COVID-19 and U.S. Disputes Over Authority, 2020-2021: Implications for the Constructionist Analysis of Social Problems.

COVID-19 is very different from the cases typically studied by constructionist analysts of social problems: it emerged quickly, spread widely, and affected many aspects of social life. As such, it offers important opportunities to reconsider the constructionist model. We focus on three issues-metrics, masks, and vaccines-where COVID-19 disputes about authority led to different alliances among several categories of claimsmakers. Our point is that COVID-19 discourse seems far messier than most of the narratives presented by constructionist analysts, and we identify several lessons from this unusual contemporary case that might help us strengthen existing social problems theory.

INTRODUCTION

The emergence of COVID‐19 in the United States combined some distinctive elements. It spread rapidly and had a very broad impact. After first being identified in Wuhan, China in November 2019, it soon had a broad geographic range: by March 2020, cases had been reported in all 50 U.S. states, and it was known to be spreading across the globe. In addition, it quickly became clear that it would affect virtually every aspect of society and the lives of nearly all individuals. These characteristics of its onset, spread, and social impacts set COVID‐19 apart from most of the cases examined by sociologists studying the construction of social problems, which often are understood as having long histories involving relatively stable institutional relationships, and with effects concentrated in particular sectors of society. Because it is an atypical case, COVID‐19 poses challenges for social problems theory.

Constructionist analysts view social problems as processes, rather than conditions. This approach has been theoretically fruitful, and there is now a substantial literature that develops the perspective. However, most of that literature involves case studies of typical, institutionally embedded social problems (Best 2015). Because the process surrounding COVID‐19 has been unusual—in its speed, the breadth of its impact, and the contentiousness of its evolution—examining the construction of COVID‐19 can expose some taken‐for‐granted features of the constructionist model.

CONSTRUCTIONISM: ITS METHODS AND ITS CRITICS

Models of social problems construction identify a typical sequence of stages, usually beginning with claims about troubling conditions and ending with reactions to social policies devised to address those claims (Best 2021b; Blumer 1971; Spector and Kitsuse 1977). These models inform case studies that trace the evolution of particular processes from initial claimsmaking to eventual policy changes. Such projects prove workable: a reasonable amount of research effort can provide the basis for a thesis, dissertation, or scholarly article. These considerations encouraged pre‐Internet researchers to focus on national claimsmaking campaigns that could be traced using printed sources, such as newspaper and magazine indexes and transcripts of congressional hearings.

This method had some unnoticed consequences. First, it favored narratives that had a beginning, but also an end—in the form of a policy impact. This meant the analyses tended to be retrospective efforts to look back in time to discover the origins and events leading toward a known ending—typically some policy. Second, such narratives tended to prioritize the media's secondary claims over primary claims (Best 2021b). Issues that are covered by the press tend to become “stylized” as they get translated and transformed through the media logic and news values that guide how news stories are organized and presented. Thus, the claims that are analyzed in constructionist research are often crafted through familiar frames that prioritize the views of professional experts and insider claimsmakers who are treated as having the authority to define what is at issue. This focus on secondary claims leads to yet another unintended consequence: secondary claims tend to gain coherence and minimize ambiguities relative to primary claims, which are generally unrefined and rather messy. Analyses that focus on secondary claims are likely to “smooth out” whatever might seem to be irrelevant to the central process described in the analysis. Failed campaigns or false starts may be considered unimportant or even ignored because they apparently had little impact on the known outcome.

The theoretical and methodological assumptions that shaped this approach help explain the most enduring critique of the constructionist approach—that it favors claims promoted by elites who are covered in elite media and receive the attention of national officials (Adorjan 2019; Adorjan and Kelly 2022; Collins 1989; Miller 1993; Woolgar 2022). Even when they study social movements of claimsmakers who viewed themselves as challenging elite privilege, constructionists tend to concentrate on the most visible, prominent, and ultimately powerful activists and SMOs. These were the claims most likely to be heard and succeed in the social problems marketplace, usually understood in terms of national centers of government and media power (Hilgartner and Bosk 1988). As a result, there are complaints that claimsmakers with less power, less visibility, and fewer resources tend to be overlooked. The unusual nature of COVID‐19 can help us understand the consequences of these assumptions.

