Moral Panics

Chas Critcher

doi:10.1093/acrefore/9780190264079.013.155

Summary

The concept of moral panic was first developed in the United Kingdom in the early 1960s, principally by Stan Cohen, initially for the purpose of analyzing the definition of and social reaction to youth subcultures as a social problem. Cohen provided a “processual” model of how any new social problem would develop: who would promote it and why, whose support they would need for their definition to take hold, and the often-crucial role played by the mass media and institutions of social control. In the early 1990s, Erich Goode and Nachman Ben-Yehuda produced an “attributional” model that placed more emphasis on strict definition than cultural processes. The two models have subsequently been applied to a range of putative social problems which now can be recognized as falling into five principal clusters: street crime, drug and alcohol consumption, immigration, child abuse (including pedophilia), and media technologies. Most studies have been conducted in Anglophone and European countries, but gradually, the concept is increasing its geographical reach. As a consequence, we now know a good deal about how and why social problems come to be constructed as moral panics in democratic societies.

This approach has nevertheless been criticized for its casual use of language, denial of agency to those promoting and supporting moral panics, and an oversimplified and outdated view of mass media, among other things. As proponents and opponents of moral panic analysis continue to debate the essentials, the theoretical context has shifted dramatically. Moral panic has an uncertain relationship to many recent developments in sociological and criminological thought. It threatens to be overwhelmed or sidelined by new insights from theories of moral regulation or risk, conceptualizations of the culture of fear, or the social psychology of collective emotion. Yet as an interdisciplinary project, it continues, despite its many flaws, to demand sustained attention from analysts of social problem construction.

Introduction

The concept of moral panic can be found in several disciplines: sociology, media studies, and cultural studies, as well as criminology. It often polarizes opinion. An early review of the concept, influential in its resuscitation, has an unequivocal conclusion. According to Thompson (1998), p. 142), “It deserves to be recognized for what it truly is: a key sociological concept.” Yet more recently has come another, slightly despairing, view (Jewkes, 2015, p. 103), to wit:

[I]t is difficult to explain why criminology—and its related fields—continue to place the moral panic thesis at the heart of studies of deviance and disorder when both sociology and media studies have more or less ignored it for decades.

Such reservations have not prevented the notion of moral panic spreading beyond academe into debates about social problems in society at large. So it is important to define exactly what this concept entails. A handy definition of moral panic is provided by the Online Dictionary of the Social Sciences, produced by Canada’s Open University:

Suggests a panic or overreaction to forms of deviance or wrong doing believed to be threats to the moral order. Moral panics are usually framed by the media and led by community leaders or groups intent on changing laws or practices. Sociologists are less interested in the validity of the claims made during moral panics than they are in the dynamics of social change and the organizational strategies of moral entrepreneurs. Moral panics gather converts because they touch on people’s fears and because they also use specific events or problems as symbols of what many feel to represent “all that is wrong with the nation”.
(Drislane & Parkinson, 2016)

This seems as good a place as any to start. This article¹ fills in the details about what moral panics entail, which topics or events they focus on, the agents strategic to their construction, and how they attract support from the public at large. The first section outlines the central tenets of the two main models of moral panics. The second section reviews what our accumulated knowledge of moral panics tells us about how and why they work. The third section details six substantial points of criticism leveled against moral panic analysis and the responses by their defenders. The fourth section explains the concept of moral regulation as a continuous societal process in which moral panics are sudden eruptions. The fifth section widens the focus of the discussion to identify those characteristics—risk consciousness, pervasive fear, and the politics of emotion—which may predispose late modern societies to increased levels and intensities of moral panics. Finally, the conclusion offers a further comment on the nomadic disciplinarily of this concept.

Original Models

Cohen’s Processual Model

Stan Cohen published Moral Panics in paperback in 1973. Based on his PhD thesis, it analyzed how British society reacted to seaside confrontations between members of two youth subcultures, mods and rockers, in the early 1960s. Cohen was interested how labeling deviants could amplify deviance, damage the identities of the labeled individuals, and invite them to embrace deviant identities and behavior. Cohen set out to test these ideas on mods and rockers but ended up in a rather different place. He had discovered a pattern of construction and reaction with wider purchase than mods and rockers—the moral panic (Cohen, 1973, 2002, p. 9):

Societies appear to be subject, every now and then, to periods of moral panic. A condition, episode, person or group of persons emerges to become defined as a threat to societal values and interests; its nature is presented in a stylized and stereotypical fashion by the mass media; the moral barricades are manned by editors, bishops, politicians and other right-thinking people; socially accredited experts pronounce their diagnoses and solutions; ways of coping are evolved or (more often) resorted to; the condition then disappears, submerges or deteriorates and becomes more visible. Sometimes the object of the panic is quite novel and at other times it is something which has been in existence long enough, but suddenly appears in the limelight. Sometimes the panic passes over and is forgotten, except in folk-lore and collective memory; at other times it has more serious and long-lasting repercussions and might produce such changes as those in legal and social policy or even in the way the society conceives itself.

Cohen stressed that these stages overlap. Progression through them can be thwarted or diverted. That is why it has been called a “processual model” of moral panics (Critcher, 2003, p. 13).

Cohen’s model is often mistakenly thought to focus mainly on mass media. However, he cast his net wider than that. He identified in moral panics four key agents: mass media, moral entrepreneurs, the control culture, and the public. The media are particularly important in the early (inventory) stage of social reaction, producing “processed or coded images” of deviance and the deviants (Cohen, 1973, 2002, pp. 44–48). Three processes are involved. The first is exaggeration and distortion of who did or said what; the second is prediction, the dire consequences of failure to act; and the third is symbolization, which in this case means that the word mod or rocker signified a threat. The media installed mods and rockers as folk devils, and newsmaking practices specified who was disrupting the social order. They employed “inferential structures” to explain what the behavior was like, who perpetrated it, and why it happened (what Cohen called “orientations,” “images,” and “causation”). They were primed for panic.

