February 28, 2003

Richard L. Garwin
Philip D. Reed Senior Fellow
for Science and Technology
Council on Foreign Relations
58 East 68th Street
New York, NY 10021
(212) 434-9663
FAX: (914) 945-4419
INTERNET: rgarwin at cfr.org
Many papers at www.fas.org/rlg

Sound Science in Support of Sound Policy Decisions:
Nuclear, Biological and Chemical Weapons as an Example

Breakfast with Members of Congress, Room S-120
March 6, 2003

Thank you for the honor of your presence. My initial remarks are limited to 15 minutes, and the subject is vast and urgent. Biological, chemical, and nuclear weapons could hardly differ more-- in the nature of their effects, time required for influence, and ease of development, production, and dissemination. I will take BW first, then nuclear explosives.


In 1969 I was a member of the President's Science Advisory Committee-- PSAC-- and also of the Panel on Biological Warfare Agents asked by President Nixon to conduct a review of the capabilities and hazards of BW agents. At that time, PSAC's 18 members met two days every month, and its dozen panels had an equally intense schedule so we were able to marshal the best of U.S. scientific and technological expertise on such questions. President Nixon issued his Executive Order banning not only possession but all offensive work on BW (and later toxins-- which are chemical compounds made by living things). You all know the active U.S. BW program of the 1950s and 60s developing and testing agents with humans, plants, and animals as targets. Note that the BW agents against humans were infectious but not highly contagious-- e.g., tularemia or Venezuelan Equine Encephalitis-- VEE. By 1972 the Biological Weapons Convention-- BWC-- was ready and signed, enthusiastically but not honestly by the Soviet Union.

President Nixon's judgment in abandoning and attempting to outlaw BW and toxins was that although the United States could no doubt do these things better than any other country, the threat to U.S. security and in fact survival would be increased by the availability of such weapons and those which could be perfected by the approaching revolution of molecular biology. Almost any country could be a potent BW power. Furthermore, the risk of accident with BW was considerably larger than with nuclear weapons, and there loomed large the prospect that aggressive or desperate nations would weaponize highly contagious microbes such as smallpox virus.

Because they are not living entities, toxins such as ricin or botulinum toxin would not have been included in a ban on BW, and President Nixon banned them specifically. They were included in the BWC.

In 1972, PSAC was presented with the proposal by the Centers for Disease Control and the National Institutes of Health to abandon vaccination against smallpox in the United States, since the disease was well on its way to eradication worldwide and was in fact eradicated in 1980. I was dead set against this, in the near certainty that some smallpox virus would have been retained by scientists or governments, and that the world would inevitably be hostage to smallpox release, which could kill hundreds of millions of people. For the U.S., deterrence was available (even if unstated) by its store of nuclear weapons, but the problem was not only nations armed with BW but terrorists.

Rather than continuing smallpox vaccination and researching safer vaccines to eliminate the one or two per million death rate among those vaccinated, we abandoned the program and are scampering now to catch up.

For whatever reason, the Soviet Union accelerated its BW program rather than abandoning it, and Russia continued with a modern program incorporating the techniques and insights of modern biology. And Russia is still far from having come clean.

Even in a world of compliance with the BWC, we are left with the non-deterrable use of BW against the civilized world-- by home-grown terrorists or Al Qaeda. Vaccine development against many of these scourges should proceed, with the involvement of capable organizations in India and elsewhere, as well as in the United States. But, as Senator Frist has emphasized, and Ralph Gomory and I have been pushing, individuals can greatly reduce the impact of BW either spread in the open air or in buildings, and especially those which are contagious. Consider:

The impact of public health measures is far greater than might be imagined by a simple reduction of infectivity by a factor ten. For a disease, for instance, for which there are 10,000 primary victims, but in which each primary victim infects three others on the average (as may be the case with smallpox), the number of victims would triple every two weeks, and unless effective immunization were available the numbers having been infected would grow to the tens of millions. But if instead if infecting three additional persons, each primary case infected only 3/10 (because of improved public health measures), this whole chain would die out with a total of some 14,000 victims. Quite a difference from tens of millions! This simple fact of course, has much in common with the growth of the nuclear chain reaction.

