We live in an era of presidential governance because of congressional abdication and delegation. Across every conceivable issue area, Congress has given the executive branch broad powers to take various actions and to make regulations and issue executive orders with the force of law. Moreover, presidents have tended to interpret these conferrals of authority aggressively, thus expanding presidential power even more. The issue obviously did not originate with Donald Trump. But Trump exercised power, or threatened to, in ways that have amplified the issue and, in some instances, have led to bipartisan calls for reform.
The chapters in this part focus only selectively on this very large problem and, in particular, on four problems that became especially salient during the Trump presidency:
There is a perennial debate in the United States about the legality and wisdom of presidential unilateralism in the deployment of U.S. military force abroad. On this issue perhaps more than any in this book, the most dangerous expansions of presidential power took place prior to Trump. However, the significance of and hazards in this expansion—especially with regard to presidential control over nuclear weapons—became crystal clear under Trump due to his undisciplined rhetoric, mercurial judgment, and indifference to facts and expert judgment. Trump, in short, was the lit match that made clear the dangerousness of the badly packed explosives in the presidential woodhouse.
Background
As background, we first summarize the legal doctrine governing the president’s use of force generally and then turn to the law, policies, protocols, and past practices related to nuclear weapons.
General Legal Principles Governing War Powers
Article II designates the president as the “Commander in Chief of the Army and Navy of the United States.” Alexander Hamilton stated in Federalist No. 69 that this designation “would amount to nothing more than the supreme command and direction of the military and naval forces, as first general and admiral of the Confederacy,” and he noted that this power fell far short of the analogous power in the king of England, who had the power to declare war and raise and regulate the army and navy. [498] Article I of the Constitution allocated these latter powers, and many others related to the conduct and control of the military, to Congress.
The Framers believed that the Constitution required presidents to receive authorization from Congress before they could use significant military force abroad, though it remains unclear whether this requirement was grounded in the Declare War Clause, or in Congress’s control of the purse, or both. The Framers also believed that presidents would retain the authority, without congressional authorization, to “repel sudden attacks,” a principle that the Supreme Court recognized in 1863. [499] This understanding of the allocation of war powers proved unstable as the United States grew in global power and as Congress approved a larger and larger standing Army and Navy (and later Air Force).
Through presidential practice, the “repel sudden attacks” rationale morphed into a broad “self-defense” rationale. Presidents on dozens of occasions unilaterally deployed military forces abroad to protect U.S. persons, property, and interests from ever-more-distant threats outside the United States. The broadest self-defense (and related) rationales came in two Office of Legal Counsel (OLC) opinions written in connection with the 9/11 attacks and the 2003 invasion of Iraq.
The first opinion was issued a few weeks after the 9/11 attacks. In it OLC collected and read expansively all the significant precedents on unilateral presidential uses of force. [500] It ruled that independent of Congress’s statutory authorization for the president to use force, the president under Article II can use force “both to retaliate for those attacks, and to prevent and deter future assaults on the Nation.” The president, OLC reasoned, was not limited to uses of force against the terrorist organizations that attacked the United States. Instead, it concluded, “the Constitution vests the President with the power to strike terrorist groups or organizations that cannot be demonstrably linked to the September 11 incidents, but that, nonetheless, pose a similar threat to the security of the United States and the lives of its people, whether at home or overseas.” The analysis in the opinion was directed at threats from terrorist organizations, but its logic applied to threats from states as well.
OLC’s second opinion was issued on October 16, 2002. It concluded that Article II authorized the president to use force against Iraq if he concluded that “Iraq’s development of WMD [weapons of mass destruction] might endanger our national security because of the risk that such weapons either would be targeted against the United States, or would be used to destabilize the region.” [501] OLC added that a preemptive use of military force against Iraq’s supposed WMD capabilities would be consistent with the U.N. Charter. The factors OLC considered were “the probability of an attack; the likelihood that this probability will increase, and therefore the need to take advantage of a window of opportunity; whether diplomatic alternatives are practical; and the magnitude of the harm that could result from the threat.”
Applying these factors, OLC concluded that “even if the probability that Iraq itself would attack the United States with WMD … were relatively low, the exceptionally high degree of harm that would result, combined with a limited window of opportunity and the likelihood that if we do not use force, the threat will increase, could lead the President to conclude that military action is necessary to defend the United States.” OLC added that “[a] proportionate response might include destruction of Iraq’s WMD capability or removing Saddam Hussein from power.”
Both of these Bush-era opinions were technically unnecessary to the matters at hand since Congress had previously authorized military force concerning the issues for which OLC claimed there was independent Article II authority. Nonetheless, although more than a dozen Bush-era OLC opinions were withdrawn, these two remain on the Justice Department books. And although the Obama administration never purported to rely on either opinion, it developed its own broad conception of the president’s authority to use military force in self-defense against terrorist organizations, including a broad notion of anticipatory self-defense, under both Article II and international law. [502]
Presidents have also used force without congressional authorization in circumstances that implicate “national interests” other than self-defense. This trend culminated in the pre-Trump years in two opinions written during the Obama administration, one concerning the military force deployed in Libya, and the other supporting military force, largely for humanitarian reasons, in Iraq. [503] The opinions recognized a number of broad national interests that would justify presidential unilateralism, such as promoting regional stability, assisting allies or strategic partners, and preventing a worsening of humanitarian atrocities. OLC recognized only one possible limit on this permissive “national interest” test for presidential unilateralism. If the planned military engagement is “prolonged and substantial,” it involves “exposure of U.S. military personnel to significant risk over a substantial period,” and the scale of the expected hostilities are large, then the action constitutes “war” that requires prior congressional authorization.
These opinions mean that, without congressional authorization, the president can use U.S. military force in any situation where he or she determines that doing so would serve the “national interest,” loosely conceived, as long as the force does not pose a great risk of American casualties or significant escalation. In addition, the Obama administration read the 1973 War Powers Resolution not to apply to the Libya conflict after that conflict exceeded the statute’s sixty-day limitation on “hostilities” not supported by congressional authorization. Many prior presidents had claimed that the resolution’s sixty-day trigger was unconstitutional. The Obama administration did not, but it instead read the “hostilities” trigger not to apply to war from a distance that did not pose a serious risk of harm to U.S. troops or a serious risk of escalation. On these understandings, the president can conduct “war from a distance” (manned and unmanned air power, long-range-missiles, and cyberattacks) without congressional authorization and outside the strictures of the War Powers Resolution.
These permissive understandings of the president’s lawful unilateral presidential authority to use force represent only the executive branch’s view and are not shared by many in Congress or by many scholars, including scholars of international law. Nonetheless, this legal understanding is highly relevant to any account of presidential war powers because presidential uses of force are not generally subject to judicial review, and thus the executive branch view of the law will govern the actual use of force absent some binding intervention by Congress that the executive branch accepts.
Against this legal background, the following controversies over unilateral presidential action related to conventional forces arose in the decades prior to the Trump presidency:
● The Bush administration embraced the unprecedentedly broad conceptions of unilateral military power described in the 2001 and 2002 OLC opinions.
