Introduction

This article is part of the Special Issue “Corona Truth Wars” guest edited by Jaron Harambam and Ehler Voss.

Until 2019, I felt free in Denmark, and I used to have a lot of trust in the authorities and thought that they all did what was best for us. And then, after COVID, that has just been turned upside down. Oh God, it’s a completely different worldview that I have now. I thought this was a science-based country where the path taken was based on science and evidence. And then I realized, oh my, there is propaganda in Denmark. There is censorship, and wow, I really had not thought that.

(Tina, 51, unvaccinated , September 2021).

Tina’s world is changed as she witnesses the data-political spectacle of the pandemic. According to Cambridge Dictionary, “spectacle” means both an “unusual or unpredicted event or situation that attracts attention, interest or disapproval” and “a public event or show that is exciting to watch” (Cambridge Dictionary 2022). We consider the pandemic a data-political spectacle in both these senses: an unexpected virus which came to change everyday life and placed data center stage in a performance attracting attention as well as (dis)approval from its audience. Tina exemplifies a particular reaction to the data-political spectacle. She also represents a new phenomenon in Denmark: a person who has not previously been politically active, but who—during the COVID-19 pandemic—has come to rethink her relationship with the authorities and the information they convey.

The COVID-19 pandemic has moved public health science and population data to the political center stage, but it has also shown that science and data do not quell conflict and doubt. Rather, data make room for new types of politics and invigorate new political actors. The aim of this article is to better understand this data political phenomenon. Based on ethnographic material and interviews with members of the general public who either support or disagree with governmental restrictions and vaccination policies, this article explores how the pandemic has interacted with what people count as true, and their reasons for subscribing to particular “truths.” We have conducted our fieldwork in Denmark—a small, peaceful, democratic welfare state in Northern Europe that has managed the pandemic with relatively low death tolls and relatively high degrees of personal freedom and of public satisfaction with the political handling of the pandemic. We use the reflections and experiences of both proponents and opponents of governmental COVID-19 measures to rethink how we may have to understand data politics and the use of numbers in governance, even beyond Danish borders.

Data politics has often been seen as a neutralizing form of power. Important work in Science and Technology Studies (STS), anthropology, sociology, and critical data studies have shown how data can be used to conceal political priorities and present political choices as governed by “necessity” or superior knowledge (Adams 2016; Espeland and Stevens 1998; 2008). The COVID-19 pandemic has challenged this conceptualization. As data came to inform politics on an unprecedented scale, ordinary citizens began analyzing data on a daily basis. The impact of this attention was not neutralizing. Even the most basic numbers, such as how many people were infected or hospitalized, were questioned. Society was not united in the combat against the virus: it was gradually divided into groups reading the same data in almost opposite ways. Gradually, some citizens came to see themselves as belonging to something akin to a resistance army, a group working to liberate society from the types of politics defined by the authorities’ use of data.

The level of societal division between people identifying themselves as opponents or supporters of pandemic measures will be seen by some foreign observers as relatively innocent in Denmark compared to the opposition in, for example, Canada or the Netherlands, where groups of opponents barricaded border crossings and sometimes violently fought their case (BBC 2022, 2021). Nevertheless, or perhaps therefore, we believe there is something to be learned from the Danish experience. It has still been possible to look for ways of bridging the divide. Denmark is a small multiparty welfare state in Northern Europe with 5.85 million inhabitants. It is a highly digitalized society with an elaborate data infrastructure and data integration—a society celebrating data (Hoeyer 2023). Even so, the pandemic moved data politics to the foreground of ordinary Danes’ lives in unprecedented ways. Denmark was among the very earliest European countries—both in terms of calendar date and number of infections per 100,000—to implement a national lockdown (Tænketanken Europa 2022; Plümper and Neumayer 2020). Throughout the pandemic the approach has been contrasted to the neighbouring country, Sweden, with fewer restrictions and higher death tolls (Mishra et al. 2021; Johns Hopkins University 2022). The lockdown was announced before there was any significant number of infections. It was decided by the Prime Minister’s office, and overruled the more moderate recommendation of the health authorities (Petersen 2021; Boswell et al. 2021). Lockdowns have taken many different forms, and the Danish version was relatively mild. It never involved curfew, people could gather with friends and family throughout (within various numerical limits), and the right to demonstrate was never questioned. At the beginning of the pandemic, there was broad agreement inside Parliament with the measures taken (Boswell et al. 2021) and in international surveys, the Danes have several times during the pandemic shown themselves to be, as a population, (among) the most satisfied with their government’s handling of the pandemic (Devlin and Connaughton 2020; Petersen and Roepstorff 2021). However, as time passed, levels of disagreement with government decisions began to rise both inside and outside Parliament (Boswell et al. 2021). According to a Danish survey, around 30% of the population in 2021 thought the government had gone too far in their pandemic response and around 20% feared for their democratic rights. However, support for public protests remained relatively low (around 15%) (Petersen and Roepstorff 2021).

