Preventing Nuclear War

Principal Investigator: Paul Slovic

Project Period: April 1, 2019 – September 30, 2020

Participating in two recent meetings on the risk of nuclear war organized by the Stanley Foundation has led me to see the need to bring my knowledge and experience from almost 60 years of research on the psychology of risk to bear on this existential threat to human survival. Specifically, these meetings allowed me to start investigating how my earlier research on psychological obstacles to action in the face of genocide and other mass atrocities might apply to decisions to use nuclear weapons.

What I found was that two key cognitive phenomena, psychic numbing and a decision-making bias called the prominence effect, were even more powerfully illustrated in warfare than in indifference to humanitarian crises. Whereas numbing and prominence has led to inaction in humanitarian crises, they have led to overt actions in war planning, designed to kill many millions of people to obtain national security objectives. In addition, dehumanization of the enemy and blaming of the victims amplifies the effects of numbing and prominence. All of these psychological factors allow military and government officials with access to powerful weapons to kill masses of noncombatants without the normal abhorrence that we would expect to inhibit such actions.

My research draws on psychological experiments and theory compatible with the pioneering insights of two of the leading psychologists of the 20th Century, Amos Tversky and Daniel Kahneman, including the work that led to Kahneman’s Nobel Prize in 2002. I strengthen my arguments by reporting supporting evidence from the historical record of planned and actual mass killings. My colleague and I conclude that we have created weapons whose consequences, if they are used, are so far beyond human experience that we cannot comprehend them; “The more who die, the less we care.” This raises critical questions in need of further study: How can we rationally weigh the costs vs. the benefits of deciding to use these weapons if we don’t comprehend the costs? If rational decision making is questionable, what are the implications for key elements of nuclear weapons policy: deterrence, disarmament, and nonproliferation?

Adding to this problem, we have no clear evidence that those with the power to use these weapons appreciate the difficulties of making this most consequential decision. To our knowledge, the President of the United States is not given any training about how to think about it. In fact, one high level commander has described an incident where the President was deliberately kept from participating in training exercises so as not to produce evidence of proclivities that could inform an adversary. The brief hearings by the U.S. Senate Foreign Relations Committee in November 2017 titled “Authority to Order the Use of Nuclear Weapons” barely touched on these decision-making challenges, providing further evidence that they are not recognized and taken seriously.

Proposed Research and Outreach. Decision Research has been awarded funds to enable my colleagues and I to focus intensively on additional laboratory experiments to test whether humans can think rationally about the high-stakes tradeoffs between the military benefits of using nuclear weapons and their vast costs. We will conduct a national survey in the U.S. to replicate and extend a prior survey by researchers at Stanford that found a shocking result: When considering the possible use of nuclear weapons in a hypothetical scenario about war with Iran, almost 60% Americans prioritized protecting U.S. troops and achieving American war aims, even when doing so would result in the deliberate killing of millions of civilian noncombatants. These findings suggest that public opinion is unlikely to be a serious constraint on any president contemplating the use of nuclear weapons in wartime. However, there were numerous limitations in the design of this survey that we believe must be corrected to determine its validity. We also wish to determine whether support or opposition for the use of nuclear weapons can be influenced by videos personalizing the potential victims of the bomb (predicted to increase opposition) or by videos personalizing the families of military personnel whose lives might be saved by use of the bomb (predicted to increase support for using the bomb).

Talks between the U.S. and North Korea are about to begin again, and they are unlikely to lead to denuclearization. Russia is scaling up its weapons and missile defense programs and the U.S. is likely to do the same. Nations such as Iran and Saudi Arabia appear to be seeking to develop nuclear weapons. Social media shortcuts diplomacy and careful decision analysis and spreads false information, adding to the decision-making problems. The threat of nuclear war is unlikely to diminish. Our psychological studies have implications for concrete steps that can be taken to reduce what are unacceptable risks to humanity.