“It is going to be about as ugly of a scene as I think you can imagine,” said the homeland security chief for the US government, on 5 September 2005. With that, Michael Chertoff understated the essence of Katrina in a nutshell. It was beyond imagination, actually imaginations, of so many. The great sadness is that this was an easy one. Sure, no one can stop a hurricane but the effects of this hurricane could have been softened, its body count so easily diminished. But someone had to have been paying attention, and the right kind of attention. Someone had to have realized the worst case possibilities.

Ah, but not just “someone.” Three conditions must hold. The someone must be possessed of clear vision and expansive imagination, she must be in a position of considerable power so that resources can be directed as needed, and he must be sufficiently decisive to try to change the course of events so that people are valued over the mindsets and interests of bureaucracies. Those were the key failures in the governmental response to Katrina and they are by turns heart rending and infuriating.

Katrina is being called a worst case, and surely it’s hard to disagree with that judgment. But in what ways was Katrina a worst case and what are the consequences of calling her one?

At this writing, 7 September 2005, we do not have a full body count. New Orleans Mayor Ray Nagin was the first to say the death toll could be in the thousands; a week after the storm skirted New Orleans he said that “it wouldn’t be unreasonable to have 10,000” corpses. For Katrina to be the worst “natural” disaster in American history it would have to top the 1900 Galveston hurricane that killed at least 8,000 people. But it is noteworthy that Katrina is being called a worst case, even before we count our dead.

Katrina will likely end up being the most expensive US hurricane—all estimates at this time are pure guesses, but the latest one I’ve seen is 150 billion US dollars—but that’s merely the cost of property, which depends on inflation, demand for real estate, and so on. Billions of dollars are important, of course, but that’s not why Katrina’s devastation is capturing our imaginations.

That’s the key to understanding Katrina as a worst case—the imagination. Events that we call worst cases are beyond the imagination, overwhelming it with images, data, noise, disorder, and sometimes violence and despair. Since the disaster, I’ve been grappling with the social, political, and physical dimensions of Katrina, trying to figure out what I think about it all. It is stretching my imagination, and that is one way I know it really is a worst case.

There are many aspects of the case that are mind-numbing, which is another way of saying that our imaginations lack the categories necessary to make sense of the tragedy. There are the usual tales of altruism and heroism coming out of New Orleans. But there’s also unspeakable brutality, stories of gun fighting and utter neglect of people’s humanity and dignity. And at one point a full third of the New Orleans police department was unaccounted for. These things are quite unusual in the aftermath of disaster. We—especially those of us who specialize in disaster research—are saying things to reporters and each other that help make sense of what can only be described as a breakdown of social order. But I doubt any of us would have predicted it.

We should be clear that the disorder is itself a symptom of official failure; the other, more important, symptom is the great loss of life. Disorder was not inevitable. The burning question is why wasn’t more done to diminish the loss of life?

One answer is that the remaining victims of Katrina were poor, and African American to boot, so officials callously turned their backs on those whom they did not care about. That may be so. After all, officials and their organizations didn’t take care of those people before the hurricane, so why would we expect them to do so afterwards? The problem with that answer is that Katrina trained all the spotlights on the officials and their organizations so they had excellent reasons to come to the rescue. It is much easier to tolerate unforgivable neglect of poor people when they are out of sight.

In fact, those in charge of the federal government had every incentive to make the response go smoothly. Approval ratings of the Bush administration, regarding Iraq, homeland security, terrorism and the economy, have been declining steadily. It is, at first glance at least, inexplicable how the Bush spin-team, led by Karl Rove and vaunted as brilliant political strategists, could miss that this would be a great opportunity for them. Even if they couldn’t predict that New Orleans would drown, there was every reason to think that there would be widespread devastation. In early August NOAA, the National Oceanic and Atmospheric Administration, warned that “this may well be one of the most active Atlantic hurricane seasons on record…”1http://www.noaanews.noaa.gov/stories2005/s2484.htm On August 27 the National Hurricane Center extended a hurricane watch for Louisiana. A hurricane watch, according to the Louisiana Homeland Security and Emergency Preparedness Department, means to board up your property and prepare to evacuate.2http://www.ohsep.louisiana.gov/factsheets/DefinitionOfaHurricane.htm Katrina slammed New Orleans and surrounding areas on Monday, August 29. The point is that the storm’s advent and its damage were utterly predictable. It was an opportunity, really, for the Bush administration to shine. But they blew it.