The Danish route has been criticized for being both too strict and too lax. Denmark was quick to lock down in March 2020, yet also—to the best of our knowledge—the first country to downgrade COVID-19 (in September 2021). Then on February 1, 2022, just after the UK downgraded the pandemic, Denmark again did the same, lifting restrictions despite high infection numbers because the Omicron variant caused relatively mild disease and did not pose a threat to the hospital system. In April 2021, Danish Health Authorities permanently removed two vaccines from Denmark’s mass vaccination program based on identification of severe, but rare, side effects (Pottegård et al. 2021). This increased the support for the vaccination program (it remained relatively high and by February 2022, 81% of the population had been vaccinated) (Hoeyer 2023), but the decision to remove these two vaccines from the program was questioned by people eager to get a COVID vaccination as quickly as possible. Others wanted to halt vaccinations altogether. Again, the path taken was seen by some as too lax and others as too strict.

We use the Danish experience to explore the games of truth and power through which the pandemic has fueled a societal division. We believe a better understanding of these dynamics is important for the future of the health services, as well as for our understanding of data politics. We begin by outlining our position with respect to existing work on data politics, data activism and the contestation of “truth” during the pandemic. We then outline our methods before presenting our analysis in four analytical sections dealing with: the pandemic as a data-political spectacle; the everyday tactics of living during the pandemic; how moral and political assessments of danger and desirability interact with perceptions of “truth”; and how activism is fueled through both offline and online interaction.

Data Politics, Data Activism and Pandemic Games of Truth and Power

Data have become ever more central to commerce (Zuboff 2019), science (Leonelli 2016) and politics (Madsen et al. 2016) in the past decade. Along with political and economic interest in data, an important form of critical scholarship has emerged, known as, for example, critical (big) data studies (Iliadis and Russo 2016; Wyatt 2021) and the anthropology of (big) data (Douglas-Jones, Walford, and Seaver 2021; Levin 2019). A common thread running through this work has been the unpacking of the norms and politics embedded in data, which are otherwise presented as neutral and unquestionable. Evelyn Ruppert and colleagues use the term “data politics” to point to the way power is materially engrained in data infrastructures that mediate the lives and rights of citizens (Ruppert et al. 2017). They emphasize the performative power of data: the ability of data to conjure the objects and relations that they are said to portray (Bigo et al. 2019). Data politics is affiliated with the social dynamics of quantification. Quantification involves some form of commensuration where differences are equalized, and things are turned into “types” that can be counted (Espeland and Stevens 1998). By leaving out differences, certain types of ignorance become possible. This can be conducive to governance. The emphasis on metrics in contemporary governance has been criticized for concealing not only differences, but also the values and political choices that go into the making of a particular data representation (Adams 2016; Mau 2019). The datafication of politics thereby becomes a form of “anti-politics machine” that conceals political priorities (Ferguson 1994), and it becomes an important task to uncover the values hidden in everyday data practices (Ruppert and Scheel 2021).

Data have also become subject to various forms of activism. Beraldo and Milan have united STS and Social Movement Studies and proposed the concepts of contentious data politics and data activism (2019). Data activism is the field in which data “re-mediate activism” (Beraldo and Milan 2019: 2). To understand the data-political spectacle, we need to understand how people interpret and employ pandemic data. Data activism involves three aspects, emphasizing: the practice in data activism; the dual understanding of data as offering both opportunity and risk; and the importance of infrastructure (Beraldo and Milan 2019: 5). Beraldo and Milan argue for understanding data activism along two analytical dimensions. The first dimension is from data as “stakes” (as objects of political struggle) to data as “repertoires” (as tools for political struggle), and the second dimension is from individual practice to collective action. In our analysis we demonstrate both the individual data interpretation practices and the collective action taking place in relation to public life and public good, and how data are both stakes and repertoires—sometimes simultaneously.

Shelton (2020) sees what he terms the “post-truth pandemic” as illustrative of the relationship between a data-centric society and post-truth, arguing: “Indeed, we might see a post-truth society as the logical outgrowth of the data-driven (or, perhaps more accurately, data-centric) society we have been living in in recent years, where individualized, decontextualized data points serve as the focal point for social and political life” (2020:3). Shelton suggests that during the pandemic, data have fed into “meta-narratives”(Shelton 2020) or what Lee and colleagues call “deep-stories”(Lee et al. 2021). These narratives tap into people’s political preference and individual and collective experience (Ogola 2021; Prasad 2021). Rather than operating in a dichotomous space of truth and falsehood (Harambam 2021), the task for STS in this situation is to explore what Ogola (2021) calls the ‘ethico-normative struggles’ of everyday life. We add to the study of the role of data in people’s pandemic lives by taking a closer look at pandemic data interpretation, practice and mobilization than do previous quantitative and online observational studies (Lee et al. 2021; Shelton 2020; Ogola 2021; Lu et al. 2021). Some research has looked at COVID-19 skepticism as science skepticism and investigated this in relation to other forms of science skepticism, as well as in relation to social characteristics and political ideology (Scheitle and Corcoran 2021; Rutjens et al. 2021). Others have argued that it is wrong to see opposition towards, for example, vaccines as science skepticism. Goldenberg (2021) finds that so-called skeptics often spend more time and effort on becoming scientifically literate than others. Goldenberg (2021) suggests exploring what people think without assuming particular elements of the public to be less rational than others. As we take up her advice, we simultaneously follow an old invitation from the field of public understanding of science that seeks to avoid measuring “publics” against a deficit of scientific understanding (Wynne 1992). Once we approach opponents and supporters symmetrically, we can see how both opponents and supporters are fiercely concerned about “truth,” though they arrive at different conclusions.