● The Bush administration unilaterally expanded Congress’s 2001 Authorization for Use of Military Force (AUMF), which was aimed at the perpetrators of the 9/11 attacks, to include all manner of al-Qaeda-associated forces. The Obama administration interpreted the AUMF to include the Islamic State, which is not an associated force of al-Qaeda. These interpretations allowed presidents to use significant military force in almost ten nations against many terrorist groups that were not in existence on 9/11.
● The Obama administration, building on past practice, engaged in humanitarian interventions in Iraq and Libya without congressional authorization. These actions did not concern the safety or property of U.S persons or the defense of allies. And they stretched the “national interest” test for unilateral war powers to its furthest point yet.
● The Bush and Obama administrations stretched the “self-defense” rationale for unilateral war powers to include broader and broader conceptions of self-defense, including a generous multifactored test for anticipatory self-defense, and a broad notion of “unit self-protection” that expands unilateral presidential authority since U.S. troops are deployed in many war zones around the world.
● Presidents have on their own authority involved the United States in various forms of military and intelligence support for other nations’ wars. A prominent recent example is the United States’ support since 2015 of a coalition of nations led by Saudi Arabia, in the war in Yemen.
There are three reasons why the nation’s military policies, as reflected in the president’s decision on when and how to use military force, came to rest almost entirely on the president’s discretion alone. First, Congress provided the president with a two-million-strong armed force (1.2 million service members on active duty) and an astonishing array of weapons systems. Whatever practical constraints inhered in a small standing army dissipated long ago, and unilateral presidential uses of forces have in general risen with the growth in the permanent military’s size.
Second, the presidency grew in size and responsibility, especially after the Civil War. As the nation also grew in power and responsibility on the international stage, presidents perceived ever-widening national interests that required deployments of U.S. troops.
Third, Congress failed to exercise its constitutional authorities to constrain presidents from using the troops and weapons it provided. As early as 1804, the Supreme Court upheld congressional restrictions on the circumstances in which the U.S. Navy could seize enemy ships during the Quasi-War with France. [504] Congress can also check or shape presidential uses of force after the fact by prohibiting or limiting spending on continuing uses of force. In 1983 and 1993, Congress forced presidents to change missions (in Lebanon and Somalia, respectively) through the use of conditional funding restrictions. Congress’s funding restrictions were also a major contribution to the end of the Vietnam War. But Congress has constrained the presidency very little in military affairs in recent decades.
One reason Congress has difficulty exercising its prodigious war powers is that after a president deploys force unilaterally, the political pressure to “rally around the troops” makes it hard for many members to oppose the action. Second, Congress must muster veto-proof majorities to reverse or check unilateral presidential uses of force. Third, and most importantly, most members of Congress do not want to take public responsibility for war powers decisions. They prefer to remain quiet so that they can take credit or allocate blame, depending on how things work out. Consider the nineteen-year war that Congress ostensibly authorized in the 2001 AUMF and that presidents have persistently expanded through interpretation. Many members of Congress have complained about these unilateral presidential extensions of the statute. And there have been numerous calls (including, notably, by President Obama during his term) for Congress to update and refresh the 2001 AUMF to take greater responsibility for how the conflict has developed. But Congress has failed to act, at least in a public way, although it has by large majorities quietly enacted appropriations that aim precisely to support the conflict against the Islamic State.
Congress’s major effort at reform of presidential war powers—the 1973 War Powers Resolution—has been a failure. The resolution requires the president to cease unilateral military actions after sixty days in any situation where U.S. forces are engaged in hostilities or likely hostilities, unless Congress authorizes the mission. It has been a failure for a number of reasons. First, most presidents have not accepted its constitutionality. Second, the executive branch has interpreted the sixty-day window before it needed to secure congressional authorization as an effective authorization to use force within that window. Third, as noted earlier, presidents have interpreted the “hostilities” trigger for requiring congressional authorization opportunistically to cover, at most, only extended engagements of U.S. troops that pose a serious risk of U.S. casualties and escalation. Fourth, Congress has acquiesced in almost every context where the executive branch has emasculated the War Powers Resolution through interpretation or defiance.
Nuclear Weapons
Nuclear weapons present a special case of the general war powers problem. They have a unique capacity to kill many millions in a single detonation, and their use, even tactically, threatens to spark an existential escalatory spiral and to create long-term and widespread damage to health and the environment. Because of their uniquely devastating power, specific law, policies, and protocols have developed to govern their use.
Law
A 2013 Department of Defense report accurately states the conventional wisdom about the president and nuclear weapons: “Consistent with decades-long practice, the President, as Commander in Chief of the U.S. Armed Forces, has the sole authority to order the employment of U.S. nuclear forces .” [505] This sentence needs unpacking. The main idea is civilian control: The president alone has authority to order deployment of nuclear weapons. A related idea, more contested, is that the president alone is currently authorized to deploy nuclear weapons.
There is no published Justice Department opinion that explains how OLC jurisprudence on Article II war powers applies to nuclear weapons. It seems clear that the president has the authority to launch nuclear weapons in self-defense in the face of an incoming nuclear attack—a much-planned-for scenario since the early days of the Cold War. Matters become more complex if the president orders a first-use nuclear strike in a situation short of an incoming nuclear attack. Some observers have reasoned that the OLC’s “national interest” opinions require congressional authorization for the use of nuclear weapons in this situation, since any such nuclear attack would be on a scale that implicated prolonged and substantial hostilities and the exposure of U.S. troops (and civilians) to significant risk. But two factors complicate this analysis.
First, the “national interest” opinions written in contexts other than self-defense (such as the recent interventions in Syria and Libya) may not be relevant when the president deploys military force in self-defense , including anticipatory self-defense. The president’s unilateral military power is at its height when he or she exercises force to protect the nation. And as noted, OLC currently takes a very broad view of the president’s authority to use force in self-defense, including anticipatory self-defense. There are a number of plausible scenarios short of an incoming nuclear attack—including verbal threats to use nuclear weapons, and intelligence indicating preparation to use such weapons—that might suffice under extant OLC opinions to authorize the president to use force. North Korea, which undoubtedly has a nuclear capacity, has taken many threatening actions toward the United States, and uttered many more threatening words. Based on the OLC opinions written to justify the invasion of Iraq, and other broad self-defense opinions, the Justice Department could conclude that Article II would authorize the president to deploy nuclear weapons necessary and proportional to the threat to eliminate Kim Jong Un’s nuclear threats against the United States.
Second, there is a weak argument that Congress has already authorized the president to use nuclear weapons at his or her discretion. The Atomic Energy Act (AEA) of 1946 (amended in 1954) provides that the president “may direct” the Atomic Energy Commission “to deliver such quantities of fissionable materials or weapons to the armed forces for such use as he deems necessary in the interest of national defense. ” [506] This language almost certainly aimed to secure civilian control over all applications of nuclear technology, and not to authorize presidential use of nuclear weapons against an adversary. [507] But it could be read as an authorization to use force, and such an authorization would have been consistent with the widespread understanding at the time that the use of nuclear weapons was a question of presidential discretion. [508] Finally, a 1989 OLC opinion cites this passage when it states in dicta that “[t]he President is authorized by the Act to require the Commission to deliver nuclear material and to authorize its use for military purposes.” [509]
A final legal uncertainty concerns international law. In the 2002 opinion sketched earlier, OLC interpreted jus ad bellum principles—ones derived from the U.N. Charter and related customary international law—to allow the president to use conventional air, naval, and ground power in anticipatory self-defense, conceived very broadly. However, OLC has never opined on how the U.N. Charter applies to the use of nuclear weapons, and there is considerable uncertainty on this point. The proportionality component of the anticipatory self-defense test would almost certainly operate differently, and with greater restriction, in the context of using nuclear weapons.
