As no commission report could ever do, the terrorist acts of September 11, 2001, have galvanized the United States. More than six thousand people died in New York—one of every thousand workers in the city. Taking over a commercial aircraft to use it as a piloted cruise missile evidently exploited a terrible vulnerability of modern society. No commercial pilot could be induced by threats to do this, but the imagination of public officials did not encompass those willing and even wishing to die to kill hundreds or thousands of others.

This instrument of kilo-terrorism is fragile; it can be defeated by not much more than a sturdy locked cockpit. The terrorist planners knew this and went to the trouble, and risked the vulnerability, of planning at least four such hijackings within minutes of one another—too little time to spread the word effectively throughout the aviation system.

My purpose here is to discuss threats and not primarily solutions, although the two are interlinked. If hijacking a passenger aircraft will no longer work, motivated terrorists will doubtless choose something else. I have heard from at least six of my colleagues proposals, for example, to modify the aircraft flight control system so that the pilots can irrevocably switch control to the ground. Or to program the aircraft to automatically land at the nearest suitable airport. Such remedies are unnecessary and distracting. It is enough to lock a strengthened cockpit door and to make all understand that it would not be opened even if hijackers kill the passengers and cabin crew. Clearly, once the aircraft radar transponder is switched to emergency or hijack mode, it should not be possible to switch it back until the aircraft has landed. There is a case to be made for improved 1990s-era crashproof recorders that will capture video and audio from the cabin as well as the cockpit.

Both smaller and bigger terror weapons exist, and their use may be expected. But even if we have seen the end of hijacked passenger jets as cruise missiles, that is not the end of their equivalent—the use of rented or stolen cargo jets as piloted cruise missiles. Opportunities range from large fleets such as those of UPS or Federal Express to the hundreds of 707s and even 747s available for lease at airports in the United States and elsewhere. More skill but less violence would be involved in stealing such an aircraft. It might be used against buildings or against operating nuclear reactors, which are not designed to withstand the impact of a jumbo jet at high speed. It is relatively simple to have an automatic radar that shuts down a nuclear reactor if a high-speed aircraft is detected within a few seconds of collision; but even in that condition a reactor might suffer a core meltdown because of interference with the emer-gency core cooling system and other engineered safeguards.

Thomas Friedman in his New York Times column of September 25 characterizes the new type of terrorist as evil, educated, and suicidal. Of course, the terrorists don’t characterize themselves as “evil” or even “suicidal.” A brief anecdote. After a few years of being involved in building and testing nuclear weapons (from 1950 on, at Los Alamos), I began to work on North American air defense, and in the mid-1950s joined several panels of the President’s Science Advisory Committee (PSAC)—among them the Strategic Military Panel. This panel met two days each month until the demise of PSAC in 1973. I had already spent a month in Korea and Japan during the Korean War. I never understood why it was a source of comfort for some US strategists that the Soviet Union had not mastered in-flight aerial refueling of bombers; most people seemed to feel more secure in the knowledge that Soviet bombers armed with nuclear weapons would not have fuel to return home after a nuclear attack on the United States. I argued that I, for one, would be quite willing to participate in a one-way retaliatory nuclear strike on the Soviet Union, and I regard myself as rational—not suicidal.

Just as I struggled for years in Washington to bring cruise missiles into the US force rather than unmanned bombers (because of the relative simplicity and increased payload of the one-way mission), so have committed terrorists found a way to exchange commitment for complexity. It is a powerful tool. During the 1960s I recall discussions with experts on terrorism in the PSAC and intelligence community who judged at the time that terrorists did not really want to kill people but to gain sympathy for their cause. Hence, it was said, they wanted to show their power to explode bombs at places and times of their choice, with a propensity to warn so as to reduce the damage.

Now that almost three times as many Americans have been killed in one day as died at Pearl Harbor, and twice as many as died in the entire history of the “troubles” in Northern Ireland, and without a demand being made, that old judgment must be definitively retired. And it doesn’t take years of preparation for a person to gain access to an aircraft and to detonate a few-pound bomb disguised as a book or a computer, thus killing a few hundred people and further disrupting the modern economy.


