Towards a Bourdieusian frame of moral panic analysis: The history of a moral panic inside the field of humanitarian aid

Introduction: Back to the future of moral panic theory

Since its inception in the early 1970s, the concept of moral panic has grown to become one of the most prominent concepts in the sociology of deviance and mass media. For more than four decades, the number of publications and conferences on moral panic has shown no sign of letting up. Whilst panics continue to inform contemporary social problems and changing mechanisms of control and regulation, moral panic remains one of the most illuminating concepts to explore the cultural dynamics of, and responses to, social anxieties and fears. As David Garland (2008: 9) puts it, ‘if Cohen had not come up with the term in 1972, it would have been necessary for someone else to invent it’.

Over the past forty years, however, the concept of moral panic has become a rather eclectic and fissured, rather than cohesive, assemblage (see David et al., 2011). Today, this sociological construct, moral panic, is facing two equally dangerous pitfalls: routinization/trivialization (due to a broad and vague use) on the one hand, and progressive abandon (due to a lack of sociological interest) on the other. For moral panics to avoid windling away, it is essential to approach the concept from different theoretical standpoints. After exploring the inherent weaknesses of moral panic research, the article will re-energize the concept of moral panic by expanding on Bourdieu’s theoretical concepts of habitus, field and hysteresis. It shows how a Bourdieusian frame of moral panic research does away with the identified pitfalls while building more theoretical rigour into the field of moral panic studies.

The concept of moral panic can be a sterile or fruitful point of departure to explore specific types of reactions to social problems, depending on whether it is used as a mere debunking tool or as a tool suited for the advancement of the sociological imagination. As Jock Young (2011) explains, the concept of moral panic originally spoke to C Wright Mill’s long-standing injunction to investigating the relations between ‘the personal troubles of a milieu’ and ‘the public issues of social structure’ (1958: 8). The three original panic studies (Cohen, 2002 [1972]; Hall et al., 1978; Young, 1971) provided evidence of sociological imagination by connecting the conditions of collective anxiety with broader processes of historical and structural changes in advanced industrial societies. At this moment of historical conjuncture, that is, post-war Britain, moral panics were a clear illustration of what Mills (1959: 11) once described:

When people cherish some set of values and do not feel any threat to them, they experience wellbeing. When they cherish values but do feel them to be threatened, they experience a crisis—either as a personal trouble or as a public issue. If all their values seem involved, they feel the total threat of panic.

Yet contemporary moral panic studies have forfeited much of the sociological imagination that was once at the heart of the concept of moral panic (Garland, 2008; Young, 2005, 2009). Today, the concept is very often merely applied rather than thought through. More particularly, the notion of moral panic is used as a negative normative judgement to describe seemingly excessive and inappropriate reactions to social problems (see, for instance, Burns and Crawford, 1999; Rothe and Muzzatti, 2004; Welch et al., 2002). What causes a particular audience to react disproportionally in a specific time and location is not thought relevant to the new project at hand, that is to say, to “liberate” people by suggesting that their reactions are just misinterpretation of the actuality of the danger (see Garland 2008). Interestingly, the academic use of the notion of moral panic, as a short-hand term for talking about surface characteristics like ‘disproportionality’, has become quite similar to its caricature version in the popular media, with both uncritically employing the term to describing collective reactions to social problems (Jewkes, 2004: 65; see also Altheide, 2009; Critcher, 2003; Hunt, 1997).

Insofar as panic scholars strive to debunk seemingly irrational reactions to social problems, then they do not concern themselves with the political values and moral standpoints advanced by their work. Durkheim, whose influence on the sociology of moral panic is direct, warned against the danger of ‘claiming at the outset the relationship of the normal state, and the contrary state … of social abnormality’ and so ‘to reach prematurely the essence of the phenomena’ (Durkheim, 1966: 91). In the Rules of Sociological Method, Durkheim (1966: 75) wrote that:

When research is only beginning and the facts have not yet been submitted to any analysis, their sole ascertainable characteristics are those sufficiently external to be immediately apparent. Those less apparent [in French in the text, ‘plus profond’] are doubtless more essential. Their explanatory value is greater, but they remain unknown at this stage of scientific knowledge and cannot be visualized save by substituting for reality some conception of the mind.

