vaccine is ineffective because individuals tested positive after vaccination, and 38% believed that the vaccine has adverse effects that would only become visible years later.

The multitude of available surveys provides a discouraging picture of the social climate during the pandemic: low levels of trust, processes of delegitimization, and, more generally, a critical, negative, and suspicious attitude toward the authorities involved in managing the health crisis. Quantitative data can be complemented by an understanding of how authorities are discursively constructed and how criticism, dissatisfaction, and expectations are publicly articulated. Survey instruments are well-suited to documenting the scale and distribution of institutional distrust; content analysis can systematically map the forms and prevalence of hostile expression in digital spaces. What these approaches are less equipped to capture, however, are the constructive and performative dimensions of delegitimizing discourse: the moral accusations that do not merely describe but enact a normative violation, the sarcastic constructions that simultaneously express contempt and position the speaker as morally competent, or the recurrent interpretative patterns through which citizens collectively contest institutional authority. Accordingly, this study draws on Discourse Analysis and Discursive Psychology—whose methodological rationale is developed in Section 2.3—to map the normative-moral structure underlying citizens’ critical positionings during the pandemic peak in Romania, attending to precisely those constructive and performative dimensions that complement what aggregate measures reveal.

1.2. Public Discourses as Mechanisms for the Assertion of Normative and Moral Structures

A natural, unmediated source for exploring these dynamics is the online environment, where individuals spontaneously and often unfiltered express their evaluations of pandemic management. Public discourses of evaluation, criticism, and blame encountered in digital spaces thus constitute cultural resources that provide insight into the “moral fabric” of a community (Tileagă 2018). They reflect processes through which normative structures, collective expectations, and modes of displaying and negotiating moral identities and ethical positions are constituted.

In a similar vein, Márquez-Reiter and Haugh (2019) argue that public denunciations function as mechanisms for affirming normative infrastructures. In most cases, such denunciations are more stigmatizing than restorative: they aim to delegitimize and undermine the moral position of the other, while simultaneously consolidating and displaying the moral standards of the accuser. From a constructivist perspective, denunciation contributes to the construction, restoration, and reaffirmation of normative order.

Practices such as reproach, accusation, criticism, or condemnation reaffirm moral standards and highlight the extent to which these standards are perceived as having been violated (Drew 1998). At the same time, from a discursive perspective, Edwards (2005) emphasizes that accusations, criticisms, and the attribution of blame are part of a broader process of managing moral order.

Starting from this understanding of the moral function of criticism, it becomes important to observe how these mechanisms concretely manifest themselves in the online environment. In such spaces, critical discourse can be expected to span a wide register, from simple evaluative formulations to aggressive and inflammatory statements, as discussed in the following section… The discussion of what constitutes aggressive communication is itself nuanced, as there is a significant number of definitions and a wide range of semantically related terms. The present research focuses on the analysis of communicative acts characterized by critical denunciation and the delegitimization of authorities. Such forms of communication may take diverse forms, ranging from insults and inflammatory language to impoliteness, incivility, more elaborate critical constructions and sarcasm.

1.3. Hostility in Online Comments

A substantial body of research has shown that online environments are particularly conducive to the amplification of hostility in its various forms. Murthy and Sharma (2019), as well as Andersen (2021), argue that hostility becomes normalized online, embedded within a broader “culture” of communication. Similarly, Coe et al. (2014) and Rossini (2020) show that incivility functions as a pervasive norm in political discussions. Moor et al. (2010) likewise report that hostile comments on YouTube are highly common. Early work by Sproull and Kiesler (1986) further suggests that computer-mediated communication tends to foster more self-centered and less regulated behavior, facilitating more extreme, impulsive, and often less socially acceptable forms of expression.

The most frequently invoked explanation for this pattern is anonymity. Suler (2004) develops a psychologically oriented theory that brings together personality factors with identity and role-play dynamics that emerge under conditions of anonymity. While acknowledging that anonymity produces a disinhibiting effect, Suler nuances this claim by arguing that it is not simply the “true self” that surfaces when anonymity is present. Rather, anonymity enables an exploratory “play” with identities and roles—an activity that becomes possible precisely because social accountability is suspended. These claims shift the discussion from a predominantly individual-psychological framework toward a social-psychological one.

Rösner and Krämer (2016), by contrast, reports experimental findings suggesting that anonymity has a smaller impact on aggressive/deviant behaviors than is often assumed and that aggression may be better explained through group norms. At the same time, Rösner and Krämer (2016) do not dismiss anonymity entirely, but rather reframes it: anonymity appears to intensify conformity to the acceptance of deviant norms. On closer inspection, this finding can be read not as a refutation of Suler’s theory, but as a bridge toward it. Anonymity, as a situational condition, may enable and amplify the emergence of aggressive norms and, subsequently, conformity to those norms. In what follows, I adopt this perspective: an aggressive normative structure is more likely to emerge and intensify within anonymizing environments such as online communication, and such a normative order can then reproduce further aggressive expression. The present study therefore seeks to evaluate the intensity and form with which online aggressive comments are directed toward authorities in Romania. This requires, first, a brief conceptual clarification of hostility and aggressive expression.

One of the early definitions, Infante and Wigley (1986), describes verbally aggressive expressions as attacks on the person (self-concept), including character attacks, competence attacks, insults, maledictions, teasing, ridicule, profanity, and nonverbal emblems. Myers et al. (2013) identify nine types of verbally aggressive messages: competence attacks, work ethic attacks, swearing, threats, character attacks, nonverbal behaviors, teasing, background attacks, and physical appearance attacks. Deliligka et al. (2017) reviewing the literature, similarly outline defining elements that can fall under this umbrella, including verbal criticism targeting character, competence, physical appearance, or origin, as well as disregard, derision, threats, obscenity, curses, loud voices, teasing gestures, and related cues.

A key concept—incivility—gradually entered political communication research as a way of capturing hostile forms of expression. For Papacharissi (2004), incivility refers to behaviors that threaten democratic deliberation through negative stereotyping, discrimination, or attacks on individuals’ rights, while hostile formulations are defined more narrowly as impoliteness. This line is extended by Anderson et al. (2014), who also include insults and flaming among behaviors that can disrupt democratic dialogue.

Subsequent scholarship reconceptualizes incivility as an umbrella term for diverse forms of verbal hostility. Santana (2014) includes inflammatory and racist attacks, vulgarity, and stereotyping; Ksiazek et al. (2015) define civility as the absence of name-calling and profanity, without equating this with politeness; Anderson and Huntington (2017) add labeling, mockery, exaggeration, and hostile emotions; and Groshek and Cutino (2016) extend the concept to threats, hate speech, racist epithets, insults, accusations of lying, sarcasm, and typographic intensity.

Against this background, Rossini (2020) proposes a useful distinction: incivility encompasses forms of verbal hostility (vulgar, harsh, disrespectful tones), whereas attacks on democratic principles—such as the rejection of pluralism and equal rights—belong to the domain of intolerance, which exceeds individualized hostility. This shift reflects a movement from defining incivility primarily through its democratic impact toward a broader inclusion of verbal hostility, including flaming—another widely used concept for mapping online hostility. Andersen (2021) defines flaming in terms of its emotional character and effects and its violation of conversational norms, including insults and typographic intensity (e.g., excessive capitalization or punctuation). The author also emphasizes the definitional difficulty of hostility, insofar as it may depend on both form and motivation. Turnage (2007), in an analysis of flaming via email, identifies a set of characteristic dimensions: hostility, aggression, intimidation, level of insult, offensiveness, and unfriendliness.