The second group was moral entrepreneurs, referring to individuals and groups who target deviant behavior. Cohen spent time and effort understanding their worldviews and actions. The third group, the societal control culture, comprised those with institutional power: the police and the courts and local and national politicians. They were made aware of—“sensitized” to—the nature and extent of the problem. Concern was passed up the chain of command to the national level, where draconian control measures (innovation) were instituted. The fourth group, the public (acting as witnesses), had to decide who and what to believe. Cohen discussed the mods and rockers problem with individuals and groups, finding that they initially mistrusted media messages, yet ultimately believed them.

The complex interplay between these four groups defined the problem, its remedies, and the proposed solution (normally, a change in the law or its enforcement). Mods and rockers provoked a strengthening of one law (about drugs) and the introduction of a new one (about criminal damage). However, especially as the threat was largely mythical, such laws were more ritualistic than effective.

Laws confirm a primary function of moral panics: the reaffirmation of society’s moral boundaries. Cohen (1973, p. 197) argued that the “affluent society” of the 1960s disconcerted traditionalists, fearful that the young were rejecting adult ideals: “the response was as much to what they stood for as what they did.” Future moral panics seemed inevitable. The pace of social change and the persistence of social inequality generate tensions that find an outlet in the identification and vilification of new kinds of deviant behavior. Cohen’s work was predictive. Six years later, Hall, Critcher, Jefferson, Clarke, and Roberts (1978) argued that a moral panic about “mugging” validated Cohen’s predictions. A distinctively American take on moral panics appeared 16 years later, as discussed next.

Goode and Ben-Yehuda’s Attributional Model

In 1994, Erich Goode and Nachman Ben-Yehuda published Moral Panics: The Social Construction of Deviance, with a second edition in 2009. Their approach, known as social constructionism, challenged the basic assumption that sociology could define, measure, explain, and ameliorate social problems. On such issues as crime, mental retardation, and homosexuality, definition and measurement foundered.

Reviewing empirical studies in the constructionist tradition, Goode and Ben-Yehuda (2009, p. 37) arrived at five defining “elements or criteria” of a moral panic.

Concern

Any moral panic involves a “heightened level of concern over the behaviour of a certain group or category” (Goode & Ben-Yehuda, 2009, p. 37) and its consequences. Indices of concern include opinion polls, media coverage, and lobbying activity.

Hostility

Moral panics exhibit “an increased level of hostility” toward the deviants, who are “collectively designated as the enemy, or an enemy, of respectable society.” Their behavior is seen as “harmful or threatening” to the values and interests of society, “or at least a sizeable segment” of it (Goode & Ben-Yehuda, 2009, p. 38, emphasis in original). Constructing such folk devils is integral to moral panics.

Consensus

In a moral panic, “there must be at least a certain minimal measure of consensus” across society as a whole, or at least “designated segments” of it, that “the threat is real, serious and caused by the wrongdoing group members and their behaviour” (Goode & Ben-Yehuda, 2009, p. 38). Consensus can be challenged by organized opposition—“counter claimsmakers.”

Disproportionality

Fundamentally, “the concept of moral panic rests on disproportion” (Goode & Ben-Yehuda, 2009, p. 41, emphasis in original). It is evident where “public concern is in excess of what is appropriate if concern were directly proportional to objective harm” (Goode & Ben-Yehuda, 2009, p. 40). Statistics are exaggerated or fabricated. The existence of other equally or more harmful activities is denied.

Volatility

Panics are by their nature fleeting, subsiding as quickly as they erupt. The same issue may recur, but individual panics cannot be sustained for long. This has been called an “attributional model” of moral panics (Critcher, 2003, p. 25) because it identifies essential characteristics. Central to this model is claims making about the problem: who makes claims, how, and why. Such claims are frequently made by social movements, who perceive and seek remedies for problematic behavior. Movements protest and demonstrate, appeal to public opinion, and gain access to the media. They may behave irresponsibly: exaggerating the threat, polarizing opinion, and vilifying opponents. Other authoritative organizations may collude with them, such as religious groups, professional associations, and the police. More often, the media, occasionally active in moral panics, are more often conduits for others’ claims making.

Goode and Ben-Yehuda assessed three competing explanations of moral panics. First, the grass-roots model identifies the source of panic as widespread anxieties about real or imagined threats. In the second explanation, the elite-engineered model, an elite group, manipulates a panic over an issue that they know to be exaggerated in order to divert attention from their own inability or unwillingness to solve social problems. Third, interest group theory argues that “the middle rungs of power and status” are where moral issues are most acutely felt. Goode and Ben-Yehuda suggested that elites are marginal. The combined forces of grass-roots feeling and middle class agitation lie behind the most effective panics. The wider explanation, however, lies in the nature of collective behavior (Goode & Ben-Yehuda, 2009, pp. 51–72).

The consequences of moral panics are twofold: institutional legacy and normative transformation. Institutionalization involves establishing new laws, agencies, or professions. Normative transformations alter ideas about the acceptability of behavior, thus redrawing society’s moral boundaries. The missing children controversy of the 1980s is an oft-cited American example of moral panic (Best, 1990) despite the skepticism about the concept expressed by one of the leading writers on the topic (Best, 2013).

Comparing the Two Models

Comparing the processual model of Cohen with the attributional model of Goode and Ben-Yehuda reveals three basic similarities and three significant differences. The first similarity is their shared view that moral panics are an extreme form of more general processes by which social problems are constructed in public arenas. The second similarity is that they both observe that moral panics are recurrent features of modern society that have identifiable consequences on the law and state institutions. The third similarity is the perceived sociological function of moral panics as reaffirming the core values of society.

On the other hand, the first difference lies in how they assess the role of the media. In the processual version, the media are strategic in the formation of moral panics. They may be the prime movers or endorse others already campaigning, but they are always actively involved. In the attributional model, despite their greater prominence in Goode and Ben-Yehuda’s second edition (2009, pp. 88–108), the media play a much more passive role. They provide an arena where different versions can compete.

The second difference is their conception of the most important agents in moral panics. In the processual model, state agencies, politicians, and legislators do not merely react to moral panics, but rather are frequently complicit in their construction. The attributional model places much more emphasis on the strategies adopted by claimsmakers. Their ability to galvanize public opinion about their case is crucial.

