Posted: January 6, 2005
This page is categorized as:    link to Crisis Communication index
Hover here for
Article SummaryOne of the reasons the Thai government neglected to warn its people in advance of the devastating December 2004 tsunami was a fear of panicking them. As if to prove this deadly decision right, a few days later there were warnings that aftershocks might produce another tsunami. Thousands of seaside residents fled to higher ground – and when it turned out there was no second tsunami, media reports said they had panicked. Jody Lanard and I had written before about the tendency of officials and journalists to misdiagnose caution as panic, but the tsunami impelled us to revisit the issue. This time we focused more narrowly on how rare panic is in response to natural disasters – or to warnings about natural disasters.

Tsunami Risk Communication:
Warnings and the Myth of Panic

Pop quiz: One of the following true stories was called “panic” by officials and the media. Can you tell which one?

  • A ten-year-old girl on a hotel beach thinks the water “started to go funny,” and tells her mother she “had a feeling there was going to be a tsunami.” The mother and hotel staff urge all the other tourists to clear the beach and go to higher ground. They do.
  • Japanese tourists on a beach notice that the tide seems to have gone out unusually far, and think it might mean a tsunami. The other beachgoers follow the Japanese to higher ground.
  • People along the coasts of southern India, Thailand, and Sri Lanka receive a government warning that another tsunami may be coming. Thousands of them flee to higher ground.

Hint: The first two stories were followed by the devastating tsunami of December 26, 2004. The third was a false alarm three days later.

Answer: Dozens of media reports and many officials referred to (c) as panic. Mainstream media headlines included: “False Tsunami Alarm Sparks Panic in India” and “Wave warning sets off panic but no damage.” In contrast, the other two were reported as heart-warming, even inspirational. Do you agree? Why/why not?

Thought experiment: Imagine yourself at the beach on a beautiful day. A bright-eyed ten-year-old tells you she has a feeling a tsunami is coming. You and a hundred other people flee up a hill. No tsunami arrives. What does the newspaper call your behavior?

Tsunami Warnings

For more than half a century, disaster experts have written articles about “the myth of panic.” Four truths are well documented. First, officials, journalists, and even the public itself all tend to expect that in emergencies and potential emergencies people will be too frightened to behave intelligently, and will give in to mass hysteria instead. Second, true group panic, though horrific when it happens, is vanishingly rare; in major emergencies and potential emergencies, people are typically resilient and resourceful in their efforts to protect themselves, and they are more pro-social than usual in their willingness to help others. Third, the public’s response to emergencies and potential emergencies, even when it is appropriate to the situation, is often labeled “panic,” especially if people take precautions different from the ones officials recommended. And fourth, the precautions people decide to take are frequently reinterpreted after the outcome is known; the same actions that are seen as wise and even prescient if the situation deteriorates are labeled “panic” if the potential emergency recedes without serious effects.

Much government policy is based on the misperception that the public is always about to panic, or is already panicking. Among the common policies based on this misperception: cover-up of alarming information, false reassurance, premature over-confidence, and phony consensus – all of which erode official credibility just when it is most needed. We have written about this phenomenon – which we dubbed “panic panic” – in connection with specific threats such as SARS and pandemic influenza. One of us (Sandman) wrote about it more generally with respect to worst case scenarios. And we covered it in detail in a column entitled “Fear of Fear: The Role of Fear in Preparedness … and Why It Terrifies Officials.”

But the recent tsunami provides an especially pristine, stunning example of the devastating effects “panic panic” can have. Thousands of people who perished in the tsunami could have saved themselves if they had been warned to move to higher ground. The fear of panicking them was one reason they weren’t warned.

It wasn’t the only reason, of course. In many cases there was no possibility of warning the public, because the appropriate officials didn’t know a tsunami might be on its way, and the faraway experts who did know couldn’t figure out whom they should tell. Even in cases where officials did get the relevant information, their decision not to warn the public may sometimes have been grounded in something other than “panic panic” – a fear of embarrassment and job repercussions, for example. There was probably also a sense that people wouldn’t know what to do even if they were warned. Without advance preparations (inundation maps, posted evacuation routes, brochures in every hotel room), would an hour’s warning have done any good? The answer is pretty clearly yes, as the stories of tourists led to safety by other tourists, and their children, amply demonstrate. Suboptimal but better-than-nothing evacuation is possible with virtually no planning: “A really huge wave might be coming. Head for higher ground now. Stay up there for several hours until we know it’s over. No time to tell you anything else. Go.”