As Rohloff and Wright (2010) observe, the ambition of panic scholars to expose what they perceive as exaggerated reactions to social problems might explain their present-centred focus at the expense of the historically structured processes that feed into moral panics. When priority is given to descriving apparently excessive reactions to social problems, panic scholars do not suscribe to the sociological imagination and deny opportunities for exploring deeper socio-cultural trends generating panics. Assessing the dimension of disproportion of social reactions, as Critcher (2008: 1138) points out, ‘is not the entire analysis itself’, but a ‘means of beginning an analysis of larger social conditions’. As Jock Young (2009: 14) emphasized,

The notion of disproportionality is something of a paradox. For the response to the event is somewhat proportional to the anxiety, otherwise it would simply not be a fully-fledged moral panic. What is disproportionate is the reaction to its immediate manifestation. It is proportional to the anxiety, not to the actual event. It is, on the surface of things, a mistake in reason, but it is not, on a more in-depth level, a mistake in emotion.

The debunking impetus of contemporary panic scholars has generated a wider disenchantment with the analytical validity of the concept of moral panic as misguided in its objectives and priorities (see, for instance, Hunt, 1999; Moore and Valverde, 2000). Eventually the progressive abandon of the concept of moral panic is strengthened by the perception that moral panic theory is no longer adequate to reflect upon contemporary societies (McRobbie and Thornton, 1995; Ungar, 2001; Furedi, 2010) and no longer coherent in the face of recent developments in social theory (Critcher, 2009; Hier, 2002; McRobbie, 2011; Thompson, 1998).

The draining away of support for moral panic research has encouraged several authors to re-engage with the sociology of moral panic.Drawing from the sociology of risk and moral regulation, Hier (2002, 2008, 2011) suggests to reconceptualize moral panics as a temporary inversion of individualizing long-term regulatory processes (for instance, drinking responsibly or monitoring children more closely in the park) into collectivized yet volatile discourses of harm posed by ‘irresponsible’ others (see also Critcher, 2009). Such discourses call for more direct defensive group reactions to ensure an inclusive sense of collective safety (for instance, punishing drunk drivers or publicly announcing the identity of sex offenders). Drawing on Norbert Elias’ social theory, Amanda Rohloff (2011; see also Rohloff and Wright, 2010) identifies moral panics as short-term moralizing campaigns against ‘less civilized’ others at a time when long-term civilizing processes of the self break down. According to Rohloff, the notion of moral panic can be equally defined as a ‘civilizing offensive’ at the intersection between civilizing and de-civilizing trends ‘with a corresponding increase in cruelty and the potential for re-emergence of violence into the public sphere’ (Rohloff and Wright, 2010: 417). Rohloff is inspired by the issue of climate change and offers the example of the ozone hole, with a dialectic form of moralizing between individualized risk management process (for instance, protecting oneself and ones’ children from the sun) and collective moralizing campaigns (for instance, calls from authorities to implement CFC products) (Rohloff 2011, p. 639).

These novel approaches are at once timely and confusing additions to current debates and discussions surrounding moral panic theory. There are clear opportunities to extend the scope of moral panic research to incorporate new theoretical strands of thought such as discourse analysis, risk society and moral regulation. As David Garland (2004: 163) points out, ‘[a] theoretical edifice that remains static over time is a monument to a mistaken view of what theory is for’. However, what is particularly remarkable in attempts to think ‘beyond’ moral panic is that they have built an entirely different version than the one originally conceived by Young, Cohen and Hall in the mid-1970s. For scholars like Hier or Rohloff, moral panic is no longer construed as the formative element, triggering broader regulatory processes, but rather as a short-term rupture in a wider project of moral regulation (Hier, 2002). The resulting transformation is a total reversion of the sequence of processes. Arguably, then, these novel accounts cannot be described as an extension of moral panic theory but as a rupture with its original goals and priorities.

From this revised vantage point, moral panics are taken to flow directly from the actions of irresponsible others, or pre-labelled folk devils. Accordingly, the reasons why certain groups are made into folk devils or why they refuse projects of moral regulation are not (anymore) considered a matter worthy of sociological examination. Moreover, these novel accounts expressly evacuate the disproportional character of moral panics (Rohloff and Wright, 2010: 419). Equally problematic, they suggest unpredictable periods of moral panics; the likelihood of moral panics is always present, appearing and disappearing from outside broader regulatory processes without evident cause. While the original focus of moral panic is on broader historical changes in social structures, the focus of contemporary panic research is on ‘fleeting events, peripheral disturbances in an otherwise regulated universe’ (Young, 2009: 4).