The third difference involves how to conceptualize the language of moral panics. In the attributional model, the emphasis is on the rhetoric of claims making (i.e., how campaigners adopt particular styles of argument). The processual model emphasizes moral panics as activating ideological discourses, such as those around law and order.

These differences are not insuperable. In an underrated effort, Klocke and Muschert (2010) created a composite model capitalizing on the strengths of the original theories. For a new research project, this might be a more useful tool than settling for just one of the original theories and thus omitting the insights of the other. Still, even in their present form, moral panic models have revealed a good deal about the topics that they have highlighted.

Accumulated Knowledge

Five Topic Clusters

Moral panics have been studied extensively over the last 40 years. The accumulated findings will be reviewed here in the context of five clusters of topics: child abuse, drugs and alcohol, immigration, media technologies, and street crime. Cohen (2002, p. xv) has a slightly different list of seven clusters. For each of these clusters, we know quite a lot about how moral panics operate and the impact that they have.

Child abuse. We know that exceptional cases of physical or sexual abuse become drivers of child protection policy, regardless of their typicality or alternative evidence from social work agencies. The original construction of the pedophile as the dangerous stranger was subsequently modified by revelations about pedophiles in the priesthood and among celebrities. But their presence in and around the family is still rarely acknowledged. Organizations apparently dedicated to saving children have their own self-serving agendas, while the media and public are highly susceptible to images of innocent children being damaged or corrupted (Jenkins, 1992; Kitzinger, 2004; Krinsky, 2008; Warner, 2015).

Drugs and alcohol. We know that consciousness-changing substances used for pleasure are a constant target for legal action because they allegedly jeopardize either the health of those who enjoy them or general law and order on the streets. Laws governing the sale, possession, or consumption of such substances are tightened continuously but can be enforced only selectively to avoid criminalizing broad swathes of primarily young people. A huge and permanent disjunction exists between the policies and prognostications of police, politicians, and the media and the views and practices of young people as a whole. Drug advice agencies find their expertise consistently undervalued (Parker, Aldridge, & Measham, 1998; Jenkins, 1999).The most recent examples include methamphetamine (Linneman, 2010; Omori, 2013), mephedrone (Collins, 2013; Alexandrescu, 2014), and novel psychoactive (“designer”) drugs (Miller et al., 2015).

Immigration. We know that a serial moral panic is likely to recur whenever people migrate to another location to live alongside the indigenous population, especially if the immigrants are of a different color. Accusations against these newcomers are invariably that they bring alien cultures and refuse to integrate with the mainstream culture; that they make excessive demands on welfare, education, and housing systems; and that they are excessively involved in crime. These negative assumptions occur in very different contexts, ranging from resistance to refugees from war in the Middle East in Britain or the European Union to American hostility to migrant labor from Mexico. Antipathy to migrants is routinely reproduced by the popular press, is confirmed by politicians of all shades, and has reinvigorated the extreme right in Europe (Welch, 2002; Kubrin, Zatz, & Martinez, 2012; Philo, Briant, & Donald, 2013).

Media technologies. We know that advent of any new medium of communication produces disquiet among guardians of childhood and culture. They fear that it will encourage children to seek pleasure in narratives of little intrinsic merit that proffer dangerous role models of violent or otherwise antisocial behavior. Children enter a fantasy or virtual world where they can act out roles and emotions that were otherwise prohibited. Such fears are often based on ignorance about the medium’s actual capacities or usage. Moralizing organizations, often religiously motivated, commonly advocate censorship that the media industry resists, while parents remain concerned but lack coherent control strategies (Barker & Petley, 1997; Livingstone, 2002). Cyberbullying (Milosevic, 2015) and sexting (Draper, 2012) are recent manifestations of this type of development.

Street crime. Interpersonal crime has always been a central concern of modern mass media. Coverage expands dramatically if new types or patterns of crime emerge, especially involving increased violence or the use of weapons. This sustains the belief that crime is out of control, so fear of being randomly attacked on the street by violent young men is prevalent in modern cosmopolitan societies, even among those least likely to become victims. This pattern prevails for such crimes as mugging and knife- or gun-related crime. Social causes, in the deprived structural position and limited cultural options of inner-city youth, are occasionally recognized, but the solutions are repressive: more vigorous policing and longer sentences for the actual or potential use of violence (Chambliss, 1995; Jewkes, 2015).

These five groupings are by no means exhaustive. Others could easily be added. Panics about welfare dependents exist, but they may be better analyzed as an ideologically motivated and increasingly successful attack upon the basis of the modern welfare state. Reaction to inner-city riots could be seen as moral panics but may simply represent what the state has always done when confronted with insurrection: come down immediately with the full force of the law, without preamble or apology.

Modern terrorism, one product of Islamic fundamentalism, is in a league of its own. An explicitly political type of deviance, it is an international confrontation that goes way beyond the confines of the nation-state, indivisible from the Western view of, and actions toward, the Muslim world. An initial view might be that the moral panic concept seems ill equipped to deal with its global nature, but it has nevertheless been occasionally applied. Early on, Rothe and Muzzatti (2004) employed both models to demonstrate that terrorism fitted them exactly. Walsh (2016) argued that terrorism remains a moral panic, if an exceptional one, since terrorists as folk devils deliberately provoke overreaction. Other analysts advocate substantial changes to the models to make them applicable (Morgan & Poynting, 2012; Shafir & Scairer, 2013). They may be stretching a point, however.

Other topics with the potential for a moral panic never develop. Both Levi (2009), for white-collar crime, and Jenkins (2009), for Internet child pornography, have explored how complex issues, that are not routinely visible and difficult to detect and featuring perpetrators who are not instantly recognizable, fail to attract the attention of moral entrepreneurs, the media, or enforcement agencies, except on rare and fleeting occasions. Domestic and sexual violence against women is another category that fails to spark a moral panic. What is absent from moral panics may be as significant as what is present in them. We now know a great deal about the latter.

Generalizations

On the basis of the research to date, we can make some reasonably robust empirical generalizations about moral panics, as follows:

•

In capitalist democracies, moral panics appear to be endemic; it is not a question of whether, but when, the next one will appear.

•

The relationship between an alleged problem and its actual occurrence or significance ranges from almost total fabrication through exaggeration of a relatively minor problem to systematic distortion of a major one.