Moreover, there is another body of international law—the law governing the conduct of war, known as the law of war, or jus in bello —that governs the legality of targeting. The Justice Department has not said much about how this body of law applies to presidential uses of force. But the Defense Department takes this body of law seriously. It has made clear that “[t]he law of war governs the use of nuclear weapons, just as it governs the use of conventional weapons”; that “nuclear weapons must be directed against military objectives,” not civilian targets; and that “attacks using nuclear weapons must not be conducted when the expected incidental harm to civilians is excessive compared to the military advantage expected to be gained.” [510]
Nuclear Declaratory Posture
The Nuclear Posture Review (NPR) is a document prepared by the Defense Department that assesses and states U.S. policy and posture on nuclear weapons. (It is notable that an executive branch agency and not Congress prepares this review.) Modern NPRs began in 1994, during the Clinton administration, to assess the United States’ nuclear posture after the end of the Cold War. The George W. Bush, Obama, and Trump administrations also promulgated nuclear posture reviews. (The Bush and Obama NPRs were mandated by Congress.) Among other things, NPRs aim to examine and explain the role of nuclear weapons in U.S. security and to tell the American public, allies, and adversaries about the U.S. nuclear deterrent posture and the circumstances in which the United States will deploy nuclear weapons.
Of most relevance to legal reform are the NPR’s stated circumstances under which the United States will deploy nuclear weapons. President Obama was famously committed to reducing U.S. nuclear stockpiles and U.S. reliance on the use of nuclear weapons. The key elements of his declaratory policy were the following:
● The United States will continue to strengthen conventional capabilities and reduce the role of nuclear weapons in deterring non-nuclear attacks, with the objective of making deterrence of nuclear attack on the United States or our allies and partners the sole purpose of U.S. nuclear weapons.
● The United States would only consider the use of nuclear weapons in extreme circumstances to defend the vital interests of the United States or its allies and partners.
● The United States will not use or threaten to use nuclear weapons against non-nuclear weapons states that are party to the Nuclear Non-Proliferation Treaty (NPT) and in compliance with their nuclear nonproliferation obligations. [511]
This policy maintained the option of using nuclear weapons in a first-use situation. It also maintained the possibility of using nuclear weapons to deter “non-nuclear attacks,” and declined to spell out the circumstances under which the United States might do so. But it was noteworthy for expressing the aims of eventually eliminating this role for nuclear weapons and limiting their use to deterring nuclear attacks. The Obama NPR also continued U.S. ambiguity on what “vital interests” of the United States would justify in extreme circumstances the use of nuclear weapons. President Trump’s NPR declaratory policy was somewhat different:
The United States would only consider the employment of nuclear weapons in extreme circumstances to defend the vital interests of the United States, its allies, and partners. Extreme circumstances could include significant non-nuclear strategic attacks. Significant non-nuclear strategic attacks include, but are not limited to, attacks on the U.S., allied, or partner civilian population or infrastructure, and attacks on U.S. or allied nuclear forces, their command and control, or warning and attack assessment capabilities. The United States will not use or threaten to use nuclear weapons against non-nuclear weapons states that are party to the NPT and in compliance with their nuclear non-proliferation obligations. Given the potential of significant non-nuclear strategic attacks, the United States reserves the right to make any adjustment in the assurance that may be warranted by the evolution and proliferation of non-nuclear strategic attack technologies and U.S. capabilities to counter that threat. [512]
The Trump NPR maintains the option of first-use and is explicit about it: “To help preserve deterrence and the assurance of allies and partners, the United States has never adopted a ‘no first use’ policy and, given the contemporary threat environment, such a policy is not justified today.” It is also explicit about the strategic ambiguity that has historically characterized U.S. nuclear policy: “It remains the policy of the United States to retain some ambiguity regarding the precise circumstances that might lead to a U.S. nuclear response.” The Trump NPR departs in other notable ways from the Obama NPR. One is that the Trump NPR spells out the types of “significant non-nuclear strategic attacks” for which the United States is prepared to use nuclear weapons. The Trump NPR also walks back somewhat the pledge not to threaten or use nuclear weapons against NPT parties in light of the development of “non-nuclear strategic attack technologies,” including, presumably, cyberattacks.
Process
A central concern during the Cold War was that the U.S. government needed to be able to retaliate in the face of a Soviet first-strike before the incoming Soviet missiles destroyed U.S. land-based missiles. To respond to such a first-strike, and also to deter it, the United States needed a credible way to respond during the approximately twenty-five to thirty minutes it would take from launch for a Soviet missile to reach the United States.
The executive branch thus developed an elaborate nuclear command-and-control system—much of it classified—that consists of people, procedures, capabilities, communication systems, and protocols to ensure that the president can make the decision to use nuclear weapons quickly in a variety of circumstances, and also with assurance that it is the president who issues the order. Simplifying a great deal, the system works as follows. The president is accompanied at all times by military aides who carry the Emergency Satchel (or “football”), which contains strike options and a system of secure communications. The order is communicated from the president to the National Military Command Center at the Pentagon, which authenticates the message and then communicates the order securely, through Strategic Command, to the entity launching the intercontinental or submarine-launched ballistic missile.
The system places a premium on authentication and speed. It is not designed to check unreasonable, irrational, or impulsive presidential commands. The existence of any such check would be antithetical to the quick and definitive decision-making that is necessary to meet the incoming attack, and that is central to credible nuclear deterrence. (It is an open question, discussed below, whether and to what extent such decision-making speed is needed in situations short of an incoming nuclear attack.)
There is, however, one potential check in the system that has risen in prominence during the Trump era: legality. Military personnel in the chain of command have a duty, on pain of court martial or termination, to comply with a lawful presidential order. But the military subordinate—be it a combatant commander or the captain of a nuclear-armed submarine—also has a duty not to obey an illegal order. The decision lies in the subordinate’s discretion, and that decision’s validity can be fully adjudicated only after the fact in a court martial. In the meantime, the president can fire a uniformed officer who fails to carry out an order that the officer believed was unlawful. And the president can keep firing people until he or she finds someone to carry out the command. The president remains in control as a legal matter. But people in the chain of command can use deliberations and objections related to legality as a basis for slowing the implementation of an exceedingly unreasonable presidential order related to nuclear weapons. [513]
There are at least four drawbacks to relying on legality as a possible basis for slowing presidential decision-making related to the use of nuclear weapons. First, subordinate deliberation about legality is antithetical to the needed speed of attack. Second, the subordinate might be wrong about illegality in a situation in which the president has good reason to order the deployment of nuclear weapons. Third, empowering or encouraging those in the chain of command to make independent legal judgments about presidential orders to use nuclear weapons could hinder the effectiveness and credibility of U.S. nuclear forces—especially since many areas of international law related to the use of nuclear weapons are unclear or contested. Fourth, not every unreasonable presidential order to use nuclear weapons would be plausibly illegal. As noted earlier, an order to fire tactical nuclear weapons at military targets in North Korea in response to a serious provocation would not obviously be illegal under legal positions that currently prevail in the executive branch.