1.4. A Discourse Analysis Approach to Online Hostility

Reviewing these operationalizations and definitions highlights the diversity of forms through which hostility and aggression can be produced and the multiple ways in which attacks and grievances are articulated. An equally important approach, however, is to describe and interpret the discursive functions these hostile forms fulfill within specific discursive contexts.

Infante and Wigley (1986) and Deliligka et al. (2017) distinguish between positional attack, argumentativeness, and aggressive communication. In their view, arguments involve attacking a position rather than a person and are therefore not considered forms of verbal aggression, insofar as they are oriented toward the substance of the message rather than the individual. In the present study, I treat the definition of verbal aggression—particularly from a discursive perspective—as more nuanced. Specifically, positional attacks and so-called “constructive critiques” may include hostile elements. Moreover, a person’s ideas and arguments cannot easily be separated from the person who produced and communicated them. Even when an argument is presented without explicit personal attacks, the dismantling and criticism of positions, ideas, and arguments may still perform injurious and delegitimizing functions.

As Gee (2010) argues, meaning—and thus the function of an utterance—resides in context. Within discourse-analytic methodology, it is context that enables the classification of communication as hostile, rather than the utterance “in itself.” Sarcasm provides a clear example: it may appear civil and non-hostile on the surface while operating as a particularly harsh form of aggression. The intent to wound is not contained in the form alone, but in the contextualized meaning of communication. This nuance allows for a significant expansion of what counts as online aggression and hostility when such discourse performs functions of attack, delegitimization, or institutional disengagement.

Anderson and Huntington (2017) define sarcasm as a subtler form of attack that uses irony to provoke and that becomes intelligible only through context, while performing complex discursive functions. Sarcasm enables face-saving through plausible deniability, facilitates affiliation with like-minded audiences, and simultaneously marks distance from those positioned outside the in-group—thus functioning as a mechanism of social differentiation as well as a means of venting frustration (Ducharme 1994). Through sarcasm, speakers may define unacceptable behaviors that fail to meet normative standards, as well as undesirable circumstances and situations.

Within this framework, the central objective of the study is to map how criticisms, dissatisfaction, expectations, and a normative–moral order are discursively articulated in relation to authorities.

Operational Objectives

The operational objectives guiding this analysis are:

O1. Identify the main types of discourse used in public online comments, considering both the targets of the comments and the nature of the issues addressed.

O2. Analyze the rhetorical strategies and devices (e.g., sarcasm, rhetorical questions, metaphors, narratives, and framing techniques) used to challenge or support institutional decisions, actions, and recommendations, as well as their discursive functions.

O3. Identify the extent to which the analyzed discourses can be grouped into interpretative repertoires concerning key institutions—their roles, decisions, and actions.

2. Materials and Methods

2.1. Data Selection and Analytical Procedure

In order to adequately describe the public discursive profile emerging in the context of a major public crisis, public online comments were selected as the primary units of analysis. The selection of comments involved several stages aimed at reducing and refining the data corpus.

The first step consisted of delimiting the temporal scope of the analysis. The study focuses on the period between 20 and 26 October 2021, when Romania experienced the most severe week of the COVID-19 pandemic, marked by the highest recorded rates of infections and deaths (average number of infections ≈ 14,516 per day). This was a critical period characterized by intense emotional responses and virulent public reactions, reflected in a heightened engagement with news flows and unfolding events. In such moments of acute crisis, individuals are particularly active in attempting to make sense of decisions and actions taken by others—most notably by institutions and authorities responsible for crisis management. Expressing evaluations through online comments in this “heated” context not only reflects public sentiment but also re-instantiates—and in some cases actively defines—systems of norms and values through judgments of institutional actions and decisions to an exceptional degree.

A second step involved the selection of public online spaces in which comments were posted. The top-ranked Romanian news and analysis outlets were identified using audience data from the Romanian Transmedia Audit Bureau (BRAT); among the three most-visited websites at the time of data collection, only HotNews permitted user comments on its articles and was therefore selected. Comments were manually downloaded, and after removing replies and unintelligible entries, a total of 627 comments—representing nearly all comments posted during the reference period—were included in the analytical database. A second data source consisted of television news channels disseminating content on YouTube, on which viewer comments are enabled. The three most-watched Romanian news and analysis channels were selected—Antena 1, România TV, and Digi24—ensuring coverage across the range of editorial orientations shaping public pandemic discourse. News videos whose titles included one or more of the keywords vaccination, COVID, and/or pandemic were identified, and a database of all comments excluding replies was created using comment-scraping software. The comments were then ranked according to the number of likes received, and only comments with more than ten likes were retained for analysis, resulting in a corpus of 467 comments. This threshold was applied as a criterion of discursive salience: comments that accumulate engagement on YouTube are those surfaced most prominently by the platform’s algorithm and encountered by the largest number of users, thus constituting the most visible layer of public discourse in the digital space. The two datasets were thus constructed using different filtering procedures, adapted to the distinct display logics of each platform. The unit of analysis throughout is the comment in its entirety, treated as a bounded discursive act.

Comments were analyzed as relatively autonomous discursive acts, without systematic reference to the specific journalistic materials that occasioned them. Within a discourse-analytic framework, journalistic materials were treated as activating stimuli for pre-existing attitudinal positionings, in view of identifying interpretative repertoires that operate at a level of generality exceeding the local occasion of any individual comment. The breadth of sources and the high-intensity temporal window selected ensure sufficient stimulus variation to capture the range of critical positionings structuring public discourse during the reference period.

2.2. Analytical Criteria and Variables

For the analysis of online content, the following criteria were employed.

The first criterion concerned the target of the comments. Who are online communicators addressing? This criterion was considered essential, as the social entities that become the objects of public discourse—whether political decision-makers, institutions, or other citizens—are simultaneously the targets of evaluation. The attribution of responsibility for crisis management indicates toward whom the public directs its expectations and whom it ultimately blames.

The second analytical variable referred to the basis of evaluation. Specifically, the analysis distinguishes between evaluations grounded primarily in moral considerations—such as fairness, justice, and ethical legitimacy—and those grounded in epistemic considerations, such as competence, professionalism, or expertise. While the former operates mainly within a moral register, the latter are more explicitly epistemic in nature.

The third criterion differentiated among three forms of engagement: argumentative (position-based critique that does not display any apparent form of aggressiveness); critical (expressing disapproval and potentially marked by a hostile tone, yet without escalating into direct insult or inflammatory attack); and aggressive (including insulting or inflammatory comments).

2.3. A Discursive Perspective

In addition to coding, counting, and grouping comments according to the criteria outlined above, this study seeks to offer an additional analytical dimension. Beyond quantitative categorization, it is essential to understand how expectations are discursively structured and how normativity and moral order emerge as processes of institutional (de)legitimation. This includes the discursive construction of authorities as well as of the speakers themselves.

Discourse Analysis (DA), and more specifically Discursive Psychology (DP), provides a complementary methodological and epistemological approach. Within DA, language is not treated merely as a vehicle for expressing opinions or internal states (Coulter 2005, p. 80), but as a form of social practice through which people act, judge, contest, and confer legitimacy (Potter 2003; Billig 1987; Wetherell and Potter 1988). DP, in particular, adopts an explicitly anti-cognitivist stance, rejecting the assumption that language simply reflects underlying mental processes or attitudes (Potter and Wetherell 1987; Wetherell 1998).