Resistance against government responses to COVID-19 have often been connected with conspiracy theories (e.g. Gruzd and Mai 2020; Islam et al. 2020; Fuchs 2021; Prasad 2021). We do not seek to theorize conspiracy theories and we do not categorize those of our informants who oppose restrictions or COVID-19 vaccinations en bloc as conspiracists. Opponents, just as supporters, are very diverse. They become a “group” through naming rather than homogeneity (Anderson 1991; Harding 2016). Fuchs suggests viewing conspiracy theories as a reaction to alienation (Fuchs 2021: 69). But we, rather than explaining people’s beliefs, seek to understand the values underpinning these beliefs; and rather than suggesting labels for the opposition, we look for similarities and differences across people supporting and opposing official policies and data interpretations.

Methods

At the beginning of the pandemic, things developed rapidly. When on March 11, 2020, Danish Prime Minister Mette Frederiksen announced a general lockdown, it came as a surprise—even a shock—to most citizens. We decided this historic moment needed to be documented. Hoeyer therefore, in collaboration with epidemiologists at our department, initiated a collection of quantitative and qualitative material (Clotworthy et al. 2021). A questionnaire was used in the general population, and 37 of the respondents (22 females, 15 males) agreed to also be telephone-interviewed qualitatively (six of them twice) about their COVID-19 experiences. In this article we are working only with the qualitative parts of this material, collected between March 2020 and February 2021. We refer to these 37 respondents as the “survey-recruited subsample.”

As more people started protesting against the government’s pandemic response, it became clear to us that we would not get to talk to the people who were protesting in the streets and online by recruiting through the questionnaire only. Therefore, we started ethnographic fieldwork directed at restrictions resistance. This fieldwork involved online observation primarily in three Danish Facebook forums centered around resistance towards restrictions, as well as through the profile of a Danish doctor, Yvonne [pseudonym], who was consistently referenced by informants encountered during demonstrations and in the Facebook forums. We also conducted offline participant observation (street demonstrations) as well as interviews with 11 individuals (six females, five males). We interviewed two of these online and nine in-person, which was legal throughout the period and done to respect the preferences of the interviewees. These 11 individuals who were recruited during street demonstrations and through network and snowball sampling considered themselves critical towards the government’s handling of the pandemic —and are here referred to as the “restrictions resistance subsample.”

When vaccines were introduced, new forms of opposition evolved. Here again á Rogvi conducted online participant observation focusing on vaccines, for instance, through inviting people in a Danish vaccine-critical forum on Facebook to share, through comments, their experiences of mass vaccination. She also recruited for interview eight individuals (seven females, one male) who had refused COVID vaccination. We refer to these interviewees as the “unvaccinated subsample.” Some in the “unvaccinated subsample” were also more generally critical of the COVID-19 response, but not all. We deliberately refrain from describing the latter as “anti-vaccine” or “anti-vaxxers” as they also include people who are not generally against vaccines.

The three subsamples represent how interviewees were recruited, and therefore should not be considered distinct groups existing prior to our categorization. There are many degrees of resistance to the governmental COVID-19 response. The sense of “two groups,” for and against, is a rhetorical figure looming large among our informants, but we have come to see people’s sense of facing an “Other” as a product of societal division rather than a label for homogenous groups. When talking to informants regardless of their position towards COVID-19 governance, we assumed the role of an interested audience. Symmetry in relation to post-truth and conspiracy has been debated within STS (Sismondo 2017; Lynch 2017; Holman 2020). To us, symmetry meant approaching informants without a priori looking for different kinds of explanations dependent on the person’s (dis)agreement with the COVID-19 response. We did not correct what we believe were misunderstandings. We wanted to understand what the data spectacle looked like from the position of interviewees, not to evaluate their perception against our own. Finally, as the role played by media was mentioned several times by our informants, we interviewed editors at two national media channels about their COVID-19 data representations in June 2021. Figure 1 gives an overview of the components of our study.

Before each interview, interviewees were informed that the study purpose was to understand their views of and experiences with the pandemic, that they would remain anonymous and retain a right to withdraw. Interviews were transcribed verbatim and Facebook posts were saved concurrently. Interviewees, street demonstrators and the medical doctor we followed on Facebook are given pseudonyms. Translations from Danish were made by the authors. All transcripts were read by both authors who discussed themes and formulated questions used for subsequent thematic coding (Madden 2010). In writing the article we moved back and forth between these codes and the research questions and theories used to theorize our material in an abductive analysis (Timmermans and Tavory 2012).

