The Limits of the Madman Theory

Even though they have already witnessed four years of a Donald Trump presidency, analysts have little certainty about how Trump plans to approach most countries in his second term. That’s exactly how he likes it. Since first running for president in 2016, Trump has positioned himself as an unpredictable leader and argued that doing so strengthens U.S. foreign policy. During his first campaign, Trump argued, “We must as a nation be more unpredictable.” At a dinner in 2018, when discussing negotiations with North Korea, he called himself a “madman”—only partially tongue in cheek. Asked during his 2024 campaign how he would respond to a Chinese blockade of Taiwan, Trump responded, “I won’t have to, because [Xi Jinping] respects me and he knows I’m fucking crazy.”

Other politicians and commentators have made similar claims. “Trump is, as his detractors and his supporters would say, unpredictable,” Vice President JD Vance said in June. “I am 100 percent certain that unpredictability redounded to the benefit of the United States.” The Washington Post columnist Shadi Hamid recently argued that Trump’s madman strategy pressured Israel to accept a Gaza cease-fire. And a report from the America First Policy Institute claimed that Putin did not invade Ukraine during Trump’s first term because “Putin could not be sure how Trump would respond.”

Trump’s first term was, indeed, an exercise in unpredictability. He threatened North Korean leader Kim Jong Un with “fire and fury like the world has never seen.” But he then became the first American president to meet a North Korean leader and declared that the two “fell in love.” He dropped the U.S. military’s largest conventional bomb on militants in Afghanistan yet also started peace talks with the Taliban. He ordered a strike on Iran and then canceled it. He imposed tariffs on some American allies, such as Canada, while sparing others, such as Australia, often for no rhyme or reason. According to Axios, during trade negotiations with South Korea, Trump ordered his negotiating team to say, in reference to him, “This guy’s so crazy he could pull out any minute.”

Trump is hardly the first leader to explicitly embrace an erratic foreign policy. For decades, heads of state around the world have deployed what is known as madman theory: the idea that by acting in a highly volatile way, they can frighten opponents into conceding. During the Cold War, for instance, some strategists suggested that by appearing unstable, a U.S. leader might prompt communist states to take U.S. nuclear threats more seriously.

But Trump should be careful when it comes to madman theory, because scholarship suggests that it is very difficult to wield successfully. In fact, it is rare that a reputation for madness actually pays off internationally. Modern leaders and heads of state who have tried to seem mad often fail to convince their adversaries. Others succeed, only to find that their reputation for madness persuades opponents that they cannot be trusted to maintain peace. Trump will thus need to walk a fine line, persuading other states that he is mad enough to make good on his threats yet stable enough to stand by the agreements he makes. History suggests that won’t be easy.

CRAZY LIKE A FOX

The basics of madman theory date at least to 1517, when the political philosopher Niccolo Machiavelli argued that, under certain circumstances, “it is a very wise thing to simulate madness.” But the modern notion was developed during the mid-twentieth century, when the advent of nuclear weapons posed a novel challenge for threat credibility. Because nuclear war would result in mutually assured destruction for both sides and conventional conflict could easily escalate to the nuclear level, the U.S. and Soviet governments had strong disincentives to fight each other. This meant that threats by one against the other could easily be dismissed. Why, after all, would a leader of either country do something potentially suicidal in response to nonexistential threats?

The answer: by being crazy. A rational U.S. president or Soviet premier would seek to avoid a nuclear confrontation at all costs. But a “convincingly mad” one, as the RAND defense strategist Daniel Ellsberg put it in a 1959 lecture, could credibly threaten large risks. Other experts agreed. The economist and nuclear theorist Thomas Schelling wrote that a “paradox of deterrence is that it does not always help to be, or to be believed to be, fully rational, cool-headed, and in control of oneself.” The military theorist Herman Kahn argued that when dealing with a nuclear-armed leader who appears to be “stark, staring mad,” opponents must either yield to his demands or “accept the possibility of being annihilated.”

These strategists’ ideas soon gained a prominent adherent: U.S. President Richard Nixon. In fact, according to the memoirs of former White House Chief of Staff H. R. Haldeman, Nixon gave the concept its name. “I call it the Madman Theory, Bob,” the incoming president said in 1968, when discussing how to compel North Vietnam to surrender. “I want the North Vietnamese to believe that I’ve reached the point that I might do anything to stop the war.” Nixon suggested that his advisers could “slip the word” to the North Vietnamese that “we can’t restrain him when he’s angry—and he has his hand on the nuclear button.” Shortly after entering office, Nixon and his National Security Adviser Henry Kissinger attempted to do just that. Following Kissinger’s instructions while visiting Moscow, the White House adviser Leonard Garment told Soviet officials that Nixon was “a dramatically disjointed personality” who was “capable of barbaric cruelty,” “more than a little paranoid,” and “predictably unpredictable.”

