This is a copy of the statement Tom Collina, Director of Policy at Ploughshares Fund, delivered in testimony before the Senate Armed Services Committee (SASC) at a hearing on United States nuclear deterrence policy and strategy. His remarks, prepared jointly with former Secretary of Defense William J. Perry, focused on ending the current policies of sole presidential authority to use nuclear weapons, first use of nuclear weapons, maintaining intercontinental ballistic missiles (ICBMs) on high alert, and continuing the Ground-Based Strategic Deterrent program.
You can watch the hearing here.
We will take steps to reduce the role of nuclear weapons in our national security strategy, while ensuring our strategic deterrent remains safe, secure, and effective and that our extended deterrence commitments to our allies remain strong and credible.
Mr. Chairman, members of the committee, thank you for inviting me to speak about US nuclear policy. This prepared statement is on behalf of myself and former Secretary of Defense Bill Perry, who regrets he could not be here. I request permission to submit this statement for the record.
Many of the ideas I will talk about today are based on the book Dr. Perry and I co-wrote and released last year called The Button: The New Nuclear Arms Race and Presidential Power from Truman to Trump. The main conclusion of the book is that US nuclear policy is focused on the wrong threat. And by focusing on the wrong threat, we have adopted the wrong policy.
US nuclear policy has, for decades, been built on one central assumption: that Russia might launch a disarming first nuclear strike—a bolt from the blue—against the United States. Thus, the president has sole authority to launch hundreds of Intercontinental Ballistic Missiles (ICBMs) deployed on high alert. This strategy is expensive, requiring large numbers of deployed weapons. At the same time, Moscow makes the same assumption about Washington.
Looking back at the Cold War, we found no compelling evidence that either side would have launched a surprise attack. As former CIA Director and Defense Secretary Bob Gates wrote, “In fact, very few in Washington thought there was even a remote chance that the Soviets would suicidally throw the dice that way.” 2 STRATCOM Commander Adm. Charles Richard recently said that “A bolt out of the blue is unlikely.”
Yet by preparing for this unlikely threat, US policy creates a greater danger: the possibility that these forces could be used by accident, through a political or technical miscalculation. It is time to shift the focus of US nuclear policy to the greatest danger: blundering into nuclear war by mistake.
We strongly believe that the risk of an intentional attack was and is significantly smaller than the risk of blundering into a nuclear war. Throughout the Cold War and still today the two superpowers have been focused on the wrong threat. We have been undermining our own security.
A nuclear war by accident or mistake would be just as deadly as one by intent. The size and lethality of US and Russian forces ensures that a nuclear exchange—regardless of why it started—could result in the end of our civilization. Starting a nuclear war by mistake is the greatest existential risk to the United States today.
This is not just a theoretical possibility. We came, in fact, very close to blundering into nuclear catastrophe several times during the Cold War. And the advent of cyber threats to nuclear forces and warning systems only increases the risk of false alarms and mistakes.
A. Buying the President More Time
Preparing for a disarming first strike has led us to take on dangerous policies to speed up a launch decision, such as giving the president sole authority to launch, preserving the option to launch first, and keeping land-based ballistic missiles on high alert so they can be launched on warning of attack. These options increase pressure on the president to make a quick decision and thus make a nuclear blunder more likely.
The Biden administration can modify US nuclear policy to remove quick-launch options and to give the president more decision time, while maintaining the ability to deter and respond to an (highly unlikely) intentional attack. Here’s how:
1. End Sole Authority for First Use
First, the Biden administration should end sole authority for starting nuclear war. The last weeks of President Trump’s term in office demonstrated the extreme danger of giving one person unilateral authority over launch decisions. In a state of emotional turmoil, the president could have ordered the use of nuclear weapons. The danger was so acute that House Speaker Nancy Pelosi actively looked for ways to prevent the "unstable president from… accessing the launch codes and ordering a nuclear strike." We may like to think that the military would not follow such an order, but ultimately the stakes are too high for hope to be our only safeguard.
Sole authority no longer serves US interests for the simple reason that the main benefit of that policy—quick launch—is outweighed by the risk of accidental or mistaken launch. Moreover, sole authority is not necessary for deterrence which is assured by US nuclear-armed submarines at sea. There is no realistic threat of a bolt from the blue and we can maintain deterrence without sole authority. We do not need it.
The Biden administration can end sole authority for first use in two primary ways: the president can share that authority with Congress (either all of Congress or a subset), and/or the president can declare that the US would only use nuclear weapons in retaliation.
Such policies would provide clear directives for the military to follow: A launch could be legally ordered only if the nation had already been attacked with nuclear weapons or if Congress had approved the decision, providing a necessary constitutional check to executive power. Both would be infinitely less risky—to our nation and to the world—than our current doctrine. Presidents should only have legal authority to order the use of nuclear weapons in retaliation to a confirmed nuclear attack or with the approval of Congress.