Fig. 1
figure 1

– Study components

Full size image

The Pandemic as a Data-Political Spectacle

Data are omnipresent in the discursive problematization of COVID-19 (Gjerde 2021; Vandendriessche 2020): it has been a data spectacle. Representation – in the form of data – play a key role in this spectacle (DeBord 1995). The pandemic threat was first presented through models estimating the potential death tolls of the pandemic (Kreps and Kriner 2020; Caduff 2020). The predictive models were, however, initially based on limited knowledge because the virus was new (Post, Bienzeisler, and Lohöfener 2021). The pandemic came to be monitored through surveillance data, which were also used for new predictions and policy adjustment: numbers of people infected (when and where tests became available and prioritized), numbers admitted to hospitals, numbers in intensive care units, numbers on ventilators. Different countries and media foregrounded different numbers, sometimes depending on data availability, sometimes on political prioritization (Caduff 2020).

In this section we illustrate how data are used by the Danish authorities to frame interventions and show that citizens across the subsamples support an ideal of data-based governance. In government press conferences Prime Minister Mette Frederiksen, Minister of Health Magnus Heunicke, or heads of different health authorities presented detailed surveillance data (Statsministeriet [Prime Ministry] 2022; Villadsen 2020). The health authorities were delivering daily updates in the form of data to the Prime Minister’s office (Lehmann 2021). This underscores the omnipresence of data in the pandemic response: in political planning, management and communication. Data were also widely disseminated to the Danish population through official websites and Danish media (TV, print, online) including dashboards and particular sites dedicated to COVID-19 surveillance overviews. According to our interviewees from Danish media, online articles providing overviews of COVID-19 data had, by the time of interview, been by far the most visited articles since the public radio/TV stations began registering traffic on their websites. Our media interviewees also explained that COVID-19 data presentations had given rise to more reader reactions than any other news.

Among all informants an ideal of science and data-based governance existed. Whereas in Tanzania, opposition has been associated with opposition to Western (positivist) science (Ogola 2021), this is not the case in Denmark. Informants across the subsamples agreed that interventions ought to be based on data. In the survey-recruited subsample, people reproduced the connection between the rise in numbers of infections and the need for further restrictions as presented by government and authorities. Andreas (76, survey-recruited subsample, April 2020), for example, valued the Danish pandemic response and compared it to Sweden, where high numbers of infections and deaths told him “what is in reality happening over there” with Sweden’s “nonchalant” approach to the pandemic. Interestingly, Andreas presented his political position as generally critical towards state interference, and as thinking that many things should be up to the individual. Still, he agreed with the Danish government’s approach and considered the numbers of infections and deaths as the truth about how things were going in Sweden, a situation which terrified him.

Interestingly, the restrictions resistance and unvaccinated subsamples seemed even more dedicated to evidence and “truth” than those from the survey-recruited subsample. Malik (aged around 50, restrictions resistance subsample, January 2021), who had a long history of activism, explained his belief in reason in this way:

What we should aim at is to learn from history. For the first 2,000 years we used religion to govern humanity worldwide. In the previous 100 years, the world’s been governed by politics. And it’s gone wrong both ways. People have been split up into [political] parties. People have been split up into religious convictions. And then they can no longer agree. Now we have to enter a time where reason prevails. And this reason must be global (…) then we’ll no longer need parties or religion to govern us.

The doctor Yvonne’s profile is also an example of someone among the critical voices who cherishes evidence. In a video (in Danish) posted in Spring 2021, when Denmark was still under its second lockdown, and which three days after its publication had more than 16,000 views, Yvonne stated:

This is a farce. It’s [just] a story (…) We could open the country this very moment. (…) If we based it on facts and not fear. (…) Here on social media, we have to have a counterweight to the press (…) which has to an extreme degree censored me and my facts.

She ended the video by creating a community around herself and stated: “You know that I am right. Because I focus on facts.” Yvonne articulated the ideal of data and facts in all her COVID-19-related video and posts, which often included diagrams and pandemic data visualizations. Yvonne often used official data taken directly from the public institution Statistics Denmark, but presented them differently, something that Lee and colleagues also demonstrate among American anti-mask online activists (Lee et al. 2021). We see that both Malik and Yvonne preferred reason and data as bases of governance, which they saw in opposition to current (and previous) governance that they characterized as either politics and religion (Malik), or a farce and a story (Yvonne)—things they clearly did not value. They upheld a positivist ideal, but like Tina mentioned above, they were disappointed. Pelle (aged around 40, restrictions resistance subsample, January 2021) who was interviewed just before his participation in a demonstration, described it in this way: “I’ve found out that truth is irrelevant. Truth has died. It doesn’t matter how much evidence you have. (…) It won’t help, if nobody pays attention to it.” As the data spectacle unfolded, some observers lost faith.

The Everyday Tactics of Living in a Data-Political Spectacle

In this section, we show how data and data politics were read and interpreted as part of daily living during the pandemic. We show that informants across all subsamples came to doubt the meaning of pandemic data. However, the ambiguity of data did different things for people as they planned their own response. People differed in what de Certeau (1984) has called ‘everyday tactics’: the degree to which they questioned, tinkered with, and resisted pandemic data governance.

