SOCIOHISTORY OF THE FIELD
Descriptions of calamities exist as far back as the earliest human writings, but systematic empirical studies and theoretical treatises on social features of disasters appeared only in the twentieth century. The first major publications in both instances were produced by sociologists. Samuel Prince (1920) wrote a doctoral dissertation in sociology at Columbia University that examined the social change consequences of a munitions ship explosion in the harbor of Halifax, Canada. Pitirim Sorokin (1942) two decades later wrote Man and Society in Calamity that mostly speculated on how war, revolution, famine, and pestilence might affect the mental processes and behaviors, as well as the social organizations and the cultural aspects of impacted populations. However, there was no building on these pioneering efforts.
It was not until the early 1950s that disaster studies started to show any continuity and the accumulation of a knowledge base. Military interest in possible American civilian reactions to post-World War II threats from nuclear and biological warfare led to support of academic research on peacetime disasters. The National Opinion Research Center (NORC) undertook the key project at the University of Chicago between 1950 and 1954. This work, intended to be multidisciplinary, became dominated by sociologists as were other concurrent studies at the University of Oklahoma, Michigan State, and the University of Texas (Quarantelli 1987, 1994). The NORC study not only promoted field research as the major way for gathering data, but also brought sociological ideas from the literature on collective behavior and notions of organizational structure and functions into the thinking of disaster researchers (Dynes 1988; Dynes and Tierney 1994).
While the military interest quickly waned, research in the area obtained a strategic point of salience and support when the U.S. National Academy of Sciences in the late 1950s created the Disaster Research Group (DRG). Operationally run by sociologists using the NORC work as a prototype, the DRG supported field research of others as well as conducted its own studies (Fritz 1961). When the DRG was phased out in 1963, the Disaster Research Center (DRC) was established at the Ohio State University. DRC helped the field of study to become institutionalized by its continuous existence to the present day (having moved to the University of Delaware in 1985). In its thirty-six years of existence DRC has trained dozens of graduate students, built the largest specialized library in the world on social aspects of disasters, produced over six hundred publications and about three dozen Ph. D. dissertations (see http:// www.udel.edu/DRC/homepage.htm), continually and consciously applied a sociological perspective to new disaster research topics, initiated an interactive computer net of researchers in the area, and intentionally helped to create international networks and critical masses of disaster researchers.
DRC was joined in time in the United States by two other major social science research centers (both currently headed by sociologists). The Natural Hazards Center at the University of Colorado has as part of its prime mission the linking of disaster researchers and research-users in policy and operational areas. The Hazards Reduction and Recovery Center at Texas A & M University has a strong multidisciplinary orientation. The organization of these groups and others studying disasters partly reflects the fact that the sociological work in the area was joined in the late 1960s by geographers with interest in natural hazards (Cutter 1994), in the 1980s by risk analysts (including sociologists such as Perrow 1984; Short 1984) especially concerned with technological threats, and later by political scientists who initially were interested in political crises (Rosenthal and Kouzmin 1993). More important, in the 1980s disaster research spread around the world, which led to the development of a critical mass of researchers. This culminated in 1986 in the establishment within the International Sociological Association of the Research Committee on Disasters (# 39) (http:// sociweb.tamu.edu/ircd/), with membership in over thirty countries; its own professional journal, The International Journal of Mass Emergencies and Disasters (www.usc.edu/dept/sppd/ijmed); and a newsletter, Unscheduled Events. At the 1998 World Congress of Sociology, this committee organized fourteen separate sessions with more than seventy-five papers from several dozen countries. The range of papers reflected that the initial focus on emergency time behavior has broadened to include studies on mitigation and prevention, as well as recovery and reconstruction.
CONCEPTUALIZATION OF “DISASTER”
Conceptualizations of ”disaster” have slowly evolved from employing everyday usages of the term, through a focus on social aspects, to attempts to set forth more sociological characterizations. The earliest definitions equated disasters with features of physical agents and made distinctions between ”acts of God” and ”technological” agents. This view was followed by notions of disasters as phenomena that resulted in significant disruptions of social life, which, however, might not involve a physical agent of any kind (e.g., a false rumor might evoke the same kind of evacuation behavior that an actual threat would). Later, disasters came to be seen as crises resulting either from certain social constructions of reality, or from the application of politically driven definitions, rather than necessarily from one initial and actual social disruption of a social system. Other researchers equated disasters with occasions where the demand for emergency actions by community organizations exceeds their capabilities for response. By the late 1980s, disasters were being seen as overt manifestations of latent societal vulnerabilities, basically of weaknesses in social structures or systems (Schorr 1987; Kreps 1989).
