Risks of Launch-on-Warning of U.S. Ballistic Missiles
Lead author: Frank von Hippel, January 2020
U.S. Strategic Command maintains virtually all its intercontinental ballistic missiles (ICBMs) and about half of its submarine-launched ballistic missiles (SLBMs) at sea on alert so that they can be launched within minutes on warning of an incoming nuclear attack. The purpose of this option is to be able to get the ICBMs launched before they can be destroyed and, more generally, to launch a U.S. counterstrike while the U.S. nuclear command, control and communication (NC3) system is intact.
Launch on warning is controversial, however, because the warning may be false due to an equipment failure, human error or a hack into the warning system. Also, keeping the missiles launch ready increases the danger that they could be launched by mistake, malfeasance or hacking. This is, of course, a hugely important issue because mistaken launch by either the United States or Russia could result in a global disaster involving a billion deaths.
Responding a warning would have to be almost by thoughtless reflex. The longest flight time of a ballistic missile attacking the United States from Russia or China would be about 30 minutes and the flight time from an offshore ballistic-missile submarine could be as little as ten minutes. Subtracting the times required to determine the trajectories and to implement any decision for response before the estimated arrival time would leave essentially minutes at most for consideration and debate.
There were four documented cases of false warnings of incoming Soviet attacks on the U.S. during 1979 and 1980 and two known cases of false warnings in the Soviet Union and Russia in 1983 and 1995, respectively. We don’t know how many false warnings have not been publicly reported.
The launch-on-warning option has been debated within the U.S. government since the Kennedy Administration but apparently was adopted by the 1970s. Because of his concern about accidental nuclear war, President Reagan wanted to negotiate an agreement with the Soviet Union to eliminate ballistic missiles in favor of bombers. Although not without risk (see the classic film, Dr. Strangelove), nuclear bombers can be launched on warning because their launch is reversible.
Both Presidents G. W. Bush and Obama came into office proposing to take the ICBMs off “hair trigger.” Once in office President Obama pursued the issue but retreated in the face of opposition from Strategic Command and Russian disinterest in doing so reciprocally.
In 2018, Congress requested that the Department of Defense (DOD) “contract with a federally funded research and development center to conduct a study on the potential benefits and risks of options to increase the time the President has to make a decision regarding the employment of nuclear weapons." The DOD selected the Institute for Defense Analyses (IDA), which produced a report that appears to in substantial part reflect Strategic Command’s thinking on the subject.
The contribution of launch on warning to deterrence
The report acknowledges the danger of mistaken launch and suggested ideas to reduce it but argues against de-alerting the ICBMs, citing the importance of the option to deterrence of a Russia attack:
“If Russia were confident that it will be able to destroy U.S. ICBMs on the ground, along with fixed bomber, fighter-bomber, and SSBN bases, it would be more optimistic about its chances of leaving the United States with a small nuclear force after the attack and no choice but to capitulate. Second, the capability to launch ICBMs promptly adds an element of speed and unpredictability that makes adversaries less self-assured that they can favorably manage escalation. Before conducting a large-scale nuclear strike on the United States, Russia would likely require confidence that it could convince U.S. leaders to back down, rather than comprehensively retaliate against Russia. However, if Russia knows that the U.S. president has minutes to contemplate ordering a nuclear strike, potentially with hundreds of nuclear weapons, the possibility of an escalatory U.S. response would make it less likely to risk a disarming first strike.”
The idea that Russia would be tempted to mount an attack on U.S. nuclear forces in the belief that the United States might have “no choice but to capitulate” because would have left with only 10 ballistic missile submarines at sea carrying about 800 warheads with 100-500 kiloton yields (4-12 Hiroshimas each) reflects thinking that McGeorge Bundy, President Kennedy’s national security advisor, criticized in his 1969 article, “To Cap the Volcano”:
“There is an enormous gulf between what political leaders really think about nuclear weapons and what is assumed in complex calculations of relative ‘advantage’ in simulated strategic warfare. Think-tank analysts can set levels of ‘acceptable’ damage well up in the tens of millions of lives. They can assume that the loss of dozens of great cities is somehow a real choice for sane men. They are in an unreal world. In the real world of real political leaders-whether here or in the Soviet Union-a decision that would bring even one hydrogen bomb on one city of one's own country would be recognized in advance as a catastrophic blunder; ten bombs on ten cities would be a disaster beyond history; and a hundred bombs on a hundred cities are unthinkable.”