By thinking ‘beyond’ moral panics, these scholars miss many important context-specific, historically-situated explanations behind moral panics. In the manner of the grand concepts of grand theory, their revised concept of moral panic provides ready-made answers to all possible queries about the when, why, who and whither of moral panic. Particularly relevant here, Pierre Bourdieu (1984: 160–161) observed that: ‘[t]hose who pay instant attention to the instant, which, drowned in the event and the emotions it arouses, isolates the critical moment, and thus constitutes it as a totality containing within itself its own explanation’.

Casting the intellectual project of moral panic research in Bourdieusian language

There is a need for the concept of moral panic to remain dynamic and evolve in the face of recent developments in social theory. This article submits that the social theory of Pierre Bourdieu would provide greater insight into the cultural dynamics and social conditions of moral panic without evacuating the concept’s original equivocations. The original moral panic toolkit (folk devils, disproportionality, deviancy amplification, etc.) remains central to understanding the inherent dynamics and processes that animate moral panic. However, the sociology of moral panic would gain a lot in terms of theoretical and methodological depth by relying on Bourdieu’s concept of field, habitus, and hysteresis. Accordingly, this article does not supplant but rather complements the under-developed theoretical grounding of moral panic research by relying on Pierre Bourdieu’s conceptual framework.

To cast moral panic in a Bourdieusian language contributes to more than just introducing new concepts in the sociology of moral panic. From the outset, there has been a tension between the structural and the psychological forces behind moral panics. As it will be explained below, the analysis of the relationship between habitus and field, as relatively autonomous structures, helps to overcome the classical structure/agency dilemma that continues to plague moral panic studies. It is the nature of the relation between these relational structures that provides the key for understanding moral panics. Overall, a Bourdieusian model of moral panic would bring to bear the tools of structural analysis and historical explanation in a dialectical relation to the cognitive and behavioural dimensions of moral panics.

The treatment of Bourdieu’s concept of field opens new ways of inquiring into moral panics. It allows moral panic researchers to delimit the form and contour of a given moral panic. Bourdieu (2005: 32, emphasis in original) defines a field as ‘a kind of small universe caught up in the laws of functioning of the larger universe, but nonetheless endowed with a relative autonomy within that universe and obeying its own laws, its own nomos’. Each field has its own set of shared assumptions, values and practices, or doxa, which appears as natural and self-evident for those inhabiting the field in question. The doxa is shaped within and beyond the perceptions of the agents who are active and acting in the field through the medium of their habitus. The notion of habitus can be defined as a disposition that informs perceptions of the field as a meaningful world; the habitus is cognitively structured by the field, which is a social world endowed with sense and meaning (Bourdieu, 1988).

Importantly, this approach locating moral panickers inside a given field does not suggest isolating the field in question from its wider environment and excluding the activities of external actors. Eventually a fully fledged moral panic could not emerge without essential connections and transactions with, and at the intersection between, other fields, including the media, the academic and the political fields. External actors do play a crucial role in propagating moral panics, but only because of underlying crisis in a given field or social space, out of which a moral campaign can be built.

However, a Bourdieusian reading of moral panic would locate moral panickers inside the structure of a given field, within which they can experience effects of hysteresis, a concept that is explored later. One considerable advantage of thinking in terms of field is to avoid what Bourdieu (1988) calls the ‘short-circuit fallacy’—in the case of moral panic, the tendency of passing directly from the activities of some external agents such as politicians or the media to moral panics, without exploring the specific mediation of the structure of the field in question.

Further, the notion of a field is not restricted to a geographical viewpoint. Fields present themselves in various economic, cultural and social dimensions, whether at the local, national, international or supranational level. The general population feature of traditional panic research does not allow the possibility of moral panic happening in a more microcosmic environment (like in the case studied here, i.e. the field of humanitarian aid) or within larger social spaces. A field approach to moral panic opens new and interesting avenues of moral panic research (e.g. moral panic about doping in the field of sport, see Goode, 2011, moral panic around child abuse in the child welfare social work field, see Smith, 2006, moral panic about new wars in the field of international politics, see Duffield, 2001).

As Bourdieu points out, social processes of accordance or match between subjective dispositions (habitus) and objective conditions (field) usually take place gradually along the established trajectory of the previous field configuration. In case of rapid and abrupt field restructuring, the habitus takes some time to adjust, generating moments of discordance or mismatch with the new environment into which social actors have strayed. Bourdieu uses the term hysteresis to describe the effect of the past into the present, whenever ‘changes in objective structures are so swift that agents whose mental structures have been molded by these prior structures become obsolete and act inopportunely (à contre-temps) and at cross purposes’ (Bourdieu and Wacquant, 1992: 130). Effects of hysteresis emerge from a structural disadjustment between field and habitus, as the result of a rupture between the routine adjustment of subjective and objective structure.