•

The media play a crucial role in moral panics, but there are important differences between types of media: local and national, press and television, or upmarket and downmarket.

•

Moral panics can easily be exploited by party politicians.

•

Threats to children or from youth heighten emotional tension in moral panics.

•

Factors identified as causing the decline of a moral panic include:

Its displacement by other more novel and dramatic problems, especially in the media;

Its apparent or symbolic resolution by legal and related measures;

A decline in the symptoms of the problem as a result of social control initiatives;

The emergence of counterclaims that challenge or discredit the originators of the moral panic.

The knowledge accumulated by the numerous studies of moral panics has also confirmed the following basic features of the models as recurrent:

•

The strategic roles occupied by identifiable groupings: pressure groups, accredited professionals, mass media, and politicians.

•

The nature of the institutional legacies that they leave behind, especially changes in the law, though often symbolic.

•

The distorting effect of panics on the quality of public debate about social problems.

•

Their apparent function, in times of rapid or unsettling social change, of reaffirming the basic moral values of society.

Qualifications

Three features vary considerably across moral panics. The first is the status of the folk devil. Three of our five clusters have a clear folk devil, in their purest forms—pedophile, illegal immigrant, and mugger. But in the two others, the problem is an object, such as an Ecstasy pill or a computer game. Moral panics do not require a clear-cut folk devil, although they may be more effective when they have one.

The second is an apparent variability in the level and intensity of public (as opposed to elite) engagement. Threats from immigration, child abuse, and street crime seem to provoke collective emotional recognition, unlike those from new media technologies or recreational drugs. The reasons for this, such as the visibility of the deviance, require further investigation.

The third is the role of notorious cases with exceptional symbolic power, often with children or teenagers as victims. Variously conceptualized as “key events” (Kepplinger & Habermeier, 1995), “signal crimes” (Innes, 2003), or “scandals” (Butler & Drakeford, 2005), they attract media coverage, political response, and public opinion. These are not necessarily the triggers for the panic, which can create its own cases. Heightened sensitivity to an issue can transform an otherwise routine event into an emblematic one. Visual images of victims, especially if they are children, can become culturally iconic.

Both recurrent and variable features may be found in moral panics at different times and in different places. This point is discussed in the next sections.

Moral Panics in Time and Space

More and more information about the geography and history of moral panics has gradually emerged. Studies originally emanated from Anglophone countries (the United States, Canada, Australia, New Zealand, and the United Kingdom), with supplements from Scandinavian countries (Denmark, Norway, and Sweden). This is now changing. Moral panic analysis has gained a foothold in Central and Eastern Europe, South America, and the Far East. Krinsky (2013a), for example, includes contributions from Brazil, Argentine, Poland, and Japan. A scholar of Japan (Toivonen, 2013, p. 265) notes that

though most of the relevant literature focuses on Europe and North America, non-Western societies with modern media apparatuses are similarly susceptible to episodes of moral panic.

Social problems and their associated moral panics may be exported from Western societies, especially the United States (Best & Furedi, 2001). We can now recognize national variations in sociological characteristics crucial to moral panics. Systems of politics, media, religion, and law enforcement may appear similar in principle but differ in practice. It remains true that moral panics everywhere emerge from the interactions of the five Ps: press, politics, pressure groups, police, and public. But such systems vary across nations, with other variables appearing, such as the hegemony of organized religion over moral issues.

As the geographical perspective on moral panics widens, so does the historical one. Hostility to those perceived as deviant is intrinsic to the histories of Jewish and Romany people. The greatest moral panic of all time, the pursuit of witches, happened in Europe in the Middle Ages. A study of a later appearance of the phenomenon among the earliest settlers in the United States greatly influenced moral panic pioneers (Erikson, 1966). Moral panics can now be identified in the eighteenth and nineteenth centuries (Lemmings & Walker, 2009).

Such studies indicate that moral panics beyond religious or ethnic persecution are largely a product of modernity. Preconditions for moral panics include a formally free press, a government willing to respond to popular pressure, campaigners able and willing to organize to bring about legal change, and an elite belief that social stability depends upon the maintenance of a secular moral order. These conditions did not exist in England or the United States before the late eighteenth century and do not exist today in many societies. Moral panics cannot occur in closed societies. An impervious political system may create its own scapegoats but will not permit, nor respond to, independent agitation against moral evils. Societies ceasing to be totalitarian, as in Eastern Europe, are likely to start producing moral panics. They are, paradoxically, a product of culturally open societies.

We are progressing toward “a comparative sociology of moral panic that makes comparisons within one society and also between societies” (Cohen, 2002, p. xxii). Appreciating the geographical breadth and historical depth of moral panic analysis assumes that the enterprise remains worthwhile. This is, however, a far from universal assumption. Its principles and practices continue to be challenged.

Critiques of Moral Panic Models

Reservations about moral panic analysis are many and varied. Some are based on a single case study, such as LSD (Cornwell & Linders, 2002) or AIDS (Miller & Kitzinger, 1988). Others discuss clusters of panics, such as those concerning crime (Jewkes, 2015). At a different level are those that consider the presuppositions and internal consistencies of the models as totalities (Garland, 2008). The debate is further confused by the fact that the target of criticism is usually only one of the two main models, and what applies to one may or may not apply to the other. Because it has the most influence in Britain, the debate is more intense there than anywhere else.

There are more complex maps of the past and present of moral panic analysis than can be considered here. Hier (2011a) divides analysts into three camps: the conventional, the skeptical, and the revisionist. Krinsky (2013b) perceives two waves of moral panic development, early and late. Rohloff, Hughes, Petley, and Critcher (2013) outline fundamental conceptual issues and contemporary debates. The approach here is more akin to that of David, Rohloff, Petley, and Hughes (2011), who enumerated the basic issues at stake. They are here divided (and thus fragmented) into six main lines of criticism: loose terminology, disproportionality, outdatedness, rigidity, political partisanship, and assumed media effects.