Given these variants about the concept, it is not surprising that currently no one formulation is totally accepted within the disaster research community (see Quarantelli 1998 where it is noted that postmodernistic ideas are now also being applied). However, there would be considerable agreement that the following is what is involved in using the term ”disaster” as a sensitizing concept: Disasters are relatively sudden occasions when, because of perceived threats, the routines of collective social units are seriously disrupted and when unplanned courses of action have to be undertaken to cope with the crisis.
The notion of ”relatively sudden occasions” indicates that disasters have unexpected life histories that can be designated in social space and time. Disasters involve the perceptions of dangers and risks to valued social objects, especially people and property. The idea of disruption of routines indicates that everyday adjustive social mechanisms cannot cope with the perceived new threats. Disasters necessitate the emergence of new behaviors not available in the standard repertoire of the endangered collectivity, a community, which is usually the lowest social-level entity accepted by researchers as able to have a disaster. In the process of the refinement of the concept, sociologists have almost totally abandoned the distinction between ”natural” and ”technological” disasters, derived from earlier notions of ”acts of God” and ”man-made” happenings. Any disaster is seen as inherently social in nature in origin, manifestation, or consequences. However, there is lack of consensus on whether social happenings involving intentional, deliberate human actions to produce social disruptions such as occur in riots, civil disturbances, terrorist attacks, product tampering or sabotage, or wars, should be conceptualized as disasters. The majority who oppose their inclusion argue that conflict situations are inherently different in their origins, careers, manifestations, and consequences. They note that in disaster occasions there are no conscious attempts to bring about negative effects as there are in conflict situations (Quarantelli 1993). Nevertheless, there is general agreement that both conflict- and consensus-type crises are part of a more general category of collective stress situations, as first suggested by Allan Barton (1969).
MAJOR RESEARCH FINDINGS
While the research efforts have been uneven, much has been learned about the behavior of individuals and households, organizations, communities, and societies in the pre-, trans-, and postimpact time periods (Quarantelli and Dynes 1977; Kreps 1984; Drabek 1986). A separation of the disaster-planning cycle into mitigation, preparedness, response, and recovery phases has won partial acceptance at some policy and operational levels in the United States. However, international usage of the terms is far from total and there also is disagreement regarding what should be considered under mitigation. Therefore we will continue to discuss findings under the older ”time” breakdown.
Preimpact behavior. Individuals and households. Most residents show little concern about disasters before they happen, even in risk-prone areas and where threats are recognized. Citizens tend to see disaster planning as primarily a moral even more than a legal responsibility of the government. Very few households ever plan in any concrete way for possible disasters. Exceptions to these passive attitudes are where there are many recurrent experiences of disasters as occur in some localities, where disaster subcultures (institutionalized expectations) have developed, and where potential disaster settings (such as at chemical complexes or nuclear plants) are the focus of activist citizen groups.
Organizations. Except for some disaster-oriented groups such as police and fire departments, there usually is little organizational planning for disasters. Even agencies that plan tend to think of disasters as extensions of everyday emergencies and fail, according to researchers, to recognize the qualitative as well as quantitative differences between routine crises and disaster occasions. In disasters the responding organizations have to quickly relate to more and different groups than normal, adjust to losing part of their autonomy to overall coordinating groups, apply different performance standards and criteria, operate within a closer-than-usual public and private interface, and cannot function well when their own facilities and operations may be directly impacted by the disaster agent.
Communities. Low priority usually is given to preparing localities for disasters, and when there is some effort it is usually independent of general community development and planning. This reflects the reactive rather than proactive orientation of most politicians and bureaucrats and the fact that the issue of planning very seldom becomes a matter of broad community interest as would be indicated by mass media focus, discussions in the political arena, or the existence of advocacy groups. Efforts to initiate overall disaster preparedness often are hindered by prior organizational and community conflicts, cleavages, and disputes. Starting in the 1990s, major systematic efforts from the top down have been made in a few countries to push for the implementation of local mitigation measures. For programs to be implemented, however, people must accept the realistic criticism that while a disaster may be a high-impact, it is a very low-probability event.