From this perspective, online comments do not constitute merely spontaneous reactions, but rather forms of public action through which citizens evaluate, justify, and contest institutional authority. They become expressions of an emergent moral and epistemic order, within which legitimacy is discursively examined and negotiated. A central concept within this approach is that of interpretative repertoires, understood as relatively coherent sets of expressions, metaphors, argumentative patterns, and rhetorical frameworks through which people make sense of and evaluate social events (Wetherell and Potter 1988). Repertoires do not belong to individuals; rather, they circulate culturally and are reused and recontextualized according to the discursive demands of specific situations.

In online comments, interpretative repertoires can function as tools of contestation, enabling the articulation of accusations, the expression of suspicion, or the legitimation of indignation. At the same time, they contribute to the construction of moral and epistemic identities—such as the vigilant citizen, the critical expert, or the abused victim—thereby shaping the broader discursive landscape of institutional (de)legitimation.

By identifying recurrent repertoires, rhetorical strategies, and their discursive functions, the analysis allows for a deeper understanding of how institutions are evaluated and redefined from the standpoint of citizens’ everyday experiences. In the context of the pandemic, these discursive processes reveal the dynamics of trust and delegitimation, as well as the mechanisms through which public authority is reconstructed, ironized, or rejected through language.

2.4. Coding Procedure and Reliability

The entire empirical material was coded exclusively by the author, who developed all categories, subcategories, and interpretative repertoires used in the analysis. Coding and quantitative analysis were conducted using the qualitative data analysis software Quirkos (version 2.5.3). Given the exploratory and interpretative nature of Discourse Analysis, formal inter-coder reliability procedures were not applicable in the conventional sense. Internal consistency was pursued through a structured iterative process: the coding scheme was developed inductively across multiple passes through the corpus, with categories refined and recalibrated each time ambiguous cases were encountered. Borderline coding decisions were documented and systematically revisited in subsequent review cycles to ensure coherent application of analytical criteria. The categories and subcategories were finalized only after the scheme reached saturation—that is, when additional passes through the material no longer produced revisions to the existing framework. The absence of a second coder nonetheless remains a methodological limitation, as it precludes the formal verification of coding consistency through inter-rater agreement measures; this constraint is discussed further in the Limitations section.

2.5. Ethical Considerations

With regard to ethical standards, this study relies exclusively on publicly available comments posted on open-access platforms and therefore does not fall under regulations governing research involving human subjects. Comment authors are anonymous or use pseudonyms, none of which are reproduced in the text. All excerpts analyzed were translated into English, eliminating the possibility of re-identification through online searches. No interaction with users took place, and the analysis focuses on collective discursive patterns rather than on identifiable individuals. Consequently, informed consent was not required, and the research does not pose risks to participants.

3. Results

3.1. Introduction

The present analysis is based on a total corpus of 1094 online comments collected from digital platforms during the COVID-19 pandemic. Of these, 118 comments were identified as argumentative positionings without a critical orientation, while 158 comments could not be reliably classified due to ambiguity or lack of semantic clarity. These categories resulted in 276 comments being excluded from the critical analysis.

The remaining 818 comments display, to varying degrees, a critical or aggressive tone, ranging from direct insults to more elaborate forms of criticism. These comments were analyzed and categorized according to both the target of criticism and the form of criticism. Although efforts were made to assign each comment to a single category within the coding framework, this proved challenging, as a proportion of comments overlapped across multiple categories (see Appendix A for the post-coding distribution of comments).

Among these, analytical attention focused on the 509 comments explicitly directed at authorities, as these contribute most directly to the discursive construction of institutional (de)legitimacy. In the final stage of coding, 457 comments were included in the discourse analysis distributed across five main categories: inflammatory discourse (126), moral criticism (176), epistemic criticism (78), sarcastic moral formulations (42), and sarcastic epistemic formulations (35).

Initially, rhetorical questions were also coded as a rhetorical device, resulting in 52 comments being assigned to this category. However, these largely overlapped with sarcastic formulations. In order to reduce analytical ambiguity—and because rhetorical questions would require a dedicated and more fine-grained analysis—they were ultimately excluded from the analysis presented here.

The following section examines how online comments discursively construct the authorities involved in managing the pandemic. The analysis aims to capture both the types of criticism articulated and the interpretative repertoires and rhetorical strategies mobilized by users in evaluating institutions and public decisions. This approach enables a deeper understanding of how the identified discourses and repertoires reveal recurring structures in public engagement with authority in a context of crisis.

3.2. Morally Oriented Discourses Directed Against the Political–Administrative System (176)

The largest cluster of comments consists of morally framed accusations directed at the political–administrative system. Rather than forming a homogeneous body of criticism, these comments organize themselves into distinct interpretative repertoires through which dissatisfaction is articulated and justified. Each repertoire mobilizes a relatively stable rhetorical logic for constructing blame, legitimacy, and moral evaluation.

3.2.1. Denunciations of Hypocrisy, Inconsistency, or Contradiction (69)

(Moral Critique Repertoire 1—MC1)

One of the most frequent discursive constructions functioning as an interpretative repertoire relies on a demonstrative rhetorical structure aimed at denouncing problems in administrative action. This rhetorical pattern indexes a contradiction or inconsistency—often framed as hypocrisy—in the decisions and actions of authorities. Its function is to expose conflicting interests, justify the withdrawal of trust and legitimacy, and demonstrate that citizens are knowledgeable and unjustifiably treated as naïve or incapable of understanding.

They are closing schools again//while leaving malls and bars open. I don’t understand how Romanians can endure such humiliation for the third time. They are openly mocking us. They say schools should remain open and be the last to close, yet they are always the first. And then they complain that we don’t trust them.

This comment signals a moral discrepancy by juxtaposing schools—institutions with an educational and socially valued role—with malls and bars, associated with leisure or even vice. The discursive construction evaluates the government’s inability or unwillingness to prioritize institutions according to their moral and social importance. Romanians are portrayed both as victims and, rhetorically, as partially responsible for tolerating the situation. The government is further accused of insincerity—an attempt deemed naïve, as its intentions are presented as transparent to the vigilant commenter.

3.2.2. Conspiratorial Accusations (35)

(Moral Critique Repertoire 2—MC2)

The second most prominent subcategory consists of comments with a conspiratorial tone, suggesting the existence of parallel interests pursued by institutions and authorities. The relatively high prevalence of such comments indicates a profound lack of trust in the good faith of institutions and those who govern them. This distrust is generalized, unspecified, and non-falsifiable: even beneficial decisions are interpreted as grounded in illegitimate motives. Institutions are perceived as irredeemably corrupt, and the delegation contract through which citizens entrust decision-making power to authorities is considered invalid.

Such comments do not seek to demonstrate, argue, or complain about poor pandemic management. Instead, they signal what is already “known.” The focus is not on evaluating the quality of decisions, but on asserting a pre-existing belief that governmental and institutional actors pursue interests that are parallel or divergent from the legitimate public mandate.

Depending on the type of interest invoked, these comments fall into two main subtypes: those referring to economic interests and those referring to population control.

Discourses on Economic Interests

They want vaccination at any cost. Whether vaccination actually provides protection no longer matters. What matters for those in power and their offshore accounts is the number of doses purchased and the push for vaccination. Follow the trail of the ants to find the anthill.

Here, the government’s interests are constructed as decoupled from the public interest and therefore from legitimacy. Regardless of apparent convergence with public goals, governmental motivations are portrayed as fundamentally immoral and profit-driven. The government is thus not depicted as merely opportunistic, but as wholly corrupt. Trust is entirely absent: even when outcomes appear beneficial, the underlying motives are framed as illegitimate.