Besides these other factors, a major reason for not warning the public was the false premise that warnings provoke panic. This misperception will impair decision-making in future crises as long as it remains received wisdom. We want to help stamp out the myth of panic.

There is at least one documented and acknowledged example of this from last month’s tsunami. Government officials in Thailand knew about the earthquake, speculated amongst themselves about a possible impending tsunami, and decided not to alert the public. Writing in The Financial Times, reporter Amy Kazmin describes this tragic decision:

Thailand’s meteorological department knew by 8:10 a.m. (local time) on Sunday about an hour before the first waves hit that a powerful earthquake had struck near Sumatra, and they discussed the possibility that the quake could cause large sea disturbances. The department had already distributed information pamphlets several years ago explaining the risks of tsunamis around southern Thai beach resorts.

But without definitive proof of an imminent tsunami, the meteorological department dared not issue a national warning lest it be accused of spreading panic and hurting the tourism industry if the disturbances did not materialize.

“Not every earthquake that occurs in the sea will cause a tsunami; it is very difficult to know,” said Sumalee Prachuab, a seismologist at the department. “If we issue a warning about the possibility of a tsunami, people will panic very much.”

Financial Times, December 29, 2004

Some hurricane warnings along the Atlantic coast, and some tsunami warnings along the Pacific, have turned out to be expensive and inconvenient false alarms. Most screening systems are imperfect; false positives are part of the price you pay for warnings. But tourists still flock to Hawaii, Florida, and the Caribbean. The beaches remain popular despite the false alarms. And true alarms have saved tens of thousands of lives. Purely in economic terms, these sorts of warnings pay for themselves, not just because they reduce the damage and loss of life, but also because they give people confidence that if an emergency arises they will be warned. In other words, false alarms damage tourism a little; actual emergencies damage tourism a great deal more; emergencies that come without warning do the most damage.

Hurricane warnings in the Atlantic and tsunami warnings in the Pacific normally lead to fairly orderly public responses, and thus to mutual confidence between leaders and citizens. If urged to evacuate, most people do so calmly. If the evacuation turns out unnecessary, most people go home calmly. This is part of the social compact – if we know that officials are trying to protect us, and we know that they see us as resilient and able to cope, we are more likely to forgive their well-intentioned “mistakes” on the side of caution. Just as beachgoers need some basic knowledge about tsunamis, like that provided by ten-year-old Tilly Smith in the example above, government officials need to know the data about false alarms. False alarms do produce some bad economic days at the beach, of course, and they lead to some warning fatigue (a tendency to ignore future warnings). But people and places recover quickly from false alarms. They seldom lose faith in the efficacy of the warning system. And they almost never panic.

Was the Thai meteorological department unique in thinking otherwise? We don’t know, but the odds are against it. How many other experts and officials in the affected countries learned of the earthquake, thought of the tsunami possibility, and were afraid to act? How many experts along the Pacific Rim were similarly afraid to contact CNN or other broadcasters when they couldn’t figure out how to reach officials in Indian Ocean countries? How many journalists were in fact contacted, but also hesitated to issue a warning that wasn’t officially sourced and could turn out mistaken? The widespread failure to act had many motives, among them fear of embarrassment, job loss, and economic repercussions – all valid concerns, though hardly reason enough to put so many lives at risk. Among the motives, undoubtedly, was the unjustified fear of panicking the public.

What do we call it when an expert, an official, or a journalist urgently warns people about a risk that never materializes? The same thing we call it when people respond to such a warning by taking precautions that turn out unnecessary. We call it, pejoratively, “hitting the panic button.” This is an extremely damaging fallacy: judging the wisdom of a precaution according to whether it turned out to be needed. In hindsight, an appropriate response to a potential emergency is often inappropriately labeled “panic” if the emergency doesn’t happen.

