At your peril: deterrence is still key

Having a nuclear ‘deterrent’ is not the same as conducting ‘deterrence’. The former is the physical instrument, the latter the policy for its use. The essence of deterrence is to persuade another state (‘the deterree’) to refrain from taking actions that you (‘the deterer’) wish to avoid.

This can be done in two ways. Deterrence by ‘punishment’ does not try to prevent the action itself, but instead holds out the promise of a retaliation, the costs of which would outweigh any gains to be had from the original action contemplated.

Deterrence by ‘denial’, otherwise known as defence, demonstrates an ability to defeat the unwanted action itself and therefore the futility of undertaking it.

In the nuclear age only deterrence by punishment has been held to be effective, as a worthwhile defence could not be feasible on cost or technical grounds. When Britain had to deter a hostile, nuclear-armed superpower (the Soviet Union) it could do nothing to defeat a nuclear attack. Instead, the UK maintained an ability to cause unacceptable damage in retaliation, which led to the so-called ‘Moscow Criteria’ — the ability to devastate the Soviet capital.

Today the bipolar Cold War is but a memory. In the ‘Second Nuclear Age’ nuclear weapons have become less numerous, but more widespread. The UK’s very survival is not threatened in the way that it was during the Cold War, but could be again if a major strategic threat were to arise in a future ‘Third Nuclear Age’. Instead, we face a less threatening but more diverse and uncertain world, one that is still, and will remain, nuclear-armed.

There is a single, nuclear-armed superpower, the United States. Most, though not all, second-tier powers are also nuclear-armed — Britain, France, Russia, China and India. Of these, Russia retains a superpower-scale nuclear armoury. Japan and Germany could ‘go nuclear’ if they felt threatened and were unable to rely upon the ‘extended deterrence’ currently provided by the United States.

There are a few strictly regional powers with nuclear capabilities — Israel, Pakistan and North Korea. Iran is likely to become the world’s tenth nuclear power under a regime almost as unpredictable as that of North Korea. Other states could acquire nuclear weapons to face regional threats — South Korea, Taiwan, Saudi Arabia, Egypt, or Turkey, to mention just some speculative possibilities.

Finally, NATO contains six ‘quasi-nuclear powers’ with forces trained and equipped to deliver US nuclear weapons — Germany, the Netherlands, Belgium, Italy, Greece and Turkey. Complexity is a dominant feature of the second nuclear age.

The most obvious role for nuclear weapons is to deter the use of other nuclear weapons. Some argue that this is their sole credible and appropriate function. Even a ‘limited’ use of nuclear weapons would be sufficiently catastrophic as to warrant the scale of effort required to maintain a deterrent against it.

But nuclear weapons might have a wider utility. Britain, like all western nations, has foresworn other WMD — chemical, biological and radiological weapons. In the face of these threats the UK can try to defeat them with conventional forces and passive defence measures, or threaten retaliation with the only WMD we have — nuclear weapons.

Although included in the generic WMD category, nuclear weapons are of a different order. No other WMD have the immediate impact and enormous heat and blast effects of a nuclear detonation. But the large-scale civilian casualties inherent to almost any nuclear use makes this form of retaliation appear disproportionate — and a disproportionate threat is not a credible deterrent.

Whilst the UK government will not explicitly rule out a nuclear response to other forms of WMD attack, for fear of encouraging just such an event, it cannot be assumed that nuclear deterrence will deal with all potential WMD threats. A nuclear response to terrorism appears wholly inappropriate, except where the terrorist threat is both WMD-armed and identified with a known state.

The December 2006 white paper makes this connection explicit. Terrorist groups do not operate in a territorial vacuum and, especially if they are acquiring WMD, require substantial financial and physical resources. A WMD terrorist threat is more likely to be state-sponsored than not. They may not be deterrable, but their sponsors probably are.