Societies. Generally, disaster planning does not rank very high on the agenda of most societies. However, increasingly there are exceptions in developing countries when major recurrent disasters have had major impact on the gross national product and developmental programs. Also, in developed societies certain even distant catastrophes such as a Bhopal or Chernobyl can become symbolic occasions that lend impetus to instituting preparedness measures for specific disaster agents. Increasingly too, attention to national-level disaster planning has increased as citizens have come to expect their governments to provide more security in general for the population. Also, mitigation or prevention of disasters is being given higher priority than in the past.
Transemergency Period Behavior. Individuals and households. When disasters occur, individuals generally react very well. They are not paralyzed by a threat but actively seek relevant information and attempt to do what they can in the emergency. Victims while usually very frightened, not only act positively but also show little deviant behavior; they extremely seldom break in panic flight; they do not act irrationally especially from their perspective; and they very rarely engage in antisocial activities, although stories of such contrary behavior as looting may circulate very widely. Prosocial behavior especially comes to the fore, with the initial search and rescue being undertaken quickly and mostly by survivors in the immediate area. Most sudden needs, such as emergency housing, are met by kin and friends rather than official relief agencies. Family and household relationships are very important in affecting immediate reactions to disasters such as whether evacuation will occur or if warnings will be taken seriously, because mass media reports are filtered through primary ties.
Organizations. As a whole, organizations do not react as well to disasters as do individuals and households. But while there are many organizational problems in coping with the emergency time demands of a disaster, these difficulties are often not the expected ones. Often it is assumed that if there has been organizational disaster planning, there will be successful crisis or emergency management. But apart from the possibility of planning being poor in the first place, planning is not management and the former does not always translate well into the latter in community disasters. There typically are problems in intra- and interorganizational information flow, and in communication between and to organizations and the general public. Groups initially often have to struggle with major gaps in knowledge about the impacts of a disaster. There can be organizational problems in the exercise of authority and decision making. These can stem from losses of higher-echelon personnel because of overwork, conflict regarding authority over new disaster tasks, and clashes over organizational jurisdictional differences. Generally, there is much decentralization of organizational response which in most cases is highly functional. Organizations operating with a command and control model of response do not do well at emergency times. There often too are problems associated with strained organizational relationships created by new disaster tasks and by the magnitude of a disaster impact.
Communities. Since disasters almost always cut across formal governmental boundaries, problems of coordination among different impacted political entities are all but inevitable. The greater the impact of a disaster, the more there will be the emergence of new and adaptive community structures and functions, especially emergent groups (that is, those without any preimpact existence). The greater the disaster also, the more organized improvisations of all kinds appear accompanied by pluralistic decision making. In addition, the mass convergence of outside but nonimpacted personnel and resources on impacted communities, while functional in some ways, creates major coordination problems.
Societies. Few societies ignore major disasters, but this sometimes occurs especially in the case of slow and diffuse occasions such as droughts and famines, especially if they primarily affect subgroups not in the mainstream of a developing country. In responding to domestic disasters, typically massive help is provided to impacted areas. Increasingly, most societies, including governmental officials at all levels, obtain their view of what is happening in their disasters from mass media accounts (what has been called the ”CNN syndrome”); this also affects what is often remembered about the occasions. There is also a spreading belief, so far unsupported by research, that new technologies—especially computer-related ones—will allow major improvements in disaster planning and management.
Postimpact Behavior. Individuals and households. Overall, there is little personal learning as a result of undergoing a single disaster. While the experience of a major disaster is a memorable one from a social-psychological point of view, there are seldom lasting and widespread negative behavioral consequences. Disasters very seldom produce new psychoses or severe mental illnesses. They often, but not always, generate subclinical, shortlived and self-remitting surface reactions, such as loss of appetite, sleeplessness, and anxiety. More common are many problems in living that stem more from inefficient and ineffective relief and recovery efforts of helping organizations rather than from the direct physical impacts of disasters. However, not all postimpact effects are negative; sometimes, the experience of undergoing a disaster results in positive self-images and closer social ties among victims.
Organizations. Organizational changes, whether for planning for disasters or for other purposes in the postimpact period, is not common and selective at best. Most modifications are simply accelerations of noncrisis-related changes already planned or underway. Postimpact discussion of how to improve disaster planning seldom gets translated into concrete actions (unlike civil disturbances which in American society in the 1960s led to many changes in organizations). However, overall, both in the United States and elsewhere, there have been in recent decades the growth of small, locally based, formal social groups primarily concerned with emergency time disaster planning and management. Partly because of seemingly constantly escalating economic losses, certain businesses in such sectors as banking and insurance have increasingly become interested in disaster preparedness and recovery.