The connection between launch-on-warning and counterforce
In an interview in 1998, four years after he retired as the first commander of Strategic Command, General George Lee Butler warned that, if it was believed that an incoming attack had been detected, the briefing that would be given by Strategic Command to an inevitably inadequately prepared President on her/his options would be strongly biased toward launch on warning.
General Butler indicated that Strategic Command’s plans for a response to a major attack required launching its alert ballistic missile warheads. The ballistic-missile submarines that were not on station on alert could constitute a strategic reserve as a deterrent against further attack by either Russia or a third country. Russia apparently has its forces in a similar posture.
Butler said that, to assure that the President would agree with its priority, Strategic Command had “built a construct that powerfully biased the president’s decision process toward launch before the arrival of the first enemy warhead. And at that point, all the elements, all the nuances of limited response just went out the window. The consequences of deterrence built on massive arsenals made up of a triad of forces now simply ensured that neither nation would survive the ensuing holocaust.”
It would not only be the two countries exchanging missiles that would be devastated. The climate and economic effects of a large nuclear exchange between Russia and the US would turn the nuclear exchange into a global catastrophe. Large fractions of the populations of the non-target countries would probably starve with the survivors taking decades to rebuild some semblance of modern civilization. The horrors of an US-Russian nuclear war would dwarf those World War II.
Somehow, the imperative of not allowing a Russian first strike to reduce the effectiveness of its retaliation has resulted in Strategic Command not treating the danger of a mistaken launch with similar seriousness. The fact that U.S. ballistic missiles are not equipped with systems to allow their destruction in flight indicates the overwhelming priority that Strategic Command has given to ensuring retaliation in case of attack.
De-alerting ballistic missiles
The only instance we are aware of when some U.S. ICBMs were de-alerted was in 1991 when the Soviet Union was beginning to collapse. President George H. W. Bush ordered the de-alerting of 450 U.S. Minuteman II missiles that were to be retired under the START Treaty and President Gorbachev stated that Russia would reciprocate.
In a 1998 meeting with a group that included the author, General Eugene Habiger, then Commander in Chief of Strategic Command, asserted that Russia had not fulfilled Gorbachev’s de-alerting commitment, rejected the idea of an accidental launch, and described how demoralizing it had been for the young officers keeping watch in the Minuteman II launch-control facilities to have their launch keys taken away from them. Four years later, however, in 2002, after he had retired, General Habiger, testifying in favor of the Strategic Offensive Reductions Treaty urged that “as Senator Nunn has said, we have to find a way to move more weapons off alert status and give leaders more decision time…" And: “I recommend that the President begin by order an immediate stand-down of the nuclear forces we plan to reduce under the treaty. This includes four Ohio class Trident submarines and all 50 Peacekeepers [the 10-warhead MX ICBMs]. This would advance our own security and help build confidence in our intentions.”
Apparently, General Habiger was more free in retirement to think for himself. Also, he may have had the illusion that nothing bad could happen on his watch while he was not so sure about his successors.
Ex-generals can be helpful to civilian governments confronting military orthodoxies. President G. H. W. Bush’s national security advisor, General Brent Scowcroft, may have made it possible for the president to overcome Strategic Command’s resistance to de-alerting at the end of the Cold War – at least de-alerting of the missiles that were soon to be scrapped.
De-alerting was rejected in the Trump Administration’s Nuclear Posture Review, which repeated an objection made in the Obama Administration nuclear posture review:
“The de-alerting of U.S. ICBMs would create the potential for dangerous deterrence instabilities by rendering them vulnerable to a potential first strike and compelling the United States to rush to re-alert in a crisis or conflict ...”
One might question that a “dangerous deterrence instability” might be caused by 400 U.S. warheads being vulnerable to a massive attack that would deplete Russia’s strategic forces more than those of the US when 800 or more U.S. warheads would be invulnerable in eight to ten U.S. ballistic missile submarines hidden in the vastnesses of the North Atlantic and North Pacific oceans. Would a Russian leader be willing to accept retaliation by 800 U.S. warheads but be deterred by the possibility of launch under attack of an additional 400?