It is doubtful that the terrorists had confidence that the World Trade Center towers would collapse—which they did, not from the impact but from the softening of the steel from the intense fire provided by the aircraft fuel. In fact, the amount of fuel in the form of paper in filing cabinets on a given floor is comparable with that delivered by the plane, but it is more difficult to ignite and easier to put out. It is entirely feasible to build into tall buildings features that would be adequate for fighting such fires and furthermore to equip buildings with means for rapidly bringing firefighters to any floor. This latter might be done on a World Trade Center–like structure by having a number of pulleys of twenty-ton capacity projecting from the roof, with a lead line down near ground level. Firefighters could snag the line and with a ground-based winch pull up a heavy cable which, in turn, could be used to carry platforms, hoses, and pumps to the floors involved. But it would be preferable to have dispersed foam nozzles in a hardened sprinkler system.

Terrorists have other means of turning the strength and assets of American society against itself. These include targeted attacks on chemical plants, but even more important, on shipments of industrial chemicals such as chlorine, which are transported in tank cars or trucks. The terrorist driver might apply for a job with the intent of fitting the tank truck with detonators and exploding it in a community; or such a truck might be ambushed and the material dispersed by a rocket-propelled grenade. While the use of nerve gas or other material would give a far higher number of casualties per ton of material, the vast amounts of dangerous chemicals that move in commerce make this a significant problem.

Some failures to protect particular vulnerable points would cause tremendous damage and inconvenience to modern society—at the major bridges and tunnels, for instance. Not only destruction but radiological contamination of tunnels could be very disturbing, even if it killed few people.

Detonating thousands of tons of ammonium nitrate loaded on a ship in a harbor would have the impact of a small nuclear explosion. Three hundred tons (0.3 kilotons) of ammonium nitrate apparently exploded in France on September 24, killing twenty-nine people and injuring more than 2,500.

Terrorist acts are possible that would be less significant in damage but highly significant in causing terror and weakening perceptions of American strength. Attacks on spectators in a sports stadium seem a particular hazard, especially in the case of events shown on TV. Such an attack could combine explosives and chemical agents; it could even be made by diving a small aircraft loaded with fuel into the stands. The attractiveness of such tactics to terrorists might be reduced by a several-second delay in TV transmission, so that there would be no broadcast, even if thousands of people were killed and several times that number injured.

Concerning nuclear and biological terrorism, the largest amount of damage would be caused either by a nuclear explosion in a city or by a biological warfare attack. It is abundantly clear that the same nineteen terrorists who hijacked the aircraft and destroyed buildings and thousands of lives in what seemed an instant would not have hesitated to detonate a nuclear explosive if they had acquired one. A first-generation (10-kiloton) nuclear explosive would kill at least 100,000 people in a typical urban environment. The theft and detonation of one of the 500-kiloton strategic weapons would probably kill a million people in an instant and flatten 100 square kilometers of buildings. Fifty-five years of development of technology and spread of knowledge make it relatively simple to build a 20-kiloton nuclear weapon if sufficient highly enriched uranium were available, of which there is a thousand-ton surplus in Russia. Much excess plutonium that was developed for making weapons is available as well, although it is somewhat more difficult to use. We must give the security of such materials the attention it deserves.

As for biological warfare, many tons of anthrax may still exist in Russia. Infection can be prevented by prior vaccination; but anthrax is extremely durable as a spore and kills 30 percent or more of the people who have been infected, if there is no adequate prolonged treatment with antibiotics. Even more potentially dangerous are biological warfare agents that are contagious, by contrast with those, such as anthrax, that are simply infectious. High among the contagious agents is smallpox. Although legitimate stocks of smallpox have existed only in two places—in the US and in Russia—it is not precluded that other stocks may have survived the smallpox eradication campaign.


Even though some of these threats are ill defined and it may be hard to prevent their being carried out, some nonspecific solutions are eminently practical. None will give 100 percent protection, but 99 percent protection could be the difference between a million deaths and ten thousand deaths. In a war, that is a great difference. To the 99 percent, it is the difference between life and death. And some of the solutions can be implemented by individual families, corporations, or localities.

The first and most practical defense against biological warfare attack is to maintain “positive” pressure of filtered air within buildings. It takes a very small capital expenditure and a very small expenditure in power to provide a positive pressure so that normal winds will not infiltrate a building, and the anthrax spores or other microbes will be kept out. To do this the air intake to a normal building—whether an office building, an apartment building, or a private house—should be provided with a small blower that delivers air through a High Efficiency Particulate Airfilter (HEPA) at a rate that exceeds the leakage of air in or out of the building. Such “makeup” air will then produce excess pressure in the building so that air flows out through any cracks or apertures, blocking any inflow of unfiltered air. If no form of air intake exists, a window or a portion of a window can be removed to make one. It is interesting to note that any normal building, no matter how tightly closed, will have the same exposure to a biological warfare agent as it would if the windows were wide open—it takes longer for the agent to enter, but it stays there a much longer time. Positive-pressure filtered air largely eliminates this problem.