Discourses on Population Control

We don’t have medicines because they won’t give them to us—they send it back so that we die, because there are too many of us.

In this case, the government is accused of controlling population size through the deliberate withholding of medication. Authorities are indirectly constructed as criminal or even genocidal.

3.2.3. Lack of Responsibility and Institutional Passivity (30)

(Moral Critique Repertoire 3—MC3)

A further subcategory comprises discourses accusing the state of passivity and failure to assume responsibility. The state is constructed as indifferent, dehumanized, and inert. The core violated expectation underlying these critiques is that of an active state, engaged in addressing the problems generated by the pandemic.

The state is weak//Politicians downplayed the problem, saying we had defeated the pandemic. They lacked the courage to take tough and unpopular measures in time. And from all this, conspiracy theorists conclude that we should fear… DICTATORSHIP.

Here, the state is defined as weak and lacking resolve. Politicians are depicted as having minimized the pandemic and as lacking courage. The comment highlights a paradox: conspiracy theorists accuse the state of dictatorship, while the state is constructed precisely as insufficiently authoritative. Temporality is also problematized—the state either rushed to declare the pandemic over or failed to act decisively in time. The result is a portrayal of a weak, desynchronized, and hesitant state.

3.2.4. Accusations of Authoritarian or Dictatorial Tendencies (26)

(Moral Critique Repertoire 4—MC4)

Another category consists of discourses accusing decision-making authorities of abusive, dictatorial behavior that violates fundamental rights. These comments primarily target restrictive measures imposed during the pandemic. Such measures are often hyperbolized, with the state discursively associated with historical dictatorial regimes in order to intensify the accusation.

The central charge is abuse, grounded in the violation of human rights and dignity. These decisions are frequently depicted as lacking justification, as demonstrated discursively within the comments themselves. As a result, restrictions are framed as gratuitous abuses. The state is not merely portrayed as adopting abusive decisions, but as inherently abusive as an institution, given its willingness to impose unjustified restrictions.

Yeah//soon even 100% will be considered a low vaccination rate… interesting how something that started as a certificate to ease travel within the EU ends up being a basis for banning basic rights… it resembles quite well the star imposed by a madman a few decades ago, supposedly for the good of a large part of the population… otherwise, why should some people (those who get tested) be forced to expose themselves to infection from vaccinated people who are not asked anything—not even whether they have symptoms, as happens here?

The comment opens with irony, functioning as a hyperbolic explanation of state abuse. The argument highlights moral and logical contradictions to delegitimize restrictions imposed on the unvaccinated. An indirect allusion is made to fascist regimes and anti-Semitic restrictions, constructing current measures as abusive through historical analogy. Authoritarianism is further amplified through rhetorical questions, which function to expose perceived contradictions.

Overall, this category of morally oriented critiques constructs institutions and decision-makers in several recurring ways: as hypocritical and inconsistent, treating citizens as naïve; as driven by interests divergent from legitimate public goals, often amplified through conspiratorial narratives and accusations of corruption; as abusive and authoritarian; or, finally, as passive and indifferent, insufficiently concerned with human suffering.

3.3. Inflammatory and Insulting Discourses Directed Against the Political System (126)

This category includes comments characterized by direct insults, abusive language, and curses as forms of overt aggression, as well as violent reproaches and admonitions addressed to authorities. Inflammatory discourses display the highest level of explicitly expressed verbal violence. They tend to contain a weaker critical and argumentative component: in most cases, there is little or no attempt to mitigate the target, as the comments operate as direct and often unfiltered expressions of maximal aversion. The relatively high prevalence of this category—ranking second in frequency—indicates intense emotional tension and strong hostility toward political authorities.

Within this category, however, several internal subtypes were identified: (1) insults and curses as direct aggression (I1), (2) violent evaluative labeling that nevertheless retains an argumentative component (I2), and (3) aggressive admonitions that are moralizing in tone but entail a lower degree of outright rejection (I3).

3.3.1. Insults and Curses (75)

(Inflammatory Repertoire 1—I1)

Below are two illustrative examples, the first containing insults and the second curses:

Insults

I agree!//It’s absurd that these thugs can do whatever they want on the state’s money. If they mess up, we’re the ones who pay again.

Curses

Because you carried out an “information campaign” fit to put children to sleep, you are not capable—go to hell!.

3.3.2. Evaluative Labeling (31)

(Inflammatory Repertoire 2—I2)

The second subtype consists of accusations articulated through negative evaluative labels such as liar or incompetent. These insulting terms are attributed either on epistemic grounds (competence, knowledge) or on moral grounds (honesty, integrity). Epistemic labeling most often refers to incompetence, whereas moral labeling focuses on accusations of lying or hypocrisy.

A representative example of moral labeling—also the dominant form (approximately two thirds of this subtype)—signals intense indignation at perceived violations of moral norms, with the central accusation being dishonesty. The violated expectation is that authorities should act with honesty and fairness:

Lies everywhere—where are the researchers and the specialists in the field?

By contrast, roughly one third of inflammatory comments focus on the alleged lack of competence, accusing the government of being unable to manage the challenges raised by the pandemic. Here, the violated expectations concern competence and expertise:

Incompetent//Aren’t we tired of this? Why aren’t they fired if they don’t leave on their own?

3.3.3. Reproaches and Admonitions (19)

(Inflammatory Repertoire 3—I3)

The final subtype comprises expressions framed as reproachful or admonitory statements, intended to scold authorities for violations of moral norms and unmet expectations. A recurrent formula is:

Shame on you for what you did!

These comments display a moralizing character. Rhetorically, reproach implies a form of attachment—disappointment, a betrayed expectation, and, in some cases, an attempt to elicit behavioral correction. Pragmatically, this form is less radically distancing than “pure” insults, as it presupposes an orientation toward norms that the target is expected to recognize. At the same time, public reproach can also perform a punitive function by invoking collective opprobrium and signaling a flagrant breach of publicly shared moral standards.

Overall, inflammatory and insulting discourses radically delegitimate authorities. They are dominated by direct discrediting strategies, with a lexicon marked by aggression. Yet the category also contains internal variation—from “pure” insults to evaluative labeling and moralizing reproaches—indicating gradations in positioning toward authority: from total rejection to criticism that still presupposes unmet expectations.

3.4. Epistemic Critical Discourses Directed Against Authorities and the Political System (78)

This category includes critical discourses oriented toward the political system on epistemic grounds, targeting the competence, intelligence, and knowledge of authorities. As in the other categories, several subtypes (or repertoires) were identified. Below, I outline those repertoires as recurrent ways in which commenters evaluate authorities.

3.4.1. Lack of Knowledge or Clarity Regarding Laws or the Virus (25)

(Epistemic Critique Repertoire 1—EC1)

Nah! This is a trick. It’s written neither legally correctly nor logically, and from this article it’s not clear that we’re generally not allowed in the mall. The interpretation comes from Arafat, but in the government decree this isn’t properly mentioned; this article does NOT say you can’t go to the mall without a certificate. Either we no longer know how to write a coherent legal bill! Anyway, the decree is unconstitutional as a whole—but who cares, let’s just be healthy. For 30 days we’ve been stuck with this decree and Arafat’s interpretations.

The author adopts a colloquial style that conveys ease and confidence in dismantling juridical incompetence, while also attributing manipulative intent (“this is a trick”) to authorities. The critique develops by highlighting both irrationality and lack of clarity in legislative drafting. The commenter constructs themself as legally competent, distinguishing between legal text and legal interpretation, the latter portrayed as enabling potential abuses. The author positions themself among vigilant citizens able to readily detect legislative incompetence and ambiguity. The critique is further escalated beyond unclear drafting toward claims of unconstitutionality, i.e., illegality. The discourse closes with a resigned tone, combining epistemic superiority with the stance of a competent yet powerless citizen-victim.