COVID‐19 AS AN ATYPICAL SOCIAL PROBLEM

In contrast to the processes depicted in most constructionist case studies, the social construction of COVID‐19 seems remarkably messy. As soon as COVID‐19 emerged, there were dozens, if not hundreds of competing claims about problematic aspects of COVID‐19. Most of these can be grouped into broad categories (each containing many distinct claims, some of which overlapped into other categories):

Medical concerns (e.g., COVID‐19 as a disease; its impact on those it infects, sickens, and sometimes kills; the challenges it poses to medical practice and institutions; the development and administration of vaccines, the establishment of medical and public health policies);

Policy concerns (e.g., aspects of COVID‐19 policies that extended beyond purely medical concerns, including lockdowns, quarantining, social distancing, and masking measures, efforts to ameliorate the sickness, death, and other direct damages caused by the disease's spread, as well as the impacts of policies designed to address the epidemic);

Economic concerns (e.g., including lost productivity and disruption in the economy; increase in unemployment; economic hardships for various categories of people, and the role of economic issues in shaping social policies);

Structural concerns (e.g., COVID‐19's many relationships to race, class, gender, and age; its impacts on social institutions other than medicine and the economy such as the family and education; and prospects for COVID‐19 leading to enduring structural changes); and

Cultural concerns (e.g., the ways COVID‐19 affected cultural practices and their impacts on people's lives, as well as speculation about whether these effects would last).3

Obviously, this schema is imperfect. Some claims overlapped two or more of these categories, and no doubt other claims might be grouped under additional categories. COVID‐19 posed multifaceted, seemingly all‐encompassing challenges. But a central point should be clear: there was never a COVID‐19 social problem; rather, there were many issues relating to COVID‐19 simultaneously being constructed as social problems. Lots of claimsmakers made lots of competing claims in lots of arenas (Hilgartner and Bosk 1988).

In comparison, most of the topics constructionist analysts have chosen for case studies have been more narrowly focused, and therefore more analytically manageable. Certainly, critics can argue that there are other, even bigger problems—capitalism, say, or inequality—each with its own vast constellation of specific claims; however, these are understood to be problems with very long histories. As soon as COVID‐19 emerged, diverse claims about its many problems began.

Notice as well that there is another temporal difference between most constructionist analyses and discussions of COVID‐19. Case studies usually involve an analyst (and often claimsmakers) looking back in time, examining how a particular social problem emerged. In contrast, the early claims about COVID‐19—as well as the initial efforts to analyze those claims—were basically commentaries on current events—if anything, they tended to imagine what the future might bring (Best 2020b). Moreover, case studies typically edit their accounts in favor of presenting coherent (i.e., simplified, straightforward) narratives. In other words, constructionists generally do not favor the study of emergent and ongoing cases, opting instead to focus on retrospective social histories of more established issues.

This suggests that the earliest stages of social problems claimsmaking often may be more confusing than the edited summaries presented in case studies imply. A troubling condition may initially have been marked by competing interpretations about what was wrong and what ought to be done, but this incoherence tends to disappear in analysts' retrospective accounts, after the eventual consolidation around a particular interpretation of the problem leads to rival approaches disappearing from the central narrative.4 The seemingly unusual complexity, confusion, and incoherence of COVID‐19 claims may be an artifact of its recent emergence. It seems likely that future analysts will eventually construct a more straightforward, smoother COVID‐19 narrative. Still, this ought to lead us to suspect that the relatively tidy accounts presented in other case studies may obscure a good many (what have come to be forgotten) complexities.

This paper seeks to explore the complexities of COVID‐19 claimsmaking by focusing on disputes over authority. While this is not the only angle from which these complexities might be approached, it is arguably an especially important one: in any effort to construct a social problem, it is a considerable advantage to be deemed an authoritative voice, able to define what is at issue.

DISPUTES OVER AUTHORITY REGARDING COVID‐19

As the epidemic spread, aspects of COVID‐19 became the subjects of bitter disagreements about what could and should be done; claims inspired counterclaims. Many of these disputes centered on issues of authority, over who ought to be able to define what was and should be happening.

In retrospect, this might seem surprising. COVID‐19 was an epidemic disease—an instance of a familiar sort of social problem. There were established organizations (such as public health agencies) staffed by designated experts (epidemiologists and other public health and medical specialists), who could be guided by established procedures for dealing with emerging epidemics (Krieger 2013). It is easy to forget just how frequently new epidemic diseases emerge and are addressed; the threat of a new, epidemic infection is not at all unusual, and there is a more or less standard playbook for responses.5 In most recent cases, on most matters, the press, the public, and policymakers deferred to those expert authorities.

In other words, constructionists might have anticipated that COVID‐19 would follow a typical trajectory for a new epidemic disease, with its construction being largely shaped by an alliance of authoritative experts and policymakers. Instead, issues of authority proved to be unusually contentious. Table I identifies four categories of claimsmakers who played prominent roles in the construction of COVID‐19 problems.

Table I

Key Claimsmakers in COVID‐19 authority disputes

	Experts	Non‐experts
Government Officials	Professionals (trained in medicine, public health, epidemiology, biology) working at federal agencies such as the Centers for Disease Control and the Food and Drug Administration, and their state and local counterparts	Persons elected or politically appointed to federal, state, and local offices (e.g., the President and Vice President, members of Congress, and appointees administering federal agencies, and their state and local counterparts)
Non‐government Actors	Professionals (trained in medicine, public health, epidemiology, biology, data science) working in academia, commercial laboratories, science journalism, etc.	Commentators in the media, social media, academia, religious institutions, corporations, etc.

