When something goes wrong, we look for someone to blame, in the hope that by finding and punishing a culpable individual we can prevent a repetition. Sometimes this is little better than scapegoating, which is my reaction to the search for someone to blame for the failure to detect the September 11 plot or to discover that Saddam Hussein had abandoned his weapons of mass destruction. Intelligence failures, even ones that seem gross in retrospect (like the failure to anticipate Japan's attack on Pearl Harbor), are typically the result of the inherent limitations of intelligence rather than of culpable negligence. In the case of the botched response to Hurricane Katrina, however, it is easy to find culpable individuals at all levels of American government. This is fishing in a very well-stocked pond. Yet there is a danger that the excitement of the sport will cause systemic problems to be overlooked that may make the affixing of blame useless, except to vent indignation. For suppose there was never a reasonable expectation that the government would respond competently to Katrina. Suppose that the American government is incapable of dealing competently with certain distinctive challenges of modernity, including the challenge of catastrophe.
A hurricane, however fierce, might, like the Indian Ocean tsunami of last year, not seem a distinctly modern menace. But that is not correct. The economic, demographic, and technological changes associated with modernity have increased the probability, and in some cases the lethality, of certain types of catastrophe. Even if global warming is not a factor in the increasing virulence of storms (it may well be, but this is still uncertain), there is little doubt that the vulnerability of New Orleans to hurricanes has increased because of engineering and construction activities, and because of population growth. When Katrina struck, the city, already below sea level, had for many years been gradually sinking because flood control prevented the Mississippi River from depositing sediment to renew the subsiding silt that the city is built on. And the wetlands and barrier islands that provide some protection against the effects of hurricanes had been disappearing as a result, in part at least, of land development for commercial and residential purposes.
As another example, consider how the scale and the rapidity of modern global transportation have increased the risk of epidemics, such as the avian flu that is now feared, and how modern transportation and modern communications have facilitated both the organization and the concealment of global terrorist networks. Consider how advances in the technology of destruction may equip these terrorists, and outlaw nations (or even individual madmen), with a varied arsenal of weapons of mass destruction, including nuclear bombs, "dirty" bombs (conventional explosives coated with radioactive materials), poison gas, and biological weapons. To these should be added the possibility (analogous to the dirty bomb) of using conventional explosives as a triggering device for massive destruction. Had the New Orleans levees been destroyed by terrorists rather than by nature, there would have been no warning and therefore far greater loss of life.
The other way in which modernity exacerbates catastrophic risk is-- paradoxically--by giving us tools for eliminating or mitigating risks that formerly had to be accepted in a spirit of fatalism. An example is the risk (far greater than most people realize, though small) of a disastrous asteroid strike. We now have the scientific competence to spot an impending asteroid strike years in advance and to deflect the asteroid from its collision course with the Earth. It is only when we have the tools to avert catastrophe that the issue of competence in their use arises. But the more tools we have, and therefore the more potential catastrophes that beg for our attention, the more dispersed and distracted our attention is likely to be. And so the more likely we are to be surprised and unprepared.
After the attacks of September 11, the entire nation--and not just a few voices crying in the wilderness, such as Richard Clarke, Gary Hart, and George Tenet--realized that powerful, far-flung terrorist groups were targeting the American mainland for attacks of unprecedented ferocity, perhaps utilizing weapons of mass destruction. So we knew that an American city might undergo at the hands of terrorists an attack that would render the city, or large parts of it, uninhabitable. And the experts among us knew that an effective defense against attacks, including "attacks" by nature, has four dimensions. They are threat assessment, or identifying vulnerable high-value targets; intelligence designed to identify impending attacks on those targets in time to prevent them (the National Weather Bureau is to hurricanes what the intelligence services are to terrorist attacks); measures to harden potential targets against attack; and all else failing, emergency response measures designed to minimize the losses from a successful attack or a natural catastrophe.