During the weeks of anthrax letters of 2001, it was widely reported that the lethal dose of anthrax was some 5000 spores, as if a person who had ingested 2000 would be safe. Far more likely is that of 5000 people each ingesting one spore, one would contract the disease. But there are microbes which are far more efficient than anthrax in causing disease.

Never mind. If the attack is of such a magnitude as to kill the maximum number of people, there will be many who would be perfectly well if their exposure were cut by a factor two or ten. The important point is that non-specific public health measures such as hand washing and simple masks (such as this one) could reduce exposure by a factor 10 or 20, and would save many lives. An important benefit, too, is that such a mask inhibits the all too common habit of putting fingers to the mouth or nose, effectively transferring germs from outside to inside.

Scientific knowledge is required; but also the translation and validation of such knowledge into behavior. And it is a matter also of refining the informative messages to protect the public. The Washington Post is preparing such a self-help guide.

More scientific and engineering knowledge is needed. For instance, we are quite ignorant about the hazard of resuspension of spores which have settled out onto surfaces. Furthermore, quite benign foam (from Sandia National Laboratories) is claimed to be highly effective in destroying BW agents. This needs to be further tested and, if satisfactory, deployed.

And the notion of collective action in infection should be settled once and for all. If a single spore of anthrax conveys on the average 1/5000 probability of infection among the general population, that should be better understood. If (as seems unlikely) there is some threshold before there is any damage, that is equally important to know. The same is true for many other possible BW agents. Animal experiments should clarify the general question.

More practical experiments with simulants is required in order to guide remedial measures, such as the efficacy of HEPA filtering, or for that matter, the utility of ultraviolet "tunnels" irradiating portions of the room in order to destroy BW agents. Many of these measures will be beneficial for the general public health. Furthermore, practical surveys of the efficacy of expedient masks, such as the $1 N95 masks which are widely available would be most helpful, together with further analyses as to the utility of such measures, which might reduce exposure "only" by a factor 10, in reducing fatalities and illness.

Of course, general disease-fighting methods, such as bacteriophage, should be further explored, and might have enormous public health benefit in the world.

With the burgeoning of knowledge and technique in molecular biology and genomics, there is much promise also in looking at "small molecules" specifically to interfere with or inactivate pathogens. The large pharmaceutical corporation model for bringing these to market, though, is not appropriate for diseases which are rare in the United States, but might be used as BW agents against us. Still, in the mode of the anti-AIDS program initiated by the Bush Administration, such vaccines and disease-fighting agents could further U.S. foreign policy as well as help to protect against terrorism.


We have been fighting proliferation of nuclear weapons for more than 50 years, but not as if our lives depended on winning this battle. Which they do. Because it is no longer a matter of disruption of international security if two neighbors in some other part of the world acquire nuclear weapons and use them on one another. In almost all cases we could deter the use of nuclear weapons against the United States.

The problem now is actual use of nuclear weapons by terrorists who simply want to destroy our society and as many of us as possible.

This sets the standard much higher than when we could ask of a proliferant nation, "What would they expect to gain by striking us?"

Now more than 50 tons of weapons plutonium and hundreds of tons of reactor-grade plutonium are still vulnerable to diversion or theft by terrorists who, in some cases, have tens of millions of dollars in support. Unlike BW, for which a tiny trace of starter for the fermentation can result in kilograms of fermentation product adequate to destroy 100,000 people or more if dispersed in a practical fashion, the plutonium or highly enriched uranium-- HEU-- must be made by once-esoteric processes. The problem is that, once made, HEU is quite easy to handle. And even in 1945 there was no secret in making a gun-type bomb with 60 kg of HEU.

HEU is difficult to detect by its weak radioactivity. Detection would not help at all if the chosen delivery means were by missile, and our 1998 Rumsfeld Commission Report stated clearly that short-range cruise missiles or ballistic missiles from ships would be a more accurate and easier way to deliver nuclear weapons against U.S. cities than would the building of an ICBM. In any case, terrorists are unlikely to have an ICBM at their disposal, and a country whose territory might be used for such activities would bear in mind the return address provided by the strike.