The upper left‐hand cell of Table I contains government officials with expertise in dealing with epidemics. At the federal level, this included the epidemiologists and other medical, public health, and scientific professionals working at such agencies as the Centers for Disease Control (CDC) and the Food and Drug Administration (FDA). These official experts had counterparts in state and local public health agencies. Taken together, these are the experts who constructionists might expect to be the most authoritative and influential claimsmakers in constructing social problems related to an epidemic. Of course, they focused on COVID‐19's medical aspects.

The upper right‐hand cell contains non‐expert government officials, particularly elected politicians, their staff members, and those they appoint to oversee government agencies. In 2020, these figures at the federal level included members of the Trump administration as well as members of Congress. In addition, this category included political leaders at state and local levels. Whereas officials with expertise tended to view the epidemic as a medical problem, non‐expert officials were likely to view it as involving a broader set of political, economic, and social concerns.

The lower left‐hand cell contains experts employed outside the government. These included physicians, epidemiologists, and biologists as well as data scientists and science journalists who worked in academia, corporations, and other non‐governmental institutions. Like their official counterparts, they tended to medicalize the issue. They played an unusually important role in COVID‐19 by calling into question claims by both expert and non‐expert officials.

Finally, the lower right‐hand cell contains a large, diverse collection of people who lacked either official positions or expertise yet offered claims on coronavirus issues. These ranged from prominent journalists and media personalities with large audiences, to religious and business leaders, to individuals exchanging views in face‐to‐face conversations or posting on social media. People in this category outnumbered those in the other three cells, and they devised claims concerning a diverse array of topics.

Table I sorts claimsmakers by expertise and official positions. This is not meant to imply that the members of particular cells had unified or homogeneous views. Many of the experts in the left‐hand cells shared similar professional training and disciplinary perspectives so that there may have been a fairly broad “expert consensus,” but even here we should not be surprised to discover dissenting voices. But the right‐hand cells were wildly diverse, containing people with incompatible viewpoints, who often engaged in loud, dramatic disputes.

The analysis considers the interplay of claims within and across these cells by examining disputes over three issues: (1) metrics (attempts to determine the extent of the coronavirus problem, such as counting people infected by, hospitalized with, or dying from COVID‐19); (2) masks (whether masks should be used or required as part of efforts to contain the spread of the disease); and (3) vaccines (the role vaccinations should play in controlling the outbreak). All three topics proved contentious, and they involved different patterns of alliances and disputes among the four categories of claimsmakers identified in Table I. However, before exploring these disputes, we need to consider some additional aspects of the context within which claimsmaking occurred.

BACKGROUND: CHANGES IN THE CLAIMSMAKING ENVIRONMENT

The foundational models (Blumer 1971; Spector and Kitsuse 1977) and debates within constructionism (e.g., Holstein and Miller 1993; Woolgar and Pawlutch 1985) appeared before the Internet—or even cable television—began having broad social impact in the United States. Early constructionists envisioned claims being made and evaluated in relatively centralized, usually national arenas, such as evening network news programs and weekly newsmagazines. It was really not until after Hilgartner and Bosk (1988) published their seminal analysis about the challenges of attaining access to key claimsmaking arenas that technological developments, especially the dramatic increase in the audience and number of channels for cable and later satellite television, followed by spread of the Internet, and then the further emergence of blogs, social media, and the like, led to a vast proliferation of arenas where claims could be advanced and debated.6 In general, these changes involved a shift from fewer arenas aimed at large, heterogeneous audiences, to far more arenas that tended to address smaller audiences that were more demographically, ideologically, and socially homogeneous. It became simultaneously easier to advance a claim, yet harder to reach a broad, diverse audience and thereby attract a broad base of adherents.7

These media developments’ effects on COVID‐19 disputes are reflected in the commentators’ (lower right‐hand) cell in Table I. Contemporary societies offer a broad array of online platforms where claimsmakers can construct social problems. Our earlier observation that hundreds of COVID‐19‐related problems have been constructed reflects the ease with which people can create and promote claims.

A second, related change has been increased political polarization and the resurgence of populist political leadership. The declining audience for mainstream news media has reflected both the increase in partisan news outlets and the growing tendency of people to consume news and commentary on cable news channels or especially online, where the algorithms that govern search engines and social media platforms steer viewers to content that reaffirms the views they favor.8 These developments have coincided with growing partisanship (cf. Kollman and Jackson 2021) that, in turn, has encouraged political populism (cf. Gerbaudo 2021). In the United States, the presidency of Donald Trump, who of course was called upon to address COVID‐19, reflected these trends, with him having a generally low general approval ratings, yet an enthusiastic base of support. Trump hoped to base his reelection campaign upon a strong economy and became focused on the economic damage from the public health measures recommended by official experts (Mayer 2020; Woodward 2020).

