Testimony of Gary Milhollin
Professor, University of Wisconsin Law School and
Director, Wisconsin Project on Nuclear Arms Control
Before the House Committee on Armed Services
Subcommittee on Research and Development
April 30, 1992
I am grateful for this opportunity to appear before the Armed Services Committee’s Subcommittee on Research and Development, and to discuss the subject of nuclear arms proliferation.
The Subcommittee has asked me to address the question of how a nuclear attack on the United States might be carried out by means other than a long-range ballistic missile.
Before considering a number of plausible delivery methods, we should first ask which countries have nuclear weapons now, and which countries are likely to have them in the years just ahead.
Britain, France, China and several of the former Soviet republics have nuclear weapons–these are the declared nuclear weapon states. India, Israel, Pakistan and South Africa also seem to have manufactured nuclear weapons but have not openly admitted deploying them. These are called the undeclared nuclear weapon states. Argentina, Brazil, Iraq and North Korea probably could make nuclear weapons within the next few years if they wanted to do so badly enough, but they have not yet done so. These are the threshold states. Iran and Libya would also like to make the bomb but do not yet possess the means, and South Korea and Taiwan have made attempts in the past but now seem in a period of quiescence. These I call the potential nuclear weapon states. The countries are listed in Figure I.
You will notice that none can now reach the United States with an ICBM, except China and the former Soviet Republics. However, a number could deliver a bomb to the United States by what I have called “low-technology delivery,” a term that I will take up in a moment.
If we look five years down the road, only one of these countries, Israel, seems likely to be able to build an ICBM capable of reaching the United States. A number of additional states, however, could become capable of low-tech delivery during this period. So, if one is concerned about additional countries getting the bomb, and one considers how a new nuclear weapon state might actually deliver a bomb to the United States, one sees that the threat of low-tech delivery is the one that is increasing. The threat of ICBM delivery seems to be constant or possibly even decreasing if Russia proceeds with deep cuts in its arsenal, and the other former Soviet republics get rid of their missiles as they have promised.
Manufacturing a first-generation bomb today is easier than it used to be. From secret documents discovered in Iraq, we know the design of the bomb that the Iraqis were building. We also know that Iraq fabricated a bomb part–a natural uranium reflector–that matched exactly the design of the reflector described in the secret documents.
The Iraqi design is strikingly similar–both in size and configuration–to the one China tested successfully in 1966 and gave to Pakistan in the early 1980s. After getting the design, Pakistan scoured Europe for steel spheres very close to the size of the sphere that the Iraqi design now uses. U.S. government analysts speculate that China’s design probably became known to Pakistan’s suppliers, and through them to the Iraqis. If so, China’s giant act of proliferation is still having its aftershocks. A sketch of the Iraqi design is attached as Figure II.
There are two things to keep in mind about the design: first, it works; and second, it is generally available to proliferant countries. This means that a country trying to make the bomb is already starting part way up the nuclear mountain. It can begin its effort with a design that works and then try to buy or make the parts. The main bottleneck, therefore, is acquiring the plutonium or weapon-grade uranium to fuel the bomb. This material is not easy to manufacture, and as far as we know, bomb quantities of it have not been for sale on the world black market. To stop proliferation, we must keep this material out of the hands of the countries we are worried about, or convince them to allow international inspection of any of it that they happen to acquire.
What, then, are the low-technology ways of delivering a bomb to an American target? We will assume that the bomb is about the size of a desk, and that it weighs about a ton. We can also assume that it will yield about twenty kilotons, which was the power of the Nagasaki bomb and also the estimated power of the planned Iraqi bomb.
Some of the possible means of delivery are listed in Figure III. I have tried to rank them in order of decreasing technical difficulty. The first is the classic case of the smuggled bomb. There is no reason why this would not work if done carefully. Bombs can be taken apart and put together safely.
One advantage of this option is that the bomb could be positioned wherever desired; one need not worry about the accuracy of a missile guidance system. Some of the other delivery methods are more limited with respect to positioning the bomb. I will go into that in a moment. The last entry of Figure III deals with missile technology, which is the most demanding technically. It could become more attractive, however, as cruise missile technology spreads. All of these options are within the means of a new nuclear weapon state.
To show the impact of a twenty-kiloton bomb on Washington, D.C., I have prepared Figure IV. The red circle has a radius of .6 miles, within which one could expect up to 90% immediate fatalities, based on the experience in Nagasaki and Hiroshima. The black circle has a radius of 1.6 miles, within which one could expect 30% fatalities and 30% additional injuries. Between 1.6 and 3.1 miles one could expect 1 to 2% fatalities and 10 to 25% injuries.