3.4.2. Inability to Anticipate Absurd or Severe Consequences (18)

(Epistemic Critique Repertoire 2—EC2)

Shops should have extended hours//It is irresponsible stupidity to reduce the opening hours of grocery stores—it only produces crowding that facilitates the spread of COVID. These individuals should be held criminally accountable for such absurd measures.

The comment opens in a strongly normative and vehement register. The critique is developed aggressively through invective (“irresponsible stupidity”), which intensifies distancing and condemnation, alongside an argument about unintended consequences framed as the outcome of incompetence. The final sentence introduces a normative demand for legal accountability, positioning decision-making as not merely mistaken but punishable. By referring to authorities as “these individuals,” the speaker further de-institutionalizes them, discursively reducing them to generic, anonymous persons of negligible competence. The call for criminal prosecution constructs authorities not only as incapable and irrational, but also as dangerous—potentially even criminal—through the consequences of their decisions.

3.4.3. Noticing Contradictions or Inconsistencies in Communicated Data (14)

(Epistemic Critique Repertoire 3—EC3)

We’re grinding our teeth, but we “reach the plateau”//How? Simple. We stop counting. Brilliant, right? We cap the number of tests—we got a plateau! We have 1700 ICU beds? Perfect. We capped the number of severe cases: we only have 1700. And so on. Real statistics? Next year, when excess mortality for 2021 is calculated compared to the 2010–2019 decade.

From the outset, the author highlights perceived manipulation by authorities. A colloquial register is used to diminish institutional prestige. Through rhetorical questions in a quasi-pedagogical tone, the comment stages an “unmasking” of the alleged strategy, positioning the author in a stance of epistemic superiority, reinforced by irony. The rhetorical question “Real statistics?” again signals distrust toward official information, framed as distorted by authorities’ self-interested efforts to protect their public image.

3.4.4. Direct Accusations of Incompetence with Suggestions for “What Should Have Been Done” (13)

(Epistemic Critique Repertoire 4—EC4)

Absurd! You don’t do vaccinations at the peak of the pandemic! This should have been done in the summer; now we only stimulate virus mutations.

This discourse adopts a critical stance toward health and governmental authorities, suggesting that their actions are not merely ineffective but dangerous. Incompetence is initially foregrounded through the direct attack “Absurd!”, which conveys indignation and frames the vaccination strategy as fundamentally wrong. The legitimacy of the action is then rejected outright, and the author concludes with a simple, “common-sense” recommendation for what should have been done. The elementary character of the reasoning further amplifies epistemic distancing and severe critique. The commenter positions themself as epistemically superior, maximizing the distance from official rationales for vaccination decisions.

Across epistemically oriented critical discourses, authorities are predominantly constructed as incompetent, irrational, or manipulative, thereby placing their legitimacy under suspicion. These comments express a marked epistemic distance from state institutions, as commenters frequently position themselves as more knowledgeable, vigilant, and competent. Authorities are portrayed as incapable of making informed and effective decisions, unable to anticipate the consequences of their own measures, or responsible for unclear and ambiguous legislation—sometimes framed as unconstitutional and conducive to abuse—standing in contrast to “competent” citizens who can readily detect such shortcomings. Finally, some voices accuse authorities of methodological manipulation of statistical data, thereby justifying the withdrawal of trust in official information.

The implicit expectations underlying these discourses are that the state should be transparent, coherent, competent, and rational. When these expectations are perceived as unmet, the public reaction is not limited to disappointment but escalates toward delegitimation. Consequently, the state is no longer constructed as a legitimate source of authority, but rather as a suspect and/or incompetent actor.

3.5. Sarcastic and Ironic Discourses Directed at the Political System and Authorities (77)

A distinct category of discourses identified consists of those that use irony and sarcasm to articulate indirect critiques of authorities. The primary rhetorical device is sarcasm, which, beyond irony, has a more pronounced hostile character (Tannen 2005) and is often used to undermine the credibility of a social actor (person, institution, official discourse). Both sarcasm and irony generally mark the speaker’s superiority, as the speaker adopts a position of intellectual or moral distance. In addition, sarcasm allows for the maintenance of plausible deniability and thus of the speaker’s polite and morally intact public image, enabling face-saving (Jorgensen 1996). Moreover, sarcasm may at times serve a function of mitigating damage to the negative image produced. Both epistemic and moral forms of sarcasm were identified.

3.5.1. Epistemic Sarcasm (35)

Below I present the main types of epistemic sarcastic repertoires identified.

Direct Sarcasm Through Hyperbolization and Contrast or Other Rhetorical Devices (14) (Epistemic Sarcasm Repertoire ES1)

Organized chaos!//“My guv’ment” works, “my guv’ment” doesn’t think!

The comment contains two sequences: a paradoxical statement (“Organized chaos!”) and a repetitive ironic formula with a binary structure (“my guv’ment works/doesn’t think”). The use of the colloquial, nonstandard form “guv’ment” signals a strategy of linguistic caricature, suggesting provincialism and lack of sophistication attributed to the authorities. The alternation between “works” and “doesn’t think” creates a sarcastic antithesis: governmental activity is portrayed as lacking reflection, rationality, and strategy, amounting instead to chaotic and unproductive agitation.

Unmasking Lack of Expertise and Corruption Through Inconsistencies (9) (Epistemic Sarcasm Repertoire ES2)

Hey, Gheorghiță, //are you sure you’ve got enough vaccines? You should have played the “essential workers” card for dose three first. Moderna works well as a booster for Pfizer and vice versa, in case you didn’t know. Go full speed ahead before the West—full of responsible intellectuals—goes into lockdown with a vaccination rate of 123.

The comment opens with “Hey, Gheorghiță,” a colloquial and overly familiar form of address that signals disrespect and functions to demystify authority by lowering the addressee to the level of an ordinary, culturally marginal figure. The comment is constructed simultaneously as advice and mockery, in which the author “helps” ironically by offering seemingly logical suggestions delivered with evident condescension. Beneath the surface of technical information (“Moderna works well as a booster for Pfizer and vice versa, in case you didn’t know”), the presumed ignorance of the authority is ironized, with the implicit assertion of the commenter’s epistemic superiority.

Unmasking Data Manipulation by Pointing Out Contradictions or Inconsistencies (7) (Epistemic Sarcasm Repertoire ES3)

According to which study does it decrease twentyfold? // I see, we don’t need to know such details. We just carry out orders from the colonel.

The comment is divided into two parts: a provocative question followed by an ironic statement that implicitly answers it. Through the sarcastic formulation doubled by a rhetorical question—an expected-answer structure—the speaker constructs the figure of the citizen as a victim of opacity, deliberately denied access to knowledge. A clear opposition is thus established between the rational, inquisitive citizen and an opaque, authoritarian, hierarchical state. The reference to “the colonel” functions as an ironic metonymy, evoking the image of a militarized state in which orders are issued from above and are not open to discussion, thereby intensifying the critique of an abusive and authoritarian form of governance.