Nuclear weapons have been held to have a general war-prevention role. It is argued that the presence of nuclear weapons makes major war between nuclear-armed states highly unlikely for fear of where hostilities might lead. The prevention of major war has always been part of NATO strategy and was re-iterated in the recent white paper.

War prevention is related to so-called ‘existential’ deterrence — the belief that nuclear weapons deter simply by their sheer awfulness, and irrespective of particular strategies for their use. There is some obvious merit in this view, given the general caution that the presence of nuclear weapons induces. But a reliance purely on the existential awfulness of nuclear weapons means the person most likely to be deterred is yourself — in which case others are unlikely to be deterred.

Successful deterrence depends upon credibility — the belief in others that you would, under extreme circumstances, ‘push the button’. After all, it is the deterree, not the deterer, who decides whether deterrence is successful.

The need to be credible means that a nuclear state must have the intent, will, capability and plans to use nuclear weapons in order that its nuclear threats are believed. In that way its weapons will have the desired deterrent effect and so not actually be used. That is the essential paradox of nuclear deterrence. This means that the distinction sometimes drawn between weapons for deterrence and weapons for war-fighting is a quite false one.

Only a ‘usable’ weapon is credible as a deterrent, as an ‘unusable’ weapon will deter no-one. Development of more accurate, lower-yield weapons is designed to meet just this problem. During the Cold War deterrence was based on a threat to inflict as much damage as possible on the adversary’s society. Today the credible deterrent may be the one that promises the least collateral damage. Unless one’s very existence as a society is in question, a threat to devastate another society is neither proportionate nor credible.

In today’s more diverse nuclear world, nuclear relationships are much less symmetrical than during the superpower stand-off of the Cold War. This asymmetry is not just a mismatch of capabilities but also of intent, the stakes at hand and even moral scruple. In regional conflicts, in which western powers may be involved, the local states may have a good deal more at stake (including regime survival) than other parties to the conflict. They may also have regimes less mindful of their own populations than are western governments. Future opponents may be willing to inflict, and suffer, greater costs than we are willing either to inflict or to endure. Self-deterrence in the face of less scrupulous and more highly motivated foes can thus be a real possibility. Development of more discriminating deterrent capabilities is therefore essential.

It is not possible to predict with any confidence future deterrence needs. We are left with what Sir Michael Quinlan calls the ‘To Whom It May Concern’ requirement. To this is added a further ambiguity — the desire not to rule in, or out, any particular option for nuclear use.

Drawing ‘lines in the sand’ invites others to step right up to the line. If they then step over it you either have to respond when in the circumstances it may not be appropriate or risk having your future credibility fatally undermined. The ‘threat that leaves something to chance’ has much to commend it, and is endorsed in the 2006 white paper.

Against this, however, must be the danger that an ambiguous, existential deterrent threat is insufficiently explicit to actually deter the actions one wishes to prevent. A general deterrence stance may need to be made a good deal more specific in particular circumstances. This needs to be done in ways that are not in themselves escalatory.

The reliability of nuclear deterrence is hotly contested. Deterrence failure is self-evident but success can be attributed to other factors. After all, when deterrence works, nothing happens.

Today’s more diverse nuclear world and the absence of any meaningful ‘strategic stability’ makes deterrence alone inadequate. Deterrence needs to be fully integrated with non-proliferation, which seeks to minimise the threats that may need to be deterred.

There may also be a new role for defence (or deterrence by denial). In the face of less numerous, less sophisticated threats a defensive capability based on modern sensing and computing technologies can do two things. It can add a denial element to the existing punishment deterrence capability, thereby raising the nuclear threshold. And in the event of deterrence failure it can limit the consequences.

The requirements of deterrence in the ‘second nuclear age’ are less serious, but more diverse, than we have been accustomed to. Nuclear weapons, whatever our aspirations to the contrary, are here to stay and their continuing salience assures a healthy future for deterrence. Better to be a ‘Have’ than a ‘Have-not’.

Dr Jeremy Stocker, Consultant to the International Institute for Strategic Studies.