... I recently spoke with Tom Nichols. He is a professor of national security affairs at the U.S. Naval War College and also teaches at the Harvard Extension School. Nichols is the author of seven books, including "No Use: Nuclear Weapons and U.S. National Security" and “The Death of Expertise: The Campaign Against Established Knowledge and Why It Matters.”

A longer version of this conversation can be heard on my podcast, which is available on Salon’s Featured Audio page. Nichols' opinions and comments are his own and do not reflect those of the United States government.

Given Trump's unpredictability in respect to foreign policy, the concept of the "madman theory" has been used by journalists and others to describe his supposed strategy. Can you provide some context for the madman theory? What are its origins?

The madman theory is a notion of deterrence in which you want to convince your opponent that you're more reckless than he is. Deterrence in its simpler version is like a game of chicken. So if that's the paradigm, the madman theory is like downing a bottle of vodka in front of your opponent and saying, "You know what, dude, I don't care if I live or die. I'm crazy. I'm nuts." On that concept, the other guy is going to swerve first, because he is going to say, "My opponent just isn't rational." The problem with the madman theory is that it also makes your opponent very nervous and prone to want to attack first, because you cannot be reasoned with. The biggest problem with Trump has been that if you tell people, "Hey, I'm doing the madman thing," then you've blown the whole idea. You don't publicly say, "Hey, I'm just trying to appear unreasonable." Because then your opponent says, "OK, not only are you reckless, you're stupid."

How have foreign policy professionals responded to Trump's apparent embrace of the madman theory?

Well, in my limited conversations with my fellow policy experts on this, I do not think anybody takes it seriously or that Trump is getting over on the North Koreans this way. I do think the North Koreans see him as unpredictable. They're trying to figure out who is in charge of the U.S. government. But I think all that is bad. I think deterrence is not well served when your opponent can't figure out who's running the show. The president of the most powerful country in the world should not say, "Hey, I'm unstable. I could do anything. I'm kind of a madman."

But I think one of the reasons you hear the madman theory discussed so often now is because President Trump's supporters are desperate to reverse-engineer some reason for what he's doing. They’re always trying to present his impulsiveness as a plan. So when he says, "Hey, Rocket Man, I could end you tomorrow." They say, "Ah, don't worry, it's the madman theory. He's playing four-dimensional chess." When in fact he's just tweeting or blurting stuff out. But his supporters have a hard time just admitting that the guy is impulsive. So they come up with elaborate theories about why he's doing what he's doing. I don't think anybody is really buying that, here or anywhere else in the world.

During the campaign Trump said, "Well, I don't want America's enemies to know what we're going to do. ... I want to surprise them. That's how we're going to win." That's how we're going to make America great again." 

The great Thomas Schelling, who pioneered a lot of this work back in the '60s, once said that deterrence rests on the threat that leaves something to chance. The problem is that the way Trump and his mouthpieces are taking it is they're saying that being totally unpredictable and opaque and nutty is the way to do stuff. Well, that's just wrong. Deterrence theory would tell you to say to your opponent, "Hey, I'm not interested in war, but you're setting in motion some things here that can go bad. I can't tell you that I can stay in control of everything happening. It's not a perfect world, but I can't control everything." This is far preferable to saying, "I'm totally unpredictable." ...