What’s the Problem with Dictators Having Nuclear Weapons? Accidental Detonation

Q&A with Scott Sagan, a senior fellow at FSI’s Center for International Security and Cooperation, the Caroline S.G. Munro Professor of Political Science, and the Mimi and Peter Haas University Fellow in Undergraduate Education. Written with Katy Gabel Chui.

Photo by Getty Images.

How should we handle situations where personalist dictators get ahold of nuclear weapons? What does this mean for how the U.S. deals with North Korea? Nuclear expert Scott D. Sagan explored this question in a recent piece for Foreign Affairs. In this Q&A, he explains a new theory for deterring “armed and dangerous” dictators.

You argue in your new Foreign Affairs article that “nuclear-armed personalist dictatorships” are the most dangerous regimes that might acquire nuclear weapons. What exactly is a personalist dictator, and why might they be more dangerous than other leaders?

Personalist dictators differ from other autocratic leaders because they hold such dominant personal power that other institutions — such as the party, or the military, or a politburo — are not able overturn any decision made at the top. This means that such dictators can make crucial decisions about going to war, or using nuclear weapons in a war, on a whim. They can make decisions based on faulty assumptions or deeply flawed decision processes.

This is a huge challenge for the whole concept of nuclear stability. Political scientists sometimes say that successful deterrence requires pristine rationality on the part of leaders. That is not correct. A leader may be irrational, but deterrence can still hold if the state has sufficient checks and balances so that crazy decisions are not implemented and rash statements are ignored.

So what are examples of such dangerous leaders?

There are a number of historical examples of personalist dictators who sought nuclear weapons but failed to get them: Saddam Hussein and Muammar al-Qaddafi are the most dramatic cases in point. Both leaders surrounded themselves with yes men, who told them only what they thought the leader wanted to hear. Both made major life and death decisions, such as Hussein invading Kuwait in 1990 or Qaddafi ordering terrorist attacks against American targets in Europe in 1986 — basically on their own. We are very fortunate that Iraq and Libya failed to build the bomb, though they clearly were trying to do so.

The strongest example today is Kim Jung Un, who has nuclear weapons now and the ability to deliver them with high confidence against targets in Japan and South Korea (including American military bases there) and with less confidence against the mainland United States. But Kim will be hard to deter because he has consolidated power, purging rivals and making crucial decisions on his own, for better or worse.

What specific scenarios for war with North Korea worry you the most?

War by accident or false warning. Remember the false warning incident that happened in Hawaii back in January? A Hawaii Emergency Management Agency system operator, after having been told by a supervisor that an exercise was not a drill, posted a warning that Hawaii was under attack by North Korea. Throughout the islands, citizens panicked, some running for the beaches, while others (more appropriately), sought shelter inside their homes. At the national level, however, no one panicked for three reasons. First, the United States has sophisticated redundant warning sensors, and none of them reported that a missile had been launched. Second, the U.S. has highly professional civil defense officials and military officers who reported up that chain that there had been a mistake. Third, no high-level official believed that Kim Jong Un was about to launch an unprovoked nuclear attack on Hawaii.

But imagine what would have happened had a similar incident occurred in Pyongyang rather than Honolulu. The North Koreans lack sophisticated and redundant warning systems, relying on primitive radars. Moreover, if military officers made a serious mistake, like reporting falsely that the U.S. or South Korea was attacking North Korea, it is unlikely that such officers would correct themselves in a crisis. After all, you don’t lose your job for making mistakes in North Korea — you can lose your life. Finally, if U.S.-North Korean relations turn sour in the future, because of their failure to denuclearize, we are likely to reenter a crisis atmosphere like 2017. If that happens, which I think is likely, Kim Jong Un is likely to believe that the U.S. might just launch a surprise nuclear or conventional attack against him. Why? Because President Trump has so often said exactly that in the past.

So what can we do about a nuclear North Korea if Kim is so hard to deter?

He is hard to deter, but it’s not impossible. Indeed, I think deterrence can still work if the U.S. is smarter about its threats and promises. First, avoid threatening North Korea with aggressive statements hinting that the U.S. may well start a preventive war if they do not denuclearize. In May, Trump said “In Libya, we decimated that country,” adding that “that model would take place [in North Korea] if we don’t make a deal, most likely.” Today Trump has been saying that he has “fallen in love” with Kim. But that will change if and when it becomes clearer that Kim is failing to give up his nuclear arsenal. If and when that happens, it is crucial that the U.S. softens its rhetoric.

Second, we need to have a smarter and more ethical nuclear deterrence doctrine. The U.S. could adopt a “no first use” doctrine, promising that we would not be the first state to use nuclear weapons on the Korean peninsula. Moreover, we do not need to threaten massive nuclear retaliation against North Korea even in the event of war. To deter personalist dictators, we need to make tailored threats against regimes and individual leaders, not to attack their people or industry.

Finally, we should emphasize conventional retaliation against the vast number of targets in North Korea that could be destroyed with our powerful conventional bombs and missiles. We may need a small number of nuclear weapons to threaten to destroy the small number of targets that are buried or otherwise hardened, but even then we should plan to use the lowest yield nuclear weapon possible to reduce collateral damage as much as humanly possible.

Does that mean that the U.S. needs to build and deploy new tactical nuclear weapons?

We should consider building new lower-yield weapons or redesigning more existing warheads to have what is called a “dial-a-yield” capability. But I object to the commonly used term “tactical nuclear weapons.” Any use of a nuclear weapon, regardless of its size or target, would be a strategic, not a tactical, decision.


Faculty views expressed here do not necessarily represent those of the Freeman Spogli Institute for International Studies or Stanford University, both of which are nonpartisan institutions.