An unstable president issuing an exceedingly unreasonable order to use nuclear weapons can be slowed in another way. There are many human beings in the loop in the chain of command between the president who gives the order to fire a nuclear weapon and the men or women who launch the weapon, and any one of them, starting with the secretary of defense or the head of Strategic Command, could simply act in an insubordinate manner, on risk of termination or court martial. In 1974, for example, President Nixon’s secretary of defense, James Schlesinger, reportedly ordered the chairman of the Joint Chiefs of Staff not to execute the president’s military orders without checking with him, or (in a different version) ordered military commanders to “check with either him or Secretary of State Henry Kissinger before executing” a nuclear launch order from Nixon. (The historical record on this point remains contested.) [514]
Past Presidents
Presidents prior to Trump often spoke about the dangers of nuclear weapons but rarely spoke publicly about using or threatening to use nuclear weapons.
On November 30, 1950, in the midst of the Korean War, President Harry S. Truman had the following exchange with a journalist during a press conference:
The President: We will take whatever steps are necessary to meet the military situation, just as we always have.
Q: Will that include the atomic bomb?
The President: That includes every weapon that we have.
Q: Mr. President, you said “every weapon that we have.” Does that mean that there is active consideration of the use of the atomic bomb?
The President: There has always been active consideration of its use. I don’t want to see it used. It is a terrible weapon, and it should not be used on innocent men, women, and children who have nothing whatever to do with this military aggression. That happens when it is used.
Afterward, the White House issued a press release that stated that while “[c]onsideration of the use of any weapon is always implicit in the very possession of that weapon, … only the President can authorize the use of the atom bomb, and no such authorization has been given.” [515]
In a 1954 press conference, President Dwight D. Eisenhower spoke implicitly about using nuclear weapons:
[T]here are weapons now in being that give more than ever to the attacker a tremendous advantage, the man who attacks by surprise. The element of surprise, always important in war, has been multiplied by the possibility of creating such widespread destruction quickly. Therefore, any president should be worse than impeached, he should be hanged, I should say, if he didn’t do what all America would demand that he do to protect them in an emergency. [516]
In an effort to end the Korean War, Ike also communicated privately to the Chinese and North Korean governments in May 1953 that he was authorized and prepared to use strategic and tactical weapons. Ike and his subordinates also threatened to deploy nuclear weapons in a dispute with the People’s Republic of China over Taiwan. Indeed, Secretary of State John Foster Dulles gave a speech entitled “The Strategy of Massive Retaliation” that publicized the Eisenhower administration policy that “atomic weapons would be used at the outset of a general war.” [517]
President John F. Kennedy often spoke about the threat or danger of nuclear weapons, including the possibility of their use “by accident or miscalculation or by madness.” [518] On October 22, 1962, at the height of the Cuban Missile Crisis, Kennedy stated to the nation:
Nuclear weapons are so destructive and ballistic missiles are so swift, that any substantially increased possibility of their use or any sudden change in their deployment may well be regarded as a definite threat to peace …. We will not prematurely or unnecessarily risk the costs of worldwide nuclear war in which even the fruits of victory would be ashes in our mouth—but neither will we shrink from that risk at any time it must be faced …. It shall be the policy of this Nation to regard any nuclear missile launched from Cuba against any nation in the Western Hemisphere as an attack by the Soviet Union on the United States, requiring a full retaliatory response upon the Soviet Union. [519]
President Richard Nixon never publicly threatened to use nuclear weapons. He once stated in a meeting with members of Congress that “I can go in my office and pick up a telephone, and in 25 minutes, millions of people will be dead.” [520] And in the summer and fall of 1969, he and National Security Adviser Henry Kissinger did self-consciously deploy a “madman” theory against the Soviet Union that implicitly relied on that possibility. They let it be known to the North Vietnamese government that they would “take measures of the greatest consequence” if the North Vietnamese did not take steps toward peace. “We’ll just slip the word to them that ‘For God’s sake, you know Nixon is obsessed about Communism,’” Nixon told White House Chief of Staff H. R. Haldeman, explaining the theory. “‘We can’t restrain him when he’s angry—and he has his hand on the nuclear button.’” Nixon also ordered his secretary of defense to conduct a series of alert measures for U.S. strategic forces as part of the bluff.
None of the other presidents in the nuclear age—Johnson, Ford, Carter, Reagan, Bush, Clinton, Bush, or Obama—ever overtly threatened to use nuclear weapons, at least outside the official U.S. statement of its nuclear declaratory policy. (President George W. Bush came close when he was asked if his claim that “all options are on the table” regarding Iran implied the possibility of a nuclear strike. Bush reiterated, “All options are on the table.” [521] ) These presidents made general statements on the danger of the weapons, on the need for non-proliferation, and sometimes on the need for nuclear weapons reduction.
The Trump Era
Against the baseline of Obama administration practices, the Trump administration ramped up the use of drones and loosened the rules of engagement in the fight against the Islamic State. This resulted in geographic victory over that terrorist organization, though the threat clearly persists. Trump twice used force against Syria for its use of chemical weapons, but the OLC opinion in support of the second strike was basically in line with legal precedents for humanitarian intervention set by Trump’s predecessors. Trump’s targeted killing of Iranian Gen. Qassem Soleimani was high stakes and dangerous in the sense that it easily could have sparked a much broader conflict with Iran (though as of this writing, it has not). The killing probably did not exceed the generous precedents established by prior presidents, though Trump’s failure to vet the strike fully, the high risk of the strike, and the executive branch’s confused public explanation for its legality led to widespread criticism and concern. Trump continued U.S. military and intelligence support for a coalition of nations, led by Saudi Arabia, in the war in Yemen. He also tried, but had only middling success, in reducing the numbers of U.S. troops in Afghanistan, Syria, and Iraq.
The Trump administration has not developed novel legal theories on the use of conventional forces. But Trump’s ad hoc decision-making process, his apparent indifference to facts, and his disrespect for senior military figures, whom he reportedly attacked as “losers” and “a bunch of dopes and babies,” caused concern about the scale of unilateral presidential control over the use of military force. [522] Congress tried to push back against some of Trump’s military decisions. It twice enacted legislation, vetoed by the president, to block billions of dollars in sales of arms to Saudi Arabia in connection with its war with Yemen. And after the strike against Soleimani, Congress passed a statute that required the president to seek congressional authorization before further uses of force against Iran. President Trump vetoed this law as well, on May 6, 2020.
Trump caused most concern, however, when he threatened to use nuclear weapons. Most notably, between January 2017 and July 2018, he made a number of menacing statements toward North Korean leader Kim Jong Un. During the first half of 2017, Trump issued numerous veiled or abstract threats to halt or reverse North Korea’s nuclear weapons capability. [523] On August 8, 2017, Trump pledged that North Korean threats “will be met with fire and fury like the world has never seen.” [524] On August 10, he said that if Kim attacks Guam, “it will be an event the likes of which nobody’s seen before, what will happen in North Korea.” [525] A few days earlier, Trump’s national security advisor, H. R. McMaster, implied that the United States was prepared to launch a preventive strike against North Korea in the face of its nuclear threats. On September 19, Trump first referred to Kim as “Rocket Man” in an address to the U.N. General Assembly in New York City. [526] The vague threats continued throughout the fall, as did threats by North Korea to use force against the United States.