According to the logic of Strategic Command and its Russian counterparts, however, if nuclear war seemed inevitable, destroying as much as possible of the enemy’s forces before they could be used might save some lives and infrastructure that could be used to rebuild after a nuclear holocaust. This consideration supports eliminating the vulnerable ICBMs altogether rather than de-alerting them.
The vulnerability of nuclear comand-and-control
Even if the US eliminated its ICBMs and moved their warheads onto its ballistic missile submarines, however, the vulnerability of the U.S. nuclear command, control, communication (NC3) system would still result in pressure for launch on warning.
This is the concern that the U.S. nuclear command could be “decapitated” by attacks on the White House the nuclear command posts in the Pentagon and elsewhere, and on key communication links through which launch commands would be transmitted. The attacker might thereby be able to avoid U.S. retaliation.
The IDA report warns, “If a president feared that the NC3 system was fragile, he or she might feel pressure to order nuclear strikes before the system degrades, particularly in a situation where a president thought that he or she might not survive.”
The IDA report argues that preserving the option of launch on warning deters Russia from exploiting the vulnerability of the U.S. NC3 system,
The idea that the United States would be disarmed by such a first strike again ignores the U.S. ballistic missile submarines at sea. Even in peacetime, the submarines have done patrols as long as 140 days. This would provide ample time for them to connect with a reconstituted U.S. leadership. Also, U.S. ballistic missile submarines carry their launch codes with them locked in safes for which unlock codes would have to be transmitted from a central command post. They are likely to have equipment on board, however, that could be used cut open their safes if the crews agreed to do so. It would be foolhardy for Russia to assume that there would be no U.S. retaliation after a decapitation attack.
Nevertheless, the vulnerability of command and control would be destabilizing in a crisis so severe that nuclear war seemed like a serious possibility. This is why Russia has invested in its “Dead Hand” system and why U.S. analysts who worry about accidental nuclear war urge greater investment in U.S. nuclear command and control.
Conclusions and recommendations
A major source of Strategic Command’s objection to dealerting continues to be that described by General Butler: not wanting to lose warheads that are key parts of its nuclear strike plan. That can be mitigated by eliminating U.S. ICBMs.
A second source is the vulnerability of the U.S. nuclear command, control and communication system to attack. That system can be made more robust but, if the attack were real, many of the officials who might have been involved in a launch on warning might be dead.
The most important issue, however, is preventing a nuclear war in the first place and a launch on warning posture may be a significant contributor to the probability of such a catastrophe happening. The increased danger introduced by the option of launch on warning far exceeds any perceived increase in deterrence that it provides. Thus, we recommend that the US eliminate entirely launch on warning as an option in nuclear war decision-making.
Also key is the probability of a false warning being believed and acted upon rather than being disbelieved and enough time being taken to find the cause. Belief or disbelief would be a function of the relationship between Russia and the United States. If they demonize each other as they did during the most tense parts of the Cold War, they are more likely to believe that the other country would be willing to launch a first strike. This is why Gorbachev and Reagan agreeing and repeating the mantra at their summits, “A nuclear war cannot be won and must never be fought” was so important for tension reduction.
But tension reduction requires more than summit statements. It requires engagement at lower levels as well. Arms-control negotiations provides one venue for such engagement. The United States had such negotiations almost continually throughout the Cold War. Unfortunately, the 2002 decision by the George W. Bush Administration to scrap the ABM Treaty, which limited Russian and U.S. ballistic-missile defenses has made new agreements limiting offensive nuclear weapons more difficult. Indeed, it has stimulated Russia to develop new types of offensive weapons and China to build up its strategic nuclear arsenal.
Also, at the end of the Cold War, the heads of the U.S. and Russian nuclear weapons laboratories engaged with each other as colleagues as did the heads of the two strategic commands. This has stopped, due in part to NATO expansion and the fact that the United States stopped considering Russia as an equal and in part due the paranoia of the Putin Administration and its destructive interference in the affairs of the United States and other democracies. Regardless of their adversary relationship, however, neither country wants a nuclear war and they must continue to prioritize mutual reassurance in their nuclear relationship.