The problem of accidental or inadvertent nuclear war has been couched largely in terms of superpower confrontations during a crisis. Whether the focus is on the major powers, or on developing nations with ballistic missiles and probable nuclear weapons capability, stability in those who handle weapons and effective safeguards on use are essential preventive measures. The United States and the USSR have been careful to guard against unauthorized launch. All nuclear nations have been concerned with retaining ultimate control of nuclear weapons in civilian hands; with monitoring the reliability and stability of the forces that handle the weapons; and with preventing weapons from coming into the possession of outsiders. In 1986, an analysis of the sources of human instability in those who handle nuclear weapons concluded that thousands of unstable individuals were involved in "minding our missiles."1 The present paper serves as an update on the problem and links it to potential areas of increasing risk as the world changes.

A conference was held at Stanford in May 1992 on possible cooperation between the United States and Russia on early warning of ballistic missile attack. Included in this volume is a briefing paper that was prepared immediately after that meeting.

In recent years, world attention has been drawn to the acquisition by developing countries of delivery systems for nuclear weapons as distinct from the nuclear warheads themselves. In particular, concerns have been raised about the spread of ballistic-missile systems and technologies to areas such as theMiddle East, in which there are strong regional tensions.  The extensive use of ballistic missiles in the "war of the cities" in 1988 during the Iran-Iraq War and the use of Scud missiles by Iraq against Israel and Saudi Arabia during Desert Storm highlighted the rapid proliferation of these weapons, and served as a premonition of worse things to come if and when developing states deploy ballistic missiles armed with weapons of mass destruction.  The ballistic missiles of concern have ranges of a hundred to a few thousand kilometers (km) and can carry payloads of up to one or two thousand kilograms (kg). These systems are becoming increasingly prominent in Third World arsenals, and are perceived as threatening stability and regional military balances.

This study examines the impact of ballistic-missile proliferation, focusing on sixteen regional states which have or soon could have ballistic-missile capabilities.  Specifically, the study: (1) evaluates the military effectlveness of ballistic missiles in comparison with advanced strike aircraft, (2) identifies trends in the supply of and demand for ballistic missiles, (3) identifies key technologies and systems whose control is essential to a successful missile non-proliferation regime, (4) reviews the relevancy and effectiveness of present control mechanisms including the MlssiIe Technology Control Regime (MTCR), and (5) offers policy options for strengthening controls on the acquisition by developing states of longer-range ground strike delivery systems, including ballistic missiles and advanced combat aircraft.

In August 1991, the people of Moscow and Leningrad demonstrated to the world that democracy is an idea whose time had come for the Soviet Union. As a result, the Soviet
government received a mandate for political reform. This reform, if successful, will result in a political system that will be very different from any that the people in the Soviet Union have ever known-a government of the people, by the people, and for the people.

Hand in hand with political reform, the Union and its republics will be attempting sweeping economic reforms. While recognizing how difficult it will be for the Soviet Union to achieve economic reform, it is also important to observe that it is at least conceivable now. A critically important component of economic reform in the Soviet Union is the conversion of their defense industry from a centrally managed monolith to a collection of independent, privately owned enterprises, producing civil products.

Operational arms control can take many forms, and one of the most important is direct military-to-military talks. The 1989 Dangerous Military Activities agreement, in which military officers headed negotiations for the first time, should not be considered the final step in improving U.S.-Soviet military-to-military relations.  It should be seen instead as a major step forward toward a much deeper and wider network of discussions and agreements which reduce the likelihood of misunderstandings and potential incidents between the militaries of the two states.

This conference report contains papers presented September 14-16, 1990, at a symposium at Stanford University, sponsored by the Institute of Far Eastern Studies (IFES) of the Soviet Academy of Sciences and the Center for International Security and Arms Control (CISAC).

After eight years of marathon negotiations, the United States and the Soviet Union are finally close to concluding a strategic-arms-reduction treaty (START). At the 1990 Washington summit, U.S. president George Bush and Soviet president Mikhail Gorbachev signed a communique concerning the reduction of strategic nuclear arms. Although the agreement is not the long-awaited START, the two presidents reaffirmed their determination to have the treaty completed and ready for signature by the end of 1990. The marked progress toward nuclear disarmament by the two superpowers has once again caused vast repercussions. While hailing progress, many people show more concern for the implications of the treaty for the future of arms control.

A warning system such as the Command, Control, Communications, and Intelligence system (C3I) for the U.S. nuclear forces operates on the basis of various sources of information, among which are signals from sensors. These sensors include, for example, radar receivers and satellites. A fundamental problem in the use of such signals is that the sensors provide only imperfect information. Two kinds of errors are possible: missed signals (Type I errors) and false alerts (Type II errors). Probability is therefore a key concept in the logical treatment of such signals. It is defined, in the Bayesian sense, as a degree of belief in the possibility ofeach event (in this case, attack or no attack, signal or no signal, and action or no action) (Savage 1954). One of the main problems in the use of the Bayesian method is that, because the base of evidence for estimation of these probabilities may be small, the figures may be fuzzy. A first issue is thus how to account, using probabilities, for the primary uncertainties about signals and events. A second issue is how to assess and treat uncertainties about these probabilities (epistemic uncertainties) while preserving probabilistic logic. A third issue is the relevance of these epistemic uncertainties in rational and prudent decision making when the stakes are as high as the possibility of an accidental nuclear attack.