In his speech to the United Nations on September 19, Trump also stated: “The United States has great strength and patience, but if it is forced to defend itself or its allies, we will have no choice but to totally destroy North Korea.” Addressing the South Korean National Assembly on November 7, he said: “The longer we wait, the greater the danger grows, and the fewer the options become.” [527] And on January 2, 2018, in response to Kim’s claim to have a nuclear button on his desk, Trump tweeted: “I too have a Nuclear Button, but it is a much bigger & more powerful one than his, and my Button works!” [528]
Many observers inside the United States, and presumably abroad as well, interpreted these words “to mean that the President is actively considering the use of nuclear weapons in order to deal with the threat of North Korea,” as Senator Ben Cardin put it in November 2017. [529] The Trump administration’s statements did little to assuage Cardin’s and others’ concerns: Trump’s Nuclear Posture Review called for the development of “low-yield [nuclear] options,” [530] and then–Secretary of Defense James Mattis testified in response to questioning from Senator Ed Markey that “he could imagine the president ordering a first strike” nuclear attack. [531]
These statements, combined with Trump’s apparent impulsiveness, his indifference to (or perhaps weak grasp of) military and legal advice, and his frequent inversion of traditional U.S. interests abroad on a number of fronts, led to several proposals in Congress to check the president’s control over nuclear weapons. The proposals tended not to question the president’s authority to respond to an incoming nuclear attack, where the need for reactive speed and his constitutional authority are at their height. Instead, the proposals focus on the president’s ability to use nuclear weapons in a first-use context.
The most prominent bill to restrict the president’s first-strike nuclear authorities came from Rep. Ted Lieu and Senator Markey. Their bill proposed to declare it “the policy of the United States that no first-use nuclear strike should be conducted absent a declaration of war by Congress.” [532] It then provided that “[n]otwithstanding any other provision of law, the President may not use the Armed Forces of the United States to conduct a first-use nuclear strike unless such strike is conducted pursuant to a declaration of war by Congress that expressly authorizes such strike.”
Reform: Conventional Forces
Trump’s tenure has sparked a renewed national debate about why the president possesses unilateral authority to deploy nuclear weapons without any express restraint. Trump has not obviously extended the bounds of presidential war powers with respect to conventional forces, though he has continued to exercise the broad unilateral powers staked out by his predecessors. But in Trump’s hands, these powers have for many observers seemed more worrisome than ever. We propose reforms in both contexts, starting with nonnuclear issues.
The president has too much unchecked discretion to use military force to start new conflicts or to significantly expand extant conflicts beyond originally stated objectives or geographic limits. Such broad discretion to use military force is not necessary to protect vital national interests, and in many contexts the exercise of this discretion has harmed the national interest. One person should not have plenary power to decide for the nation when and against which countries to go to war. As we explain in this section, we believe that the president should possess the authority to use military force without congressional authorization in self-defense and related scenarios. But there is no normative justification for placing essentially unchecked offensive military power in the discretion of a single person.
There is no way to reform the president’s unilateral military authority without a dramatic change in congressional responsibility and engagement to the point where veto-proof supermajorities of Congress are interested in major reform. (This is what happened in 1973 when Congress enacted the War Powers Resolution over Richard Nixon’s veto.) We do not know when, if ever, that will happen. But consistent with the aims of this book, here we set out several discrete reforms, from least to most ambitious, that Congress (and in some instances the executive branch) should consider and that together we believe would help restore the constitutional balance.
Withdraw Prior Office of Legal Counsel Opinions
The executive branch should withdraw the 2001 and 2002 OLC opinions discussed earlier. These opinions take historically, and in our view excessively, broad views of the president’s authority to use force in anticipatory self-defense. Neither opinion was written to authorize an actual use of force, since Congress had previously authorized the conflicts in question. Both were in a real sense dicta, and remarkably overbroad and under-analyzed dicta that in effect say that there are no legal limitations at all on even large-scale presidential uses of force. We believe that these opinions can be withdrawn without affecting the president’s needed self-defensive powers. The opinions are loaded guns that should be unloaded and stored safely. They should, in short, be abrogated.
Repeal the 2002 Authorization for Use of Military Force
In 2002 Congress authorized the president “to use the Armed Forces of the United States as he determines to be necessary and appropriate in order to ... defend the national security of the United States against the continuing threat posed by Iraq.” [533] The purpose of this authorization was, as the preambular language made clear, to meet the threat posed by Iraq’s past aggression, international law violations, and supposed threat of weapons of mass destruction. But the language of the authorization sweeps more broadly and has been construed imaginatively in a number of circumstances as authorization to use force that was unrelated to the threat actually posed by the government of Iraq. For example, the Obama administration invoked it as one basis for using force against the Islamic State in Iraq. And the Trump administration used it in 2019 as a basis for killing Soleimani in Iraq.
These uses of force over a decade (and in one case almost two decades) after the 2002 AUMF was enacted, for reasons entirely removed from what Congress had in mind, show that it is an undisciplined authorization for any needed use of force related to Iraq. Congress should repeal the law to make clear that presidential uses of force in and related to Iraq have to be justified by either the 2001 AUMF or the president’s Article II self-defense powers. This likely will not have a material impact on the president’s authorities in the area, since every use of force in which the 2002 AUMF was invoked likely could have been authorized independently by some combination of the 2001 AUMF and Article II—bases of authority for which we propose reforms below. But the statute has outlived its purpose and should be repealed as part of tidying up presidential military authorities after Trump.
Repeal and Update the 2001 Authorization for Use of Military Force
Congress must become more involved in the sprawling two-decade war that began on 9/11. There appears to be a general consensus in the executive branch (through action and interpretation) and Congress (through acquiescence and appropriation) that the president can use military force against al-Qaeda, the Taliban, the Islamic State, and associates. But Congress should repeal the 2001 AUMF and enact a new one in order to make this authorization express in accordance with the following principles:
Specify the Enemy
The 2001 AUMF was drafted before it was clear who was responsible for the 9/11 attacks, and so the enemy was described in vague terms as the “nations, organizations, or persons [the president] determines planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001, or harbored such organizations or persons.” [534] This lack of specificity is one factor that has allowed presidents to expand the AUMF through interpretation. Congress can and should be more specific. The replacement AUMF should name the enemies (at the moment, al-Qaeda, the Taliban, and the Islamic State), including all current associated forces (such as al-Shabab) against whom Congress wishes to authorize force.
Update Associated Forces
Congress needs to provide a rigorous mechanism for updating the list of associated forces against which the president can use force. Terrorist organizational boundaries are fluid, and related forces that are not formally or obviously part of the groups against whom force is authorized can be excluded from a too-rigid definition of “associated forces.” The worry, however, is that presidential administrations have interpreted the phrase “associated forces”—a phrase not found within the 2001 AUMF but that has been added through interpretive gloss—without transparency and on occasion opportunistically. The “associated forces” extension of the 2001 AUMF is easily subject to manipulation and abuse, and has been the basis for expanding a war far beyond what Congress formally authorized, and sometimes outside of public (or even congressional) knowledge.