It makes a considerable difference where such a bomb is set off. If it were on a boat in the Potomac River, the effect would be different from that of a device set off in a building or van at Tenth and Pennsylvania, or on a plane approaching National Airport.
It is certainly within the power of Israel, and probably within the power of India, Pakistan and South Africa to deploy bombs more powerful than twenty kilotons. A fission bomb yielding one hundred kilotons would produce 90% fatalities roughly a mile in radius.
What conclusions can we draw from all of this? First, we should realize that there is really no defense against nuclear weapons. This, by the way, has been the position of every American president from Truman through Carter. If someone with a nuclear weapon wanted to deliver it to an American city tomorrow and set if off, we probably could not stop it. Thus, it is crucial to keep track of who has nuclear weapons and who is trying to get them.
Second, the low-tech delivery option has many advantages for a country with a handful of bombs. One is cost. It is much cheaper to buy a van, a boat, an aircraft, or even an office or apartment building than to buy an ICBM system. Another advantage is anonymity. A bomb in a building or a boat or a van would not leave much evidence. An American president would be reluctant to order a nuclear strike against another country without being certain where the bomb came from. It could take some time to figure this question out, especially if the FBI building is within the red circle. Forty months after the Lockerbie case we are still not certain who is responsible, and there was extensive physical evidence at the scene.
The doctrine of nuclear deterrence depends on knowing where an attack is coming from. If we look at the list of low-tech options, we see that the lower the technology the lower the risk of being detected, and therefore the lower the probability of being deterred.
Low-tech options can also be developed quietly. For India to develop an ICBM, there would have to be a series of long-range tests. The first test would undoubtedly set off a storm of opposition, and could easily end India’s access to the World Bank and International Monetary Fund.
Finally, if a country has only a handful of warheads, it will want to be sure they reach their targets. It would be more prudent to count on one of the low-tech methods than on an untried ICBM, which would itself be vulnerable to a preemptive strike.
The lesson, I believe, is that we should redirect our defense resources to counter the threats that are increasing rather than decreasing. We have already spent–and are still spending-billions of dollars to counter a Soviet threat that is disappearing, but we are spending very little to confront the growing nuclear threat from the developing world, which is not likely to include ICBMs.
To counter nuclear proliferation we must spend our money differently. We need, for example, better intelligence. Our intelligence on Iraq before the Gulf War was far too thin, and our intelligence now on North Korea is plainly inadequate. It is not acceptable to be told that North Korea is a closed society and therefore our government has no idea how close the country may be to making an atomic bomb. If we are going to protect ourselves during the rest of this century, we will have to improve our intelligence gathering.
We also need to fully fund the International Atomic Energy Agency–or even increase its funding–so it can carry out more aggressive inspections. And we need to fully fund the U.N. inspection effort in Iraq, a process that we cannot allow to fail.
In addition, we need to reduce the amount of nuclear weapon material circulating in world commerce. The more it circulates, the higher the risk that a few critical masses will go astray. Japan’s plans for plutonium shipments need to be canceled.
Finally, we need to tighten export controls. Iraq’s fantastic success in buying nuclear and chemical weapon and missile technology cannot be repeated. If we care about our cities, we cannot let that happen again.
Possible Nuclear Threats to the United States
Able to reach United States
Able to reach United States
With Low-tech delivery
|Now||In five years||Now||In five years|
|South Africa||South Africa||South Africa||No|
|North Korea||North Korea [?]|
|Belarus||Belarus||Belarus [?]||Belarus||Belarus [?]|
|Kazakhstan||Kazakhstan||Kazakhstan [?]||Kazakhstan||Kazakhstan [?]|
|Ukraine||Ukraine||Ukraine [?]||Ukraine||Ukraine [?]|
THE LOW-TECHNOLOGY OPTIONS
* In a van or small truck and driven to the target
* On the upper floor of an office or apartment building
B. A bomb assembled elsewhere and delivered by:
* A truck or van entering the United States across the Mexican or Canadian border
* A yacht, cruise ship or cargo vessel entering port
* A cargo or passenger plane approaching National Airport and unwittingly carrying the bomb in a crate
* A private aircraft deviating from its approach to National Airport on a suicide mission
C. A missile launched from the deck of a ship:
* A Chinese-supplied Silkworm cruise missile (sold to Iran) could reach Washington from the deck of a freighter in the Chesapeake Bay, or could reach New York City or other American ports from 40 to 50 miles at sea.