Loose Terminology

Critics argue that using the term and constructing a boundary around “moral” tends to prevent links to apparently similar issues, such as health or food scares. The definition of which issues are or are not “moral” is bound to be subjective and is always fluid because almost any issue can be moralized. Likewise, the term panic raises the objection that it imputes irrationality to people who may be genuinely and logically concerned. It pits the rational analyst against the irrational participants. “It is never very clear who is doing the panicking. Is it the media, the government, the public, or who?” (Miller & Kitzinger, 1988, p. 216).

Disproportionality

The logical basis of moral panic analysis is also questioned. Goode and Ben-Yehuda (2009) assert that moral panic analysis rests fundamentally on the ability to demonstrate that a response to a perceived social evil has been or is disproportionate. This implies that we are in a position to know what a proportionate response would be. For that to happen, we need to have detailed knowledge about the real dimensions of the problem. Critics say that often, such data are unreliable or unknowable. Even if they are known, who is to say what is a “proportionate” response to, say, the rape and murder of a child? Disproportionality does not work, but without it, there are no grounds to decide what is or is not a moral panic. The principal difficulty about a moral panic lies in

establishing the comparison between the scale of the problem and the scale of response to it … Conceptually the notion of a moral panic lacks any criteria of proportionality without which it is impossible to determine whether concern about any … problem is justified or not.
(Waddington, 1986, p. 246)

Outdatedness

A different set of criticisms stems from the datedness of Cohen’s model. Produced 40 years ago, it reflected the British political system, cultural assumptions, and media structure of that time. Each of these has become much more fluid. The advent of new and social media in particular has changed the locus of definitional power so that many more voices are heard than was previously the case. Those in power can no longer be confident that their definitions of issues will prevail:

The proliferation and fragmentation of mass, niche and micro-media and the multiplicity of voices, which compete and contest the meaning of the issues subject to “moral panic,” suggest that both the original and revised models are outdated insofar as they could not possibly take account of the labyrinthine web of determining relations which now exist between social groups and the media, “reality” and representation.
(McRobbie & Thornton, 1995, p. 561)

Rigidity

The model itself is alleged to be mechanistic. “The most serious flaw of the concept of moral panic … is its lack of agency” (Miller & Kitzinger, 1988, p. 216). The panic follows a prescribed script that robs those involved of any agency and the issue itself of any specificity. The model is rigid; that is, it does not permit variations in processes or outcomes. If cases depart from the anticipated sequence, the model founders and cannot explain such deviations. It is thus a self-fulfilling prophecy. Only those events that approximate to the model are recognized as moral panics.

Political Partisanship

For some, moral panic too readily becomes a term of political abuse. It is a way of discrediting conservative claimsmakers. Liberal or radical campaigns on social issues are not accused of mounting moral panics. This sleight of hand disguises essentially political judgments as intellectual ones. “Moral panic is an invidious label” that is “driven largely by unstated ideological assumptions” (Best, 2013, pp. 42–43).

Assumed Media Effects

In addition, there are allegations that this approach assumes that the public necessarily believes what the media transmit. This invalidates independence of judgment and experience—the idea that people may have a good foundation for their beliefs other than learning them from the media. Moral panic scholars “presuppose that in finding consensus on certain issues, audiences are gullible and that they privilege mediated knowledge over direct experience” (Jewkes, 2015, p. 100). There is a built-in resistance to investigating media audiences empirically (Miller & Kitzinger, 1988).

Proposed Revisions

The critiques given in the previous sections are a formidable array, sufficient for some to advocate abolishing the approach altogether or limiting its application to a narrow range of cases. Best (2013) argued that the term moral panic should be strictly confined to media-led campaigns against perceived deviance among young people. McRobbie and Thornton (1995, p. 572) argued more than 20 years ago that “the model of moral panic is urgently in need of updating precisely because of its success.” They explained why it should be done, but not how to do it. Garland (2008) wished to retain the essentials but advocated conceptual refinement. Hier (2002) sought to recharacterize moral panic as a special instance of a wider process of moral regulation. Ungar (2001) argued that in a risk society, new kinds of anxiety, especially about possible environmental catastrophes, displace traditional moral panics. In a much-neglected critique, Watney (1998) insisted that by concentrating on discrete episodes, moral panic analysis inherently fails to grasp the more continuous and profound ideological struggle over representation.

Counterarguments

Defenders of the models, including Cohen (2002), Goode and Ben-Yehuda (2011), and Critcher (2003), have responded to these criticisms. They admit that the terminology is not wholly satisfactory, but they contend that it is better than any alternatives. There are social problems that directly involve questions around basic societal norms—“the expression of outrage at the violation of a given absolute value” (Goode & Ben-Yehuda, 2011, p. 21)—in a way that is demonstrably not the case for other categories of problems, such as health concerns or food scares. Cohen (2002) concedes that panic is a problematic term because of “its connotation with irrationality and being out of control” (p. xxvii) but insists that it “still makes sense as an extended metaphor” (p. xxvi), and that similarities remain in terms of psychological mechanisms with urban myths and natural disasters. The concept of panic may seem to overstate the case, but it remains appropriate to indicate how, at the height of social reaction, collective emotion overwhelms individual reason.

Disproportionality is probably the most contested judgment inherent in moral panic models. The originators remain adamant that usually such claims can be assessed for proportionality. As Cohen (2002, p. xxviii) puts it, using the example of myths about asylum seekers, “the core empirical claims within each narrative can usually be reached by the most rudimentary social science methodology.” Goode and Ben-Yehuda (2009) argue, more elaborately, that disproportionality is evident when any one of five conditions is met: (a) that statistical claims are demonstrably false; (b) the putative problem is nonexistent; (c) absurd rumors flourish; (d) some problems are emphasized at the expense of similar, but more significant, ones; and (e) media and related attention increases despite no change in the rate of behavior.

Being outdated is more applicable to Cohen’s work from the 1970s than to Goode and Ben-Yehuda’s later and then revised model. Neither model tackles how the rise of the Internet and digital technologies, especially social media, has altered communication patterns during moral panics. While undeniable in principle, the practical effects of digital communication on moral panics have yet to be empirically proven. In some instances, social media have merely intensified the level of vitriol directed at identified deviants. In other examples, such as designer drugs, the Internet was the vehicle for spreading knowledge about and access to them. It was also possible to correct inaccuracies in the mainstream media’s reporting of supposedly drug-related deaths (Forsyth, 2012). But when legal action was taken, consumers had no voice.