One solution to this problem is for Congress to expressly authorize force against associated forces but to define the term and require the president to notify Congress and the public when new groups are added to the list. The definition that three presidents have used for associated forces, and that we think is appropriate, is as follows: An associated force is one that “has entered the fight alongside” and “is a co-belligerent with” al-Qaeda, the Taliban, or the Islamic State “in hostilities against the United States or its coalition partners.” [535]
As for notice, we draw on an idea in the AUMF that Senators Corker, Kaine, Flake, Coons, and Nelson proposed in 2018. [536] That bill authorizes the president to identify new groups of associated forces, but it also requires him or her to notify Congress of the designation within thirty days. It then provides for an expedited review mechanism by Congress and guaranteed congressional debate on whether the designation should be repealed. To be sure, repeal would be hard for Congress to accomplish since it would need to overcome a presidential veto. One could thus argue for the opposite presumption: The president should have to return to Congress and get affirmative approval before adding any terrorist organization to the list. We think ex post notice, transparency, and debate would discipline the president’s addition of associated forces and hold him or her accountable for those extensions, and we doubt that Congress would ever embrace responsibility for fine-grained approval of associated forces in the first instance. But reasonable minds can differ.
Sunset Provision
Presidents have interpreted and stretched the AUMF for two decades to meet the changing terrorist threat. Congress has failed to revisit the 2001 AUMF with regard to use-of-force issues during this period, despite these many presidential changes. In light of the irresponsibility and lack of discipline this failure reflects, we think it is vital that, when Congress finally musters the capacity to revisit the 2001 AUMF, it establishes a forcing mechanism to ensure its future engagement and review.
The forcing mechanism we propose is a two- to three-year sunset clause. A sunset clause is a provision that in advance terminates, or “sunsets,” the authority in question and thus requires its affirmative renewal. A sunset in a new AUMF would commit Congress and the president to revisit the nature and scope of the war against Islamic terrorists on a regular basis. It would require the president to publicly explain and defend the nature of the conflict and the reasons why it must continue (and how). It would force Congress to exercise its constitutional and democratic responsibilities to deliberate about and vote on (or at least face) the issue. And it would prevent the situation that has prevailed for two decades: a sprawling, lengthy, morphing, long-term war without congressional involvement beyond appropriations.
A sunset clause in an authorization to use military force poses risks. The main one is that Congress will not renew the authorization, or not renew it to the president’s liking, thus forcing the president to rely on his or her less-certain and politically riskier Article II authority to continue the conflict, or to narrow the conflict further than he or she would like, and perhaps end it. We believe it is unlikely that Congress will refuse to renew authority if the president makes the case forcefully that it is necessary to meet a real threat to U.S. national security. If history is any guide, political pressures tend to push Congress toward rather than away from authorizing force that a president requests. And in any event, the nation should not shy away from the possible consequences of democratic deliberation on a vital question like war.
Some critics of sunset clauses have claimed that they send a signal to the enemy of weakness or a lack of resolve. We disagree. The constitutional requirement that congressional military appropriations last no longer than two years does not signal that Congress or the nation lacks the resolve to continue a fight. All that a sunset signals, and means, is that Congress must exercise its constitutional responsibilities and reconvene to assess and update the president’s authorities to use force every few years in light of new conditions.
An alternative to a sunset clause is the forcing mechanism in the Corker-Kaine bill. The senators proposed to require Congress to have a floor debate on a regular basis (every four years) about the continued validity of the AUMF but do not make continued authority under the AUMF turn on affirmative approval. This is better than the current situation because Congress would at least have to deliberate about the shape of the conflict as a body every four years. But it is far short of the type of engagement that Congress should have in an ongoing war, and so in our view it does not suffice.
Revise or Replace the War Powers Resolution
As noted earlier, the 1973 War Powers Resolution, Congress’s last and indeed only effort to assert its constitutional responsibilities related to war in a general way, has been a failure. The fundamental reason for its failure is that Congress eschewed responsibility for direct oversight of presidential uses of force, and instead tried to create an acontextual, self-enforcing legal mechanism that, without discrete congressional engagement, required a president to cease using force after sixty (or, with an extension, ninety) days.
We believe that Congress must go beyond such a passive mechanism by placing affirmative limits on presidential unilateralism and by enforcing those limits with appropriations restrictions. Abstracting away from many important details, we believe that any such revision should embrace the following general principles:
Define and Limit Inherent Presidential Power
C ongress should define the circumstances in which the president can use force under Article II without congressional authorization and limit his or her authority to those circumstances. Trump and other presidents may disagree, but it is important for Congress to put down a clear marker. We propose, based roughly on two bills from then-Senator Biden in the 1990s, the following authorized bases for unilateral presidential use of force, which are basically variations on the self-defense theme:
In the absence of a declaration of war or statutory authorization for a specific use of force, the President, through powers vested by the Constitution of the United States, is authorized to use force abroad only in the following circumstances—
(1) to repel an attack upon the United States or its armed forces;
(2) to respond to a foreign military threat that severely and directly jeopardizes the vital national interests of the United States under emergency conditions that do not permit sufficient time for Congress to consider statutory authorization or a declaration of war;
(3) to extricate citizens and nationals of the United States located abroad from situations involving a direct and imminent threat to their lives;
(4) to forestall an imminent act of international terrorism directed at citizens or nationals of the United States or to retaliate against the perpetrators of a specific act of international terrorism directed at such citizens or nationals; and
(5) to protect internationally recognized rights of innocent and free passage in the air and on the seas in circumstances where the violation, or threat of violation, of such rights poses a substantial danger to the safety of American citizens or the national security of the United States. [537]
Prohibit Humanitarian Intervention Without Congressional Authorization
Humanitarian intervention is the constitutionally least-justified basis for the president to use the U.S. military, since by definition it is not designed to assist or protect U.S. persons or property and thus is furthest away from the president’s core power to use military force in self-defense. For the same reason, humanitarian intervention is also the ground on which congressional control is constitutionally most sound. Even though a morally powerful case for humanitarian intervention can be made in particular circumstances, it is a vast expansion of the president’s unilateral power to deploy force, it often makes the situation on the ground worse, and it always has unforeseen consequences.
The availability of this form of unilateral presidential action also opens the door to an administration’s assertion of humanitarian aims as rhetorical cover for pursuit of other objectives. Even in the most straightforward cases, this kind of intervention can begin with the most honorable intentions and end in a major and sustained military presence overseas. While humanitarian intervention is implicitly prohibited by the preceding principles, Congress should expressly rule it out. We do not oppose humanitarian intervention in all cases and do not believe the American people do either. But should the president determine such an intervention is required as an expression of American values and standing on the world stage, he or she should take such action only after making the case to Congress and receiving explicit authorization.
Consultation
In situations (1) and (3)–(5) above, where the president is authorized to use force, but not in situations where an emergency exists that does not permit sufficient time to consult Congress (such as in situation (2)), the president must inform and seek the advice of a specially identified group of members of Congress before any use of force abroad.
Reporting
Not later than forty-eight hours after commencing a use of force abroad, the president shall submit to the speaker of the House of Representatives and to the president pro tempore of the Senate a report stating (1) the objective of such use of force and (2) in the absence of a declaration of war or specific statutory authorization for such use of force, the specific sections of this law that authorize the use of force.