Trump is not the first leader to embrace an erratic foreign policy.

Other international leaders have also tried to put the theory into practice. In the 1950s and 1960s, Soviet Premier Nikita Khrushchev deliberately cultivated an image of insanity that, at least initially, some U.S. officials believed. Secretary of State John Foster Dulles said that Khrushchev was “obviously intoxicated much of the time,” “essentially emotional,” and “could be expected to commit irrational acts” without considering the consequences. Saddam Hussein was also viewed as a madman by foreign leaders, although there is no evidence that he cultivated this reputation deliberately. A 1991 psychological profile by a U.S. analyst said that the Iraqi dictator’s personality was characterized by “messianic ambition for unlimited power, absence of conscience, unconstrained aggression, and a paranoid outlook.” U.S. President Ronald Reagan called Libyan leader Muammar al-Qaddafi a “mad clown” and an “unpredictable fanatic.” Reagan’s defense secretary, Caspar Weinberger, believed that Qaddafi possibly suffered “from an incurable venereal disease” that caused “occasional bouts of madness exhibiting hysteria, braggadocio, and extreme theatricalism.”

More recently, Russian President Vladimir Putin has developed a reputation for madness after invading Ukraine. In February 2022, Marco Rubio, then a senator and now the secretary of state, said that Putin appeared “to have some neuro/physiological health issues”—warning that the Kremlin’s risk calculus had changed in dangerous ways. Former British Prime Minister Boris Johnson said that Putin may be “an irrational actor” and “thinking illogically” about the effects of sanctions and the risks of war. Putin’s repeated threats to use nuclear weapons have only played into this idea.

Assessments of North Korea’s Kim Jong Un have also raised questions about madness. When asked in a 2017 television interview if the United States could rely on mutually assured destruction to deter North Korea, then National Security Adviser H. R. McMaster said that he was unwilling to “bet the farm, or a U.S. city” on Kim’s rationality. In a leaked 2017 conversation with then Philippine President Rodrigo Duterte, even the unpredictable Trump expressed uncertainty about whether deterrence would work with the North Korean leader because Kim “could be crazy.”

IS IT BAD TO BE MAD?

Madman theory may be popular, but historically it has done little for its adherents. Nixon failed to persuade the North Vietnamese and their Soviet allies of his madness, and he ultimately had to withdraw from Vietnam. During the Soviet Union’s efforts to seize control over all Berlin, Khrushchev threatened to use nuclear weapons and often seemed to lose control of his emotions when meeting Western officials—yelling, gesturing, and turning red in the face. Yet he was unable to compel the United States to retreat. He later blinked during a standoff with Washington over placing nuclear missiles in Cuba, pulling them from the island. Saddam’s and Qaddafi’s perceived madness led them to worse fates: they were removed from power by U.S. military interventions and then killed. Putin’s nuclear threats have succeeded at limiting Western military assistance to Kyiv, but only to a point. After strict initial restrictions, Washington and its allies are now providing the Ukrainians with many kinds of advanced weapons, including long-range missiles, tanks, and fighter jets.

The most obvious reason for this shoddy track record is that it is hard for any leader, and especially leaders of nuclear-armed countries, to convince adversaries that their behavior is fundamentally irrational. But plenty of leaders do manage to cultivate a reputation for instability or extreme unpredictability, and they still fail. That is because by engaging in such behavior, leaders have trouble persuading opponents that they will stand by any commitments. Persuading leaders of foreign governments to yield to a demand, after all, requires not just convincing them that resisting will be punished. It also requires convincing them that yielding will actually prevent punishment.

Consider, for example, Putin’s threats to retaliate against the West for supporting Ukraine. In addition to invoking Russia’s nuclear arsenal, they feature threats to conventionally directly attack NATO countries. In November, for example, Putin said he might strike “the military facilities of those countries that allow their weapons to be used against our facilities.” Yet despite such drastic threats, Putin has failed to change Western policy. One reason might be that Western analysts doubt that Putin is prepared to strike a nuclear-armed adversary (NATO). But equally important may be their doubts that complying with Putin’s demands will lead to peace. Russia hawks in the West have repeatedly argued that if Putin prevails in Ukraine he will hardly stop there; he will go on to attack other countries, including NATO members. The West, they say, is thus better off resisting Putin now than later.