2. Declare Sole Purpose
Second, the Biden administration should declare sole purpose. The United States should never initiate nuclear war, but only use these weapons to deter or respond to a nuclear attack against us or our allies. With US conventional superiority, we believe that no rational president would use nuclear weapons first, in any scenario. Against a nuclear-armed state like Russia or China, first use would invite a devastating retaliation. Against a nonnuclear state, first use would go against fifty years of US nonproliferation policy. How can we possibly hope to convince other states that they do not need nuclear weapons if the United States itself says it needs them for nonnuclear threats?
The Biden campaign stated that “the sole purpose of the US nuclear arsenal should be deterring—and if necessary, retaliating against—a nuclear attack. As president, [Biden] will work to put that belief into practice, in consultation with our allies and military.” We agree, and regardless of what this policy is called (sole purpose, no first use or something else), such a policy should: 1) clearly prohibit the US from starting a nuclear war; 2) specifically rule out preemptive nuclear attacks, which have a high risk of starting nuclear war by mistake and should not be considered under any circumstances; and 3) prohibit launching nuclear weapons on warning of attack, as such launches increase the risk of starting nuclear war in response to a false alarm.
Establishing a sole purpose policy will require consultations with allies, but they cannot have a veto over US policy. US allies need to be reassured that a policy of sole purpose does not undermine Washington’s commitment to their security.
3. Take ICBMs Off Alert/End Launch-on-Warning
Third, the Biden administration should take land-based missiles off alert. If early warning sensors indicate that enemy missiles are en route to the United States, the president would have to consider launching ICBMs before the enemy missiles arrive, known as “launch on warning.” Once ICBMs are launched, they cannot be recalled. The president would have less than 10 minutes to make that terrible decision.
In such a situation, a president would be under great pressure to “use them or lose them” and launch ICBMs before the attack can be confirmed. If the president orders a launch and the attack is a false alarm, the president would have started nuclear war by mistake.
False alarms have happened multiple times, and in an era of cyberattacks on US command-and-control systems, the danger has only grown. ICBMs on high alert are a nuclear catastrophe waiting to happen. We should take ICBMs off alert and end the policy of launch-on-warning.
B. Maintaining Deterrence for Less
As the Biden administration extends presidential decision time, it can also deter an intentional attack with a smaller and more affordable nuclear force than currently planned. Deterrence depends on a credible second-strike capability, which is provided by US nuclear-armed submarines at sea and backed up by nuclear-armed bombers. We can and should move to a sole purpose, deterrence-only nuclear posture. The United States does not need ICBMs to deter nuclear war.
The Biden administration can make a sole purpose policy more credible and further reduce the risk of accidental launch by retiring the ICBMs. ICBMs are most likely to be used first, in response to a false alarm; they are highly unlikely to ever be used in retaliation, as most would be destroyed in any Russian nuclear attack. Thus, ICBMs have no logical role in US nuclear policy and would have no place in a Biden administration that adheres to a sole purpose, deterrence-only policy.
The United States can move to a smaller, more secure second-strike nuclear force whose sole purpose is to deter nuclear attack. We do not need to spend hundreds of billions of dollars more in a dangerous and futile attempt to “prevail” in a nuclear conflict. Those funds should be spent on higher priority projects.
The Biden administration can cancel the Ground-Based Strategic Deterrent (GBSD) program and save much of the projected $264 billion lifetime cost. At a minimum, the GBSD program should be delayed while the administration explores new arms control negotiations with Russia. In the meantime, the administration can refurbish the existing Minuteman III missiles at a fraction of the cost of buying GBSD.
In the context of retiring the ICBMs, it would be prudent to increase US investments in submarine and bomber survivability and nuclear command and control. The president must have high confidence that the United States could launch a devastating retaliation after absorbing a (highly unlikely) first strike.
To conclude, these two principles—reducing the risk of accidental war and maintaining deterrence for less cost—translate into clear policy priorities for the Biden administration. To reduce the risk of blundering into war, the United States should seek to increase presidential decision time by ending the current policies of sole presidential authority, first use, and ICBMs on high alert. And to maintain deterrence for less, the United States should cancel the Ground-Based Strategic Deterrent.
By making these important policy shifts, the Biden administration can save hundreds of billions of dollars, reduce the risk of nuclear war, and still protect the United States and its allies.
1. Perry served as secretary of defense in the Clinton administration and Collina is director of policy at Ploughshares Fund. These recommendations are based on their book, The Button: The New Nuclear Arms Race and Presidential Power from Truman to Trump, published in 2020. https://www.ploughshares.org/thebutton
2. The Button, p. 27
Photo: Jefferson Memorial, Washington, DC. Reflection Of Trees In Water. Steven Heap-EyeEm/Getty Images