The Effects of Nuclear War

EXECUTIVE SUMMARY

At the request of the Senate Committee on Foreign Relations, the Office of Technology Assessment has undertaken to describe the effects of a nuclear war on the civilian populations, economies, and societies of the United States and the Soviet Union.

Nuclear war is not a comfortable subject. Throughout all the variations, possibilities, and uncertainties that this study describes, one theme is constant—a nuclear war would be a catastrophe. A militarily plausible nuclear attack, even “limited,” could be expected to kill people and to inflict economic damage on a scale unprecedented in American experience; a large-scale nuclear exchange would be a calamity unprecedented in human history. The mind recoils from the effort to foresee the details of such a calamity, and from the careful explanation of the unavoidable uncertainties as to whether people would die from blast damage, from fallout radiation, or from starvation during the following winter. But the fact remains that nuclear war is possible, and the possibility of nuclear war has formed part of the foundation of international politics, and of U.S. policy, ever since nuclear weapons were used in 1945.

The premise of this study is that those who deal with the large issues of world politics should understand what is known, and perhaps more importantly what is not known, about the likely consequences if efforts to deter and avoid nuclear war should fail. Those who deal with policy issues regarding nuclear weapons should know what such weapons can do, and the extent of the uncertainties about what such weapons might do.

FINDINGS

1. The effects of a nuclear war that cannot be calculated are at least as important as those for which calculations are attempted. Moreover, even these limited calculations are subject to very large uncertainties.

   Conservative military planners tend to base their calculations on factors that can be either controlled or predicted, and to make pessimistic assumptions where control or prediction are impossible. For example, planning for strategic nuclear warfare looks at the extent to which civilian targets will be destroyed by blast, and discounts the additional damage which may be caused by fires that the blast could ignite. This is not because fires are unlikely to cause damage, but because the extent of fire damage depends on factors such as weather and details is of building construction that make it much more difficult to predict than blast damage. While it is proper for a military plan to provide for the destruction of key targets by the surest means even in unfavorable circumstances, the nonmilitary observer should remember that actual damage is likely to be greater than that reflected in the military calculations. This is particularly true for indirect effects such as deaths resulting from injuries and the unavailability of medical care, or for economic damage resulting from disruption and disorganization rather than from direct destruction.

   For more than a decade, the declared policy of the United States has given prominence to a concept of “assured destruction:” the capabilities of U.S. nuclear weapons have been described in terms of the level of damage they can surely inflict even in the most unfavorable circumstances. It should be understood that in the event of an actual nuclear war, the destruction resulting from an all-out nuclear attack would probably be far greater. In addition to the tens of millions of deaths during the days and weeks after the attack, there would probably be further millions (perhaps further tens of millions) of deaths in the ensuing months or years. In addition to the enormous economic destruction caused by the actual nuclear explosions, there would be some years during which the residual economy would decline further, as stocks were consumed and machines wore out faster than recovered production could replace them. Nobody knows how to estimate the likelihood that industrial civilization might collapse in the areas attacked; additionally, the possibility of significant long-term ecological damage cannot be excluded.

2. The impact of even a “small” or “limited” nuclear attack would be enormous. Although predictions of the effects of such an attack are subject to the same uncertainties as predictions of the effects of an all-out attack, the possibilities can be bounded. OTA examined the impact of a small attack on economic targets (an attack on oil refineries limited to 10 missiles), and found that while economic recovery would be possible, the economic damage and social dislocation could be immense. A review of calculations of the effects on civilian populations and economies of major counterforce attacks found that while the consequences might be endurable (since they would be on a scale with wars and epidemics that nations have endured in the past), the number of deaths might be as high as 20 million. Moreover, the uncertainties are such that no government could predict with any confidence what the results of a limited attack or counterattack would be even if there was no further escalation.
3. It is therefore reasonable to suppose that the extreme uncertainties about the effects of a nuclear attack, as well as the certainty that the minimum consequences would be enormous, both play a role in the deterrent effect of nuclear weapons.
4. There are major differences between the United States and the Soviet Union that affect the nature of their vulnerability to nuclear attacks, despite the fact that both are large and diversified industrial countries. Differences between the two countries in terms of population distribution, closeness of population to other targets, vulnerability of agricultural systems, vulnerability of cities to fire, socioeconomic system, and political system create significant asymmetries in the potential effects of nuclear attacks. Differences in civil defense preparations and in the structure of the strategic arsenals compound these asymmetries. By and large, the Soviet Union is favored by geography and by a political/economic structure geared to emergencies; the United States is favored by having a bigger and better economy and (perhaps) a greater capacity for effective decentralization. The larger size of Soviet weapons also means that they are likely to kill more people while aiming at something else.
5. Although it is true that effective sheltering and/or evacuation could save lives, it is not clear that a civil defense program based on providing shelters or planning evacuation would necessarily be effective. To save lives, it is not only necessary to provide shelter in, or evacuation to, the right place (and only extreme measures of dispersion would overcome the problem that the location of safe places cannot be reliably predicted), it is also necessary to provide food, water, medical supplies, sanitation, security against other people, possibly filtered air, etc. After fallout diminishes, there must be enough supplies and enough organization to keep people alive while production is being restored. The effectiveness of civil defense measures depends, among other things, on the events leading up to the attack, the enemy’s targeting policy, and sheer luck.
6. The situation in which the survivors of a nuclear attack find themselves will be quite unprecedented. The surviving nation would be far weaker—economically, socially, and politically—than one would calculate by adding up the surviving economic assets and the numbers and skills of the surviving people. Natural resources would be destroyed; surviving equipment would be designed to use materials and skills that might no longer exist; and indeed some regions might be almost uninhabitable. Furthermore, prewar patterns of behavior would surely change, though in unpredictable ways. Finally, the entire society would suffer from the enormous psychological shock of having discovered the extent of its vulnerability.
7. From an economic point of view, and possibly from a political and social viewpoint as well, conditions after an attack would get worse before they started to get better. For a period of time, people could live off supplies (and, in a sense, off habits) left over from before the war. But shortages and uncertainties would get worse. The survivors would find themselves in a race to achieve viability (i. e., production at least equaling consumption plus depreciation) before stocks ran out completely. A failure to achieve viability, or even a slow recovery, would result in many additional deaths, and much additional economic, political, and social deterioration. This postwar damage could be as devastating as the damage from the actual nuclear explosions.

---