Even a superficial program of threat assessment would have quickly identified New Orleans as a highly vulnerable highvalue target for nature or for terrorism. And we could rely on the Weather Bureau for timely "intelligence" about the natural threat. The next question was whether to "harden" the target by strengthening the levees or taking other measures to reduce the likelihood of disastrous flooding. A study in 1998 estimated that New Orleans could be made safer against flooding caused by powerful hurricanes, at a cost of $14 billion, by restoring and sustaining Louisiana's coastal ecosystem. The Army Corps of Engineers had estimated an annual probability of almost one in three hundred of a disastrous flood, which mounts up to a near 9 percent probability over a thirty-year period. That is a high probability, but maybe not high enough to justify, in strict cost-benefit terms, a $14 billion expenditure that would have made New Orleans safer but not completely safe. Suppose the damage inflicted by such a flood were estimated at $200 billion; then, as a first approximation, preventive measures that reduced the probability of the flood from, say, 9 percent to 3 percent would not be cost- justified if they cost more than $12 billion (6 percent of $200 billion--the expected benefit of the investment).
Once it became clear that the $14 billion was not going to be spent--and in any event it would have been spent over thirty years, during which time the city would remain highly vulnerable to flooding--the emphasis could be expected to shift to the fourth type of protective measure: emergency response. Such a measure would be necessary in any case, since New Orleans, like any other American city, is a potential target of a terrorist attack that could cause massive human and economic damage without precipitating a flood. A biological, chemical, nuclear, or dirty-bomb attack could well require evacuation of the city with little or even no warning.
Of all the measures for dealing with catastrophic risk, the simplest and cheapest is planning the emergency response--not that the cost could be limited to the actual drafting of the plan. Some pre-positioning of supplies (including vaccines and chemical-protection clothing) and equipment (including means of transportation) would be necessary, along with some expansion or improvement of escape routes, occasional or even frequent rehearsals of the plan, the creation of a backup communications network, and the creation of a standby staff to take control and direct evacuation or other response measures. In particular, since unity of command is essential in an emergency situation, one individual (not necessarily a federal official) would have to be designated to take command of all federal, state, local, and private entities in a position to respond effectively to the emergency. The plan would not only designate this individual and his staff, but it would also specify, again by analogy to the military setting, how the various responders would be deployed and supplied and how they would communicate with one another and with the commander. It would be a mobilization plan.
The emergency, if great enough, would have a long tail: the acute phase might end in days, but the reconstruction might take months or years. There would have to be a plan for that period, too. Were there no plan, so that the reconstruction had to be improvised, there would be enormous waste. In advance of any catastrophe, therefore, contingent contracts would have to be made with the various companies that would furnish essential goods and services for the reconstruction. An allocation of costs among different levels of government, and between government and the private sector, would also be necessary, and perhaps also compulsory insurance (for example, flood insurance in areas vulnerable to flooding) to lighten the cost to taxpayers.
We have learned something important from the response to Hurricane Katrina. We have learned that four years after September 11, and two and a half years after the creation of the Department of Homeland Security, the federal government had yet to devise an executable plan for responding to a catastrophic event in New Orleans--or, I imagine, in any city in the United States.
This seems incomprehensible. The planning that I have just described would not be costly. It would not step on any big political toes. The need for emergency planning was not only obvious (a series of articles in the Times- Picayune in 2002 had explained the risk of a disastrous flood in New Orleans in detail); it was explicitly acknowledged at every level of officialdom. So why did nothing happen?
Some bad answers have been offered: the Department of Homeland Security worried more about terrorist attacks than about natural disasters; placing a federal official, especially a military person, in charge of emergency response would violate the principles of federalism. The first answer is bad because the kind of plan that I have described should have been prepared in order to deal with the aftermath of terrorist attacks, and would then have been available for natural catastrophes that had similar consequences. New Orleans would have been flooded whether the levees were breached by man or by nature.
The second answer--which drags in its train frequent references to the "Posse Comitatus" law enacted in 1878 to dramatize the end of Reconstruction by making it a crime to use the Armed Forces for law enforcement unless an Act of Congress expressly authorizes such use--is merely an excuse for inaction. An act of Congress known as the Stafford Act, passed in 1974, authorizes the use of the Armed Forces to help out in emergencies, and surely this help can include fending off looters. More fundamentally, in conditions of great danger legalistic limitations fall by the wayside. There is even an argument that, if interpreted to prevent the president from responding effectively to a major emergency, the Posse Comitatus law would be an unconstitutional limitation on sovereign power and executive prerogative. The Constitution really is not a suicide pact. No one, as far as I have been able to determine, has ever been convicted of violating the Posse Comitatus law.
As for the principles of federalism, they teach that government responsibilities should be pushed down to the lowest level at which they can be performed effectively--but not lower. If a terrorist attack or a comparable natural disaster paralyzes local government, spills across state lines, and for these or other reasons exceeds the capacity of local and state government to respond, then responsibility shifts upward to the federal government. The federal government must be prepared for such an eventuality. There are certain essential tasks that only the federal government can do.