Across the three subsamples, people questioned whether the people infected with SARS-CoV-2 who had died in Denmark died of or with COVID-19. Not even doctors agreed on how to count one and the other (Kielgast, Hecklen, and Møller 2021). Mathias (83-year-old, survey-recruited, first interview, April 2020) wondered: “I might be wrong, but (…) many die of [old] age, some of cardiovascular disease, blood clots, and what not. They have to go anyway, so it need not be COVID’s fault.” John (60, restrictions resistance, May 2021), like Mathias, questioned the validity of the number of deaths—whether people died from COVID-19 or “would have died anyway.” He criticized what he saw as lack of contextualization of mortality data. For him, this could have changed the picture of what was an appropriate pandemic response.

People across the three subsamples also problematized the numbers of people infected as these were related to the numbers of those tested. The positive percentage is the percentage of PCR-tested whose test has shown positive for COVID-19, and it is used to balance assessment when there are fluctuations in numbers of tests. The positive percentage is an interesting example of how some opponents trust the validity of official data, but disagree with its political meaning and consequences; some online activists accept the number, but reverse it to direct attention to the high percentage—in some periods 99%—of PCR-tested who were not infected by SARS-CoV-2.

People from the survey-recruited subsample overall believed that government measures were based on science, and trusted authorities and government to know and understand the pandemic situation and to respond appropriately. Barbara was one such person who relied on the government’s expertise to interpret data:

I follow the government’s advice and restrictions and so on because I assume they know what they’re talking about. And I assume that there’s a reason that they’ve applied them. (…) And whether there are 300 or 280 admitted to hospital, it makes no difference to me. I don’t have enough expertise to tell the difference.

Barbara (27, survey recruited, February 2021) 

Like the majority of the people in the survey-recruited subsample, she had decided to trust the government. In contrast, the individuals in the restrictions resistance and unvaccinated subsamples wished to see for themselves, or expressed “constructive informational needs” (Post, Bienzeisler and Lohöfener 2021). This led to fierce engagement with data, data analysis and interpretation. Some spent hours searching for information that they did not find in the “mainstream media.” They often referred to this as “fact-checking” and “source-checking.” 53-year-old Lisa (restrictions resistance, March 2021) described how she searched for knowledge regarding COVID-19 and COVID-19 vaccines:

It’s my hobby to sit and search on PubMed and then I dig into the sources. I don’t have a university degree for nothing. […] And then I get a lot from Twitter, where I follow researchers with another position than the prevailing one.

The differences in tactics between opposers and supporters is clear when we counterpose Lisa with 54-year-old Anders from the survey-recruited subsample. When asked about whether he had been disagreeing with the COVID-19 response, Anders (54, survey-recruited subsample second interview, November 2020) responded:“No. I am super-naïve. When somebody says ‘Jump!’ I just ask, ‘How high?’ That hasn’t caused me trouble before.” For Anders, it felt safe to trust the authorities. For Lisa, and most of those opposing the official response, this was associated with danger. It called for scrutiny.

How Moral and Political Assessments of Danger and Desirability Interact with Perceptions of “Truth”

We have shown above that pandemic data were contested across the subsamples, but their ambiguity was settled in different ways. People framed the problem of the pandemic differently as to what kind of threat it posed to individuals and to society. Still, there are also similarities across the subsamples in relation to perceived danger: people feared losing agency. Among the interviewed supporters, however, most people thought of the virus as potentially depriving them of agency, while the opponents we interviewed thought of restrictions or vaccines as threats to agency. These differences often reflected previous life experiences.

Informants in the survey-recruited subsample did think of human and economic costs of lockdown and restrictions, but the virus remained their main concern. Some in the survey-recruited subsample thought of the danger personally: if they themselves (or their dear ones) fell ill of COVID-19. Katrine (65, survey-recruited subsample, second interview, November 2020) expressed how she valued protection against disease, and thereby biological life, above all:

I fully understand the lockdown and I don’t understand those who are upset, like really upset, about it. Well, I do understand, okay people lose some money or some salary or things like that. But what if you become ill and lose your life?

Already in March 2020, when little was known about the virus and Denmark was in its first lockdown, 64-year-old Karen (survey-recruited subsample) explained specifically how she feared getting very ill with COVID-19 and losing her agency: “Thinking about being admitted to an intensive care unit and lying there for three weeks, where I lose my agency. (…) Losing grip on one’s own life, I think that’s terrifying.” Others feared how the virus might overwhelm the health sector. These fears were in line with the official scenarios delivered at press conferences.

For people in the restrictions resistance subsample and some in the unvaccinated subsample, submission to unconstrained governmental power was the greater danger of the pandemic. To varying degrees, they saw the pandemic as a means to govern the people, to implement political changes or even to introduce a (Communist) dictatorship. The similarity with fear of losing agency mentioned above is intriguing. As Karen imagined her own possible experience of COVID-19, so was the lived experience of loss of freedom an element of Tatiana’s imagination. Tatiana (around 50, restrictions resistance) was interviewed during a demonstration in February 2021:

I’ve lived in a marriage where my boundaries were moved little by little, until I realized I was in fact living in a prison. And that’s the same experience I’m having now, that we’re little by little being trained to accept that we’ve been deprived of something, and all of a sudden we have nothing.