Effects of hysteresis can be the source of moral panic. In other words, moral panics take place during critical moments of transition, when the doxic complicity between field and habitus breaks down and the prevailing system of meaning undergoes historical transformation. It is as a result of the internalization of the doxa that rapid and brutal field changes become unsettled, generating episodes of moral panic. This approach recalls the traditional conceptualization of moral panic. Yet it also augments its original equivocation by maintaining the analytical integrity of habitus and field as relatively autonomous structures—with the former in need of being constantly adjusted to the latter. As Bourdieu (2000: 160) reminded,

Habitus is not necessarily adapted to its situation nor necessarily coherent. It has degrees of integration—which correspond in particular to degrees of ‘crystallization’ of the status occupied. If they are particularly frequent, however, they must not be treated as a universal rule.

Moral panic is a particular form of orthodox strategy at a moment where the ruling doxa has been destabilized. These moments of critical transition, when a field displays a high degree of volatility, are more favourable to moral entrepreneurs united by affinities of habitus, who are largely preoccupied with the task of defending the doxic order from the previous state. Importantly enough, the folkways proposed by moral entrepreneurs do not offer a reflexive way out from the crisis of meaning, what Bourdieu (1984: 191) refers to as ‘the myth of the moment of awareness’, but, rather, propels a reactionary move that is aimed at covering up the misfiring of the habitus. As Bourdieu (1984: 191) points out, ‘this work is doubtless more important in times of crisis, when the meaning of a social world less totalizable than ever vacillates’.

If the term hysteresis may well conjure the image of hysteria, it would be misleading to understand effects of hysteresis (and then moral panics) as merely the result of cognitive biases or stimulus. As Cheryl Hardy (2008: 133, emphasis in original) points out,

[Bourdieu] breaks with a pre-existing moral position and suspends moralistic judgment in the interest of objectivity. He therefore chooses a word that supports a more scientific view of the relationship between society and the individual, between the subjective and the objective, hysteresis as a field condition affecting individuals within this social space. ¹

Bourdieu (2000: 160) often illustrates effects of hysteresis by citing the Don Quixote’s syndrome. While ‘tilting at windmills’ appeared to his audience as mysterious, irrational and fantastic, it is nonetheless a creative response to a divided habitus, or a ‘time lag’ between ‘the idealized chivalric rules of the aristocratic feudal habitus and the actual world of transition to commercial capital, where money rules supreme’ (Fowler, 2002: 47). This Bourdieusian approach to moral panic also provides an insightful explanation of the tendency of moral panic scholars to ascribe negative pejorative connotations to social reactions, as ‘irrational’, ‘exaggerated’ and ‘disproportionate’. As Bourdieu (2000: 148–149) observes,

when an insignificant, apparently fortuitous event unleashes enormous consequences that are bound to appear disproportionate to all those who have different habitus, one should not say that a historical event determined a behaviour but that it had this determining effect because a habitus capable of being affected by that event conferred that power upon it.

Moral panics do not affect social actors in isolation, whether individuals, organizations or groups; they alter a structured totality that is visible only comprehensively rather than selectively. That being said, a field is also a social arena of constant struggle and competition over specific types of capital. Until now, moral panic scholars did not feel bound to develop a theoretical model that would allow them to analyse differential reactions to social problems. Arguably, the original idea of moral panic pointed to cases of relative indifferentiation and generalized social reactions/positions on a topic. In other words, it is assumed that all and every moral panickers react in the same way and with the same level of energy to a given moral panic. By contrast, a field analysis allows exploring differentiations of reactions to social problems. For instance, humanitarian actors do not get swept into the moral panic about humanitarian insecurity in undifferentiated ways, but rather varies in accordance to positions, dispositions and position-taking inside the field of humanitarian aid. Understanding differential reactions towards dangers and insecurity (e.g. whether aid agencies should accept of refuse armed convoys or private security companies) requires exploring structural positions (relations) of aid organizations in the field of humanitarian aid, together with related capital formations.

The next section explores the crisis that unfolded inside the field of humanitarian aid in the post-Cold War era, and how this generated a moral panic about the dangers humanitarian aid workers face. This crisis may appear surprising to any external observer, given the massive expansion of the sector (in terms of institutional recognition, growth, professionalization, public attention, etc.). As Barnett and Weiss (2008: 9) wonder:

Why … do those in the sector act as if they are slipping past their golden years rather than enjoying their golden age? What are the forces that have so thoroughly shaken the humanitarian sector that they have unleashed this ontological insecurity? Why is there such anxiety amid all these accomplishments? What, precisely, has changed?