Although the George W. Bush and Obama Administrations had worked to establish a contingency plan for dealing with epidemic diseases, it would become clear that the federal agencies responsible for implementing these plans had not fully modernized their methods of data collection or analysis, and they also had modest influence on steering the Trump Administration's COVID‐19 policies (Gottlieb 2021; Lewis 2021). The result was that, far from the official experts dominating the construction of coronavirus problems and policies, they found their authority under dispute. This paper seeks to understand this outcome. It draws upon a broad range of primary sources—including press coverage, government documents, and social media, as well as the already vast body of secondary analyses by scholars and journalists generated during the epidemic's first 2 years, and seeks to understand how and why there were such vigorous disputes over who ought to have the authority to define the COVID‐19 problems.

We have chosen to examine three focal points for dispute: metrics (i.e., efforts of measure aspects of COVID‐19); masks as a policy for impeding its spread; and vaccines. These data provide the basis for an inductive analysis, which will allow us to seek to better understand, not just why and how COVID‐19 claimsmaking became characterized by disputes over authority, but also how these findings can allow us to critically assess taken‐for‐granted assumptions that underlie the constructionist approach toward studying social problems.

DISPUTES OVER METRICS

The emergence of COVID‐19 in China initially attracted relatively little American media attention, although public health officials recognized that the disease might spread to the United States. By February 7, President Trump had not only been briefed on the threat, but also understood that public health officials believed it was serious; he said: “It goes through the air, … that's how it's passed. … It's also more deadly than even your strenuous flus” Woodward (2020: xix). At the same time, he was reluctant to implement the social distancing policies recommended by public health experts because they would inevitably increase unemployment and slow the economic activity that he considered essential to his reelection prospects.

In general, public health officials argued that COVID‐19 posed a serious danger warranting urgent action—stockpiling medical supplies, developing diagnostic tests, and taking steps to prevent or at least slow the disease's spread (Gottlieb 2021; Woodward 2020). They wanted to control travel, to develop testing and contact tracing measures, and to adopt social distancing policies that would reduce the transmission of infection. In contrast, the Trump administration worried that these steps would harm the economy.

In the spring of 2020, these disagreements inspired disputes over metrics, ways to measure the current risks and project the future impacts of the epidemic. Inevitably, early information was imperfect; there were relatively few hospitalized patients diagnosed with COVID‐19 and even fewer deaths (Best 2020a). Some epidemiologists projected that things could get much worse relatively quickly even as administration supporters were arguing that the threat was not all that severe, and that economic harms from implementing social distancing policies might do more societal damage than the disease itself. On February 24, a prominent talk‐radio host assured his audience: “The coronavirus is the common cold, folks” (Chiu 2020), while one law professor insisted the United States would have less than 500 (later amended to 5,000) fatalities (Epstein 2020). The result was that questions that might have been answered using relatively straightforward measures became contentious. In particular, President Trump continued to offer a series of optimistic predictions that warned against adopting policies that might do more harm than good (Doggett 2021).

At the same time, federal public health officials had some visible failures. The CDC initially insisted that it ought to create the standard test for coronavirus infection, yet the first test it released was ineffective, and the eventual release of a better test proved too late to effectively catch the disease before it had spread widely (Gottlieb 2021; Lewis 2021). In the meantime, other experts outside the federal government were devising their own tests, but the FDA proved reluctant to license these (Boburg et al. 2020; DePillis 2021).

In an apparent effort to maintain access to and hopefully be able to influence the administration's policymaking, public health officials made statements that would damage their credibility. On February 29, trying to reduce public demand for scarce medical supplies, the U.S. Surgeon General tweeted: “Seriously people—STOP BUYING MASKS!” (Asmelash 2020), which was echoed on March 8, by Dr. Anthony Fauci (head of the National Institute of Allergy and Infectious Diseases, who quickly emerged as the leading official public health spokesperson) stating on 60 Minutes: “There's no reason to be walking around with a mask” (Farmer 2020), while on March 25, Dr. Deborah Birx (who had been appointed White House Coronavirus Response Coordinator) praised President Trump, declaring “He's been so attentive to the scientific literature and the details and the data” (Weiland and Haberman 2020). Experts outside the government reacted with skepticism to such claims.

Meanwhile, official procedures to collect basic data on the number of hospitalizations and deaths proved slow and imperfect. It might take several days for reports from individual hospitals to reach federal agencies' centralized data collections where the results could be compiled and then circulated so that official statistics lagged behind the events they were meant to measure—a critical shortcoming with a rapidly spreading virus. In response, researchers in many states began assembling their own programs to collect, collate, and report data from their hospitals to state agencies—supplying more accurate information faster than the federal government. One outgrowth of these efforts was the COVID Tracking Project (2021) website, which began collecting and posting spreadsheets and graphs showing daily shifts in the numbers of infections, hospitalizations, and deaths.9 Media and even official agencies began to treat these unofficial statistics as authoritative.