Sixty-Day Requirement
The president may not under any circumstances continue the use of force beyond sixty days unless one of the following three circumstances exists: (1) Congress has declared war or provided specific statutory authorization for the use of force abroad beyond such period; (2) the president has requested that Congress enact a joint resolution constituting a declaration of war or statutory authorization but such joint resolution has not been subject to a vote in each House of Congress; or (3) the president has determined and certified to the speaker of the House of Representatives and the president pro tempore of the Senate that an emergency exists that threatens the supreme national interests of the United States and requires the president to exceed such period of limitation.
Congress should be required to debate the appropriateness of time limitations on any authorizations to use military force in the face of presidential requests. And if the president certifies that an emergency exists and extends the deployment beyond the original sixty days, he or she must within this additional sixty days seek and receive congressional authorization to continue using force. The request will be subject to expedited consideration, and the use of force would be terminated if Congress then failed to authorize it. We think it unlikely that Congress would not approve force in a situation that threatens supreme national interests, and in any event, in such a context it is unclear whether the termination would be constitutional in the face of the president’s core Article II self-defense powers.
Funding Restrictions
No funds made available under any provision of law may be obligated or expended for any use of force abroad inconsistent with the provisions of this act.
Reform: Nuclear Weapons
The United States may not suffer the misfortune of another president as unstable, mercurial, and indifferent to facts and informed advice as Trump. But as noted in the introduction to this book, reforms of the presidency must take into account the serious possibility of demagogic, norm-breaking presidents as far as the eye can see. This brand of politics does not necessarily entail erratic speech and action related to the most destructive weapons on the planet. But it may involve a dangerous blend of inexperience, aggressiveness, and theatrical bravado that is far from ideal in leaders with nuclear weaponry at their command. If future presidents are much less likely than Trump to possess his aberrational attitude toward nuclear weapons, this has implications for reform because, as we explain in this section, placing checks on a president’s ability to authorize the use of nuclear weapons can have adverse implications for the effectiveness of U.S. nuclear deterrence. To the extent that future presidents are likely to possess Trumpian volatility, the costs to deterrence of checking an impulsive and irrational president may be warranted by the need to prevent unjustified nuclear war.
There are three basic scenarios to consider in the context of reforming presidential authority to launch nuclear weapons.
First is the classic Cold War worry: a response to an incoming nuclear attack. This scenario may seem less likely now than during the Cold War, but it is not. Russia still possesses a huge array of nuclear weapons and often threatens to deploy them in a first-use, North Korea is pursuing the capacity to threaten the U.S. homeland with nuclear weapons, China will soon have ballistic missile submarines, and Iran is attempting to develop nuclear weapons and ballistic missile capabilities. [538] Nuclear threats clearly remain. And since they are diffused among several countries with various and changing interests, they might be more worrisome today than in the Cold War, which was characterized by bilateral and relatively stable and predictable threats.
For these reasons, it remains vital that the president be able to quickly, effectively, and unilaterally authorize the deployment of nuclear weapons in response to an incoming attack. Such unfettered authority is needed to respond to the attack itself but also, more importantly, to establish credible deterrence of any such attacks in the first place. This scenario is not one in which an unstable or irrational president is waking up the military to launch a difficult-to-understand attack in the face of no incoming threat. It is the opposite—here the military is waking up the president and asking for permission to respond to an incoming attack. In this situation both the functional need for exclusive presidential authority and the constitutional arguments for unilateral presidential power are at their height. It is thus no surprise that no one in Congress has seriously proposed to eliminate presidential authority in this context.
Scenarios two and three involve variations of first-use. The second is at the other extreme from a response to an incoming nuclear attack: The president orders a nuclear attack that is outside U.S. declaratory policy and that seems to reasonable observers to be outlandish and inappropriate. For example, Trump is well known to have an obsessive hatred of Ukraine and its government. Imagine that, in a fit of pique over some nonthreatening military operation or troop movement by Ukraine, Trump orders the use of nuclear weapons against that country. Such an order would not come (as declaratory policy requires) in “extreme circumstances” and would not address “the vital interests of the United States, its allies, and partners.” The attack would also likely exceed presidential authority under even the permissive OLC opinions.
This is the scenario in which the case for reform is obviously strongest. The nation should not have to rely on the courage and systemic noncooperation of military officials to prevent or slow an entirely illegal and unjustified nuclear war. And yet a congressional prohibition on the use of force in just this circumstance is definitionally tricky. Later in this section we propose a statute that would outlaw the use of nuclear force in this and other circumstances. Note that a congressional prohibition on the use of nuclear weapons in even these extreme circumstances still depends on executive branch bureaucrats for its efficacy. If a president is unstable or irrational enough to order deployment of nuclear weapons in the imagined circumstances, he or she likely won’t be dissuaded by a legal prohibition on the weapons’ use. But perhaps such a prohibition would embolden actors in the chain of command to resist.
Scenario three falls in the broad spectrum of possibilities between the extremes of scenarios one and two. In this scenario, there is at least a plausible strategic or legal case for the use of nuclear weapons, and the use is not obviously ruled out by U.S. declaratory policy, but it would be controversial nonetheless. This scenario is also characterized by time for deliberation—many hours and possibly days or weeks. But these cases are also the hardest to regulate as a matter of substance. Consider two possible fact-patterns:
● There is a real threat—perhaps nuclear, perhaps conventional—that is growing but has not yet culminated in a launch or an attack. Perhaps Iran is moving masses of troops in a threatening way that portends a huge conventional attack on U.S. or allied civilians. Or perhaps the United States has unambiguous intelligence that North Korea plans to launch a nuclear attack on South Korea next week. In response, a president orders the use of nuclear weapons on the adversary’s military assets to prevent the threatened attack.
● A president orders a nuclear attack on North Korea in response to one of Kim’s verbal provocations but in the absence of concrete intelligence of plans for a North Korean attack on U.S. or allied vital interests. Such an order—against the background of North Korea’s nuclear build-up and many threats to use nuclear weapons against the United States—could conceivably be deemed lawful under the domestic law principles articulated by OLC. It might also be deemed lawful under OLC’s view of the U.N. Charter and might depend on whether the use of nuclear force was proportionate to the threat. And under jus in bello principles, the primary legal question under Defense Department doctrine would be whether the use of nuclear force was aimed at military targets and whether the expected harm to civilians was justified by the military advantage gained.
The first fact-pattern above presents a stronger case, legally and otherwise, for the use of nuclear weapons. But neither is obviously ruled out by U.S. nuclear declaratory policy. The United States has always declined to define the circumstances in which it would deploy nuclear weapons in a first-use against nuclear states.
Such strategic ambiguity about whether and when it would use nuclear weapons in first-use against nuclear states is thought to be central to the deterrence impact of nuclear weapons. At a 2017 hearing on the president’s authority to use nuclear weapons, Senator Marco Rubio posited that “calculated ambiguity” about the use of nuclear weapons was an important element of deterrence in the face of the Warsaw Pact’s conventional forces superiority, and might have deterred Saddam Hussein from using chemical and biological weapons in the First Gulf War. [539] He asked whether the concept was still relevant. Gen. Robert Kehler, the retired commander of Strategic Command, stated that calculated ambiguity remained crucial because it “enhances our deterrence to have some doubt in the mind of an adversary about under what conditions we would use a nuclear weapon.” Peter Feaver, who worked on nuclear policy in the Clinton and second Bush administrations, agreed. He noted that “President Obama, who was no fan of nuclear weapons …, nevertheless left in place calculated ambiguity in precisely these scenarios.” Brian McKeon, who was chief Democratic counsel on the Senate Foreign Relations Committee for a dozen years, and a senior official in the Obama administration in several positions related to nuclear policy, concurred.