The rise of a moral panic in the field of humanitarian aid

Over the past two decades, insecurity for aid workers has become a matter of great concern and a central issue within the aid community. Perhaps the most emblematic expression of this growing mood of anxiety is the bunkerization of aid compounds, where segregated living became accepted as the most reasonable way to deter attacks (Dandoy, 2013; Duffield, 2010) For experts and commentators, this widespread anxiety is due to the fact that international humanitarian organizations are facing a type of threat that is different from anything they confronted in the past, while operating in increasingly hostile environments (Bruderlein and Gassmann, 2006; UNHCR, 2010: 1). For example, it is argued that ‘[a]ttacks on humanitarian workers in the world’s danger spots represent an alarming trend that is likely to accelerate in the years ahead’ (Rogers, 2001); ‘[o]ver the past decade, threats against the safety and security of international aid workers and peacekeeping forces have escalated at an unprecedented pace’. Forced to operate in increasingly dangerous environments and in complex emergencies, the mortality and distress rates of field staff have increased dramatically’ (Currier, 2003); and so on. While the main danger was once being ‘in the wrong place at the wrong time’, it is argued that aid workers are now increasingly, deliberately targeted. This claim about quantitative and qualitative shifts in the nature of the threat to aid workers is repeated to the point of monotony in countless reports and policy documents on aid worker security.

While these catastrophic claims are viewed with some dismay across the aid community, they present a number of features that need to be considered (Dandoy and Pérouse de Montclos, 2013). For one thing, the lack of information about the geographical spread of local and international staff across different regions is a serious obstacle to measure rates and to assess the dangerousness of different operational contexts. Without more precise information, the probability of being killed in a given location remains unknown. Moreover, experts and commentators on humanitarian insecurity tend to present absolute rather than relative figures on aid worker security incidents. Yet it is important to consider the increasing number of aid worker fatalities over the past two decades in regards of the dramatic increase of the aid worker population during the same period. In their last report, the authors of the Aid Workers Security Database pointed out that:

an over-emphasis on the yearly totals [of attacks] can obscure more meaningful indicators of aid worker security, i.e. rates of violence (…) Taking the number of aid worker victims against their estimated population reveals a modest rise in the long-term trend in attack rates.

Considering the growing number of humanitarian personnel in the field, which increased by an estimated 77 per cent from 1997 to 2005 (Stoddard et al., 2006: 1), there is little evidence to suggest that their mortality rate has radically increased since the end of the Cold War.

In any case, counting aid worker security incidents over the last 20 years or so is too short to assess trends in humanitarian insecurity in the long run. One can even wonder whether, historically, casualties have increased in absolute numbers. During the invasion of Ethiopia by the Fascists in 1936, the Italian bombings killed a total of 47 patients and employees from the International Red Cross movement (Baudendistel, 2006). In a single country and for one single organization, this represents, in less than one year, a figure that is comparable in absolute number to the annual average recorded today for an aid worker population that is much greater worldwide. In this regard, it makes sense to look at the annual reports from the International Committee of the Red Cross (ICRC), which is the only organization that has kept relatively constant data on staff level and security incidents since the Second World War. According to various estimates, 76 aid workers were killed in 2003, and a total of 271 for the period between 2000 and 2005 (Stoddard et al., 2006). But the national societies of the Red Cross paid a higher tribute during the Second World War, from 1939 to 1945. In five years, the Italian Red Cross recorded 275 fatalities, while its French counterpart lost 414 employees and volunteers. Taken alone, both national societies recorded an average of 138 deaths per year. This average is almost twice as highthan fatalities for the whole aid community in 2003. Of course, it can be argued that the Second World War was an exceptional case. But when it comes to making sense of violence against relief organizations, then each armed conflict or other situation of violence must be explored separately through empirical research rather than assuming an a priori temporal, deteriorating shift across the 20th century and beyond.

