Looting After a Disaster: A Myth or Reality?
The Myth and the Realities:
Keeping the “Looting” Myth in Perspective
Not all findings about looting reported by disaster researchers have been correctly understood. Important distinctions and qualifications about the phenomena have sometimes been ignored. Thus some demythologization of the looting myth is necessary.
The word “looting,” which comes from Sanskrit (lut, to rob) entered into European languages centuries ago to refer to the plundering undertaken by invading armies. But until recently, contemporary and historical accounts of disasters have not used the term. The first systematic professional use of the word appears to have been in a well-known National Opinion Research Center (NORC) study of the 1952 Arkansas tornado.
This modern usage probably developed because the U.S. military, which initially sponsored social science studies of disasters in the early 1950s, was concerned that, in the face of atomic bombing, America would socially disintegrate and people would engage in antisocial behavior. This ignored the findings of the strategic bombing surveys of wartime Germany and Japan, as well as of British studies of their civilian populations, which showed that looting was not a serious problem after massive air
Although no formal definition of looting was ever advanced by the earliest researchers, the NORC studies, field work by Harry Moore, and research supported by the National Academy of Sciences did look at looting phenomena, generally viewed informally as the illegal taking of property. The conceptual problem of studying looting has been compounded by the fact that “looting” is not a criminal category in American penal codes, except in a handful of states that have legally formalized the term relatively recently.
A consistent observation of the early studies was that instances of looting in the disasters examined (few of which occurred in metropolitan areas) were nonexistent or numerically very rare. This contrasted with a parallel observation that stories about looting were widespread in mass media accounts and among affected populations (58% reported hearing such stories and 6% thought they had been looted in the Arkansas disaster—a finding repeated over and over again in other studies).
In the 1960s, the many civil disturbances in large American cities were studied by disaster researchers. While to this day there is no agreement that riots should be conceptualized as conflict or willful disasters, the researchers found that looting was very pervasive in the riots studied and that the pattern of the looting behavior significantly differed. In natural disasters looting was very rare, covertly undertaken in opportunistic settings, done by isolated individuals or very small groups, and socially condemned. In contrast, looting in the riots was frequent, overtly undertaken, aimed at specific targets, participated in by very large numbers of individuals often in social networks, and was socially supported.
Semi-systematic studies of looting that continued into the 1970s in the United States did not challenge the overall picture that researchers had earlier developed. Mostly anecdotal reports in other developed countries were consistent with the American experience. This view was later generalized to the proposition that looting was not a problem in modern, developed countries and that in the rare instances when it occurred it had the distinct social characteristics found by the pioneer disaster researchers. However, absent systematic studies in developing countries to this day, and using mostly anecdotal accounts and mass media reports, the best that can be said is that major looting in developing countries sometimes appears on a massive scale, such as after the recent earthquake in Pakistan, but that at other times, such as after the 1985 Mexico City earthquake, looting is an infrequent problem.
Furthermore, from the 1970s to the present day there have been occasional large-scale community crises after which researchers studied mass looting. One was the 1977 New York City blackout during which selective neighborhoods experienced massive looting illustrating the distinctive conflict situation pattern found in the 1960s. However, before “obvious” implications are drawn, one should note that similar blackouts in 1968 and in 2003 did not generate mass looting.
Crucial to any discussion of looting is what happened in St. Croix in the U.S. Virgin Islands when that city was hit by Hurricane Hugo in 1985. After that event, the University of Delaware’s Disaster Research Center undertook three different field studies, including a systematic quantitative survey of all businesses in the major shopping centers. The looting in St. Croix was massive. Not only were all consumer goods in sight taken, but there was even stripping of electrical and wall fixtures and of carpets. The largest mall (with about 150 shops) and two others were heavily hit, with less than 10% of the businesses reporting they were not totally looted.
The looting was initiated by pre-impact organized gangs of delinquent youths who first targeted stores with large quantities of consumer goods. A second stage occurred when other participants with noncriminal lifestyles began looting other kinds of stores (e.g., hardware stores). Finally, an even larger number of people joined, targeting stores with basic necessities (e.g., food supermarkets) and generally not looting items taken by the first two categories. Overall, the looting pattern was what earlier researchers had found in civil disturbances. However, contrary to widespread rumors, there was not a single authenticated case of the looting of private residences, schools, hotels, the one industrial complex with valuable equipment, or even resort restaurants. The looters used no physical force and, at worst, made only unfulfilled verbal threats.
A possible explanation for this atypical occasion of mass looting was that it involved a major catastrophe rather than a lesser disaster—with a concentration of disadvantaged persons exposed to everyday perceptions of major differences in lifestyles; a subculture tolerant of everyday minor stealing along with everyday organized youth gangs engaged in serious crime, such as drug dealing; and a local police force widely seen as corrupt and inefficient (early in the event, officers themselves had openly engaged in looting—not the usual pattern in civil disturbances).
A case can be made that what happened in New Orleans after Hurricane Katrina repeated, on a smaller scale, what had happened in St. Croix. The New Orleans event was smaller because in St. Croix a majority of the population probably participated in the looting, the looting did not last as long in New Orleans, and percentage-wise, far more stores were looted in St. Croix. But the overall pattern of mass looting, as well as the social conditions generating it, were the same in both cases.
To conclude, looting of any kind is rare in certain kinds of disasters in certain types of societies. The pattern of looting in natural disasters is different from what occurs in civil disturbances. There are occasional atypical instances of mass lootings that only emerge if a complex set of prior social conditions exist.
Disaster Research Center, University of Delaware