“As to dangers arising from an irrational American president, the best protection is not to elect one.” – General Maxwell D. Taylor, letter to author, March 14, 1976
Amid current controversies about military command disobedience, little has been focused on the single most urgent question: “What if America’s designated military commanders receive an order from President Trump to fire nuclear weapons against a particular adversary”? This is no longer a fictional or contrived scenario. Though nothing scientific could be estimated about pertinent probabilities (such estimations must be based on the frequency of relevant past events), it remains conceivable that this president could act precipitously and/or irrationally.
As an academic specializing in such matters for over half a century, I first began to wonder about nuclear command decisions as a generic rather than US-specific problem. More precisely, while working on an early “nuclear book” back in the mid-1970s, I reached out to retired General Maxwell D. Taylor. On 14 March 1976, in response to my query regarding presidential debilities and unlawful orders, the former Chairman of the US Joint Chiefs of Staff offered some authoritative answers. Most noteworthy in his handwritten letter (attached hereto) was a warning about presidential irrationality. In the closing paragraph, Taylor cautioned, such irrationality would best be dealt with during the election process. Prima facie, a potentially unstable presidential candidate who somehow managed to get elected could pose an unprecedented and “apocalyptic” problem.[1]
What does this have to do with current political disagreements on disobeying unlawful military orders? What is the current US governing situation regarding this overriding national security issue? Here are some well-researched (more than fifty years) answers.
Structural protections against unauthorized firings are built into any presidential order concerning nuclear weapons. These protections include multiple and substantial redundancies. But virtually all such mutually-reinforcing safeguards – including “psychological reliability” assessments – become operative only at lower (sub-presidential) nuclear command levels. These safeguards do not apply to the Commander-in-Chief, that is, to the elected President of the United States or to any cabinet-level official in the assigned chain of command.
All derivative meanings are straightforward. In both law and practice, there exist no readily decipherable grounds to determine the legality of a presidential order to use nuclear weapons. Inevitably, the codified military obligation to disobey an unlawful order would be shrouded in uncertainties. While certain senior personnel in the chain of command could at some point choose to invoke “Nuremberg Obligations” (i.e., obligations to disobey an unlawful order), any such time-urgent invocation would almost surely be rejected.
There is more. If an American president operating within the bewildering chaos of his own making should issue an irrational or seemingly irrational nuclear command, the only way for the Secretary of Defense, the Chairman of the Joint Chiefs, the National Security Adviser and several possible others to obstruct this order would be sorely problematic. Under the best circumstances, informal safeguards might manage to work for a time, but accepting a “best case scenario” on such existential issues would hardly represent a reasonable approach to US nuclear decision-making. It follows, inter alia, that Americans ought promptly to inquire about predictable and promising institutional impediments to a debilitated US president.
On such unique matters, the US is navigating in uncharted waters. While President Kennedy engaged in personal nuclear brinkmanship with the Soviet Union’s Nikita Khrushchev in October 1962, he allegedly calculated the odds of a nuclear war as “between one out of three and even.” This seemingly precise calculation, corroborated both by JFK biographer Theodore Sorensen and by my own later private conversations with former JCS Chair Admiral Arleigh Burke (my colleague and roommate at the Naval Academy’s Foreign Affairs Conference of 1977) suggests that President Kennedy was either irrational in imposing his Cuban “quarantine” or acting out untested principles of “pretended irrationality.” In any event, JFK operated with the assistance of serious and capable strategic advisors; he did not propose to transform the Department of Defense into the Department of War.
Plausibly, the most urgent threat of a mistaken or irrational U.S. presidential order to use nuclear weapons would flow not from any “bolt-from-the-blue” nuclear attack – whether Russian, North Korean, or Chinese – but from an uncontrollable escalatory process. Back in 1962, Soviet premier Nikita Khrushchev “blinked” early on in the “game,” thereby preventing mutual and irrecoverable nuclear harms. Now, however, virtually any escalatory initiatives undertaken by President Trump could express unstable decision-making processes.
Soon, Donald Trump should understand the grave risks of being locked into an escalatory dynamic from which there could be no choice apart from abject capitulation or nuclear war. Though this American president might be well advised to seek “escalation dominance” in selected crisis negotiations, he would simultaneously need to avoid any catastrophic miscalculations.
Whether we like it or not, and at one time or another, nuclear strategy is a bewildering game that US President Donald Trump will have to play. To best ensure that this president’s required strategic moves will be rational, thoughtful and cumulatively cost-effective, it could first become necessary to enhance the formal decisional authority of his most senior military subordinates. At the same time, Trump’s Secretary of Defense, Chairman of the Joint Chiefs of Staff, National Security Advisor and one or two others in nuclear command positions could become an integral part of the problem rather than the solution. Plausibly, relying on Pete Hegseth and Stephen Miller to save the United States from an irrational presidential decision would be irrational.
Currently, arguments about disobeying unlawful orders (that is, arguments between President Trump, Senator Kelly and others) are not directly related to matters of nuclear command authority. Still, the obligation to disobey unlawful orders is authoritatively rooted in national and international law, and could at some point affect pre-nuclear crises. Classical legal writers firmly rejected the plea of “superior orders” as a defense against the charge of war crimes. The German Code of Military Law operative during World War II provided that every soldier must execute all orders undeterred by the fear of legal consequences, but that this obligation would not excuse him in cases where he “must have known” that the order was illegal.
This view was upheld by a landmark 1921 decision of the German Supreme Court in Leipzig. Any subordinate who obeyed the order of a superior officer was liable to punishment if it was known to him that such an order contravened international law. Later, the defense of “superior orders” was rejected at the Einsatzgruppen Trial conducted by American military tribunal. According to the tribunal: “The obedience of a soldier is not the obedience of an automaton. A soldier is a reasoning agent. It is a fallacy of widespread consumption that a soldier is required to do everything his superior officers order him to do. The subordinate is bound only to obey the lawful orders of his superior.”
All this being said, an American presidential order to fire nuclear weapons could represent the ultimate test of law-commanded disobedience. Exactly how such a scenario would play out is logically beyond any pre-crisis prediction or understanding. Nonetheless, to ignore this narrative’s critically-underlying issues could express the ultimate failure of US citizen responsibility.
[1] The title of my related book was Apocalypse: Nuclear Catastrophe in World Politics (The University of Chicago Press, 1980).