In short, the public health officials who might have been expected to guide official responses to the emerging epidemic found that their presumptive authority was in fact ineffective. They were not always able to steer the White House's response, and they failed to perform some of their key responsibilities. Other experts working in academia, industry, and some state and local public health agencies lost confidence that they could count on the administration, or even the federal public health agencies. For the most part, these outsider experts questioned whether the federal government was treating COVID‐19 as seriously or as effectively as it should.

At the same time, there were countless commentators who lacked credentials in biomedical professions yet were willing to pontificate about the causes and nature of the threat. In the early weeks, they often insisted that COVID‐19's significance had been exaggerated. These rosy scenarios received support from the president who repeatedly suggested that the crisis had been overblown and would soon fade. Moreover, doubts about the accuracy of the numbers being reported made it difficult to provide evidence‐based prognostications to counter these dismissive claims.

Thus, within the first 2 or 3 months of the recognition that COVID‐19 was spreading within the United States, the basic lines along which authority over the epidemic would be disputed had been drawn. Public health officials might have imagined their expertise would gain broad recognition and acceptance, but their claims were subject to questioning, challenges, and criticism. There was no consensus about how serious the threat might be, nor could people agree about the likely consequences of various policy options.10

Disputes over metrics would continue. By early summer of 2020, official statistics showed COVID‐19 had killed tens of thousands of Americans. President Trump and other Republican officials argued that these numbers exaggerated the threat. At his June 20 campaign rally in Tulsa, the president declared: “We've tested now 25 million people. … Here's the bad part. When you do testing to that extent, you're going to find more people. You're going to find more cases. So I said to my people, ‘Slow the testing down, please!’” (Samuels 2020). Other politicians argued that hospitals had an interest in diagnosing patients with COVID‐19; thus, Republican Senator Joni Ernst: “said she, too, is ‘so skeptical’ of the official numbers. ‘These health care providers and others are reimbursed at a higher rate if COVID is tied to it, so what do you think they're doing?’” (Kellman 2020–for clarification of this claim, see Fichera 2020).

The insistence that official statistics were exaggerating the scope and seriousness of the epidemic ignored what epidemiologists call “excess deaths” (basically, the actual number of deaths minus the number of expected deaths based on death rates in recent years). The total number of U.S. excess deaths in 2020 was over 522,000—with about 72% attributed to COVID‐19 (Woolf et al. 2021). It is all very well to argue that some deaths attributed to COVID‐19 “actually” had some other cause, but this cannot explain why so many more people died during the epidemic's first year. As the election approached, there were reports of struggles between expert officials in federal public health agencies and political appointees assigned to administer those agencies: the former promoted interpretations that emphasized the seriousness of the threat posed by the epidemic, while the latter sought to use the agencies to downplay the risks and proclaim that schools, workplaces, and other institutions should be reopened (Best 2021a).

After President Biden's inauguration, the death toll continued to mount, but federal officials' efforts to dispute these figures declined. In some states with Republication governors, efforts to present data downplaying the crisis continued. In addition, commentators continued to argue that the threat posed by COVID‐19 was exaggerated. This skepticism had an enduring effect. In November 2021, the Kaiser Family Foundation (KFF) released a poll on public acceptance of eight types of COVID‐19 misinformation (Hamel et al. 2021). The most widely held of these beliefs was “The government is exaggerating the number of COVID‐19 deaths”: 38% of respondents said they had heard this and believed it to be true, and another 22% said they had heard it and did not know if it was true.

Disputes over metrics—measuring the numbers of infections from and deaths caused by COVID‐19—might seem easy to adjudicate; after all, they are about quantitative data collected by professionals. But, even here, the authority of these statistics was being called into question for several reasons:

During the epidemic's early weeks, there were good reasons to be skeptical of measures of the numbers of infected. During this period, infections and deaths were probably being undercounted; meanwhile, there were various predictions that the final death toll would be modest.

The erratic statements by the president and his administration, the efforts of experts working at the official agencies to discount those statements, and later efforts by members of the administration to alter statements devised by the official experts led to doubts about the reliability information coming from the federal government—both its political officers and the experts employed in its agencies. As a consequence, outside experts devised and circulated their own metrics.

These difficulties were replicated in many states where political leaders sought to report statistics that seemed to justify their policies.

With experts apparently disagreeing about metrics, there were opportunities for all manner of commentators to offer their own challenges regarding metrics.11 These included claims promoting unproven cures and conspiracy theories.

In short, even issues of metrics—where expert authority might seem weightiest—led to numerous vocal challenges and disputes over authority. Far from dominating the discourse over metrics, official experts found themselves fending off critical counterclaims from experts outside the government, as well as non‐experts both within and outside the government.

DISPUTES OVER MASKS

A second issue involving authority disputes can be seen in competing claims about the need to wear a mask to cover one's mouth and nose to reduce the chances of both becoming infected and, once infected, of spreading the disease to others. By the end of March 2020, the CDC seemed torn between urging people to wear masks and worrying that this might cause a run on the masks needed by healthcare workers (Goodnough and Sheikh 2020). The agency's experts soon contradicted Dr. Fauci's earlier televised statement that masks were not needed, and this inconsistency would continue to be recalled as evidence that the official experts could not be trusted.