Experts and commentators on humanitarian insecurity also argue that security incidents are more often intentional rather than peripheral side-effects of warfare. For instance, the security director of World Vision International claimed that: ‘By far the most alarming development is the fact that, for the first time in their history, NGOs are now being targeted’ (Rogers, 1998). While intentional violence often lies behind aid worker fatalities (Rowley, 2007; Sheik et al., 2000), there is little evidence to suggest that intentional targeting is a new feature of contemporary warfare. Indeed, the 1864 Geneva Convention was violated during the Franco-Prussian war of 1870, when many nurses and doctors were deliberately killed on the battlefront. At that time, already, the Red Cross emblem did not protect from the enemy shootings. On the contrary, it rather attracted them so that the staff were quick to fold their flag up to cover themselves during the siege of Paris by the Germans (Keeler, 1873). The development and diffusion of the International Humanitarian Law in subsequent decades did not alter the pattern of intentional violence against humanitarian actors. Consider, for example, the case of the vice-president of the Swedish Red Cross, Count Folke Bernadotte, who was killed in an ambush in Jerusalem in 1948. Likewise, in 1943, the ICRC delegate in Borneo, Dr Matthaeus Vischer, and his wife were accused of spying and beheaded by the Japanese as they were counting and giving food to prisoners.

Discussing disproportionate reactions to an objective threat does not suggest diminishing the seriousness of the threat to aid workers nor to ascribe a normative judgement about the appropriate proportionality. According to the US Department of Labour (2004, cited in Stoddard et al., 2006: 4), aid work held the fifth position on the list of the most hazardous civilian occupations in the USA, after loggers, pilots, fisherman and structural iron and steel workers and was the only profession where the death toll was largely due to intentional acts of violence. With a total fatality rate of 242 international aid worker victims in 2011 (Stoddard et al., 2013), the existence of a real physical threat to humanitarian actors is incontestable. Despite this, it appears that there is nothing radically ‘new’ about the dangers that humanitarian workers face. Hence the questions panic scholars should ask: why did the aid community react so remarkably to humanitarian insecurity? What is telling about this particular time period (post-Cold War)? And what does this episode tell us about the changing nature of humanitarian action in the last two decades?

Aid organizations are far from accepting mass casualties, as was the case for instance during the Second World War. Mark Duffield (2010) emphasizes changing subjectivity across the aid community away from the NGO voluntarism of the 1960s and 1970s towards militarized risk-aversion and defensive posture. Likewise, Frank Furedi (2010) observes that ‘even the noble mission of saving lives has been subordinated to the dictates of risk-aversion’. The drastic reduction in the level of acceptance of risk should not be seen as a result of new threats for aid workers. Rather, the rise of a culture of risk withinthe aid community is directly related to the growing anxiety that accompanied the reconfiguration of the field of humanitarian aid over the past two decades (Dandoy, 2013). The Bourdieusian frame of analysis developed in the previous section is useful to to understandthe rise of a moral panic about aid worker insecurity.

The claim that humanitarian actors have entered a period of moral panic over the past two decades requires first an exploration of the origins and contours of the global field of humanitarian aid. Only a few scholars have explored humanitarian practices and values from a meso-level of analysis (Calhoun, 2008; Krause, 2009). Most studies of aid organizations focus on the various parts of a common humanitarian enterprise, as an aggregate of disparate entities that exist in isolation rather than in relation to each other. On the contrary, the strength of a field’s approach is the location of the overarching principles that structure the relations between aid organizations. All humanitarian actors share a ‘specific doxa, a system of presuppositions inherent in membership in a field’ (Bourdieu, 2005: 37)—i.e., a set of working assumptions and cultural commitments about the stakes at play in the ‘humanitarian game’, that is, to ‘save lives at risk’.

Of course, there are distinct differences between the philosophy, values and practices that underpin US-based and European-based organizations, as well as often profound differences between European countries (Scandinavian or French-based organizations). These are rooted in historical, cultural and political traditions, producing different humanitarian practices. Insofar as humanitarian practices are explored in field terms, as Monika Krause (2009) observes, the focus must also be on conflict and competition amongst and between humanitarian actors. Hence the internal debates and disagreements about how humanitarian actors must deliver aid take place against a relatively settled consensus around the stakes at play in the humanitarian game.

This system of presuppositions inherent to membership in the humanitarian field has been gathered over a long period of time, starting with the creation of the International Committee of the Red Cross (ICRC) in 1863. By 1960 the field of humanitarian aid had a settled institutional structure organized around the principles of impartiality, independence and neutrality. In the post-Cold War period, however, this consensus over the ends of humanitarian action collapsed and marked the beginning of a deeper period of turbulence. The collapse of faith in humanitarianism was heralded by a series of watershed events in the Balkans and the Great Lakes region of Africa. In Bosnia, the conspicuous absence of the ICRC and the inability of the United Nations and relief agencies to protect refugees from Serb attacks led to intense debates about the worthiness of helping people that will be ultimately killed—also referred to as the ‘well-fed dead’ (Cohen, 1997). The declining support for humanitarian aid reached a turning point in the aftermath of the Rwandan genocide, when many relief agencies realized that they were serving as bait, inadvertently fuelling the conflict and rearming the Hutu génocidaires in the camps of Goma (De Waal, 1997; Terry, 2002). As Barnett (2005: 725) pointed out, ‘Rwandan tragedy and other events caused the entire community to undergo painful introspection that raised troubling questions regarding the legitimacy and effectiveness of humanitarian action.’