And in a more slow-motion impending crisis, people’s intense preparations are often called “panic” before it is known whether they’ll turn out prescient or over-cautious. When citizens dash out to buy food and batteries before a predicted storm, for example, the media customarily run slightly mocking human-interest stories, in which words like “panic” and “hysteria” may feature prominently. Once in a while the storm is huge, the power goes down and travel is impossible for days, and the news turns to the awful plight of those caught without food or batteries. The rest of the time the mocking coverage stands unrebutted, as if the only precautions worth taking were the precautions that turn out to be needed.

Three days after the December 26 tsunami, the Indian government warned people in the stricken area that earthquake aftershocks might lead to another tsunami. Though the risk seemed real, it soon became clear that the aftershocks were too weak to pose a serious threat. We can easily understand why experts and officials who had under-reacted three days before might have overcompensated. The public took a similar “once burned, twice shy” approach: People moved quickly to high ground. There was some crowding, some jostling, a few traffic accidents. The Associated Press reported:

Tens of thousands fled their homes Thursday, panicked by a false [government-issued] alarm that a new tsunami was about to hit following aftershocks in the Indian Ocean…. [T]housands of people [made] a desperate run for higher ground in southern Tamil Nadu state…. Hundreds of refugees fled relief camps set up near the coastline, jostling to get into trucks and other vehicles, one police constable said.

Associated Press, December 30, 2004

Just because you are desperate doesn’t mean you are panicking. Many of the videos of people fleeing the actual tsunami showed rational desperation – not panic, just appropriate (if belated) flight behavior, often accompanied by efforts to help others. What appears irrational in those videos are the people standing still, taking pictures of the water rising around them. (This is denial, or perhaps shock.)

Nor was the misdiagnosis of panic corrected in later coverage. On January 3, The Age (Australia) wrote: “The Indian Government came under heavy criticism several days after the original quake for issuing a tsunami warning following several aftershocks. The alerts caused extensive panic even though there had been no danger.” This article contains no hint of what the reporter means by “extensive panic.” By January 3, it was received wisdom that “extensive panic” had occurred when people fled to higher ground following the false alarms.

You can just picture the experts, officials, and journalists who didn’t warn anybody on December 26 saying to themselves: “See! Just as we thought! When you warn people they panic.”

Officials absorb the wrong impression when reporters inaccurately tell them their citizens are panicking. If they regularly think their people are panicking when they are not, policymaking will continue to be based on assumptions about the public’s fragility and incompetence. In September 2004, a New York Times article stated that people in Thailand were in a “frenzy” about bird flu. We were in Bangkok that week, interviewing laypeople about their knowledge, attitudes, and feelings about bird flu. We found no signs of frenzy, and very little concern. When we told the reporter so, he wrote back that the frenzy was actually among bird flu experts. This was closer to the truth. Certainly the experts were (and are) more worried than the public. But none of the Thailand-based bird flu experts we know were frenzied; they were intensely at work – just as the reporter was. No panic in sight.

Thai officials have a recurring fear that citizens and tourists will panic – a fear that led them to overstate their readiness to cope with SARS, to minimize the threat of Thai terrorism, to cover up early outbreaks of bird flu, and now to suppress a tsunami warning that could have saved thousands. The Thai government, more than most, needs to examine the decades of research showing that people are much more resilient than they think. But all governments, and all journalists, could do with a briefing.

What Is Panic – and Why Is It Rare in Disasters?

Some common definitions of “panic” offer clues about how officials view the public:

  • “A sudden strong feeling of anxiety or fear that prevents reasonable thought and action.”
  • “Unreasoning fear, causing individuals or groups to lose control of themselves.”
  • “Sudden frantic fear that often impairs self-control and rationality, differentiated from other degrees of fearfulness, such as terror, fear, and alarm.”

There are three criteria here for actual panic. First, you were very, very frightened. Second, your fear kept you from behaving as you would have if you were calm. And third, what you would have done if you were calm is a lot wiser than what you actually did. Strong fear itself isn’t enough. It’s not panic if you behaved wisely – no matter how terrified you might have felt. And it’s not panic if you would have behaved just as unwisely even if you’d stayed calm.

Individual panic is fairly common: Your dinner catches fire on the stove, and in your panic you forget to cover the skillet, turn off the heat, or get the fire extinguisher from the cabinet; instead, you flee to a neighbor’s house. And groups sometimes panic too, especially if alcohol is involved; it’s no accident that nightclubs and soccer stadiums are the scenes of more than their share of genuine and often fatal panics. What is striking in the literature is the rarity of panic in response to disasters – that is, widespread public emergencies. It seems like common sense that the bigger the catastrophe, the greater the risk of mass hysteria. But the evidence is overwhelming that in this case common sense is wrong.