Against this background, we turn to reform possibilities. First we assess the case for no reform. One might think that any constraint on the president in this context is an unnecessary overreaction. Trump is an unusually inexperienced, impatient, and temperamentally volatile president the likes of which we may never see again. Perhaps we should not tie his successor’s hands to meet a problem that may not recur, especially since doing so might weaken the U.S. strategic deterrence posture. The whole point of strategic ambiguity in the hands of the president is to achieve maximum deterrence and thereby to ensure that nuclear weapons are never used. On this view, the best check against a low-probability, deeply unreasonable action by a future president that at the same time preserves strategic ambiguity is the men and women in the chain of command who can invoke legal or other principles in resisting or delaying a seemingly irrational presidential order to use nuclear force.
Our answer to this fair challenge is that it requires only one president who is poorly equipped to assume these vast responsibilities to produce a catastrophic result. Once that president is elected, it is too late to impose reasonable restrictions. It requires the most optimistic of outlooks to imagine that all presidents in the future will have the capacity for rational deliberation and good judgment that is critical to the wise management of all elements of such an awesome responsibility. And we believe that prudent reforms to presidential control over the use of nuclear weapons can be achieved to temper the dangers of the worst-case scenarios without adversely affecting U.S. nuclear deterrence.
At least three major considerations inform possible reform, in the guise of legal limitations, on the president’s first-use nuclear authority: (1) what, precisely, the limitations should be, (2) what impact such limitations would have on strategic ambiguity and nuclear deterrence, and (3) how effective such legal limitations would be.
The case for more robust reform to constrain presidents’ uses of nuclear weapons can take many variations—too many to analyze comprehensively. One possibility we do not consider here is an absolute ban on first-use of nuclear weapons in any circumstances. This is a perfectly coherent position with many adherents. But at the moment, it lacks broad support in both Congress and the defense community, and it has profound implications for U.S. nuclear policy and strategic nuclear deterrence that go far beyond our present concerns.
Major reforms that pose less of a threat to traditional U.S. nuclear policy build on the fact that in every version of scenario three, there is time, and perhaps significant time, for deliberation before using nuclear weapons. In this context, the reform argument goes, the consequences to the nation and the world are too great for there not to be some form of legally required check on the president’s unilateral discretion. Many reform proposals require a sign-off within the executive branch. For example, Richard Betts and Matthew Waxman propose a requirement of certification from the secretary of defense that the president’s launch order is valid (that is, from the commander in chief) and from the attorney general that it is lawful. [540]
It is unclear whether this proposal is constitutional or why it would help check the president’s discretion in one of the hard scenario-three cases, especially since current executive branch understandings of the law are so permissive. Nor is it clear how much this check would help in scenario-two cases involving a mad president ordering an irrational launch of nuclear weapons. The secretary of defense, without or without certification, could, under the current arrangement, slow execution of the launch order by (among other means) consulting the attorney general on whether the launch is lawful. Of course, a mad president might have a defense secretary and attorney general under his or her control, or might fire them during the deliberation period in order to find other more pliant officials.
Another potential reform, one involving a truly adversarial institution, would require the president to consult with Congress and seek and receive congressional authorization for certain presidential uses of nuclear weapons. The strongest case for this check occurs when the Constitution demands authorization for the use of nuclear weapons. Unfortunately, the circumstances in which this is so are contested. Even the pro-executive OLC opinions, however, likely would not permit unilateral presidential authorization for using nuclear weapons in a fit of pique against Ukraine, as in the earlier example. Congress could define the circumstances in which it believes the Constitution permits unilateral action and the circumstances in which it requires congressional authorization prior to a use of nuclear weapons, and in the process could calibrate the check to minimize impact on U.S. deterrence. Especially if it did so in a way that limited the requirement to non-self-defense situations, this approach might raise few constitutional questions and have little impact on deterrence.
Here is our proposal to serve all of these goals, and to bar a president from ordering nuclear attacks in the exceedingly unreasonable context of scenario two. The basic idea is for Congress to enact a statute that acknowledges the president’s unilateral power to use nuclear weapons in ways that track the common ground in the Obama and Trump nuclear declaratory policies. Simplifying a great deal, the elements of such a statute would be as follows:
● Congress notes, consistent with the War Powers Resolution, that the president has the constitutional power to introduce “United States Armed Forces into hostilities, or into situations where imminent involvement in hostilities is clearly indicated by the circumstances” pursuant to “a national emergency created by attack upon the United States, its territories or possessions, or its armed forces. ”[541] This would acknowledge that the president has Article II authority to respond to an incoming nuclear attack.
● Congress further notes that the president’s constitutional power to defend against an attack on the United States, its territories or possessions, or its armed forces includes the power to use nuclear weapons in self-defense in extreme circumstances to defend the vital interests of the United States or its allies . This would acknowledge that the president’s constitutional power matches current U.S. declaratory policy without specifying the degree of imminence that is needed before using these weapons. This approach aims to impose limits on the president’s use of nuclear weapons with minimal damage to U.S. strategic ambiguity and deterrence. The self-defense standard here is open ended and for some would be controversial, but it is the current U.S. posture to which Congress has acquiesced to date.
● Congress specifies that only it can declare or authorize war in the constitutional sense and specifies further that use of nuclear weapons beyond a response to an incoming nuclear attack or in self-defense in extreme circumstances to defend the vital interests of the United States or its allies would constitute an act of war in the constitutional sense. In short, any use of nuclear weapons beyond the elements of U.S. nuclear declaratory policy common to Obama and Trump would be unlawful absent congressional authorization.
● Congress could impose a consultation requirement. It could, for example, declare it the policy of the United States that whenever the executive branch receives intelligence indicating extreme circumstances demanding the use of nuclear weapons in defense of the vital interests of the United States or its allies, it must consult with identified senior congressional leadership before authorizing the use of such weapons.
Even such a statute that aims to limit the uses of nuclear force to current declaratory policy might have an adverse impact on U.S. deterrence by slowing the circumstances under which the president can order a launch. On the plus side, this statute embeds U.S. declaratory policy in statutory form with congressional buy-in and authority, and it makes it clearly unlawful to use force outside that policy.
Of course, Congress can if it wishes try to limit presidential authority beyond current declaratory policy. It could, for example, limit presidential uses of nuclear forces to responses to actual or imminent attacks, and could define imminence to make the notion tighter than current executive branch understandings. Such an approach would raise harder constitutional questions about Congress’s power and would involve a much greater disruption to U.S. nuclear deterrence.
Finally, we suggest that Congress make clear that Section 6 of the Atomic Energy Act is not a blanket authorization to use nuclear weapons as the president sees fit. This will tamp down a tad on possibly unwarranted overconfidence in the executive branch’s legal authority to use nuclear weapons outside of self-defense scenarios.