I’m not terribly consistent when it comes to promoting, at LGM, pieces that I’ve published elsewhere. Unfortunately, a Foreign Policy article that I wrote with Dani Nedal on Trumpian unpredictability has aged well.
I suppose it’s fitting that we wrote this piece back in April, when various enablers last claimed that Trump’s unpredictability would solve the North Korea problem.
Dani and I spend some time working through the differences between the so-called “Madman Theory” of deterrence and Trumpian unpredictability.
Some commentators link Trump’s championing of unpredictability to the so-called “madman theory” of Richard Nixon’s attempt to persuade rivals — including the North Vietnamese and the Soviet Union — that he was impulsive and unpredictable. Neither Hanoi nor Moscow was ever entirely convinced by Nixon’s stance. But the madman theory also wasn’t about Trumpian unpredictability. Nixon wanted to convince his adversaries that he was irrational, but consistent, when it came to calculating the downsides of using force.
The strategy was attractive, in large part, because some of the situations Nixon faced did not lend themselves to standard solutions. In the context of nuclear deterrence and coercion — which was central to Nixon’s calculations — the textbook approach is to make a nuclear response more or less automatic. Such policies are ways of approximating the act of “throwing the steering wheel out the window” in a game of chicken. They show your opponent that you can’t swerve out of the way — that you will, metaphorically or literally, fight to the death.
There was no guarantee that the United States would go nuclear over Berlin, but the U.S. troop presence in the city made clear that Washington would be under enormous pressure to “do something” following thousands of American deaths. It left multiple pathways through which an attack on Berlin might spiral out of control. As famed nuclear theorist Thomas Schelling noted of the garrison in Berlin, “What can 7,000 American troops do, or 12,000 Allied troops? Bluntly, they can die. They can die heroically, dramatically, and in a manner that guarantees that the action cannot stop there.”
The “tripwire” of an outmatched U.S. presence in Berlin therefore enhanced deterrence. By placing its troops in a place where they might be easily sacrificed, Washington showed it simply had no other option than escalating the conflict. While we might associate such behavior with a crazy person, it is the exact opposite of unpredictability. Throwing the steering wheel out the window makes the outcome of failing to swerve totally predictable.
To some degree, Trump’s talk of “fire and fury” is more consistent with all of this than with being unpredictable (note that Nixon’s gambit didn’t actually work). But it would need to be conjoined to some indication of what might trigger him. Trump’s original stipulation—”North Korea best not make any more threats to the United States”—is so obviously not credible that Pyongyang has already blown right past it. Given this, it’s not surprising that Mattis tried to rewrite Trump’s statement as a more traditional threat: the regime should stop its pursuit of nuclear weapons and return to diplomacy, as North Korea has no hope of defeating the United States and American allies.
Indeed, as we wrote:
Similarly, leaks from the administration suggested that if Pyongyang tested a nuclear device last weekend, then the United States would launch military action against North Korea. Other members of the administration walked back those threats, creating — at least in public — significant ambiguity about possible American actions. On Monday, Vice President Mike Pence warned that North Korea should not test American resolve but that the United States is open to talks. Let’s say that Trump does, in fact, intend to retaliate if North Korea tests another nuclear device. The unpredictability of the situation likely makes Pyongyang more, not less, likely to initiate a test. After all, it cannot be sure that Trump would, in fact, use force.
There are situations where this might benefit American policymakers. If Washington wants to deter an adversary, but does not actually want to use force, then leaving the threat ambiguous reduces the political costs of backing down, stopping opponents at home from accusing you of chickening out of enforcing a supposed red line. If the goal is to keep an adversary from taking any provocative steps — even those short of what you consider worth using force or imposing sanctions over — then introducing some unpredictability about what would trigger a response might be a good idea.
The problem is that ambiguity might encourage the adversary to probe your resolve and test the limits of your interests while making it more difficult to clearly signal that a particular move is a step too far and will credibly invite retaliation. For example, in the absence of clear signals about what the United States is and is not willing to tolerate, and faced with mixed signals about American interests, Pyongyang might be tempted to initiate a series of low-level incidents designed to test the limits of U.S. tolerance. It is easy to imagine one of those actions, like the downing or seizure of a naval vessel or drone, crossing a line that prompts a forceful response to the perceived affront. The irony in such a scenario is that Pyongyang might steer clear of these actions if it could predict with some confidence how the United States would react.
As we go on to argue, the problems with Trumpian unpredictability are compounded by the fact that the United States enjoys, at least for now, an extensive network of alliances and strategic partnerships.
Thus, for the United States, unpredictability carries enormous risks. That’s true for Nixonian calculated irrationality, too, but much more so for Trumpian unpredictability. Rivals and allies can easily interpret mixed signals from different voices in the administration and frequent high-profile policy reversals as evidence that the president does not mean what he says, that he has no idea what he is doing, or that he can change his mind on a whim. Intentionally fostering uncertainty reduces the credibility of existing commitments.
Unraveling the American alliance network by undermining confidence in Washington is probably the worst way to implement an America First policy. It undercuts a major source of American strength without gaining the benefits that might follow from strategic retrenchment — that is, of making deliberate decisions about what commitments are key to American security and which can be shed, while taking steps to ensure that unwinding those commitments don’t harm vital interests and alliances.
Trumpian unpredictability creates more problems than solutions. Playing crazy may sometimes be an attractive strategy, especially for weaker actors that have a narrow set of minimalist goals — like survival or autonomy. But if a state has more expansive goals, and ample resources to pursue them, as does the United States, unpredictability is a poor approach to grand strategy. It is hard for others to follow your lead when they don’t know what your goals are.
You should read the whole thing when you get a chance.