In a section about “misconceptions about disaster behavior” in their “Blueprint for Community Emergency Management,” LaValla and Stoffel define panic as “a sudden, unreasoning, hysterical fear which spreads to other people rapidly. In terms of disasters it would imply behavior such as irrational flight to escape.” They note that the most common initial response to disaster is denial, not panic. Panic, they say, “requires three conditions which rarely exist:

  • Seeing the threat with a perception of possible entrapment (escape route blocked);
  • Feeling of powerlessness or impotency;
  • Feeling of social isolation or sole dependency on one’s self.”

This third factor helps explain why panic becomes less rather than more likely when the disaster is huge and shared. In 1961, Charles E. Fritz wrote a paper entitled “Disasters and Mental Health: Therapeutic Principles Drawn from Disaster Studies.” Fritz notes that large public disasters create a “community of sufferers.” At least temporarily, that sense of community can reduce pre-existing sources of conflict as well as some of the psychological damage done by the disaster itself; it discourages antisocial behavior and encourages cooperation. The longest section of Fritz’s essay is entitled “Therapeutic Features of Disaster”: threats are from the outside and clear; remedial needs are immediate and imperative; the shock of disaster opens people up to personal and social change; the shared fate reduces social distinctions and prejudices; people can identify with the transcendent goals of disaster response; etc.

On panic, Fritz comments: “One of the major ghosts in the attic of popular thought about disaster is the occurrence of ‘mass panic.’ A few citations from the historical and contemporary literature may help to place this ghost in its proper resting place.” (More than 40 years later we’re still trying to put the ghost to rest.) Fritz summarizes the empirical research of the 1940s and 1950s – much of it about the English, German, and Japanese response to World War Two bombings. He also cites some older historical examples.

Fritz quotes Frank Loesch, recalling the Chicago Fire of 1871:

We all realized that haste was necessary to get away somewhere out of reach of the flames which were shooting high above the blazing business district and by the light of which we were moving about inside as well as outside the houses, but frankly I saw no evidence of disregard of other’s rights in the confused moving to and fro…. It was the best-natured mass of people I was in the midst of.

And Jack London, recalling the San Francisco earthquake and fire of 1906:

[R]emarkable as it may seem, Wednesday night, while the whole city crashed and roared into ruin, was a quiet night. I passed Wednesday night in the path of the advancing flames, and in all those terrible hours, I saw no one woman who wept, no one man who was excited, no one person who was in the slightest degree panic-stricken…. Never, in all San Francisco’s history, were her people so kind and courteous as on this night of terror.

Similar experiences were described by people who evacuated the World Trade Center in New York, on September 11, 2001.

A major public disaster, in short, brings out the best in most people. It also brings out the worst in a much smaller but highly visible group of people. In the ten days since the tsunamis struck, episodes of looting and prank false alarms have occurred. In a few cases, relief vehicles have been mobbed by hungry and angry crowds, impeding the safe distribution of food and the evacuation of injured people. Ghastly reports of abuse, rape, and kidnapping of orphans are emerging. Stories like these inevitably and deservedly get a lot of attention, but they add to the sense that panic is everywhere. It is important to weigh the frequency of these occurrences against the much higher frequency of rational, cautious, sometimes desperate, often anguished, and still mostly pro-social coping behavior – and the ever-surprising but reliably high frequency of sharing, group spirit, altruism, and even heroism. It is important to remember that horrible though they are, crimes are evidence of desperation, greed, or sociopathy, not panic. And it is crucial to remember that even when panic happens, it is not likely to be caused by warnings.

Despite decades of evidence, officials still assume people will panic in response to a disaster … or even a disaster warning. They’re usually wrong. And on December 26 the myth of panic killed people.

Copyright © 2005 by Peter M. Sandman and Jody Lanard

For more on crisis communication:    link to Crisis Communication index
      Comment or Ask      Read the comments
Contact information page:    Peter M. Sandman

Website design and management provided by SnowTao Editing Services.