Similarly, 39-year old Mikkel (restrictions resistance, February 2021) saw the virus as a political weapon “making people do anything, and then concentrating more and more power and authority, and taking rights away from the citizens.” He also feared losing agency, but of a political nature. From our interviews it seems that in assessing the danger of the pandemic, personal biographies (violent marriage, previous experience with work rehabilitation authorities, having business or relatives negatively affected by restrictions) and data (lack of particular data and doubt about data predictions) intersect and form the building blocks of different worlds with different dangers. Biography and personal experience with external power and authority matter for the reading and interpretation of pandemic data, and thereby which pandemic truth one subscribes to (Ogola 2021; Prasad 2021).

The degree to which people fear power concentration, as well as whether they believe the pandemic is planned, varies. Beate was 24 years old, and truly believed the pandemic was planned. By whom specifically, however, she was not so sure—maybe a distant elite—an idea she shared with Mikkel and a few more informants submitting to more extreme conspiracy theories that, for instance, saw COVID vaccines as depopulation tools. Other informants in the restrictions resistance and unvaccinated subsamples referred to ideas of conspiracy or a sinister plan, but presented it as if they neither believed in it, nor fully rejected it. Beate (24, restrictions resistance, February 2021) elaborated on the relation between health data, the pandemic and a surveillance dystopia:

I believe that one day we’ll all have such a [silicone] chip in our hand, where the state can see 100% where we are, and what we do, all our data. And I don’t like that. It’s one thing that we have a CPR number (civil registration number), that’s okay. We have to. But then I don’t think we should be more owned than that. And I think that COVID is a step towards that.

Beate was here describing a connection between (health) data surveillance and being owned by the state. To her, COVID was a step towards an omnipresent state controlling every movement of the individual. Tatiana (around 50, restrictions resistance, February 2021) similarly said:

From today [with the new legal Act on Epidemics] we’re deprived of our ability to decide for our own body. Our body no longer belongs to us. Then it’s not “ours” anymore. Then it’s the state deciding what you’re allowed to use it for and what it should be subjected to. (…) We’re individual beings with our individual soul and mindset, and [yet] we’re treated like soldiers now. And that goes against who I am.

To understand the mobilization and activism around COVID-19 it is important that we recognize this experience of violation of the individual. Some people genuinely fear becoming a governmental tool, simply a means to an end, “soldiers.” Their fear of losing their civil rights inspires activism.

How Activism is Fueled Through Both Offline and Online Interaction

In this section we take a closer look at how people manifest their pandemic-related political stance in relation to society, and mobilize in collective action. Their activism operates at a spectrum from individual practice (or everyday tactics) to collective action (Beraldo and Milan 2019). We show that people agreeing with, and also those opposing, the government position are similar in their sense of duty or solidarity towards their fellow citizens. However, because of their different ideas about what constitutes the primary danger, their data activism takes different form and direction. Online and offline activism intersect, but social media and other news sources mediate the aspirations and activities of supporters and opponents differently.

Among people supporting the government’s pandemic response, several forms of activism quickly evolved. Some citizens set up Facebook groups where people could offer assistance to people afraid of entering shops. People’s words and actions quickly “went viral”: people were inspired to place signs in windows with either encouraging messages (“Everything will be good again”), or with hard-edged moralizing warnings (#StayTheFuckHome). Along many paths, painted stones stating “everything will be okay again” or other encouraging sentences began appearing, again illustrating a mundane form of activism.

The mobilization against the government’s COVID-19 handling initially also took a very grass-roots form. Mobilizing partly online, a diverse group began convening daily in front of the Danish Parliament, and having colorful and noisy parades through pedestrian streets, mainly in the capital, Copenhagen. They included a variety of people from both the political left and right. However, it was not until more violent demonstrations, such as the one referred to below, surfaced in December 2020 that the resistance to the government’s COVID-19 response appeared in the national Danish media. Attending one of these demonstrations, á Rogvi noted:

It is dark. There are a lot of torches. Someone rolls out a banner saying “RESISTANCE DRESSED IN BLACK” on the gravel. The group sings their version of Bella Ciao: “Mette [Danish Prime Minister’s given name] ciao.” Two men stand on a podium. One of them is talking about showing Copenhagen that resistance exists. He asks everybody to walk for those who have not shown up today but support the cause. A couple of kids sit on the shoulders of their parents. Most participants are men. Many are dressed in black and a few wear “anonymous” masks (…) The demonstration starts moving. Someone lights emergency flares and the smoke turns red. There is music playing in the street—something rarely heard these days due to the ban on public gathering.

(Fieldnote recorded at Men In Black arranged demonstration, January 2021)

In Denmark there has also been activism where citizens opposing the pandemic response have organized and, for instance, booked free PCR tests or vaccination slots without intending to attend, so as to block the system. It is something the activists themselves call “civil disobedience” (Lassen 2021). It is, however, not a conflict between right and left wing politics. In all subsamples we have met people from both sides of the political spectrum, though there might be a tendency for opponents in general to have moved away from the political center towards the poles.