Absurd or Unclear Decisions with Absurd Consequences (5) (Epistemic Sarcasm Repertoire ES4)

Dedeman/Hornbach//fine, but one essential question remains: if I am unvaccinated, live alone in an apartment, and my water pipe bursts, do I have to stay at least 30 days with a broken pipe in my home because I am not allowed access to Dedeman or Hornbach to buy repair materials?! day by day, these people are getting stupider…

The author exposes restrictive measures as absurd and irrational through a concrete scenario constructed ironically and amplified by a rhetorical question. The comment begins with an ironically framed phrase—“one essential question remains”—which parodies, through hyperbolization, the importance attributed to a specific practical issue. This is followed by a hypothetical domestic problem that escalates into a situation constructed as absurd if the strict letter of the regulation is followed. The concluding statement (“day by day, these people are getting stupider”) marks a shift from implicit irony to explicit judgment and insult, expressing accumulated exasperation. Authorities are discursively constructed as inadequate, disconnected from everyday reality, and incapable of anticipating the direct effects of their measures on ordinary citizens.

As can be observed, the comments can be grouped into four subcategories. The most frequently ironized feature (14 of 37) is governmental agitation devoid of rationality: authorities are portrayed as active but incapable of thinking, while official discourse is treated as a sequence of absurd or empty formulas. The second most frequent target (9 of 37) is lack of expertise, often linked to presumed complicity with commercial or geopolitical interests: authorities are ridiculed for their claim to scientific authority while allegedly ignoring empirical realities or merely executing “orders.” The third ironized feature (7 of 37) concerns data manipulation and numerical incoherence, with sarcasm emerging as a reaction to contradictions between reported facts and the basic logic of scientific communication. Finally, a distinct category (5 of 37) consists of absurd decisions poorly adapted to everyday life, where sarcasm takes the form of domestic or quotidian scenarios pushed to absurdity in order to highlight the rupture between normative regulation and lived reality.

This typology once again indicates a clear public expectation of a rational, coherent, and epistemically transparent state that grounds its decisions in genuine knowledge rather than in its simulation. The ironic citizen positions themselves as a rational, critical, and informed actor who not only questions the expertise of authorities but discursively replaces it through logic, observation, and sarcasm. The speaker implicitly claims the right to understand, to question, and to refuse blind obedience in the absence of rational justification. Consequently, epistemic sarcasm expresses a form of resistance to authority lacking epistemic legitimacy and becomes a means through which citizens revalidate their own capacity for judgment and construct themselves as epistemically superior in relation to experts and authorities.

3.5.2. Moral Sarcasm (42)

The final category discussed here consists of comments that adopt a critical stance toward authorities while predominantly choosing a sarcastic mode of construction grounded in moral criteria.

The following subtypes of morally oriented sarcastic discourses were identified.

Hypocrisy and Decision-Making Inconsistency of Authorities (13) (Moral Sarcasm Repertoire MS1)

We are Romanians! I have been to Portugal and France. Everywhere, rules are respected. As a tourist, you feel safe. In July, you could only enter a restaurant with a certificate. In “Little Romania,” they held a congress with 5000 PNL members—as if they’d all been dropped on their heads—and a “kakao festival” with 40,000 people. Without a care in the world. No comparison with Portugal. The ten thousand deaths in these days call for an Exceptional Tribunal to judge the cases of extermination of Romanians.

The comment opens with an already ironic tone, introducing an explanation of the critical and sarcastic positioning. Multiple expressions are used to intensify the ridiculing tone, particularly toward the political class, which is constructed as hypocritical and lacking legitimate concern for the population’s well-being. Toward the end, the sarcastic tone gives way to a directly accusatory stance, with the political class defined not only as hypocritical and indifferent, but as criminal. Through external comparisons, derogatory labels, and an accusatory crescendo, the author constructs an image of Romanian authorities as negligent, hypocritical, and indirectly criminal.

Lack of Moral Integrity, Cynicism, and Indifference Toward Human Suffering (10) (Moral Sarcasm Repertoire MS2)

Sleep peacefully//if elections were held tomorrow, the structure of Parliament wouldn’t change… It’s bad, but the water is the same, already rotten.

In this comment, the author accuses the political class of insensitivity and political cynicism. The tone is one of moral sarcasm with a fatalistic orientation. In order to intensify the critique regarding the immutability and lack of sensitivity of authorities, the author resorts to a metaphorical construction that amplifies the negative image of the political class. The political elite is constructed as irredeemably compromised. Such discourses signal political disengagement and generalized social cynicism, functioning as symptoms of a chronic crisis of trust in the legitimacy of political representation.

Bad Faith and Partially Conspiratorial Cynicism (8) (Moral Sarcasm Repertoire MS3)

They should hire him at Moderna then//since he knows better than they do how vaccines work. Romania’s number one specialist deserves to be rewarded for executing all orders. If they don’t want him, he’s good with Pfizer too—he can set their expiration dates.

The author constructs the expertise of an institutional representative in a sarcastic, hyperbolic manner, while simultaneously indicating—through a conspiratorial tone—this figure’s servility toward pharmaceutical companies. The discursive construction of the accusation also signals cynicism by foregrounding the prioritization of pharmaceutical companies’ economic interests over the population’s well-being.

Accusations of Authoritarianism and Dictatorship (6) (Moral Sarcasm Repertoire MS4)

Martial with a mask//it “is” “necessary” to have “martial” “law” “COVID19” in order to efficiently combat Romanian civil society, that is, the very being of the nation—and who could ultimately carry out this noble task better than a chubby three-star general (from the same national tribe as Cîțu, that is, the contemporary nomenklatura), backed by the army and all kinds of similarly associated institutions, and by the entire huge host of diseased politruks—vae victis.

The comment constructs a discursive opposition between the political class—critically defined as diseased, with dictatorial behaviors and criminal stakes—and civil society, defined as the essence of the nation and portrayed as a victim of political assault. A militarized, belligerent tone is used to indicate the victim status of the Romanian population, depicted as being under siege by the political class. Authorities are constructed as a repressive politico-military coalition acting not to protect citizens, but to suppress civil society.

Direct Sarcasm Through Figures of Speech (Metaphors, Name Distortions, Hyperbolization) (5) (Moral Sarcasm Repertoire MS5)

We have enough new, extremely dangerous strains, such as: Arahat, Lucovid, Wuhannis, Câțu Mâțu, Ciucă, etc.

This comment is a clear example of satirical-linguistic political sarcasm, in which the author uses wordplay and the deformation of proper names to construct a critical parody of key figures involved in pandemic management. The comment functions as satirical sarcasm with a delegitimizing role, transforming politicians into “viral strains” in a linguistic parody of pandemic discourse. Sarcasm is constructed through name deformation and symbolic equivalence between virus and political power, producing an effect of undermining authority and grotesquely exposing those who are supposed to protect the population. Such discourses are not merely jokes, but forms of discursive resistance, in which citizens respond to absurdity with absurdity, while maintaining a clear critical content: the virus is not only biological—it is also political.

Quantitative and qualitative analysis of the comments highlighted both the intensity of the critical tone and the complexity of the modes through which public dissatisfaction, suspicion, and expectations are expressed. The analyzed comments were grouped into repertorial structures understood as discursive resources created and accessed in the public sphere, through which the decisions and behaviors of public actors are judged, while the identities of commenters are also constructed. In order to understand how normative order is structured, the following section groups the repertoires according to the main traits indexed and criticized by commenters, noting that particular forms of dissatisfaction are frequently expressed through multiple rhetorical forms.

3.6. The Discursive Construction of Authorities in Online Comments During the Pandemic

As argued in the theoretical framework, critical, denunciatory, and evaluative public discourses are not mere emotional reactions, but mechanisms through which people reaffirm moral standards, negotiate responsibility, affirm or reconstruct the expected normative order in times of crisis, and construct identities. Through what they accuse, through what they signal as unjust, incoherent, or abusive, commenters implicitly reveal the expectations they hold toward institutions. Within this logic, analyzing the distribution of traits attributed to authorities becomes an indicator of the moral order mobilized discursively. In the table below (

Table 1. Hierarchy of traits discursively attributed to authorities.