The Trump administration also seemed to distance itself from mask‐wearing. This president's initial reaction was not enthusiastic (“This is voluntary. I don't think I'm going to be doing it. … ” [April 3]) and he continued to vacillate between endorsing masks and dismissing them, only rarely allowing himself to be photographed masked (Farley 2020). Similarly, Republican governors often proved reluctant to require masks (Adolph et al. 2022), while their Democratic counterparts more often promoted masking.

Experts outside government generally endorsed masking, which after all was a fairly standard public health recommendation for dealing with airborne epidemic diseases (cf. Talic et al. 2021). Non‐expert commentators, on the other hand, often voiced criticisms—that masks were ineffective, even dangerous, as well as being unreasonable constraints on individual freedom. At the same time, of course, other commentators endorsed masks and criticized those who refused to be masked.

By the summer of 2020, public opinion showed a clear pattern: most people reported wearing masks on at least some occasions, although about one in eight declared they would not consider wearing a mask. Compliance was higher among females, Democrats, and college graduates—patterns that would endure (Brenan 2020).

Among those who opposed masks, rhetoric became increasing extreme. Pascual‐Ferrá et al. (2021) examined pro‐ and anti‐mask 2020 tweets and found that, while pro‐mask tweets were more common, those expressing opposition to masks contained more “toxic language.” Similarly, Bhasin et al. (2020: 934) found that, while both pro‐ and anti‐mask tweets characterized their opponents as “Karens,” the anti‐mask tweets used the term more often and expressed greater hostility: “ …  rule‐following, mask‐wearing Karens are stupid women who have fallen for a scam. They are ‘silly,’ ‘scared,’ and willing to wear a ‘diaper muzzle’.” In April 2021, Fox News host Tucker Carlson “said that anyone who sees a child wearing a mask outdoors should consider it a form of abuse. ‘Call the police immediately. … Contact child protective services’” (Joyella 2021). In the fall of 2021, many schools that had remote‐learning policies for the 2020–2021 school year sought to return to in‐person teaching by requiring students to be masked, but mask opponents denounced school boards for adopting such policies.

Overall, masks led to different patterns of alliances than metrics. Experts—both in and outside officialdom—encouraged masks, and most members of the public, as well as most Democratic governors, went along with the experts' recommendations. In contrast, Republican officials at both the federal and state levels tended to be ambivalent and sought to avoid requiring masks. And, while most commentators endorsed masks, a minority produced a substantial amount of angry opposition.

DISPUTES OVER VACCINES

During the spring on 2020, there were frequent claims that an effective vaccine offered the most promising solution to the pandemic. The Trump administration promoted “Operation Warp Speed” to make funds available to simultaneously support vaccine testing and production by multiple vaccine makers while reducing some of the red tape that typically made vaccine approval a multi‐year process. President Trump predicted that vaccines could soon be widely available; declaring on May 15: “We're looking to get it by the end of the year, maybe before” (Timm 2020).

Prospective vaccines were developed in laboratories relatively quickly—more than 50 had appeared by late March 2020 (WHO 2020). It was the established approval process requiring several rounds of testing that took time (NIH 2020). President Trump was given to predicting rapid vaccine availability, saying in September 2020: “We will have manufactured at least 100 million vaccine doses before the end of the year, and likely much more than that,” which led to critics expressing concerns that the administration might try to circumvent federal agencies' established procedures to insure vaccine safety and effectiveness (Kopp 2020). In her Vice Presidential debate, Kamala Harris declared: “If the public health professionals, if Dr. Fauci, if the doctors, tell us that we should take it, I'll be the first in line to take it, absolutely. But if Donald Trump tells us to take it, I'm not taking it” (Abrams et al. 2020).

Of course, the enthusiasm for a vaccine solution also flew in the face of what was already decades of growing opposition to vaccines (Reich 2016). A Pew Research poll in late‐April/early‐May 2020 found that 72% of respondents said they would be willing to get vaccinated (Thigpen and Funk 2020). This poll also found respondents who were older, Democrats, and more highly educated expressed greater readiness to be vaccinated—patterns that would endure (cf. Funk and Tyson 2020, 2021).

Vaccines first became available in December 2020, but demand exceeded supply into the spring of 2021. President Trump received an early vaccine, but this did not receive much fanfare; his focus was now on denouncing the election results. The Biden administration enthusiastically endorsed vaccines, and during the spring demand continued to outstrip supply. But increases in cumulative vaccinations slowed over the summer of 2021, with much of the increase due to approvals of vaccines for teenagers (May), boosters (September), and children (October). Evidence of the vaccines' efficacy in reducing infections, hospitalizations, and deaths became clear (Buchholz 2021; Scobie et al. 2021). The unvaccinated began to account for the great majority of serious infections, which were concentrated in places with lower proportions of vaccinated people (Leonhardt 2021). Still, roughly 25–30% of the population remained unvaccinated.