The charge of rigidity is denied by defenders of the models. Cohen explicitly concedes that a potential moral panic may be stalled or sidetracked. Goode and Ben-Yehuda make a consistent and careful distinction between plentiful moral crusades and much rarer moral panics. The full-blown moral panic follows a predictable path and demonstrates consistent characteristics, but there are many claims making activities that never assume the status of moral panics.

The allegation of political partisanship—that conservative groups, but not liberal or radical ones, are castigated for seeking to create creating moral panics—cannot apply to Goode and Ben-Yehuda, who have explicitly derided moral panics that have been supported by radicals: pornography, snuff movies, school shootings, and (in retrospect, perhaps unwisely) sexual abuse by Catholic priests. Moral panic analysis does not automatically exonerate so-called progressive groupings who make controversial claims. Sex trafficking could be an example where those with the best of intentions are prone to exaggerate the existence of the problem because its nature is so vile (Weitzer, 2007; Cree et al., 2012). Cohen (2002, p. xxxi) conceded that “the criticism that ‘moral panic’ is a value-laden concept, a mere political epithet, demands more complicated attention than it receives.” Nevertheless, since moral panics function to defend the existing moral order, they will be favored by conservatives and viewed skeptically by liberals.

The final point of criticism is that moral panic analysis assumes a gullible public. The reply is that it is difficult for the public to resist media messages. Goode and Ben Yehuda argued that opinion polls continuously demonstrate that the media and the claims makers they validate do set the public agenda. Cohen, whose original work explored the many ambiguities and inconsistencies in audience interpretations of media messages, never assumed that the audience was gullible, although it was inevitably media dependent.

The points at issue are substantial and complex; they have been simplified here for clarity. Some issues are amenable to empirical evidence from studies that have either discovered inconsistencies in the models or concluded that the essentials of the models can be verified. Other issues are more abstract, concerning how we conceive the nature of social processes or even knowledge itself. If nothing else, moral panic analysis has helped stimulate debate about the social construction of social problems, which is lively and ongoing. But the original models can no longer be freestanding—they need to be connected to other cognate strands in contemporary social science. One angle is to ask a simple question: What, if anything, are moral panics extreme manifestations of? One answer is moral regulation, discussed next.

Moral Panics and Moral Regulation

The concept of moral regulation holds that open societies conduct a continuous dialogue about the boundaries of morally acceptable behavior and how to regulate what is regarded as unacceptable. In principle, this can mean that activities once regarded as unacceptable are legitimated, such as homosexual relationships. In practice, though, the boundaries are continuously redrawn to cope with new kinds of moral impropriety, such as the misuse of social media.

Alan Hunt (1999) analyzed 19th-century movements for moral regulation in the United Kingdom and United States. He found moral regulation to be aimed at such traditionally immoral activities as sex, drinking, and gambling. He emphasized how organized advocates of regulation, who were usually middle class and often female, adopted common strategies. They identified and defined an immoral activity, specified who was involved in it, developed propaganda tactics, and demanded legislative action. These are essentially the same ploys that claimsmakers use today, despite huge changes in the social, economic, and political contexts. Hunt, however, expressly distanced himself from the moral panic concept.

By contrast, Sean Hier (2002, 2008) wanted to retain a version of moral panic within the framework of moral regulation as a constant struggle over the process of moralization: that is, who or what should be made morally accountable. He endorsed major criticisms of established moral panic analysis, including its reliance on cognitive, behavioral, and normative measures of the gap between the reality of the problem and its social construction. Disproportionality cannot be evaluated without any reliable “indication of what constitutes a realistic level of concern, anxiety or alarm” (Hier, 2008, p. 178).

Drawing on the Foucauldian concept of governmentality, Hier sees moral regulation as invariably requiring ethical self-formation of both would-be regulator and regulated: “projects of moral regulation reveal as much about the identity of those who seek to regulate as they do about those who come to serve as the object of regulation” (Hier, 2002, p. 328). There is no inherent limit on the scope of moral regulation: “moralization can manifest itself empirically in any number of forms” (Hier, 2008, p. 172).

Moral panics and moral regulation share two characteristics. Each involves one set of people seeking to act on the conduct of others. In both, the regulators confirm their own identities even as they try to alter others’ behavior. But the differences are significant. First, moral panics do not need any “character reformation of moral deviants” (Hier, 2002, p. 329), but only direct and coercive intervention. Second, moral panics differentiate innocent victims from culpable perpetrators more clearly than moral regulation does. They appeal to a moral economy of harm: the idea that some are injured by the activities of others. A moral panic is a temporary rupture when the routine process of moral regulation fails: “the volatile local manifestation of what can otherwise be understood as the global project of moral regulation” (Hier, 2002, p. 329, emphases in original).

Hier includes in moral regulation a wide variety of issues, including “public surveillance, crime and disorder, child allergies, bullying, teen violence amongst females,” “hockey parents,” and “myriad health concerns” (2008, p. 186); “criminality, health risk, sexual deviance and general perceptions of public/personal safety” (2002, p. 331); the “detention of asylum seekers” or “public health scandals” (2008, p. 172); and “commercially financed athletic center advertisements” or “publicly funded anti- smoking campaigns” (2008, p. 180).

Critcher (2009) objects to such a liberal extension of the scope of moralization. Following Cohen (2002), he holds that there is a boundary on moral panic topics. Food safety issues, like bovine spongiform encephalopathy (BSE) or E.coli, are excluded because the moral issues remain distinct from the political ones. In moral panics, blame is not a matter of technical or managerial inefficiency, but rather a serious moral failing. There is a difference between being corrupt and being incompetent. Critcher proposes instead assessing the construction of social problems on three criteria: moral order, the degree of perceived threat to basic values; social control, the extent to which there is identified a viable solution; and governmentality, how far moral regulation of others is represented as requiring ethical formation of the self. He constructs a typology where issues (and hence the likelihood of them transmuting into a moral panic) rate high, medium, or low for each criterion.

