Are We Safe Yet?

This is part of IEEE Spectrum's special report: What's Wrong—What's Next: 2003 Technology Forecast & Review.

We may no longer feel like we're living on the knife edge of 9/11. But perhaps we should. In terms of radiological, nuclear, and biological attacks, we're still as vulnerable as ever, as IEEE Spectrum's senior associate editor Jean Kumagai learned in a recent interview with Richard L. Garwin. A longtime advisor to the U.S. government on matters of national security, Garwin is also a respected public critic of technology and technology policy, including, most recently, efforts to contain biological and nuclear terrorism. An IEEE Senior Member, Garwin is the Philip D. Reed Senior Fellow for Science and Technology at the Council on Foreign Relations (New York City) and an IBM Fellow Emeritus at the Thomas J. Watson Research Center (Yorktown Heights, N.Y.).

Is the public safer now than we were before 9/11 in terms of nuclear terrorism?

No, we are not safer now. In fact, the Hart-Rudman Commission's Report of January 2001 on U.S. national security has been supplemented by a more recent report with a provocative and accurate title, America—Still Unprepared, Still in Danger. Despite the politically attractive formation of the U.S. Department of Homeland Security, and the more general awareness of the threat, very little has been done, on the ground, to prevent such attacks.

What is the No. 1 threat we face today in terms of nuclear security? And what will it take to minimize this threat?

Actually, there are three threats of comparable importance but different character. The first is the traditional threat of an all-out Russian nuclear strike on the United States, resulting in the total devastation of the United States (and Russia) and the deaths of hundreds of millions of innocent bystanders in the rest of the world. The difference from the bad old days is that it would happen this time by accident, inadvertence, or unauthorized launch. It would happen because Russia takes into account not U.S. intentions but U.S. capabilities, as does the U.S. military with other potential threats. And because only one or two Russian nuclear-armed submarines are at sea at any one time (and Russia doubts the survivability of these submarines), Russia is poised to launch a nuclear strike on the United States with its vulnerable silo-based ICBMs [intercontinental ballistic missiles], as well as from its submarines dockside in Russia.

In addition, Russian early-warning satellites and other means of detecting attacks have degraded substantially over the last decade, and it is possible (although not likely) that such a deterrent capability would result in an inadvertent launch.

But because the United States has a secure second strike capability against Russia, we can quite readily eliminate the threat of inadvertent launch. We can do this by explicitly putting almost all our strategic nuclear strike capability on deferred launch—known as "de-alerting." We can also place silo-cover monitors on our ICBM silos, reporting to Russia once per minute to show that the covers have not been moved. Such cooperative threat reduction would extend the decade-long activities initiated by Senators Richard Lugar and Samuel Nunn to reduce excess Russian nuclear weapon materials.

The second threat is that of nuclear terrorism. This would involve a nuclear weapon stolen, probably from Russia (or, for that matter, Pakistan), and provided to a terrorist organization. The greatest danger is the smuggling of such a nuclear weapon (or an improvised nuclear weapon) in a cargo container or light aircraft, which could land quite readily without interception in the United States. Transporting such a nuclear warhead in a car or van to a densely populated area such as Manhattan and detonating it on a workday could ensure 100 000 to 300 000 deaths from blast, fire, and local fallout. The same sort of damage could be produced by detonating a nuclear weapon in the harbor at New York, Boston, Philadelphia, or a number of other locations.

To be sure, deploying stolen nuclear weapons requires time and talent to bypass their "use control" mechanisms. But there is also the possibility of improvised nuclear explosives based on highly enriched uranium (HEU), again probably obtained from Russia. Some 60 kg of HEU would be required, but the weapon could be fabricated in a basement or office, and would predictably yield 10-20 kilotons of explosive yield (compared with 0.002 kilotons for a large conventional truck bomb). This, too, would kill 100 000 to 300 000 people.

The solution to this second threat is to take far more seriously the consolidation and blending down of HEU in Russia and elsewhere. The existing 20-year U.S. program for buying 500 tons of excess Russian HEU should be augmented with a program to blend down within the next year not only those 500 tons but the remaining 700 tons of excess weapon HEU, to 19.9 percent uranium 235. This blending down would virtually eliminate the material's possible use in a weapon and could be done far more rapidly than the current plan of blending all the way to 4.4 percent U-235, the mix needed for use in nuclear reactor fuel.

The initial blend to 19.9 percent would in no way restrict the eventual use of the material in breeder reactors or conventional light-water reactors, and should appeal to Russia, provided the money goes to those who are doing the work and not to the central government. However, the program would require an initial provision of capital from the G-7 nations, to be paid back by eventual sales and further blending down of the 19.9 percent to the ultimate 4.4 percent.