The resistance to the pandemic response was, for some, considered an information war. In this information war, data are both the arms and the shields—the repertoires in Beraldo and Milan’s terms; and the land over which the war is fought—the stakes. During a demonstration in February 2021 against new legislation, the Epidemics Act, a speech by a ex-medical doctor, Tim, was transmitted from the phone of one of the organizers:

Everybody has to wake up and join the fight, which is indeed about information (…) We must tell people that there are fewer dead in 2020 than in 2018. Look at Statistics Denmark. There are fewer dead. So, we have no pandemic.

Here, in these words, the speaker frames resistance as a fight about information. Here data act as both repertoire and stakes. The numbers of deaths available on the public administrative institution Statistics Denmark’s webpage are seen as repertoires, or means, to awake the public. At the same time, these data, he reasons, lead to the conclusion that there is no pandemic. There is no public health problem. The idea of awakening others permeates activism against restrictions. For those engaged in this form of activism it is, no less than the encouraging messages on stones, an act of solidarity. It is something they do for the common good. Resistors often understand their activism as defending every citizen’s, also supporter’s, right to decide about their own body. And also, the right of those who have not yet “awoken.”

The pandemic moved—at least temporarily—much social life online. People spent hours in front of screens during lockdowns and restrictions. More time spent online or alone should be seen in relation to the vast amount of information about COVID-19 from unauthorized sources, something WHO has called an “infodemic,” defining this as “too much information including false or misleading information in digital and physical environments during a disease outbreak” (WHO 2022). It should also be seen in relation to online “echo-chambers” where people are exposed to information echoing their own position towards COVID-19 governance (Lang, Erickson, and Jing-Schmidt 2021). Supporters of pandemic measures have called for action to control the information circulated. Facebook, where we have followed mobilization and online activism, has implemented increasingly fierce policies against what is deemed “misinformation.” One of these policies includes a list of COVID-19 related claims, which Facebook remove from their platforms, because these claims are deemed false by “public health experts” (Facebook n.d.; Rosen 2021). Just as with the lack of reporting from demonstrations, many of our informants from the restrictions resistance and the unvaccinated subsamples considered these measures censorship or a limitation of free speech. Mikkel stated:

I used to have a group [on Facebook] called “Staying awake,” like the frame we all have. And I administrated that group with my old account [which was closed by Facebook], and it was obvious that Facebook in the end did not recommend it, and they got at people and gave them warnings, and then [came] that shit with fact checking and all that crap. (…) So, they suppress intentionally (…) But that’s just proof of how they are, if it doesn’t fit their narrative, it’s banned, suppressed, or not recommended.

(Mikkel, 39, restrictions resistance, February 2021).

After Mikkel had his account and posts deleted by Facebook, he devised an alternative strategy: instead of being part of resistance groups, and so as to be able to debate issues yet still be acceptable to Facebook, he built a new personal profile where he shared information. Mikkel only knew about 5% of the 1,500 Facebook friends on his new profile, but he added people whose Facebook photo had a predesigned (but personally chosen) frame like “Stay awake” or “Stop tyranny.” He continued:

I get around 20 requests a day. If in some public [Facebook] thread, people can see “That guy, we’re on the same team” (laughs) or something like that, [then] people have become good at accepting [my friend request] automatically, because people are, like, desperate to unite and stand up and wake up.

Through the frames mentioned above, it became possible for Mikkel to recognize rapidly if people were “on the same team” as he was. Based on this enactment of COVID-19-related political identity, he built a new community centered around his profile. Through our online participant observation, we also observed other ways of trying to bypass Facebook’s policies against misinformation on COVID-19. For example, creative spelling was used to keep Facebook from realizing that a post is about COVID-19, exemplified by misspelling in inventive ways like “va((ine” or “Co√id” or including emojis in spelling.

Another reaction to Facebook’s misinformation policy among informants from both the restrictions resistance subsample and unvaccinated subsample was to move their COVID-19 information activities and activism to other platforms, where they did not experience what they called “censorship.” They, like Mikkel, had the experience that the reason for the censorship was not whether what they shared was true or not, or was based on data or (alternative) experts’ claims, but if it fitted the “official” narrative. Interestingly, the informants from the restrictions resistance subsample and some of the informants from the unvaccinated subsample perceived the censorship on Facebook and in Danish media as a confirmation of their criticism. They saw the deletion of their own or others’ posts as a sign that they were right in their suspicion of power abuse. In this way, Facebook’s policies against misinformation, its attempts to mitigate harm and secure public health, may be interpreted in ways that push people who are skeptical towards COVID-19 governance even further away from the rest of society. It confirms their fears and pushes them onto other platforms, some of which have particular ideological inclinations that can further radicalize people’s lack of trust in authorities, government and formal media.