Attributed Trait	Number of Comments	Types of Discourse Involved
Hypocrisy, inconsistency, contradictions	103	MC1 + MS1 + EC3 + ES3
Incompetence (technical, logical, legal)	61	EC1 + EC4 + ES1 + ES2
Corruption, greed, hidden interests, conspiracies	43	MC2 + MS3
Insensitivity, passivity, cynicism	40	MC4 + MS2
Authoritarianism, dictatorship, abuse	32	MC3 + MS5
Absurd decisions, inability to anticipate consequences	23	EC2 + ES4
Violent, inflammatory address with predominantly moral content	130	I1 + I2 + I3 + MS4

), repertories initially identified and described at the level of rhetorical forms were thematically regrouped. This regrouping emerged naturally from the observation that violated norms and expectations recur across discursive forms. For example, accusations of hypocrisy appear in direct moral critiques, epistemic critiques, and indirect critiques articulated through sarcasm. The hierarchy of these traits reflects, quantitatively (through comment frequency), the perceived gravity of normative violations, offering an image of the citizen–state relationship during the pandemic from the perspective of the most salient breached expectations.

The most frequent accusation concerns hypocrisy, inconsistency, and declared contradictions (103 comments), signaling heightened collective sensitivity to the moral-discursive coherence of authority. The exposure of these hypocrisies and contradictions functions as a justificatory logic—a way of demonstrating the legitimacy of a critical stance toward the state and, ultimately, of contesting its legitimacy. Because these accusations are articulated not through direct insults but through the exposure and signaling of inconsistencies, they also enable a discursive construction of the citizen as vigilant, informed, knowledgeable, and evaluative.

Second in the hierarchy is incompetence and lack of expertise (61 comments). This dimension reveals an epistemic rupture between the state and citizens, whereby authority is no longer recognized as a valid source of knowledge. In the digital era, where information circulates horizontally and rapidly, citizens claim their own sources and modes of validation. Commenters not only refuse to accept information provided by authorities, but counter it with their own observations or arguments discursively constructed as common-sense or elementary, in contrast with authorities portrayed as lacking basic logic or as intentionally manipulative.

The third most frequent trait consists of accusations of corruption and hidden interests, formulated in a tone permeated by suspicion and conspiratorial thinking (43 comments). Through these comments, authorities are constructed not merely as lacking moral competence, but as acting in bad faith with respect to their public mission. The image that emerges is that of a state captured by private or foreign interests. Accusations in this register are unfalsifiable and construct the state, in its current form, as irreparable and irredeemably corrupt. The image of the commenter oscillates between denunciatory and resigned cynicism.

Next follow traits of insensitivity, passivity, and cynicism (40), targeting the lack of empathetic and solidaristic responses from authorities. These critiques portray authorities as indifferent and cynical toward human suffering, relying on an emotionally charged construction. When authorities are perceived as distant, insensitive, or contemptuous, discursive reactions take the form of moral outrage or sarcasm.

Accusations of authoritarianism, dictatorship, and abuse (32) are linked to pandemic restrictions and regulations, interpreted as attempts at total control, often through analogies with historically repressive regimes. While exaggerated, such associations reflect structural distrust in the regime’s intentions and a perception of the state as a potentially oppressive entity.

A distinct indicator concerns absurd decisions and the inability to anticipate consequences (23 comments), reflecting denunciations of bureaucratic logic or lack of vision. Here, commenters accuse a rupture between rules and everyday reality, constructing authorities as incapable of systemic thinking or of evaluating the concrete impact of their measures.

The final category—insults, threats, and reproaches (130 comments)—does not allow for the clear extraction of specific traits. Most of these expressions are moral in nature, although some address issues of competence. They signal an affective overload of the public sphere, in which discourse becomes a vehicle for emotional release rather than argumentation. This form of expression indicates, to varying degrees, the most aggressive form of moral and epistemic disengagement from authorities and their decisions.

The criticisms directed at authorities thus reveal the normative structure of the state–citizen relationship, as well as the expectations projected onto authorities and perceived as violated. The most important expectation concerns the state’s moral and discursive coherence: hypocrisy and inconsistency are signaled not merely as errors, and their denunciation becomes a discursive form of citizen self-legitimation as lucid, vigilant, and critical actors. A second essential expectation concerns the epistemic and administrative competence of the state; its absence leads to the contestation of authority in the domain of knowledge, within a climate in which citizens prefer their own common-sense evaluations over institutional authority. A third dimension is good faith: its violation through accusations of corruption and hidden interests reconfigures the state as fundamentally illegitimate and as betraying its public mission. Other implicit norms—such as empathy toward citizens’ suffering, respect for rights, and the rationality of administrative decisions—are likewise perceived as having been breached.

4. Discussion

Through the present study, I analyzed critical discourses addressed to authorities in the online space in the context of the COVID-19 pandemic in Romania, drawing on a corpus of 1094 post-coded comments. Of these, 509 comments express, across various registers, opposition to administrative and political authorities. In the final stage of discourse analysis and the identification of the main interpretative repertoires, 457 comments were retained, taking the form of moral, epistemic, inflammatory, or sarcastic critical address.

The analysis was grounded in the assumption that online critical discourses (comments) function not merely as emotional outlets, but also as modes of articulating the expected moral order and, consequently, as practices of delegitimation and of asserting moral and epistemic positioning. From the perspective of discourse analysis and discursive psychology (Potter and Wetherell 1987; Edwards 2005; Tileagă 2018), online comments are understood as social actions through which, in this case, citizens evaluate institutions, formulate accusations, justify indignation, or claim moral and epistemic superiority. The table presented earlier, which orders the traits discursively attributed to authorities, provides an empirical perspective on the structure of critique.

Furthermore, critical discourses are organized around interpretative repertoires—relatively stable semantic structures that offer speakers culturally shared ways of explaining and evaluating authority. Within the analyzed corpus, the dominant repertoires are those of hypocrisy, incompetence, corruption, dictatorship, and institutional indifference. The findings show that hypocrisy and inconsistency are perceived as central flaws, followed by incompetence, corruption, and lack of empathy. This hierarchy reflects a cultural grammar of legitimacy. The ordering of traits attributed to authorities mirrors the moral priorities of the Romanian public sphere: coherence, honesty, and competence as expected norms. These repertoires indicate that critical discourse, as a performative form of moral regulation, defines a moral order (Tileagă 2018).

Insulting and inflammatory comments express direct, unfiltered aggression which, according to Suler (2004), is facilitated by the “online disinhibition effect”: the absence of immediate consequences, anonymity, and the lack of nonverbal cues encourage extreme expressions of negative emotion. These comments function as “speech acts” of total rejection of authority, without any claim to argumentation. By contrast, epistemic and moral critical discourses—whether direct or indirect, through sarcasm—perform functions of exposure and denunciation through which users assume a position of cognitive and moral superiority.