Another answer, this one favored by the Federal Emergency Management Agency, is that they had to wait until local and state governments called on them to act. “The way that emergency operations act under the law is the responsibility and the power, the authority, to order an evacuation rests with state and local officials,” Chertoff said in response to a reporter’s query about his organization’s lackluster performance. That is correct, but the city did order an evacuation. It is federal responsibility to provide support for the evacuation, and they failed in this.

On August 28, Sunday, Louisiana Governor Kathleen Blanco wrote a letter to President Bush, sending it “through” the regional director of FEMA, in Texas. This was the day before Katrina made landfall, although it was clear to everyone well before that time that hell was on its way. In her letter, Governor Blanco specifically noted that Katrina’s damage would be beyond the abilities of the state and local governments to deal with. It would, she declared, be a “catastrophic hurricane.” Blanco’s letter is full of specific requests to FEMA for various kinds of assistance. She said that she had put in motion the “State Emergency Plan” on August 26. She asked that “you declare an expedited major disaster for the State of Louisiana…” 3http://gov.louisiana.gov/Disaster%20Relief%20Request.pdf The claim that the state hadn’t asked for assistance is so much hot air.

As the storm moved through on August 29, the director of FEMA, Michael Brown, said that FEMA was ready. We’ve “… planned for this kind of disaster for many years because we’ve always known about New Orleans’ situation.”4http://www.cnn.com/2005/POLITICS/08/29/katrina.washington.ap/ President Bush promised that “I want the folks there on our Gulf Coast to know that the federal government is prepared to help you when the storm passes.”5http://www.cnn.com/2005/POLITICS/08/29/katrina.washington.ap/ Such statements seem to indicate a willingness at the federal level to act quickly. Bush declared a state of emergency in Louisiana, even before Katrina arrived, setting federal bureaucracies in motion to fulfill his promises. The magnitude of the devastation was knowable, even known. People had predicted New Orleans would fill with water.

We also know that the US military was prepared to drop food and other provisions earlier in the week, but FEMA failed to ask them to do so. “We know very well how to do this, and it’s just incomprehensible that we’re not,” said a consultant to the Lawrence Livermore Labs who had helped design earlier food drops.6http://www.washingtonpost.com/wp-dyn/content/article/2005/09/02/AR2005090200670_2.html That was not a local failure, but a federal one.

Not only were lines of responsibility clear and known in advance, so too were the tasks that would need to be addressed after Katrina left town. It was clear, at least to some, that multiple and many needs would require tending to after a massive evacuation. We’ve evacuated a lot more than a half a million people, which is the size of New Orleans’ population. More than 2.5 million evacuated Florida’s coasts ahead of Hurricane Frances in September 2004.

So Katrina was predicted, and the magnitude of the devastation was knowable. On Saturday August 27, mandatory evacuation orders were issued for some of Louisiana’s low-lying areas. “This is a very, very dangerous hurricane,” said National Hurricane Center Director Max Mayfield. The proportion of people who would stay behind, to “ride it out,” should not have been news to anyone. Officials could have found out, roughly at least, the numbers of people who could not afford cars, and there were no buses or free Amtrak offers before Katrina struck. Finally, the characteristics of the alleged hurricane riders were well known: they would be poor, distrustful of the police and political establishment, largely African American, with no way to get out.

To top it all off, just last summer, in July 2004, emergency officials ran a simulation, called Hurricane Pam, that included FEMA, the US Army Corps of Engineers, and other state and federal agencies. “We made great progress this week in our preparedness efforts,” said Ron Castleman, FEMA Regional Director. “Disaster response teams developed action plans in critical areas such as search and rescue, medical care, sheltering, temporary housing, school restoration and debris management. These plans are essential for quick response to a hurricane but will also help in other emergencies.”7http://www.fema.gov/news/newsrelease.fema?id=13051

In the simulation, Hurricane Pam had sustained winds of 120 mph, dumped up to 20 inches of rain in parts of Louisiana. The storm surge topped levees around New Orleans, and more than a million residents evacuated. It’s as if Katrina took lessons from the Pam simulation.