Why was the planning so poor? Federalism complicates planning for operations that require coordinating all levels of government, but the complications should not have taken four years (and counting) to unravel. Two other explanations for the failure of planning can be set to one side as examples of sheer bad luck. One was the cuts in the budget of the Federal Emergency Management Agency (fema), the other was the retention in senior positions in the agency of persons of limited ability and experience. The first was a decision about priorities against a background of acute concern about a range of other dangers--a decision that looks unwise in hindsight but may have been sensible when made by officials not gifted with prescience. As for the quality of FEMA's management, the agency had responded adequately to the numerous routine hurricanes that it had encountered in previous years. Replacing its managers on the grounds that they would be incompetent to cope with a grand mal hurricane would have required rare clarity and courage; and it would have seemed high-handed and cruel, since such a hurricane was not anticipated.
Not anticipated--this brings us closer to the fundamental problem. A democratic government (perhaps any government) is incapable of anticipating and taking effective measures against novel threats. The threats do not have to be really novel; the fact that there was no recent example of a major American city being inundated was enough to put such a threat below the bureaucratic radar. The human mind finds it difficult to think in terms of probabilities, as distinct from frequencies, and often solves its difficulty by writing them down to zero. If an event is frequent, people expect it to recur. But if you tell them that something that has not occurred might occur and if it does occur it will cause great loss, they are unlikely to be impressed. Hurricanes are frequent, but Category Five hurricanes are rare. So it is the former for which the government prepares.
Politicians or their advisers may know better, but politicians have limited horizons. Probabilities are relative to the interval over which they are computed, as in my example of the probability of a disastrous flood in New Orleans: the probability was much greater over a period of thirty years than over a period of one year. Few politicians are looking forward to thirty years in office, so they have little incentive to support measures that may be cost- justified only if the horizon of concern is pushed out that far from the present. Their motto is, sufficient unto the day is the evil thereof. Policy myopia is built into democratic politics.
A related problem is the pressure of the immediate. Officials are continuously harassed by members of Congress, by the media, and by White House staff to deal with the crisis du jour. They are not given the leisure to address future contingencies. A further impediment is rapid turnover, as officials cycle between public sector and private sector jobs. The "revolving door" is most commonly criticized as a form of soft corruption. That is not the serious problem. The serious problem is the lack of continuity in the management of government, and the consequent foreshortening of the planning horizon. An official who spends only two years in a job is unlikely to be worrying about what may happen (but more likely will not happen) decades hence. He will receive no current reward from planning to deal with contingencies, however ominous, that seem to lie in the remote future.
Another significant factor in the incompetence of American government to deal with long-term risks is cultural. Americans are not fatalists. They accept, because it is a conventional notion, that national defense requires reserve forces and standby resources such as manned missile silos, but they would find it difficult to understand the use of government funds to establish a standby disaster command whose members sit around playing cards while awaiting Seattle to be engulfed in a volcanic eruption by Mount Rainier or New Orleans to be inundated by a storm surge. Americans simply do not accept the inevitability of disaster.
And then there is the immature political culture that has given us the Department of Homeland Security (DHS) in its present form. The creation of the department in 2003 was among a number of responses to the need to "do something" in the wake of September 11 (and another example of Americans' inability to accept the inevitability of disasters). Some of these responses, such as the much-maligned Patriot Act and the invasion of Afghanistan, made sense. One that did not, and that parallels the creation of the Department of Homeland Security, is the re-organization of the intelligence system decreed by the Intelligence Reform and Terrorism Prevention Act of 2004 in response to the report of the 9/11 Commission. It was based on the notion that the way to make government agencies work is to slot in another layer of bureaucracy on top of the existing set of agencies. This is governance by cheap, showy gestures. The effect is to impose heavy transition costs, demoralize the component agencies by demoting them, intensify turf warfare (a debilitating constant of government) by giving previously independent agencies a common boss for whose favor to vie; impede policy-making; delay, deplete, and garble the flow of information to the decision-making level; slow response time; and diminish professionalism by pushing decisions to a higher political level. The added bureaucratic layer is a fog that obscures the view both from below (the subordinate agencies) and from above (the lofty superiors).