The third nuclear security threat lies in conventional attack on the 103 operating nuclear power reactors in the United States (or their spent fuel storage pools). As stated in the U.S. National Academies' 2002 report Making the Nation Safer, a meltdown could occur through either sabotage or attack by a light aircraft loaded with explosives. Such a meltdown would condemn about 10 000 people to death by cancer, and it would have disastrous consequences for the nuclear power sector and the overall economy, because 20 percent of U.S. electrical power comes from nuclear reactors. Greater attention to protecting these reactors (including, perhaps, close-in air defenses) could help reduce that threat.

How does the potential damage of a nuclear explosive compare to that of a so-called "dirty" radiological bomb?

The potential damage of a "dirty" radiological bomb pales beside that from a nuclear explosive. In general, consequences quoted for dispersion of radioactive material assume that those exposed remain forever (or for the half-life of the material) in the affected area. But if the radioactive material were cobalt 60, which has a half-life of five years, and evacuation occurred within a month, exposure would be reduced by a factor of 80 or so, and the number of resulting cancer deaths would drop probably below the background level.

This is not to say that radiological bombs are not a threat. The likelihood of such an event could be reduced by an accurate registry of radioactive material throughout the world; the provision of cost-effective security, which would result in many cases in the replacement of radioactive sources by electron-beam radiation; and widespread monitoring on streets and in ports for radioactive materials.

You've used the term "megaterrorism" in your recent writings about national security. Is this a new class of threat or something that we've just been reluctant to recognize?

By megaterrorism, I mean an event that would cause 100 000 or more deaths from biological agents or nuclear explosions. (Previously, in a November/December 1998 article in Foreign Affairs, Ashton Carter, John Deutch, and Philip Zelikow invoked the term "catastrophic terrorism" to signify 40 000 deaths or so.) In the past, experts on terrorism believed that terrorists in general did not want to kill people; they wanted favorable mention of their cause in the media. This opinion has substantially changed since 9/11, because the suicide bombers and others in this case really wanted to kill as many Westerners or modernists as possible; their main purpose was not publicity but death. Thus it is no longer believed that if Osama Bin Laden had a nuclear weapon, he would use it to blackmail; experts now believe that he would explode it to invoke maximum destruction in the United States or another Western country.

You've also written that terrorist biological attacks pose a greater threat than nuclear attacks. Why is that?

Terrorist nuclear attacks could kill some multiple of 100 000. But the United States is a big country, with 280 million people. Even if we lost five times 100 000 to terrorist nuclear explosions, it would not threaten the country's survival.

On the other hand, the loosing by terrorists of smallpox virus so as to infect 50 000 primary cases (and 150 000 secondary within two weeks, 450 000 within four weeks, and so on) could result in almost all Americans being exposed, and 30 percent dying among the infected. This would really threaten the survival of the United States as a society, and even of civilization in the world. In contrast to the days when most people lived on a farm or within cart range of a farm, modern society is extremely specialized and interdependent— consider the operation of a semiconductor chip manufacturing plant or the complex process of building any other useful item these days. It is not clear how the United States could recover from the sudden loss of 20 percent of its people.

The impact would be so huge because smallpox is both infectious and contagious. Other contagious viruses and bacterial agents would have similar effects. The biological warfare agents that the United States weaponized in the 1950s and 1960s (before President Richard Nixon's Executive Order of 1969 banned such work) were not contagious, but simply infectious. Chief among them was anthrax, a durable spore with which the United States has had experience in the last year.

Fortunately, the United States has discovered that it has more than enough smallpox vaccine (although of a rather hazardous sort) to vaccinate everyone, and thus to prevent a pandemic. If "herd immunity" is accomplished by, say, 80 percent vaccination, then 15 000 primary cases would ultimately contribute perhaps 50 000 infections, whereas without vaccination many tens of millions would be infected.

What more needs to be done to prepare the general public?

We should be prepared to disseminate, by television or newspaper, information about the nature of whatever attack and contamination has occurred and what to do about it, including such non-specific, rough-and ready remedies like wearing facemasks. Such a mechanism does not now exist. Nor has there been much analysis of the potential benefits of interrupting a pandemic by measures such as handwashing or in-room filtration with HEPA [high-efficiency particulate air] filters.

The threat and the actuality of terrorism are real and serious. The U.S. response must be as well. We need to do the following:

• Act and speak so as not to greatly augment the pool of committed terrorists.

• Identify and recruit the most effective individuals for leadership posts in the Department of Homeland Security. We need a General Leslie C. Groves, of Manhattan Project fame, to spur the development and deployment of remedies to terrorism. We need a J. Robert Oppenheimer to head the laboratory or laboratories that will provide the technical options.

• Congress and the Administration need to focus on reality, not appearance. Rather than a gradual, multiyear transformation of existing activities and programs within the Homeland Security Department, we need to spend tens of billions of dollars of additional resources, now.