Communities. There are selective longer-run outcomes and changes in communities that have been impacted by disasters. There can be acceleration of some ongoing and functional community trends (e.g., in local governmental arrangements and power structures), and generation of some limited new patterns (e.g., in providing local mental-health services or some mitigation measures such as floodproofing regulations). On the other hand, particularly as the result of rehousing and rebuilding, there can be magnifications of preimpact community conflicts as well as the generation of new ones; some of the latter is manifested in blame assignation, which, however, tends to deflect attention away from social structural flaws to mass-media-influenced search for individual scapegoats. It is also being recognized after disasters that changes in technology that create diffuse networks and systems, such as among lifeline organizations, are increasingly creating the need for regional rather than just community-based disaster planning.
Societies. In developed societies, there are few long-run negative consequences of disaster losses whether of people or property, since such effects are absorbed by the larger system. In developing societies and very small countries, this is not necessarily true; a catastrophic disaster may reduce the gross national product five to ten percent as well as producing tens of thousands of casualties. Nevertheless, changes or improvement in national disaster planning often does not occur except in certain cases such as after the Mexico City earthquake where an unusual set of circumstances existed, including a ”political will” to do something. But increasingly, in the aftermath of major disasters, to the extent that planning is instituted or improved, it is being linked to developmental planning, a move strongly supported by international agencies such as the World Bank.
UNIVERSALITY OF GENERALIZATIONS FROM AN UNEVEN RESEARCH BASE
Cross-societal and comparative research increased markedly in the 1990s. Studies have ranged from cooperative work on local mass-media reporting of community disasters in Japan and the United States (Mikami, Hiroi, Quarantelli, and Wenger 1992) and flood responses and crisis management in four Western European countries (Rosenthal and Hart 1998), to comparisons of perceptions of recurrent floods in Bangladesh by European engineers and local residents (Schmuck-Widmann 1996), and cross-national analyses of post-disaster political unrest in a dozen countries (Olson and Drury 1997), as well as methodological issues involved in cross-societal research in Italy, Mexico, Turkey, Peru, the United States, and Yugoslavia (Bates and Peacock 1993). However, this kind of comparative empirical research so far has been limited. Furthermore, although the bulk of disasters occur in developing countries, the majority of studies from which the generalizations advanced have been derived, have been done in developed societies. Thus, the question of the universality of disaster behaviors in different social systems has increasingly been raised. Some universals appear to have been found: Prosocial rather than antisocial behavior clearly predominates in responses everywhere; household members and significant others are crucial in validating warning messages, and the larger kin system is vital in providing emergency assistance; emergent groups always appear at the height of the crisis period; organizations have relatively more difficulty in adjusting to and coping with disasters than do individuals and small groups; the disaster recovery period is fraught with problems at the household, organization, and community levels; mitigation measures are given little priority even in disaster-prone localities; and social change is seldom an outcome of most disasters.
Generalizations of a more limited nature also seem to exist. There are social-system-structure-specific behaviors. For example, there often is a major delay in the response to catastrophic disasters from centralized, compared to decentralized, governmental systems. There also may be culturally specific differences. For example, reflecting cultural values, individual volunteers in disasters very rarely appear in some societies such as Japan whereas they are typical in almost all American disasters.
THE FUTURE
There is a dialectical process at work: There will be more and worse disasters at the same time that there will be more and better planning. Why more and worse disasters? Risks and threats to human beings and their societies are increasing. Traditional natural-disaster agents, such as earthquakes and floods, will simply have more to impact as the result of normal population growth and higher, denser concentration of inhabitants in risk-prone localities, such as floodplains or hurricane-vulnerable shorelines that otherwise are attractive for human occupancy. There is an escalating increase in certain kinds of technological accidents and mishaps in the chemical, nuclear, and hazardous-waste areas that were almost unknown before World War II. There are technological advances that create risks and complexities to old threats such as when fires are prevented in high-rise buildings by constructing them with highly toxic materials, or when the removal of hazardous substances from solid sewage waste generates products that contain dangerous viruses and gases. New versions of old threats are also appearing, such as the increasing probability of urban rather than rural droughts, or the potential large-scale collapse of the infrastructure of older metropolitan area lifeline systems. Finally, there is the continual development of newer kinds of risks ranging from the biological threats that are inherent in genetic engineering, to the crises that will be generated as the world increasingly becomes dependent on computers that are bound to fail somewhere at some key point, with drastic consequences for social systems. In addition, the newer threats are frequently dangerous at places and times distant from their initial source or origin as dramatized by the Chernobyl nuclear radiation fallout in European countries and smog pollution episodes such as forest fires in Indonesia which had negative effects in many Southeast Asian countries.