The forms of critical expression identified in the corpus can be situated productively within the literature on online incivility, though that literature’s own internal distinctions are necessary for an accurate positioning. From the perspective of Papacharissi (2004), who defines incivility as discourse that threatens the collective values of democratic deliberation—through hostile stereotyping, denial of rights, or attacks on individual freedoms—the repertoires identified here would not qualify as genuinely uncivil. The dominant forms of criticism—moral accusations of hypocrisy and corruption, epistemic challenges to institutional competence, sarcastic constructions that ridicule authority—are directed at the functioning of specific institutions within a democratic order, not at the legitimacy of that order itself. No anti-democratic repertoires were identified in the corpus; the commenting citizen constructs themselves precisely as the defender of norms that institutions are accused of violating in the here and now. Rossini’s (2020) distinction between incivility as verbal hostility and intolerance as the discursive denial of rights and pluralistic participation maps onto the corpus more precisely: many comments are incivil in Rossini’s sense—marked by aggressive tone, insults, and inflammatory language—without crossing into intolerance. More broadly, these findings point to a limitation of vocabulary-based definitions of incivility: a discourse-analytic perspective reveals that critical and hostile positionings can be constructed without recourse to explicit insults, through sarcasm, irony, or elaborate moral framing—forms that would evade tone-based operationalizations but perform the same delegitimizing functions. Context, in other words, is constitutive of the hostile character of an utterance, not merely its background. Taken together, these frameworks suggest that the hostility documented in the corpus is better understood as discursive delegitimization of concrete institutional actors—structured, performative, and normatively grounded—than as a pathology of democratic communication. The online comment space thus emerges not as an arena of anti-democratic incivility, but as a site of aggressive, context-bound contestation of institutional authority in a specific crisis moment.

This contestation, however, carries a deeper tension: citizens’ critical positionings emerge from acute dissatisfaction with institutional performance and express, simultaneously, a democratic claim—that institutions ought to recognize the competence and judgment of ordinary citizens. Where that recognition fails, the demand turns into delegitimization.

Thus, critical discourses do not amount merely to moral or epistemic judgments or to the withdrawal of trust; they become acts of discursive delegitimation in which the social contract is called into question, transforming authority from a legitimate source of decision-making into an illegitimate, incompetent, corrupt, or abusive entity. The study of these comments highlights the emergence of a register of contestation in which citizens construct their own moral and epistemic authority in opposition to the state. The space of online comments becomes a “site of struggle for hegemony” (Fairclough 1992) within the Romanian political context, not along class lines, but rather along an anti-system axis—state versus citizen—through which citizens resemanticize power relations in favor of their own moral and epistemic positioning.

The analysis shows that the discursive delegitimation of authorities and the identity reconstruction of the citizen are complementary processes. Moralizing repertoires (hypocrisy, corruption, abuse) delimit the space of a moral order in which the state is constructed as a deviant actor, while the citizen is positioned as the bearer of the norm. Epistemic repertoires (incompetence, manipulation, incoherence) define an order of knowledge in which authorities lose epistemic authority and the public claims competence and lucidity. Through the combination of these two registers—moral and epistemic—a dynamic of delegitimation takes shape in which institutional authority is simultaneously discredited and replaced by the critical citizen.

Thus, beyond the delegitimizing image of authorities, the corpus of analyzed comments also constructs a complex discursive portrait of the commenting citizen. This citizen (self-)positions as superior to authority—morally, cognitively, and experientially. In many comments, the citizen is constructed as vigilant, rational, knowledgeable about law, logic, and everyday reality, capable of identifying and denouncing hypocrisy, negligence, and contradictions on the part of authorities. This positioning is particularly visible in epistemic critical comments, where arguments are often supported by concrete examples, “common-sense” observations, or elaborate ironies that presuppose a clear cognitive distance from the government and its experts.

On the moral dimension, the citizen is discursively self-constructed as the holder of a clear ethical stance. They sanction the immorality of authorities through reproaches, appeals to shame, curses, or implicit condemnations, constructing themselves as the guardian of elementary moral norms. In the context of moral sarcasm and conspiratorial comments, this sense of superiority becomes even more pronounced, doubled by a fatalistic rejection of the entire political system, described as unreformable.

Another important register is that of the ironic or sarcastic citizen who ridicules authorities. This ironic citizen is cynical, yet lucid and suspicious. The use of irony, absurd scenarios, or linguistic parody places them in the position of an “intelligent observer” of perceived absurdities, reinforcing a stance of critical distance and reflexive awareness.

5. Conclusions

Online comments from the pandemic period outline a profoundly delegitimizing image of authorities, which are variously described as hypocritical, incompetent, abusive, or motivated by obscure interests. Through direct accusations, moral and epistemic critiques, as well as sarcasm or explicit hostility, users not only challenge institutional decisions but also call into question the integrity and capacity of authorities to act in the public interest. These forms of public evaluation reveal a major rupture between citizens and institutions and show how the online space becomes a central arena for the articulation and consolidation of delegitimizing repertoires in times of crisis.

The repertoires identified and grouped within the Romanian public sphere describe both the intensity of delegitimation and the normative forms and axes through which it is produced. Future crises may reactivate similar interpretative frameworks, making their analysis relevant for assessing democratic resilience in Romania.

Beyond the delegitimation of authorities, the analysis also highlights a process of discursive self-construction of the commenting citizen. Commenters do not merely denounce the deficiencies of the state; they also discursively construct their own identity as moral and epistemic actors, superior to the institutions they criticize. They position themselves as lucid subjects, capable of seeing the “truth” ignored or distorted by authorities, endowed with clarity, intelligence, vigilance, and an ethic of “normality.”

In terms of consequences, the findings point to profound forms of civic decoupling and disengagement, reflected in a wide range of delegitimizing discourses and repertoires. In the long term, the proliferation of such forms of discourse may undermine the effectiveness of public policies, especially in crisis situations where institutional trust is crucial. The present study offers not only a mapping of this crisis of legitimacy but also highlights the necessity of restoring a minimal level of trust between citizens and institutions in a context marked by fragmentation and epistemic skepticism. The analysis underscores the need for public communication that addresses the main areas of criticism: moral inconsistencies, lack of transparency, unclear decision-making, and epistemic incoherence.

At a still hypothetical but highly plausible level, the delegitimizing repertoires identified in this analysis have implications beyond the pandemic context, as the discursive patterns through which citizens reject authority can also be observed in the dynamics of political extremism’s resurgence in Europe and beyond.

6. Limitations

The validity and accuracy of discourse analysis are inevitably constrained by the depth with which context can be understood and reconstructed. Analytical efforts focus on extracting relevant contextual cues, a process that can never be fully exhaustive. The context necessary for interpretation must often be reconstructed by the researcher on the basis of knowledge and investigation of the phenomenon under study. By connecting the analysis to the realities of the Romanian cultural space and through careful reading and examination of comments, I sought to estimate as accurately as possible their meaning in relation both to the immediate context (the specific news item, surrounding comments, etc.) and to the broader context of the crisis situation.

The corpus reflects only a specific moment in public debate, marked by uncertainty and social tension, which may limit the generalizability of the findings both temporally and across different types of situations in which similar dynamics might emerge. A further limitation concerns the coding procedure. The entire corpus was coded by a single researcher, which precludes formal verification of consistency through inter-rater agreement measures. While iterative review cycles and saturation-based category finalization were employed to ensure internal coherence, and while single-coder analysis is epistemologically consistent with the interpretative orientation of Discourse Analysis, the absence of a second coder limits the replicability of the findings. Nevertheless, by combining a quantitative approach with discourse analysis, the study aimed to achieve an interpretative triangulation, whereby statistically observable trends were deepened through their discursive contexts. Future research could extend the analysis to other types of crises (economic, ecological, electoral) or to different digital spaces in order to trace the persistence and transformation of critical repertoires and would benefit from collaborative coding designs that could strengthen the replicability of the analytical framework developed here.