Hier (2011c) subsequently accuses Critcher of “relying on a set of normative assumptions about moral order, social control, and ethical self-regulation” (p. 525) and misunderstanding Hier’s project to identify the “political and moral articulation of risk, harm, and personal responsibility” (p. 539). Meanwhile, Hunt returns to his critique of moral panic models, arguing that they are limiting and limited. Moral regulation is “better able to handle the difficult and complex entanglements associated with the problematization of risks and harms” (Hunt, 2011, p. 66). A journey down the moral regulation route may sooner or later involve jettisoning moral panic. In the same article, Hunt insists that “it is crucial that attention be focused on ‘social anxiety’” (p. 67). Assessing explanations of who panics and why is our next consideration. Three possibilities are risk society theory, the culture of fear, and the politics of emotion.

Moral Panics in Social Context: Risk, Fear, and Emotion

Risk

As Carrabine (2008, p. 162) has noted about moral panics, “all of the recent attempts to update the concept are drawn, in one way or another, to the risk society thesis.” It is not hard to see why. At a very basic level, moral panics embody a sense of risk. Most obvious is the case of children at risk of abuse. But the sense of risk is palpable elsewhere as well: the risk of becoming a victim of crime; the risks posed to the indigenous way of life by the presence of immigrants; the risks to personal health and public order from alcohol or drug abuse; and the risks to the well-being of children and young people from their immersion in new media. These are clearly different types and degrees of risk, but all indicate vulnerability and the need for protection. “The global scope of the risk society, its self-reflective quality and its pervasiveness create a new backdrop for standard moral panics” (Cohen, 2002, p. xxv).

It is a more elaborate enterprise to connect the particular concerns of moral panic analysis with the more global perspective of risk theory. This body of work associated with Ulrich Beck and Anthony Giddens—to which Lupton (2013) remains an essential guide—argued that modern Western societies had become extremely risk conscious. Lupton (2013, p. 17) summarized their overall view as follows:

[T]he contemporary obsession with the concept of risk has its roots in the changes inherent in the transformation of societies from pre-modern to modern and then to late modern.

While there are real differences between risk society perspectives, everyone agrees that

(1) risk has become an increasingly pervasive concept of human existence in Western societies; (2) risk is a central aspect of human subjectivity; (3) risk is seen as something that can be managed through human intervention; and (4) risk is associated with notions of choice, responsibility and blame.
(Lupton, 2013, p. 37)

Risk operates across both everyday personal life and the bigger issues of politics and public life. An enhanced consciousness of risk, therefore, might explain why late modern societies seem to experience a greater number and intensity of moral panics. However, such a connection has yet to be made. Two factors may be important. One is that risk theory is appropriated by comparatively narrow specialisms. So analysts are interested in what risk theory says to or about child abuse, crime, drug taking, new media technology, or immigration. Few seem interested in how the risk society is likely to construct social problems as a totality. The second factor is that in some areas—crime, drug taking, and, less clearly, child abuse—risk theory has been usurped by the governmentality approach. This utterly ignores moral panic analysis and subsumes risk within its overarching theory. So early attempts to combine risk and governmentality theory in analyzing the policing of crime, such as that by Ericson and Haggerty (1997), have faded away. Ironically, governmentality rules.

Risk theory has actually been used to mount an attack on conventional moral panic analysis. Ungar (2001, p. 274) asked the question being explored here: “How will the rise of such risk society issues affect the occurrence and development of moral panics?” His answer was that the established concept is no longer relevant in the risk society. The assumptions made by moral panic analysts about which are the most salient issues, who are the most significant actors, and what are the most likely outcomes are all inappropriate for the more unpredictable, contested, and fluid course of problem identification and management in the risk society.

Those moral panic scholars who had high hopes of risk society have had them dashed. Perhaps a better alternative is a different effort to pinpoint contemporary society’s predisposition to panic—theories about a culture of fear, discussed next.

The Culture of Fear

The major works focusing on the topics and dynamics of fear are Furedi (1997), Glassner (1999), Altheide (2002), and Bauman (2007). The clearest definition of the culture of fear comes from Altheide (2002, p. 2):

[T]he pervasive communication, symbolic awareness, and expectation that danger and risk are central features of the effective environment or the physical and symbolic environment as people define and experience it in everyday life.

Each author specifies different causes for the culture of fear, but all identify the essential paradox—that Western societies, apparently more secure than any before, have produced a pervasive culture of fear, qualitatively different from anything which preceded it: more pervasive, more free-floating. It is realized most powerfully in traditional mass media (sometimes entertainment but mainly news), as well as in other public discourse. The culture of fear systematically misrecognizes social problems, producing a distorted and disproportionate response. A major consequence is hostility toward those defined as deviants. A secondary effect is to foster distrust of others, especially strangers.

The argument is in many ways persuasive. It explains a predisposition to collective overreaction to and panic about perceived threats. It includes an account of why those who objectively have least to fear subjectively experience that fear the most. It shares some deficiencies with risk theory. The geographical scope is vague, failing to explain which countries do or do not share a culture of fear. There are difficult historical questions about when and why this culture grew. Its ontological status is dubious: Do we all live out fear on a daily basis, or is it a pervasive concern of political and cultural institutions that only occasionally impinges on the private sphere? A neglected idea from disaster research is to explore how panic develops among elites (Clarke & Chess, 2008) who may have more to fear than most.

In a rare critique of the thesis, Pain (2010) has emphasized its lack of empirical evidence of fear among the general population. Using data from polls following terrorist attacks in major capitals, she showed that fear is not the main emotional reaction; that when it occurs, it is at a low level; that it declines with time and distance; that it is consistently greater in the United States than elsewhere; and that fear is most evident among marginal groups, either within the majority community or among ethnic minorities. She concludes that “much of the empirical evidence tells a different story from recent high-profile texts on the geopolitics of fear” (Pain, 2010, p. 228). Pain emphasizes that emotional reactions are more nuanced, varied, and situationally dependent than blanket profiles of whole cultures can accommodate.