The creation of the Department of Homeland Security followed a similar pattern. The need for coordination of the numerous agencies responsible for protecting the nation's borders, and of the agencies responsible for dealing with catastrophes, whether natural or man-made, did not justify placing twenty- two agencies, including FEMA, in a new department, centrally managed, hierarchical, with information flowing upward from the brontosaurus's tail to its tiny head (a handful of people, albeit some very able ones, trying to control 183,000 civil servants), and a response groping its way back down. With the department's formation, FEMA, included in it for obscure reasons, was demoted, losing much of its perceived importance. Appointments to its senior managerial jobs could now be used to pay small political debts--and for the further reason that emergency response, though a challenging specialty, is not a formal profession like medicine or law, so that there is a less definite sense of the proper credentials for the officials.
Now FEMA had to stand in line, waiting its turn for the attention of the beleaguered secretary of Homeland Security, who was struggling to assert control over his far-flung domain. A plan formulated by FEMA for responding to large-scale catastrophes would have to be approved not only by the White House (which has its own Homeland Security Council, whose role in the response to Hurricane Katrina remains obscure), but also by the secretary of Homeland Security and the White House. And the secretary was unlikely to be an expert in emergency response, given the breadth of his responsibilities. So when disaster struck, the head of FEMA, an embarrassing amateur in emergency response--in part because the job was no longer considered very important, the agency having dropped a rung in the hierarchy of government agencies--had to consult a higher official, who was another amateur in emergency response. There were too many cooks stirring the broth.
Congress created the Department of Homeland Security, and is supposed to oversee its creations. It has not done so. (And we can be certain that the congressional investigations of the debacle will not point the finger at inadequate congressional oversight.) The Senate has confirmed unqualified nominees to critical management positions, and the monitoring of the department's activities by both houses has been so lax that apparently they failed to notice that the department, despite its title, had no plan (except the abstract rhetorical shell grandly named the "National Response Plan") to deal with the aftermath of catastrophe. DHS Secretary Michael Chertoff acknowledged candidly that the department had not anticipated an "ultra- catastrophe." One might have expected the congressional homeland security committees, with their large staffs, to have noticed this planning deficit, were it not for the fact that our government is incapable not only of dealing with what is merely probable but also of creating rational bureaucratic structures. It is astonishing to learn that eighty-eight congressional committees and subcommittees share the oversight and appropriations responsibility for the Department of Homeland Security.
On the principle that in politics the word "truth" is a synonym for "blunder, " Chertoff has been criticized for stating the truism that, other things being equal, fewer resources should be devoted to preventing a terrorist attack that kills thirty people than to preventing one that kills three thousand. The criticism reflects the incapacity of our political class, and not just of the general public, to think in cost-benefit terms. For want of such thinking, we are busy squandering untold billions of dollars to rebuild New Orleans in time for the next flood, an expenditure of conscience money for the government's failure to have responded competently to the hurricane. We cannot even waste money efficiently. Instead of giving the displaced persons cash and letting them decide where and how to live, the government is buying them their housing and thus telling them where and how to live.
Underlying the systemic problems that I have identified is the overextension of the federal government. It is trying to do too much. In the face of formidable challenges to the safety of the nation, of which we have had an abundance of recent warning signs--the latest being the threat of a lethal flu epidemic with which we apparently are not prepared to cope--the government has entangled itself in contentious, emotional, and (it seems to me) distinctly secondary issues. Matters such as abortion, fertility treatments, homosexual rights, affirmative action, religious displays on public property, capital punishment, voluntary euthanasia, and the proper treatment of people in vegetative states are not appropriate issues to engage the federal judiciary, or the other branches of the federal government. (All three branches managed to get involved in the Schiavo affair.) They are not worthy problems for national government in a federal system. The regulation of abortion only became a subject of heated contention when it was nationalized by the Supreme Court in Roe v.Wade.
Federal investigations proliferate. Judicial confirmation hearings become distended and absurd. The federal government continues its quixotic campaign of trying to prevent people from consuming an arbitrary subset of mind-altering drugs. It operates massive programs of re-distributing wealth arbitrarily, mainly to elderly people, many of whom could pay their own way quite nicely. It sponsors space travel, something the private sector can do perfectly adequately. The proper business of a national government is none of those things. It is public safety, which is gravely endangered, and which our government, and our political system, excited and distracted by a bewildering variety of second- order concerns, seems incapable of taking rational measures to protect.