A great lack of imagination helps account for the pathetic federal response. Statements by high level officials suggested that they did indeed understand what they confronted. In fact they did not. When it got right down to it, to creating and holding the expectation of what would really happen should a monster storm come screaming through New Orleans, the worst case imagination was neglected in favor of the more reasonable, probabilistic one.

The infrastructure that was supposed to control the waters of Lake Ponchartrain and the Mississippi River (luckily, the Mississippi did not flood; worst cases can always be worse) was built on the assumption that a storm bigger than a Category 3 had a very low probability of occurring. “Projects designed to keep New Orleans from flooding in a hurricane prepared the city for a probable scenario, not the worst-case scenario…,” said Lt. Gen. Carl Strock, the Commander of the Army Corps of Engineers. The Corps was confident that chance would work in our favor. “We had an assurance that 99.5 percent this would be OK. We, unfortunately, have had that .5 percent activity here,” Strock said.8http://www.newsday.com/news/nationworld/wire/sns-ap-katrina-corps-spending,0,5074432.story?coll=sns-ap-nationworld-headlines

Chance very often goes against us. As political scientist Scott Sagan has said, “things that have never happened before happen all the time.” But this is decidedly not the usual or recommended way of looking at risk. The usual way is probabilistic thinking and in the modern day it has become equated with rationality itself. Possibilistic thinking, by contrast, emphasizes potential consequences over the likelihood that some event or events will happen. Probabilism says it is highly unlikely that a nuclear power plant will melt down. Possibilism wonders what happens if a nuclear power plant has a particularly bad day. Possibilistic thinking is worst case thinking.

Neglecting worst case thinking isn’t limited to officials. A retired army colonel, talking about going back to her home said “I’m going to pay someone to get me back there, anything I have to do.” She added that “a lot of these people built these houses anticipating some flood water but nobody imagined this.”9http://abclocal.go.com/ktrk/weather/090505_APkatrina_backhome.html

That people like the colonel don’t sufficiently appreciate worst case thinking illustrates how widespread is the idea that probabilism is the best way to approach risk. But it was not decisions of people who did not hold seats of power that mattered so much in New Orleans.

Where it really mattered was at the top levels of government. Secretary Chertoff claimed that “the way these catastrophes unfolded was unprecedented,” and that the sequence of events was “breathtaking in its surprise.” Back-to-back untoward events “exceeded anybody’s foresight,” he said. We already know that they did not exceed everybody’s foresight. But they surely exceeded the foresight of disaster officials. Their imaginations were not in the worst case mode. If they had been, the possibility of an “ultra-catastrophe,” as Chertoff called it, might have been thrown into bolder relief on their radar screens.

Katrina was not a perfect storm. It could have been. It could have held its category 5 fury, stalled over New Orleans for a while, took a right turn to destroy the Mississippi coast, and then looped back to make sure that the refineries and pipelines in the Gulf of Mexico were obliterated. Katrina was, however, a perfect worst case because it exceeded the imaginations of those who claim to protect us.

There will be a next time. Prudence advises that we assume the next time will come sooner rather than later. What can we do to soften the blow? There should be institutional recognition that there will be a next time. It is a question of when, not if. As important as it is to use probabilistic thinking in disaster—as when NOAA assigns probabilities to paths that hurricanes might take—it carries with it a set of blinders that can hinder appreciation of true danger. More extensive use of worst case thinking would seem to be necessary to prepare for the next time. We have to find ways to push officials and their organizations to think outside of their boxes. Lives depend on it.

Lee Clarke is a sociologist at Rutgers University. He is author of Mission Improbable and Worst Cases, both from the University of Chicago Press. He is an internationally known expert in disasters, and organizational and technological failures.