Not all leaders are perceived as equally crazy.

It is hard for leaders who style themselves as mad to resolve this paradox. Leaders with a reputation for unpredictability—or who encourage the perception that they might do almost anything, regardless of the consequences—often struggle to make credible guarantees. So do leaders whose notoriety is based on abnormal or distorted preferences, such as ones associated with fanaticism or extremism. Saddam’s insatiable appetite for conquest, for example, made him appear inherently drawn to conflict. As a result, any promise he made of peace rang hollow. Such concerns underlay the George W. Bush administration’s fear that Iraq would inevitably attack if it obtained weapons of mass destruction, helping justify its decision to invade the country.

Yet not all leaders are perceived as equally crazy. Hussein may have seemed utterly bent on domination, and Qaddafi seemed completely unpredictable. But others are viewed as having milder and situation-specific forms of irrationality. U.S. officials saw Khrushchev as a hothead who might fly off the handle in a confrontation, but no one thought that he would attack the United States at random. Some critics, including an anonymous Biden administration official quoted by Politico, have called Israeli Prime Minister Benjamin Netanyahu “crazy” for the extremes he has gone to in attacking Israel’s enemies. But even Netanyahu’s biggest foes do not think that he is mad enough to threaten historically friendlier Arab states, such as Egypt, when they condemn his government.

These leaders are likely to fare better on the international stage because they can combine a credible threat of war with a credible promise of peace. For example, Khrushchev’s unpredictable behavior during the Cuban missile crisis—during which he initially agreed to withdraw Soviet missiles, before demanding new concessions—helped coax U.S. President John F. Kennedy to remove U.S. nuclear missiles from Turkey. Kim Jong Un managed to leverage his unpredictability into a meeting with Trump that increased Kim’s prestige.

Ironically, the best example is Adolf Hitler. Although the dictator eventually revealed himself to be a genocidal megalomaniac, in 1938 he managed to convince some of his British interlocutors that he was an extremist only on the issue of establishing a greater Germany. Partially as a result, the British agreed to let him annex the part of Czechoslovakia populated by ethnic Germans rather than start a war. The rest is history.

REASONING WITH CRAZY

Trump’s foray into madman theory could prove productive in some circumstances. His reputation for unpredictability, for instance, could have a deterring effect against both China and Russia, the United States’ most powerful adversaries. Beijing is rapidly expanding its nuclear forces, and Moscow is engaged in near-constant nuclear saber rattling. It is possible that both states hope that this behavior will deter the United States from intervening if China attacks Taiwan or Russia attacks a NATO state, lest Washington find itself in a nuclear conflict. But if Trump persuades the two countries that he might be prepared to do anything in response to their provocations, he could upend such calculations and stop invasions.

Yet Trump’s approach might also fail spectacularly. If Beijing believes that Trump will arbitrarily impose sanctions or other punishment on China, even if it respects U.S. interests, it will be more likely to challenge the United States. More worryingly, in an extreme scenario, if the United States finds itself engaged in nuclear brinkmanship with Beijing or Moscow, Trump’s reputation for unpredictability could increase fears of a U.S. nuclear first strike and thus potentially lead either government to launch a nuclear attack preemptively. And even if Trump’s madman approach makes some headway with China and Russia, it could undermine his dealings with weaker adversaries. Iran and North Korea, for example, will only cling more tightly to their nuclear programs if they believe that Trump might take action to overthrow their regimes. No matter how credible Trump’s threats are, both countries are unlikely to yield to U.S. threats without a correspondingly credible promise of long-term peace and security.

For all these reasons, Trump will have to demonstrate that his madness has limits. He will need to make it clear that his foreign policy is not totally devoid of reason and that he can be trusted to uphold a deal. Such an approach will not only increase the odds of compliance with Washington’s threats. It will also reduce the risk of inadvertent escalation, and thus reassure allies nervous about Trump’s policies.

Unfortunately, however, conveying exactly the right level of madness is very difficult. Many madmen—including Qaddafi and Saddam—developed reputations that ultimately proved detrimental, because their opponents came to believe that they would not abide by peace commitments. On the other hand, leaders like Khrushchev and Nixon may not have gone far enough, given that their adversaries doubted their willingness to use nuclear weapons. Trump might be able to succeed if he can portray himself as unpredictable and unrestrained without seeming unhinged. But if Trump comes off as hopelessly irrational, he is unlikely to get what he seeks.