My book, "Megawatts and Megatons: The Future of Nuclear Power and Nuclear Weapons," with Georges Charpak (University of Chicago Press, January 2003) provides a background into the technology of nuclear weapons and, in particular, into terrorist use of nuclear weapons. Chapter 12, in fact, is largely devoted to nuclear and biological megaterrorism. And on my website http://www.fas.org/rlg (and also to a lesser extent at www.cfr.org) you will find a large paper of August 2002 on nuclear and biological megaterrorism, with detailed calculations of the effect of a one kiloton or ten kiloton nuclear explosion in a densely populated part of Manhattan.

In short, a small nuclear explosion at ground level is more destructive than at "optimum altitude", because of the hundreds of thousands of people who will die within a week or two from radiation exposure during the first few minutes after the blast. The preferred mode of delivery, therefore, is to detonate the explosive in a van or even to construct it in a basement, apartment, or office. It is far less difficult to devise a nuclear explosive that does not need to withstand the rigors of missile launch or aircraft delivery and drop, but can be detonated on the bench. A few words are in order about the particular danger of HEU gun-type weapons in this regard (oral presentation).

The use of nuclear weapons by most nations possessing them can be deterred, assuming that the United States does not menace the survival of the nation or of its dictatorial leader. And there is great value in strengthening and supporting the Non-Proliferation Treaty (NPT), by which almost all nations of the world abjure the possession of nuclear weapons.

Unlike the case of BW, however, there is little that can be done to ameliorate the impact of a surprise nuclear detonation. We have had some experience with nuclear weapon threats used for blackmail; in contrast, Al Qaeda or other serious terrorist use of nuclear weapons will occur just as soon as the capability is ready-- with perhaps a tactical delay of a day or so to ensure maximum population density and hence maximum damage from the explosion.

Many hundreds of tons of HEU are available in Russia. Of this, 500 tons are to be delivered over 20 years (by 2012) in the "megatons to megawatts" program for which USEC is the U.S. executive agent. But probably 700 tons additional exists, and incentives should be provided to Russia to blend down this material to the LEU range (19.9% U-235 or below) where it would be much more difficult to use to produce a nuclear explosion.

Many tens of tons of weapon-grade plutonium are available also in Russia, and hundreds of tons throughout the world of "reactor-grade" plutonium. Despite the name, RG-Pu can be used to make Nagasaki-type nuclear weapons which in no case would have a yield less than 1-2 kilotons; according to official U.S. government statements RG-Pu could be made into high-performance weapons of yield comparable with the 20 kiloton yeild of the Nagasaki bomb or even more advanced explosives.

The imperfect remedy is to emphasize the Nunn-Lugar program of cooperative threat reduction and to aggregate these holdings in Russia so that they can be more reliably guarded and accounted for.

Even more serious problems exist with the HEU used for Pakistani weapons-- and the weapons themselves. And if North Korea proceeds with the retrieval of plutonium from its irradiated fuel (several bombs worth over the next six months), there is little doubt that NK would be willing to sell the Pu, just as it sells its missiles. Furthermore, NK could complete its 250 MW(e) graphite reactor, which would then provide weapon-grade Pu for an additional weapon every ten days or so.

What is the role of science in countering the threat of terrorist nuclear explosives? Sadly, one can do only so much to detect metallic plutonium or HEU as it is brought into a city or even into the country. There are, indeed things that can be done to scan shipping containers, but we have not deployed such measures in sufficient number or of the appropriate type. Similarly, truck traffic can be regulated in its content and packing so as to make it relatively easy to detect such materials. But that leaves ship and boat.

The detonation of a nuclear explosive in New York or Baltimore harbor would not produce the immediate casualties that a ground-level detonation would in the most densely populated part of the city, but it would cause enormous destruction.

Although little can be done to protect buildings and people against the effects of a nuclear explosion, appropriate redundancy can reduce societal disruption. And some measure of security could be obtained by registering and providing transponders on watercraft down to the smallest.