Conclusion

The COVID-19 pandemic has generated a strong division in Danish society and elsewhere, and high levels of activism. Our study is Danish, but we believe some of the social dynamics involved hold insights of more general relevance for our understanding of pandemic responses and data politics. The pandemic has been a social as much as a virological disturbance, and unless policymakers, authorities and citizens learn to bridge the new fault lines of conflict, the social turbulence is likely to continue its influence long after the virus has stopped being a threat. In some ways, people agreeing with the government approach and those opposing it speak as if they occupy different worlds. In other ways, however, opponents and supporters of governmental COVID-19 measures are strikingly similar. On both sides, people build their worlds with data. They tend to subscribe to similar epistemic values (a strong positivist leaning) while simultaneously having become much more aware of the doubts and ambiguities surrounding data and their interpretation. On both sides, there have been forms of activist mobilization and determined efforts to perform responsible citizenship. How come, then, that opponents and supporters differ to such a degree that they (as well as the media) refer to “two groups”? While people differ in multiple ways, whether or not they see themselves as opponents or supporters tends to reflect what they see as desirable, or dangerous, in society. Their moral and political stances are coproduced with their reading of data. Their perception of danger interacts with who they see as friend or foe, and thereby from whom they get their information. To (re-)connect with people who—like Tina from the opening paragraph—now disregard the health authorities, will be an important task for the future of health policy, and we propose that to succeed in that task, it will be necessary to understand the different positions symmetrically. This re-connection would also involve dealing with the “echo chambers” that are formed around positions towards COVID-19 (Lang, Erickson, and Jing-Schmidt 2021).

We—the authors of this paper—do not think that all “truths” are equally valid. To see knowledge as socially constructed is not the same as outright epistemological relativism (Danholt and Gad 2021). Just as a house is constructed and needs to be constructed well to be robust, vaccines and knowledge about viral infections are constructed and only when constructed well do they become biologically and socially robust. The scientific community generally stands a better chance of reaching the aim of biological robustness than skeptics mobilizing in protest. Still, the data practices of the scientific community are loaded with values (Ruppert and Scheel 2021). They are not always socially robust. Data practices generate particular forms of politics (Sætnan et al. 2018). They bring particular versions of the world into being (Bigo, Isin, and Ruppert 2019). People mobilizing against the scientific community are entitled to question these values. Only when the scientific community is in dialogue with people holding different values, will the scientific interpretations become also socially robust (Oreskes 2019; Bowker 2018). When basing politics on data practices that focus on viral danger, but not political, economic or social danger, the politics would probably fare better by being honest about its priorities than by claiming a monopoly on truth.

Addressing opponents only through their deficit of understanding will not make them listen. It will, conversely, direct them away, towards people whom they respect—and who they feel respect them. It fuels the societal division. To identify similarities between opponents and supporters of a strategy is one way of establishing a dialogue. Another way is to identify something worth respecting. Opponents might do things that supporters disagree with, but the values informing them are not necessarily wrong. Opponents generally value independence. They typically oppose suppression and guard autonomy. While opponents and supporters give different relative priority to viral risk and the risk of political suppression, most of the people we have encountered on both sides value both biological and political safety. The different priorities should not be unsurmountable if we see that we share at least some values, even when we disagree about how they should be prioritized.

When first encountering the communities of opponents to Denmark’s COVID-19 measures, it was —for both authors of this article—like going down a rabbit hole to an altogether other world. Rather than rejecting this other world altogether, we decided to look at opponents and their data interpretation as expressions of “ethico-normative struggles” (Ogola 2021), which bring the politics of data (back) to the fore. After all, as Harding has stated: “Science is politics by other means, and it also generates reliable information about the empirical world” (Harding 2016: 10). We still believe that epidemiologists know more about epidemiology than the opponents; that societies that reacted in time have managed the pandemic better than those led by people rejecting the epidemiological predictions; and that vaccine developers have sincerely tried to find a solution to the problem of a pandemic new virus, and that the fast development of vaccines is a victory for science. Still, we have also come to acknowledge that there is much to be learned from engaging in a dialogue with opponents. A dialogue helps elucidate the values of data politics. Harambam (2021) suggests establishing what he calls “deliberative citizen knowledge platforms” (much akin to the Danish tradition for technology assessments, (Horst 2021)) to facilitate this type of dialogue and Walsh has argued for the potential of diverse dialogue groups to improve connections across difference (Walsh 2007). Our point, however, is more directed at the obligation of scientists and authorities to declare the values informing their analyses, and for mainstream news media to report on the full spectrum of societal positions.

The pandemic data spectacle carries an important lesson for data politics. A common observation has been that quantification and data-based arguments serve as politically neutralizing instruments (Adams 2016; Espeland and Stevens 2008; 1998; Sætnan et al. 2018). However, the COVID-19 pandemic has shown how data can also open up new forms of contestation, for activism, for conflict. Data need interpretation. The more they impact people’s lives, the more people are likely to question them. There is no neutral use of data. Data analysis presupposes an intention, or what Dewey called “an inquiry for it” (Dewey 1998). This intention is always partly normative. Our understanding of data politics therefore stands in a different place since the arrival of the pandemic. This place is not one of “post-truth,” but a place where “truth-making” demands much more frankness and clarity about the values and assumptions that inform data analyses. Data do not speak for themselves. They can tell better stories—stories that encompass more people’s hopes and concerns—if authorities and scientists dare discuss the values that inform their work, and engage a respectful dialogue also with the people who do not share their training and values.