Last weekend The Sunday Herald published allegations by a Royal Navy submariner, William McNeilly, of safety and security flaws involving British submarines which carry Trident nuclear missiles. These coincide with concerns recently raised at the Review Conference of the Nuclear Non-Proliferation Treaty (NPT) in New York about the ongoing risks of nuclear weapon use, whether deliberate or inadvertent. At this meeting, the UK acknowledged the terrible humanitarian consequences of nuclear weapon use in populated areas, but denied its Trident nuclear arsenal poses any appreciable risk. France also insisted that its nuclear arsenal poses no risk whatsoever. Such views should be questioned, for at least two reasons.
One: none of the nuclear-armed states are willing to subject their contemporary nuclear weapon safety and security records to open scrutiny. It is therefore not reasonable for them to insist that their safety claims should be simply accepted—a point McNeilly’s assertions tend to reinforce.
Two: recent research indicates that the kinds of problems McNeilly alleges are not only plausible, they are indicative of a wide range of issues in management of the nuclear arsenals of several states, including the UK—some of them very serious. In researching his recent book, Command and Control, investigative journalist Eric Schlosser found literally thousands of nuclear weapon-related safety incidents had occurred historically in the US alone, several of which almost resulted in nuclear detonations. The nuclear-weapon states argue, in effect, that this miracle indicates the effectiveness of their nuclear command and control systems. Common sense indicates quite the opposite. (Schlosser, incidentally, described Britain’s Trident as an accident waiting to happen.) Moreover, the nuclear-weapon states characterize nuclear terrorism as a major risk. In this respect, it’s interesting that McNeilly's report raises concerns about insider threat as a significant risk to existing nuclear arsenals—unauthorized personnel reaching classified information and technology.
The responsibility to secure their nuclear arsenals, facility sites and nuclear material lies squarely on the shoulders of the nuclear-weapon states. Yet a fundamental problem with nuclear weapons is that keeping them at high-alert and ready for use runs counter to the aim of keeping them as safe and secure as possible. Although improvements in physically managing nuclear weapons might reduce risk, factors like competing organizational agendas, biases, human frailty and the incomprehensibility of systems failures to their designers and operators mean that risk cannot be eliminated.
Most of the problems McNeilly details are breakdowns in routine systems and procedures. Individually these are highly unlikely to result in nuclear detonations. Yet these problems, which seem to have become routine, are still troubling. That’s because in complex tightly-coupled systems (which nuclear powered and armed submarines are), individual or component failures can cause complex interactions that are unfamiliar, unplanned or unexpected sequences not necessarily explicable or even visible to operators. For instance, McNeilly claimed that Royal Navy submarine crew would mute missile panel alarm systems because they sounded so often for reasons that could not be explained. If this is true, it mirrors what organizational experts such as Charles Perrow have observed about ‘normal’ or ‘system’ accidents of many kinds.
Of course, when missile alarm systems go off for unexplained reasons, trust in that system will also diminish in time. Moreover, in the event of an actual accident, there might be little time to react correctly due to ‘tight-coupling’. In 1980, a tool dropped by a US Air Force technician—a simple accident—hit the bottom of a silo, bounced, struck the side of a Titan II intercontinental ballistic missile, pierced the skin and caused a fuel leak. Despite a heroic effort to save the missile, it exploded, although fortunately the nuclear warhead didn’t detonate. British weapons may be more vulnerable to accidents: the Ministry of Defence revealed in 2007 that Trident warheads are not equipped with permissive action link technology—security devices to prevent unauthorized arming or detonation.
Research in 2014 by Chatham House found that the world has come Too Close for Comfort to nuclear-weapon use on several occasions. These crises occurred for the most prosaic of reasons, from a faulty computer chip leading to reports of a Soviet missile attack on the US (1980), a military exercise tape mistaken for reality (1979) and a Norwegian weather rocket mistaken for a NATO nuclear attack on Russia (1983). These were all ‘system’ accidents in which simple failures compounded and could have resulted in catastrophe. That they were not came down only to the mindful—and sometimes disobedient—actions of individuals who happened to be in the right place at the right time. The report added that incidents ‘similar to those that have happened in the past are likely to happen in the future’.
No such thing as a safe arsenal
So to those asking, yes, we should be very worried. McNeilly’s allegations are troubling and as yet unverified, yet line up plausibly with what we know historically has occurred in national nuclear weapon complexes: there is no such thing as an inherently safe nuclear arsenal. This includes the UK: one recent independent study reported 16 collisions and groundings involving British nuclear-powered submarines since 1979, and hundreds of submarine fires and other safety problems. Because the destructive power of even one Trident nuclear warhead (and British Trident submarines can carry up to 16 missiles with multiple warheads, each with a destructive yield up to the 100 kiloton range), a serious accident one day in port or a mis-launch resulting in detonation could have catastrophic humanitarian impacts and far-reaching geopolitical consequences.
The solution, some will claim, is to spend more money—on physical security, on training, psychological screening, even a new generation of Trident submarines. As systems accident theory makes clear, however, the potential for catastrophic accidents is endemic to complex tightly-coupled systems like Trident. The only way to ensure that nuclear weapon detonations don’t occur is ultimately to move toward getting rid of nuclear weapons, something to which British ministers and Whitehall officials haven’t been receptive to date. It also suggests that independent public investigation is needed to get to the heart of safety concerns about Britain’s Trident.
To comment on this article, please contact Chatham House Feedback