Other approaches that should be implemented contribute not only to the reduction of threats but to lowering the cost of reducing the threats. Such measures would include sealing at the point of departure trucks, ships, or cargo containers, so that auto parts entering Detroit from Canada, if they were inspected at the factory, would not have to be inspected individually. Electronic manifests and bills of lading could be required in advance, and shipments that comply with these efficiency- and security-based rules would incur less delay and less cost than those shipped the old way.

Similarly, people willing to carry biometric-based identification (a thumbprint plus photo, for instance) could be given “EZ-Pass” treatment. These people would have had a suitable interview and could have provided data to be kept in an electronically accessible file. Those without the EZ-Pass would be delayed longer in driving their trucks into a city or in boarding aircraft.

Thus far, I have discussed a few of the threats that might be expected from terrorists; some of these are greatly increased by the willingness to die for the cause. On the assumption that there are dozens or even hundreds of similar agents already in place, it is unlikely that their motivation can be annulled; hence the critical importance of ensuring that such attempts in the near term will not succeed. Here are some near-term measures:

· To prevent hijacked passenger aircraft being used as a manned cruise missile, strengthen and lock the cockpit door. Assign air marshals to many flights. Ensure that the radar transponder, once switched to emergency mode by a pilot who is being attacked, cannot be switched back.

· To counter the use of rented or stolen large aircraft, ensure that each aircraft landing gear is blocked by heavy concrete barriers or other means that would sound an alarm and disable the gear if moved without authorization.

· Foreign aircraft entering US air space must be subject to the same standards as US aircraft.

· To counter biological warfare, individuals, firms, government, and other organizations should consider installing a unit to provide positive-pressure HEPA-filtered makeup air to their buildings. For most establishments, these units should not be used to guard against biological warfare agents liberated within the building but against those from outside. Because of the far smaller hazard from chemical warfare or industrial chemical attack, HEPA filters should filter only particles from the air. These are typically not individual virus particles, but bacteria or viruses that are attached to some inert material in the range of diameters from about one to five microns.

· To facilitate travel and access to sensitive areas, a first-generation biometric identification pass should be made available. Those who have had an adequate interview and have information on file could rapidly be provided with a picture ID augmented by a thumbprint. This would be analogous to the EZ-Pass now widely used at tolls.

· To facilitate the movement of cargo, more use should be made of sealing at the departure point containers, ships, aircraft, or trucks, so that inspection would occur there with adequate time and space, rather than on the fly at bridges or other choke points. Electronic manifests could be sent ahead and would also accompany the vehicle. Lower customs charges for inspection and accelerated processing would be given to those vehicles and containers packed so as to facilitate high-energy X-ray or neutron scanning. Such vehicles would be proc- cessed more rapidly and at a lower cost than those without such helpful features.

There are many more potential terrorists than there are terrorists. In moving against terrorist organizations and states and others supporting terrorism, we need hardly fear that those who are implacable enemies of society will become more deeply implacable. But it would be easy enough to swell the ranks of terrorists with those who up to now have been largely passive. Accordingly, if the United States were, for example, to undertake military action against the Taliban, it should be accompanied by an effective and sincerely concerned program to relieve the plight of the people of Afghanistan.

In taking action against terrorists and their co-conspirators, it would be useful to recall that in the United States conspiracy to commit a crime is in itself an offense. While aiding and abetting the actual crime has the same penalties as the crime itself, conspiracy has a lesser penalty. But one can be imprisoned for conspiracy even if the crime is never committed. Such doctrines could be drawn on to lay the basis for the legitimacy of US action in protecting against and responding to terrorism.

I have neither tried nor succeeded in providing here a complete evaluation of terrorist threats to modern society—let alone a reasoned evaluation of the effectiveness and cost of countermeasures. For instance, cyberterrorism is a serious potential problem, and individual hackers have already caused billions of dollars of damage. It is clear, however, that acting as individuals and as a society as a whole, we will need to make considerable investments in reducing our vulnerability. If we do this wisely and make use of market incentives wherever possible, the cost in efficiency and diversion of resources should be tolerable.

—October 2, 2001

This Issue

November 1, 2001