On the other hand, there is increasing concern and attention being paid to disaster planning of all kind. The future augers well for more and better planning. Citizens almost everywhere are coming to expect that their governments will take steps to protect them against disasters; this is often actualized in planning for emergency preparedness and response. Whereas two decades ago a number of societies had no preimpact disaster planning of any kind, this is no longer the case. A symbolic manifestation of this trend was the proclamation by the United Nations of the 1990s as The Decade for Natural-Disaster Reduction. This international attention accelerated efforts to prevent, prepare for, respond to, and recover from disasters. The effect was especially notable in the increased disaster planning in developed countries.
In developed societies, a focus on disaster planning and crisis management had started earlier, partly as a result of sociological and related research. By the 1980s, social scientists were increasingly influencing policies, political agenda settings, and operational matters regarding disasters. This can be seen in a variety of ways. Social scientists were represented on almost all national committees set up for the U. N. Decade, and contributed significantly to the reports prepared to mark the midpoint of the decade. The Board on Natural Disasters, established in 1992 in the U.S. National Academy of Sciences, always has had members from sociology and related disciplines. Sociologists have had major roles in national-disaster legislation in Greece and Italy. American social science disaster researchers typically testify before state and congressional committees considering disaster-related laws and policies. Many sociological disaster researchers provide both paid and unpaid consultant services to international, national, and local public and private groups involved in disaster-related activities. In places such as the United States, Australia, New Zealand, and some European countries, the emergency-management and disaster-planning community has become more open to recognizing the practical implications of social science research.
However, a caveat is in order for the generally correct view that studies by sociologists and others have increasingly influenced policy decisions and operational activities in the disaster area. Research results and questions are sometimes more than counterbalanced by other factors: These include the vested interests of powerful professional and bureaucratic elites to maintain traditional stances, resistance to seriously questioning the viabilities and competencies of specific organizations, and an unwillingness to face up to false assumptions of some cultural beliefs and values. Research to the contrary, for example, has had little effect on the fad-like spread of the ”Incident Command System” as a model for the emergency time operations of organizations, or an ever-spreading acceptance that victims are likely to suffer posttraumatic stress disorders, or the current common belief that mitigation measures are necessarily a better strategy for disaster planning than giving priority to improving resilience and response to crises.
RELATIONSHIP TO SOCIOLOGY
Although not true everywhere, sociologists have been increasingly accepted as having an important contribution to make to disaster planning and management. In part this stems from the fact that in many countries such as Germany, Italy, Russia, and the United States, they have played the lead role among social scientists in undertaking disaster studies. While many reasons account for this, probably the crucial factor has been that much in general sociology can be used in doing research in the area.
There has been a close relationship between disaster studies and sociology from the earliest days of work in the area. In part this is because sociologists, being among the leading pioneers and researchers in the area, have tended to use what they could from their discipline. Thus, sociology has contributed to the research techniques used (e.g., field studies and open-ended interviewing), the research methodology utilized (e.g., the ”grounded theory” approach and the employment of inductive analytical models), the theoretical ideas used (e.g., the notion of emergence from collective-behavior thinking and the idea of informal and formal structures of organizations), and the general perspectives employed (e.g., that there can be latent as well as dysfunctional aspects of any behavior and that societies and communities have a social history that is not easily set aside). In the volume entitled Sociology of Disasters: Contributions of Sociology to Disaster Research (Dynes, De Marchi, and Pelanda 1987), these and other contributions to disaster theory, disaster research methods, disaster models, and disaster concepts are set forth in considerable detail.
The relationship has not been one-sided, since disaster research has also contributed to sociology. The field of collective behavior has been most influenced and this has been explicitly noted (Wenger 1987). Other significant contributions include the study of formal organizations, social roles, social problems, organizational and social change, mass communications, medical sociology, and the urban community (see Drabek 1986; Dynes, De Marchi, and Pelanda 1987; Dynes and Tierney 1994). A symposium on social structure and disaster, coattended by disaster researchers and prominent sociological theorists, examined how disaster studies not only are informed by but could also inform sociological theory; the proceedings were published in Social Structure and Disaster (Kreps 1989). It is also perhaps of interest that for several decades now, many introductory sociology textbooks have a section on disaster behavior, usually in the collective behavior topic.