The Politics of Emotion

Just beneath the surface of moral panic analysis, and sometimes briefly on the surface, is a bubble waiting to burst: “the formation of a moral panic is a thing of energy and emotion rather than a simple mistake in rationality and information” (Young, 2011, p. 255). It is present in the original studies, as Cohen chooses the term panic and cites disaster research and Goode and Ben-Yehuda emphasize hostility and volatility as dimensions of collective behavior. Emotional salience explains why some moral panics (mugging, pedophilia, immigration) galvanize the general public, while others (recreational drug consumption, misuse of social media) do not. Emotional vulnerability is implicit in the risk society thesis (if thinly disguised as ontological insecurity), but explicit in the culture of fear, where whole societies are taken to be in a permanent state of emotional alert. More recently, Hunt (2011, p. 54) has argued for the need to take more seriously the “experiential intractability” of “social anxiety,” which emerges when societies undergo rapid cultural change.

Young (2009) activated the idea of “ressentiment,” a reservoir of moral outrage seeking targets for its negative emotions. This derives from traditional sociological analysis of system strain whenever the moral equation of effort and reward becomes imbalanced. This produces, especially among the aspiring middle classes, a degree of status frustration that finds emotional expression in hostility toward those perceived as deviant. Yet, like the risk society and the culture of fear, this is all so much fine argument, with little tangible proof.

Walby and Spencer (2011) are critical of assumptions that the emotional mood of the public can be inferred from media coverage or political pronouncements. Required instead is (p. 104)

empirically investigating what emotions do, how emotions align certain communities against others, and how emotions move people towards certain (sometimes violent) actions against others whose actions pose alleged harms.

Studying the construction of child abuse in Britain, Warner (2015) has argued for understanding “emotional politics.” Child abuse in particular (but potentially any moral panic) seems to trigger moral intensity: “to make judgements requires a sense of what is right and wrong and such moral evaluations are shaped by how we feel about the issue” (Warner, 2015, p. 11, emphasis in original). Anger and contempt, outrage, and disgust are common reactions to stories about abused children. Feelings of shame give rise to a need to blame somebody for the events (Warner, 2015, p. 6):

While emotions are generally thought of as being experienced by individuals and often being brief and episodic, the emotions that are politically important are experienced collectively and embedded in political institutions; they are also enduring rather than short-lived.

Politicians, the media, and official inquiries articulate moral judgments, inviting the public to share their emotions. Despite differences in welfare systems, Warner finds similar types of emotional politics around child abuse in Australasia, the Netherlands, Sweden, and New York in the United States, as well as Britain. Emotional blame is directed in different combinations at the underclass, women, and ethnic minorities. Emotion is socially structured.

A sociology of emotions might also take advantage of insights from social psychology about how groups construct and maintain boundaries with other groups. Pearce and Charman (2011) utilized two social psychological models. Social identity theory seeks to explain the significance and dynamics of defining in-groups and out-groups. The theory of social representations examines how people construct common sense categorizations of other people and their behavior. Their study analyzed discourse in both the media and focus-group discussions about asylum seekers in Britain. They discovered a consistent congruence between media and public discourses. Asylum seekers were perceived as economically, culturally, and physically threatening, especially as their inflow was uncontrollable. Boundaries between “us” and “them” were not completely rigid, but there were strict conditions governing flexibility. The project was not designed to explore the specifics of emotional response but resentment, hostility and metaphorical imagery (inevitably of floods) were all very evident. Overall the approach revealed “the potential for social psychological theory to extend the explanatory value of moral panic” (Pearce & Charman, 2011, p. 309).

Crossing Disciplinary Boundaries

This article has outlined and explained the original moral panic models. It summarized what cumulative research has indicated as empirical generalizations about moral panics. The many and varied criticisms of moral panic models were explored, as were the rejoinders from their supporters. Subsequently, the discussion explored the debate about whether or how to rethink moral panics as exceptional moments in an ongoing process of moral regulation. The focus then shifted to the broader societal context where, in different ways, theories about the risk society and the culture of fear sought to account for the apparently increasing prevalence of moral panics. Finally, it looked at approaches that explored the social psychological dimensions of expressing collective emotions and constructing group identities.

There has not been the opportunity in this text to consider the academic status of moral panic analysis. The moral panic concept does not belong to any larger theory of the social formation. That is why it can be and has been appropriated by very different kinds of theory—by symbolic interactionism (Cohen, 2002) and collective behavior (Goode & Ben-Yehuda, 2009) in the original studies, and then by Marxism (Hall et al., 1978) and even occasionally by feminism (Gelsthorpe, 2005). Moral panic analysis may be best understood as what Robert Merton (1967, p. 39) called “middle range theory”:

Middle range theory is principally used in sociology to guide empirical inquiry. It is intermediate to general theories of social system which are too remote from particular classes of social behaviour, organization and change to account for what is observed and to those detailed orderly descriptions of particulars that are not generalized at all. Middle-range theory involves, abstractions, of course, but they are close enough to observed data to be incorporated in propositions that permit empirical testing. Middle-range theories deal with delimited aspects of social phenomena, as is indicated by their labels.

Realizing the full potential of moral panic analysis as middle range theory may require a properly interdisciplinary approach. It involves movement across disciplinary boundaries, such as those between sociology and psychology or policy and media studies. It also needs a comparative framework across space and time. While no single moral panic study can possibly be expected to incorporate all these disciplinary perspectives, that seems to be the direction in which the field of moral panics as a whole ought to go. Only time will tell.

Browse All

Share Link

Article contents

Moral Panics

Moral Panics

Summary

Keywords

Subjects

Introduction

Original Models

Cohen’s Processual Model

Goode and Ben-Yehuda’s Attributional Model

Concern

Hostility

Consensus

Disproportionality

Volatility

Comparing the Two Models

Accumulated Knowledge

Five Topic Clusters

Generalizations

Qualifications

Moral Panics in Time and Space

Critiques of Moral Panic Models

Loose Terminology

Disproportionality

Outdatedness

Rigidity

Political Partisanship

Assumed Media Effects

Proposed Revisions

Counterarguments

Moral Panics and Moral Regulation

Moral Panics in Social Context: Risk, Fear, and Emotion

Risk

The Culture of Fear

The Politics of Emotion

Crossing Disciplinary Boundaries

Further Reading

References

Notes

Related Articles