Influenced by reports highlighting the potentiality of aid ‘doing more harm’ (Anderson, 1996; De Waal, 1997), a growing disillusionment and pessimism about the role and impact of humanitarian aid began to pervade the aid community. Rather than hesitating about how to reform the humanitarian system, as Macrae (2000) pointed out, the debates and discussions questioned whether it was worth reforming it at all (see also Rieff, 2002). Eventually a group of aid organizations considered a form of heterodoxy that contested the utility of maintaining the political-humanitarian divide in highly political emergencies. In the post-Cold War period, as Slim and McConnan (1998) observed, neutrality became a ‘dirty word’. Hence the discredit of humanitarian aid precipitated a distanciation from the cherished orthodoxy of the ICRC and a shift towards a set of competing beliefs and values that would combine relief and development assistance with new objectives and missions including human rights, democracy and conflict resolution (Barnett, 2005). As Fiona Fox (2001: 284) observes,

There is a ‘new humanitarianism’ for the new millennium. It is ‘principled’, ‘human-rights based’ and politically sensitive. Above all it is new. It marks a break from the past and a rejection of the traditional principles that guided humanitarianism through the last century. New humanitarians reject the political naivety of the past, assess the long-term political impact of relief and are prepared to see humanitarian aid used as a tool to achieve human rights and political goals.

The dissolution of boundaries between humanitarianism and other activities was further shaped by the push for ‘coherence’ between humanitarian and political engagement within integrated UN missions. As humanitarian actors expanded their mandate to engage with the structural causes of conflicts, they began to identify themselves with non-humanitarian actors who shared similar ambitions, including human rights agencies, development organizations and state agencies (Collinson and Elhawary, 2012; Rieff, 2002). As Michael Barnett (2005: 727) observes, ‘The growing interaction between different kinds of agencies that hailed from different sectors encouraged a relief–rights–development linkage within a humanitarian discourse that became tied to the construction of modern, legitimate, democratic states.’ The ICRC’s hegemonic position

The rapid and brutal transformation of the field of humanitarian aid generated effects of hysteresis, providing a fertile ground for a moral panic about humanitarian insecurity to emerge. While most aid organizations sought a middle ground, emphasizing the need to tackle the structural causes of suffering, they remained uncomfortable with the idea of openly taking sides and abandoning the principle of neutrality (Collinson and Elhawary, 2012: 14). As Donini et al. (2004: 191) observe, there was ‘an uneasy sense that the community is caught up in a chain of events over which it has no control’. Indeed, Bourdieu (1977: 78) points out that ‘practices are always liable to incur negative sanctions when the environment with which they are actually confronted is too distant from that in which they are objectively fitted’. Hence the introduction of innovative practices into a reconfigured humanitarian field ensured that humanitarian actors experienced a ‘full-blow identity crisis … an anxiety associated with deep ontological insecurity’ (Barnett and Weiss, 2008: 5). Donini (2005: 11) touched on this nicely when he wrote that:

The entire aid industry is in turmoil, not just the humanitarian enterprise. The relative slow movement of tectonic plates of the 90’s seems to be morphing into cyclones of devastating proportions that are now sweeping through the noughties. Fundamental changes are taking place in the profession, the industry and the marketplace. Some are dictated by the external environment; others relate to an internal process of evolutionary adaptation. It is either ‘go with the flow’ or ‘gone with the wind’.

The rapid and brutal unravelling of the prevailing doxa and the emergence of new ways of thinking and acting ‘on’ rather than ‘in’ conflicts left aid organizations confused about how to make sense of threats that were previously handled as a matter of course. The number of professional standards and guidelines, such as the Inter-Agency Standing Committee (IASC) or the Sphere Humanitarian Charter and Minimum Standards in Disaster Response, proliferated across the aid community. As Collinson and Elhawary (2012) observe, even the most pragmatic humanitarian actors have found solace in a renewed commitment to the principles of humanitarian action. The standard principles of humanitarian action, once self-conscious and taken-for-granted, looked as if they needed to be rewritten again, as a boundary maintenance mechanism.