They offered various explanations for their refusal: vaccines were unnatural; they might have various dangerous side effects (such as infertility or altering one's DNA); the mRNA technology used in some vaccines was novel; the process of vaccine approval had been rushed; or the vaccines had not been fully approved. And there were claims promoting alternative treatments, that COVID‐19 could be prevented or cured using other drugs, such as hydroxychloroquine in 2020 and Ivermectin in 2021. Social media platforms offered a host of exotic, easily debunked claims: for instance, that the vaccines contained “LUCIFERASE (Lucifer Race) DNA in it. You will loose [sic] your salvation forever!!!” (Avril 2021); that they contained microchips that might be used to track recipients (Lee 2021); or that they made recipients magnetic (Funke 2021).12 The KFF survey published in November 2021 found that nearly fourth‐fifths of respondents either believed or were unsure about the truthfulness of at least one of eight “common falsehoods about COVID‐19” (Hamel et al. 2021).

As it became clear that substantial numbers of people were reluctant to be vaccinated, officials began trying various methods to encourage or require vaccinations. At least 18 states offered incentives for becoming vaccinated, including Ohio's weekly lottery with prizes of up to $1 million (National Governors Association 2021). Summer and fall of 2021 saw policymakers began to shift from carrots to sticks, with a series of policy announcements requiring that various categories of people—federal employees, members of the military, or employees of large businesses—either receive vaccinations or submit to frequent and regular testing. These “vaccine mandates” were replicated in many states and localities (particularly by school districts) and by some employers (Mendelson 2021).

At the same time, vaccine mandates gave opponents a new issue around which to rally by arguing that mandates constrained individual freedom. Some states governors and legislatures tried to block mandates in their jurisdictions, and various suits sought to get courts to overturn the mandates that had been approved. Some of these suits were successful; in January 2022, the Supreme Court blocked a Biden administration mandate covering employees of large businesses (Ruger 2022). In addition, all manner of commentators warned that mandates represented an unreasonable seizure of power by government officials; the most extreme claims drew explicit parallels with the Holocaust and some protesters wore yellow stars (Trestan and Rausch 2021). In addition, the growing sense that COVID‐19 might never end seemed to foster resignation: a large share of those who received vaccinations in early 2021 did not receive recommended booster shots, even as new variants emerged.

Disputes over vaccines had a different pattern of alliances than found in disagreements over metrics and vaccines. Experts both in and out of governments, federal officials in the Biden administration, and most state and local officials were allied in endorsing and encouraging vaccinations. Nonetheless, a minority of officials and a loud minority chorus of commentators continued to oppose adopting vaccines.

DISCUSSION: WHAT SHOULD CONSTRUCTIONISTS LEARN FROM THESE MESSY DISPUTES?

Compared to the relatively straightforward narratives presented in many analyses of social problems construction, the disputes over authority regarding aspects of COVID‐19 seem complicated, even messy, with lots of claims, claimsmakers, and issues, and with shifting patterns of alliances among those claimsmakers contesting the authority to define what was at issue and what ought to be done.

Constructionists often point to the influence of expert claimsmakers in identifying social problems and devising policies to address them (e.g., Pfohl 1977; Ungar 1998). Epidemic disease is a familiar category of public health problem, so we might expect that the response to COVID‐19 would have been dominated by medical and public health officials. To be sure, those officials did have a great deal of influence: their advice did guide most policymaking, most media coverage, and the actions of most members of the public. Still, there were frequent, highly visible instances of dissention. The response to COVID‐19 has been messy, and constructionist analysts have much learn from this messiness. Here, we want to identify eight lessons COVID‐19 offers:

Most Constructionist Research is Retrospective

Many analysts begin with a known outcome, and work backward in time. Often they begin by trying to explain the nature of some policy and trace the evolutions in discourse that shaped this outcome; they know how the story ends, and seek to explain that ending. In the case of COVID‐19, the outcome is uncertain. COVID‐19 remains relatively new (we worked on the paper in the winter of 2021–2022, less than 2 years before the first case was identified in the United States). While COVID‐19 gave many academics something new to analyze, it did not allow them the luxury of hindsight.

There is a Temptation for Constructionists to Smooth Their Narratives

Precisely because they are conducting retrospective analyses, analysts must decide what is relevant to the known outcome. It is not unreasonable to eliminate what seems irrelevant from the analysis; after all, every explanation is necessarily selective, requiring the analyst to draw boundaries of relevance. This sort of editing is inevitable; no narrative can include everything.