Likewise, the moral panic about aid worker fatalities can be read as an extreme form of reaction to effects of hysteresis inside the field of humanitarian aid. It is similar to the notion of a ‘crime wave’ derived from Durkheim by Kai Erikson (1966: 69), as a specific type of social reaction which serves the purpose of affirming and reinforcing the moral boundaries of social groups: ‘Whenever a community is confronted by a significant relocation of boundaries … the community begins to censure forms of behavior which have been present in the group for some time but have never attracted any particular attention before.’ As Erikson (1966: 69) argues,

In the sense that the term is being used here, ‘crime wave’ refers to a rash of publicity, a moment of excitement and alarm, a feeling that something needs to be done. It may or may not mean an actual increase in the volume of deviation.

Humanitarian actors have long attributed to external, western actors the role of causing the ‘new and growing threat’ by blurring the line between aid organizations and military forces. However, the politicization and militarization of humanitarian aid and/or agencies are not as new as suggested by experts and commentators and cannot in itself explain the rise of concern and anxiety about humanitarian insecurity (Pérouse de Montclos, 2014). Time and again, humanitarian actors have been operating alongside, if not directly cooperating with, armed forces in the delivery of aid to populations in need. During the First World War, relief organizations were completely embedded in military units and this integration was widely admitted across the aid community. During the height of the Cold War, the objectives and priorities of relief workers were largely shaped by the foreign policy agenda of their governments. In any case, statistical data do not suggest that intentional violence against humanitarian actors is more frequent in countries where the US or other western powers are involved (Stoddard et al., 2006).

Eventually the main change in the nature of the threat to aid workers is the way it has been conceptualized inside the field of humanitarian aid in the post-Cold War period. As Larissa Fast points out, ‘the response to security concerns across the humanitarian community has been somewhat schizophrenic’ (Fast 2010, p. 2). By linking the concept of moral panic with the social theory of Bourdieu, this article has shown that ‘schizophrenic’ reactions about the perceived new and growing threat to humanitarian workers can be sociologically understood, not as a collective mistake in understanding, but, rather, as a meaningful response to effects of hysteresis in the field of humanitarian aid.

Conclusion

This article has shown how the use of Bourdieusian concepts would breathe life into moral panic theory and humanitarian discussions. It has demonstrated that the concept of moral panic can remain a sophisticated and fashionable tool to illuminate contemporary social problems and changing mechanisms of control and regulation. Recent attempts to relate the concept of moral panic to other theoretical developments have been confusing additions to current debates surrounding moral panic research. Rather than engaging with the inherent weaknesses of moral panic theory, scholars like Hier and Rohloff have created a new model that bears no similarity with the one originally developed by Jock Young, Stanley Cohen and Stuart Hall. The problem is that they have brought the concept of moral panic to such a level of abstraction that no space is given to explore the socio-cultural trends underlying public issues. In other words, the reasons why certain people are ‘panicking morally’ in specific times and locations are not considered a matter worthy of sociological analysis. It appears that, whilst these novel approach are gaining increasing influence in academia (e.g. Critcher, 2009; Herdt, 2009; Sela-Shayovitz, 2011), they miss many important context-specific explanations behind moral panics.

In this article, the notion of moral panic is conceptually enriched by using Bourdieu’s social theory. This novel approach was empiricized through an analysis of the moral panic about humanitarian insecurity in the post-Cold War era. The focus was on the historical-structural processes that fed into this particular moral panic. Although the impacts or aftermath of moral panics have not been touched upon in this article, the sociology of moral panic remains extremely useful to explore changing mechanisms of control and regulation. Again, the project of moral panic studies calls for new theoretical developments to reflect upon increasingly techno-rational mechanisms of disciplining the self and the other (Arrigo, 2013; Hier, 2008). As explored elsewhere, the moral panic about humanitarian insecurity has prompted collectivized discourses of harm avoidance, expressed spatially through the fortification of aid compounds across the world. In the same process, collectivized discourses have been synchronized with long-term individualized discourses of risk management, characterized by the rise of a culture of self-care and responsabilization among humanitarian workers (Duffield, 2010; Dandoy, 2013). Arguably, the rise of a culture of security through pervading mechanisms of risk assessment and risk management denies social (humanitarian) agents’ productive and dynamic potential (Arrigo, 2013; Dandoy, 2014). Overall, a Bourdieusian sociology of moral panic can offer the aid community guidance in terms of developing a new level of reflexivity in their quest for humanitarian (security) governance.