The temptation lies in choosing to eliminate elements that may have consumed people's attention sometime early in the social problems process, but then fell out of favor as that process evolved. In 20 or 50 years, retrospective analysts of COVID‐19 will know more; they will be better able to judge what turned out to be important—but analysts in 2022 do not have that advantage. Thus, we have chosen to examine elements that may strike future analysts as irrelevant. And this suggests that other—perhaps most, if not all—social problems may have seemed messier when they first began than they would later, when analysts peered back at them.

COVID‐19 Emerged Very Quickly

Speed matters. Some disasters occur with no warning whatsoever; other troubling conditions—think gender differences—have very long histories. Most of the world had only a few weeks or months to anticipate COVID‐19's arrival. Constructionist analysts tend to take such temporal matters for granted, but they may deserve more careful attention. The speed with which an issue emerges, and claimants' abilities to depict it as more or less urgent affect the reception of claims—the challenges to mobilizing action against climate change are an obvious example (Ungar 1998).

COVID‐19 Spread Very Widely

Within just a few months, the epidemic had spread virtually everywhere people lived. Although the notion that there may be global social problems has begun to capture attention (Mazur 2007), constructionists tend to take a social problem's geographic dissemination and eventual domain for granted. The speed and sprawling spread of COVID‐19 not only as a virus, but as a focal point for multifaceted problem claims suggests how vital it is that we more deeply consider the special bounds and reach of problem effects. A problem's movement through time and across space are linked, and both deserve more attention than we usually give them.

COVID‐19 Affected Many Aspects of Social Life

Again, it is worth repeating: COVID‐19 was not a single social problem; the pandemic led to claims about hundreds of problematic aspects. The news media and the scholarly literature are filled with studies with what we might think of as intersectional titles, for example, “High‐Minority Nursing Homes Disproportionately Affected by COVID‐19 Deaths” (Weech‐Maldonado et al. 2021). The sheer range of potential topics reminds us that all analysts must choose the boundaries of their topics; and thinking about the broad range of COVID‐19 can help us appreciate these choices and how they affect the nature and scope of our analyses.

COVID‐19 Invites Analyses Comparing Places

Different nations report wildly different COVID‐19 death rates.13 No doubt we can look forward to endless multiple‐regression analyses interpreting the effects of different public policies on all sorts of outcomes. But there is also an opportunity for constructionist analysts to compare the social problems processes for COVID‐19 issues across geography. There is a temptation for U.S. scholars to assume that the peculiar nature of the Trump administration or Fox News can account for the problematic development of our public health policies (Mayer 2020). But the disputes in the United States had their counterparts elsewhere; many European countries had controversies over masks and vaccines (cf. Chadwick et al. 2021). By late 2021, for instance, both Austria and Germany were trying to institute—over popular opposition—mandates requiring vaccination (Hubbard 2021; Moulson 2021). There will be countless opportunities to compare claimsmaking across geography, to understand how and why particular issues were constructed in particular ways in particular places.

COVID‐19 Also Invites Analyses Comparing Epidemics

The literature on epidemics is filled with accounts of resistance to public health recommendations. Defoe's A Journal of the Plague Year (1968 [1722]) describes people refusing to remain in their quarantined buildings during the 1665 plague in London, and accounts of more recent American experiences with epidemics report various acts of refusal to comply with experts' recommendations (Bristow 2012; Chase 2003; Ferguson 2021). The disputes described in this paper, while they involve tweets and other contemporary means of making claims, are not uniquely peculiar reactions to an epidemic disease. Thus, there are opportunities to make comparisons across time, as well as space, to look for enduring—or changing—differences in how societies respond to these crises.

COVID‐19 Invites Constructionists to Embrace Messiness

The apparent messiness of COVID‐19 claimsmaking—the numbers of claims and claimsmakers, as well as the shifting alliances among claimsmakers for different issues—should not be understood as unique to the U.S. response to this epidemic. We might suspect that such messiness might be found in many emergent social problems, that they often involve more complexity than is revealed in many constructionist narratives.

Here we can recognize that there is some validity to the critiques of elitism in constructionist analyses. The tendency to smooth out case studies' narratives leads analysts to focus on the claims and claimsmakers that proved most influential, and we can imagine that in the social problems marketplace, it often will be easier for those with greater power, status, and visibility to have such influence. The case of COVID‐19 helps us appreciate how much conflict and confusion may be involved in the social problems process, particularly during its earliest stages. We might anticipate that many early subjects of claimsmaking will lose—or never even achieve—salience. The case of COVID‐19 offers a rich opportunity to study this winnowing process, to try to understand which claims endure or even reshape social policies, and which fade from public view.

Obviously, every analyst is forced to define the boundaries of what is being studied. For instance, here we have decided to focus primarily on the emergence and effects of three sets of authority disputes related to the pandemic; this is but one of a multitude of ways to analyze COVID‐related claims. However, embracing the complexity of COVID‐19 claimsmaking can help us make those choices consciously, rather than simply taking them for granted. Our analyses may be improved by acknowledging messiness, and treating it as a feature, rather than a bug.